Dvoretsky I Yusupov · Secrets of Endgame Technique

Pgress iifCtiess

Volume 24 of the ongoing series

Editorial board
GM Victor Korchnoi
GM Helmut Pfleger
GM Nigel Short
GM Rudolf Teschner

2008
EDITION OLMS

m
Mark Dvoretsky and Artur Yusupov

Secrets of Endgame
Technique

School of Future Champions 3

Edited and translated

by Ken Neat

2008
EDITION OLMS

m
4

Books by the same authors :

Mark Dvorelsky, Artur Yusupov, School of Future Champions

Vol. 1: Secrets of Chess Training ISBN 978-3-283-00515-3 Available

Vol. 2: Secrets of Opening Preparation ISBN 978-3-283-00516-0 Available

Vol. 3: Secrets of Endgame Technique ISBN 978-3-283-00517-7 Available

Vol. 4: Secrets of Positional Play ISBN 978-3-283-00518-4 In Preparation

Vol. 5: Secrets of Creative Thinking ISBN 978-3-283-00519-1 In Preparation

Mark Dvoretsky, School of Chess Excellence

Vol. 1 : Endgame Analysis ISBN 978-3-283-00416-3 Available

Vol. 2: Tactical Play ISBN 978-3-283-00417-0 Available

Vol. 3: Strategic Play ISBN 978-3-283-00418-7 Available

Vol. 4: Opening Developments ISBN 978-3-283-00419-4 Available

Bibliographic information published by Die Deutsche Bibliothek

Die Deutsche Bibliothek lists this publication in the Deutsche

Nationalbibliografie; detailed bibliographic data is available in
the internet at http://dnb.ddb.de.

© 2008 Edition Olms AG

Willikonerstr. 10 · CH-8618 Oetwil a. S./Zurich

E-mail: [email protected]
Internet: www.edition-olms.com

All rights reserved. This book is sold subject to the condition that it shall not. by way of trade or
otherwise. be lent. re-sold, hired out or otherwise circulated in any form of binding or cover
other than that in which it is published and without a similar condition including this condition
being imposed on the subsequent purchaser.

Printed in Germany

Editor and translator: Ken Neat

Typeset: Arno Nickel · Edition Marco, D-1 0551 Berlin

Printed by: Druckerei Friedr. Schmucker GmbH, D-49624 Lbningen

Cover: Eva Konig, D-22769 Hamburg

ISBN 978-3-283-00517-7
 5

Contents

Preface (Mark Dvoretsky) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

PART I EN DGAME THEORY

How to Study the Endgame (Mark Dvoretsky) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

The Theory and Practice of Rook Endings (Mark Dvoretsky, Artur Yusupov) . . . . . . . . . . . . . . . . . . . 38

From the Simple to the Compl icated : The Theory of Endings

with Opposite-colour Bishops (Mark Dvoretsky) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64

The Arithmetic of Pawn Endings (Mark Dvoretsky) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82

PART II EN DGAME ANALYSIS

Typical Endings with Connected Passed Pawns (Vladimir Vulfson) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88

Adventures on Adjournment Day (Mark Dvoretsky) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97

Solo for a Knight (Artur Yusupov) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 02

More about the 'Montaignian' Knight (Mark Dvoretsky) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 08

PART Ill TEC H N IQUE

Converting a n Advantage (Mark Dvoretsky) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112

Technical Proced u res i n a Grandmaster Battle (Artur Yusupov) . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . 1 53

Lessons from One Particular Endgame (Mark Dvoretsky) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 58

Grandmaster Tech nique (Mark Dvoretsky) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 65

Analysis of a Game (Artur Yusupov, Mark Dvoretsky) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 99

PART IV

From Games by Pupils of the School (Artur Yusupov) .... ....... ........... ............. ................. 2 1 2

I ndex o f Players and Analysts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225

. .
6

Mark Dvoretsky

P reface

Pfamiliar with the fi rst two books, based

erhaps, dear reader, you are already I n my bookcase there are nu merous weig hty
tomes devoted to the theory of endings. Is it
on material from the Dvoretsky-Yusu pov conceivable to assi milate and remember all
school for talented young players (Secrets the information conta i ned i n them? It turns
of Chess Training and Secrets of Opening out that it is not at all necessary to do this.
Preparation). Then you will already know the After read ing the fi rst part of this book, you
main principles by which we are g u ided . We will see that 'your' system of endgame
have held several thematic sessions of the knowledge can and should be compact,
school, devoted to a particu lar aspect of easy to assi milate and remember, and you
improvement in chess. The aim of the will learn how to develop it, by making the
sessions was not to convey specific infor­ acquai ntance of certa in i m po rta nt sections
mation - for this we had too l ittle time. It was of the theory of rook and minor piece
fa r more importa nt to discover the deficien­ endings.
cies in the pupils' play and to help them to The second part of the book is devoted to an
eradicate them, to demonstrate the most analysis of complicated practical endings.
effective ways of studying chess , and to Such an analysis helps to deepen and
acquaint them with general mechanisms, consolidate endgame knowledge, and aid
ideas and methods of play. the development of traits of character and
You now have before you the th i rd book thinking necessary to every player.
(there are five i n all). It is based on our work
The tech n ique of converti ng an advantage is
at the th ird session of the school , devoted to
a stu mbli ng-block for many players. To raise
the problem of improving endgame and
your tech n ical mastery req u i res developing
tech nical mastery.
i n yourself certa in importa nt skills i n seeking
During recent years the reg ulations for and ta king decisions, moreover not purely
tournaments and matches have changed chess , but, so to speak, 'psycholog ical­
sign ificantly - now games are hardly ever chess' decisions. The question of how to
adjourned . Previously, after taking play i nto improve your tech n ique is exam i ned i n the
an end i n g , you could investigate its subtle­ third part of the book. Here a re both the
ties in home analysis, whereas now you 'theory' of this question, and its practical
have to act directly at the board . Without an application - both by a critical analysis of
excellent knowledge and, what is even more
games played by young players , and at the
important, understanding of the laws of the
very hig hest grand master level .
endgame, it is not easy to cope with this
task, especially if account is taken of By trad ition , the final part is devoted to an
accu mulated fatigue from the preced ing analysis of games by the pupils of the
battle. And yet endgame mista kes are the school.
last in the game - it is no longer possible to This book was fi rst publ ished i n the 1 990s.
repair them! It is clear that today the During the preparation of this new edition I
importance of endgame and tech n ical mas­ checked all the games and endings on a
tery has grown sharply. computer, which , natu ral ly, gave rise to
 Preface

numerous improvements and additions. I n problem which is exceptionally i mportant for

addition , a qu ite large chapter has been every player.
added , one which was written many years Practically all the players whom I have
after the fi rst edition was publ ished . I n it trai ned have possessed good tech nique and
some instructive examples of the successful an excellent u n derstanding of the endgame.
or unsuccessfu l solving of complicated This means that the working methods
tech nical problems a re analysed - they will described in this book have stood the test of
help you to understand more deeply the time. I hope that they will also prove su itable
tech nique of converting a n advantage, a for you .
8

PART I

Endgame Theory

Mark Dvoretsky

How to Study the E n dgame

Mcomes to playing endings. They would

any young players 'flou nder' when it All endgame positions can be arbitrarily
d ivided i nto 'exact' and 'problematic' . Posi­
not be averse to improving their endgame tions which are familiar to us, i n wh ich we
mastery, but they don't know exactly how to know beforehand the evaluation and correct
do this. I n chess l iteratu re practically noth ing plan of action, we cal l 'exact' . Note that they
is said about methods for the i ndependent are familiar to us, and not to the theory of
study of endgame theory. We will now endings in genera l . Different players have
endeavour to partly fi ll this gap. d ifferent stores of exact positions.
Two main ways of improving in the endgame All rema i n i n g positions belong to problem­
can be disti ngu ished : atic. I n them we do not demonstrate our
I . The study of theory (development of knowledge, but fig ht, seek the best moves ,
erudition, en richment of the store of end­ and calculate variations - in short, we play.
game knowledge). A na"ive opinion is prevalent, identifying the
I I . Improvement in the general tech nique of mastery of endgame theory with a knowl­
endgame play. edge of nu merous exact positions. But is a
It stands to reason that these two d i rections large store of specific knowledge really
are closely inter-connected , and prog ress in necessary? After a l l , exact positions (apart
one of them i nvariably leads to progress i n from the most elementa ry) occur rather
the other. However, let us nevertheless rarely i n practice . More often a player has to
consider them separately. fig ht in problematic situations. He should
study the general endgame laws which
I. THE STUDY OF THEORY apply in them and the most com mon
To expand your store of knowledge you reg ularities, playing methods and typical
need to make a systematic study of various eval uations. All this, together, of cou rse,
types of endings. Here the traditional divi­ with the most important exact positions, is
sion by material is q u ite appropriate. By what comprises the i nteg ral system of our
successively exami n i n g , for example, pawn, endgame knowledge.
knight and queen endings we assimilate the I must once again emphasise: the store of
specific featu res of these types of endgame. positions which you need to know exactly is
 How to Study the Endgame CLJ g
comparatively smal l . Only in rook endings 1 . . . Wh7 would also not have lost, but in this
do you need to have a firm knowledge of case the defence would have been much
several dozen specific positions; in other more d ifficult.
types of endings - much fewer. When 2 l:Ig5 .l:!.b1 !
studying them it is often not necessary to
A typical rook move i n such situations - from
delve into complicated analyses - it is
here it retains the possibil ity of checking the
sufficient merely to remember the main
enemy king both along the file, and along
conclusions.
the rank.
Take, for example, rook endings with f- and
3 l:!.c5
h-pawns. They occu r q u ite rarely, but never­
theless they do occu r, so that it would be If 3 h6 it is wrong to play 3 . . l:tg 1 +? 4 Wf5
.

useful to obtain some impression of the m . .i:.h 1 5 l:tg7 + , when Black's king is forced
However, i t is hardly advisable t o study the back onto the 8th ra n k and this leads to a
entire theory of this type of endgame - it is loss. He is saved by the waiting move
just too complicated . What, the n , from this 3 .. Jla 1 !, for example: 4 .Uh5 (4 'it>f5 .Ua5+ ; 4
theory should the practical player add to his h7 .l::i. g 1 + 5 Wf5 .l::i. h 1 ) 4 .. . Wg8 5 f5 '>th7.
armoury? 3 . . . '>tf6
Above all, the i nformation that such endings 4 .l:!.c6+ �g7 !
are normally d rawn . It is useful to examine a The main danger for Black is having his king
practical ending , demonstrating the main forced back onto the 8th rank. This would
defensive ideas.
have occu rred after 4 . . . Wf7? 5 Wg5 .U.g 1 + 6
'it>f5 .l::i. h 1 7 .Uc7 + .
Gl igoric - Smyslov 5 Wg5 .Ug1 + !
Moscow 1 947 6 'lt>f5 .U a 1
7 .l:lc7+
7 .l:.g6+ Wf7 .
7. . . Wh6
8 .l:!.e7 .l::i. b 1
9 .Ue8 'it>g7
1 0 l:!.e5 l:!.a1
1 1 .Ud5 l:!.f1
Not a bad move , although it was qu ite
sufficient to keep the rook in the corner.
1 2 :d4 .U.a1
1 3 l:!.d6 :as+
1 4 �g4 l:!.a1
14 . . . l:!.b5 is also possible, retu rn ing to the
The black rook is excellently positioned on position with which we beg a n .
the 5th rank, preventing the wh ite king from 1 5 l:!.e6 .l::i. g 1 +
advanci ng. If 1 f5 there follows 1 . . . .U b 1 , 1 6 'it>f5 .l:!.a 1
threatening a series of checks from the rear. 1 7 h6+ 'lt>h7!
1 .Ug6+ 'lt>f7 ! 1 8 .l::i. d 6 l:!.a2
10 � How to Study the Endgame

19 'iii g5 .l:.g2+
20 'iii f6 'iii xh6!
21 'it>e7+ 'it>h7
22 f5 .l:!. e2+
23 l:l.e6 .l:ta2
24 f6 !:. aS!
25 'Ot>f7 '.t>h6
An important theoretical position has arise n ,
one which should have been included in o u r
system o f precise knowledge a t a n ea rl ier
stage - in the study of endings with rook and
pawn against rook.
26 .l:.e1 l:!.a7+
1 f6 l:!.a1
27 .l:!. e7 l:!.a8
2 l:l.g7+ 'it>h8
It is simplest to keep the rook on the eig hth
rank, not allowing the wh ite king there. B ut 2 . . . 'it>f8 3 h6 followed by the u navoidable
also possible is 27 . . . l:l.a 1 28 Wf8 'iiig 6 29 f7 h6-h7 .
'iiif6! 30 <it>g8 .l:tg 1 + ! with a d raw. 3 'it>g6 .l:!. g 1 +
28 .l:td7 'iii h7 4 'Ot>f7 .l:.a1
29 .l:td1 .l:!. a7+ 5 .l:!. g8+ 'Ot> h7
30 <it> e6 .l:.a6+ 6 .l:!.e8 l:!.a7+
31 l::!. d 6 .l:!.a8 7 <it> f8
32 .Ud4 'Ot> g8 The next move will be 8 f7 (the h5-pawn
deprives the black king of the g6-sq uare).
33 .l:!.g4+ 'it>f8
It is sufficient to play through th is variation
Draw.
just once on the board - there is no need to
An examination of such an ending helps us memorise it, especially since Wh ite also has
to draw certain general concl usions. We other ways to win.
now know where Black should place his This is probably a l l that the practical player
rook. And the king, as was shown by l lya needs to remember about the g iven type of
Maizelis, is best kept at f7 u ntil there is a ending. As you see , not really so m uch and
danger of it being driven onto the back ran k . not really so d ifficult!
Then i t can stand a t g7 and su bsequently
even at h6, attacking the wh ite pawn.
Let us see how another, rather more ex­
It stands to reason that by no means all
tensive section in our system of endgame
positions with f- and h-pawns are d rawn .
knowledge is constructed - the theory of
The most important exception has already
endings, in which a rook fights against
been mentioned several times - when
pawns. A basis for study can be provided by
Black's king is cut off on the back rank, he
any endgame manual , for example, l lya
normally loses.
Maizelis's monograph Ladya protiv peshek
(Rook against pawns), published in 1 956 (in
contrast to opening books, those on the
endgame hardly date at a l l ) . Here some 400
 How to Study the Endgame ltJ 11

positions are examined . Clearly we are not C utti ng off of the king
able to study and remember all this i nforma­
tion . We need to select the most impo rta nt
key endings for practical purposes.
But how to choose the most important
material? This is the main problem . What
tells here is the player's i ntel lect, and his
abil ity to work with books , to general ise, and
to draw conclusions. He is also helped by
the knowledge (even if incomplete) that he
already has, and by his own practical
experience in the g iven field .
The play i n endings with rook against pawns
is dynamic in character, and every tempo
has a decisive i nfluence on the outcome. It
follows that here there is no large-scale Wh ite wins, by playing 1 .l:tg5 ! . When the
strategy, battle of plans, or deep regularities pawn reaches a3, it ca n be eliminated by
(as, say, in endings with opposite-colour llg3 (or with the pawn on a2 - by l:.g 1 and
bishops). There a re also hardly any exact l:r.a 1 ) . If it is Black to move , then 1 .. .'it>b5(c5)!
positions, by relying on which we could
leads to a d raw - as it is easy to see, cutti ng
avoid the need for concrete calculation. The
off the king along the 4th ran k by 2 l:tg4 does
main role is played by a knowledge of typical
not ach ieve anyth ing.
techniques, which help the correct move to
be found more qu ickly and variations to be
calculated more certa i n ly. Promotion of the pawn to a knight
The procedu res a re best mastered with the
help of elementa ry positions, in which they
are employed and where their action is not
obscured by extraneous analytical details.
Su bseq uently the exact pattern of the
position my be forgotten , but the impression
of the tech n ique will remain . Sometimes
such a position - the conveyor of the
technique - is simu lta neously an exact
position which is i mportant for us; in this
case, of cou rse, we should memorise it.
So, using some very simple schemes, let us
examine the main ideas which apply i n
endings with rook agai nst pawns.
1 l:.h2+ 'it>c1 2 'it>c3 b 1 lD+ ! 3 'it>d3 lb a3 4
l::ta 2 lbb1 1 with a d raw, but not 4 . . . lbb5? (in
endings with knight against rook, the knight
should not be separated from the king).
A d raw also results from 1 . . . 'it>b1 2 'it>b3
'it>a 1 ! 3 l:txb2 - stalemate. But with a
12 � How to Study the Endgame

bishop's pawn or a central pawn only the

promotion to a knight saves Black.
However, with a rook's pawn this idea no
longer helps.

It is hopeless to play 1 . . . a2? 2 .l:.b8+ 'it>a3 3

'it>c2 ! a 1 4J+ 4 'itc3 'it>a2 5 .Ub7 with
zugzwang. The only way to save the game
is 1 ... 'it>b2! 2 .l:!. b8+ (2 .l:.h2+ 'it>b3 ! , but not
2 . . . 'it>b1 ? 3 'it>c3) 2 .. . 'it>c 1 ! 3 .l:ta8 'it>b2 4 'it>d2
1 'it>b4(c4) a2 2 'it>b3 a 1 4J+ 3 'it>c3, and a2 5 ktb8+ �a 1 ! .
Black is in zugzwang. As you see, when learning new ideas one
It is useful to note that if Black also had a can sometimes repeat material that has
pawn on b5, all the same this would not save been covered earlier (in this case - promo­
him: 3 . . . b4+ 4 'it>xb4 4Jc2+ 5 'it>c3 tt:Je3 6 tion of the pawn to a knight).
'it>d3 4Jd5 7 .Uh4 'it>b2 8 .Ud4, and the knight,
which is separated from the king , is soon An i ntermediate check to gain a tempo
lost. 6 l:.h4! (instead of 6 �d3) wins even
more quickly: 6 . . . �a2 (6 . . . 4Jd 1 + 7 'it>d2 4Jb2
8 .l:i.b4 'it>a2 9 �c2 'it>a 1 1 0 l:.b8 ; 6 . . . 4Jd5+ 7
�b3 'it>c 1 8 .l:!.c4+ �b 1 9 .Ud4) 7 .l:.a4+ �b 1 8
.Ue4 4Jf5 9 .Ue5 4Jd6 1 0 <;i;>b3 <;i;>c1 1 1 .i.:i.c5+
'it>b 1 1 2 .Ud 5.

Stalemate
We have already examined one very impor­
tant practical instance of stalemate . Here is
another example (wh ich , incidentally, consti­
tutes one of the few 'exact' positions that it is
useful to memorise).

The diagram position a rose in the game

Korchnoi-Kengis (Bern 1 996). Black re­
sig ned , after calculating the fol lowing forced
variation .
 How to Study the Endgame tZJ 13

1 ...'it12 2 .l:tf8+! 1 . . . a5? does not work i n view of 2 .l:!.h5! - we

2 'it>d3? g3 3 l::tf8+ 'it>e 1 ! leads to a d raw. already know this idea. But 1 . . . 'it>b5? is also
2... 'it>e2 3 l::tg 8! 'it>f3 bad : 2 W a5 3 'lt>e6 'it>c4 (3 . . . a4 4 'it>d5) 4
.l:ta8 ! 'lt>b4 5 'it>d5 a4 6 'it>d4 'lt>b3 7 'it>d3 a3 8
Thanks to the intermediate check, the king
.l:tb8 + . The only saving move is 1 . . . '1t>c5! ,
has been driven one square back - from f2
preventing the approach of the enemy king
to f3.
to the pawn.
4 Wd3 g3 5 l:!.f8+ 'lt>g2 6 'lt>e2 etc.

Outflanking
'Shoulder-charge' The ideas of 'shoulder-charge' and 'out­
flanking' are vividly expressed in a famous
study by Richard Reti ( 1 928) .

1 l:!. h2+ 'lt>a3!

By not allowing the enemy king to approach
the pawn , Black gains a d raw. It is i ncorrect 1 l:td2(d3) 1 ! d 4 2 .l:i.d1 ! 'it> d 5 3 'it>d7 ! , and
to play 1 . . .'1t>b 1 ? 2 'lt>b3 a 1 lD+ 3 'lt>c3. Black is i n zugzwang: if 3 ... '1t>c4 4 'lt>e6, or
3 . . .'1t>e4 4 'lt>c6.
Now let us examine a slightly more compli­
cated example. 1 .l:td 1 ? is a mistake : 1 . . . d4 2 'it>d7 (2 'lt>f7
'lt>e4 3 'it>e6 d3) 2 . . . '1t>d5! (Black prevents the
outflanking) 3 �c7 'it>c5! (3 . . . '1t>c4? 4 'it>d6!
d3 5 'it>e5) , and it is Wh ite who ends up in
zugzwang .

Let us now turn to positions in which a rook

fights against two con nected passed pawns.

Mate th reats to the opponent's king

If the pawns a re far advanced (two black
pawns on the 3rd rank, or one on the 2nd
ran k and the other on the 4th ) , then the rook
is unable to stop the m . However, sometimes
14 � How to Study the Endgame

it is possible to save the game, by pursu ing In this position Herman Fridstein resig ned
the opponent's king when it is pinned to the against Anatoly Luti kov ( Riga 1 954 ) . He
edge of the board . considered the variation 1 �xb3 c2 2 l:.b4+
'it>d5 3 .l:tb5+ 'it>d6 4 .U.b6+ �c7 , but did not
B. Horwitz, J. Kling notice the saving i ntermediate check 1
.U.b4+ ! .
1 85 1
I should mention, i ncidentally, that d ifferent
players can single out d ifferent ideas and
ru les, depending on their experience and
knowledge. I n the above example attention
should be paid to the manoeuvre with which
the black king escapes from the checks
(after 1 l:.xb3? ) . But you can also disregard
it, if this idea is a l ready wel l known to you .

The best position for the rook i s beh i nd

the more advanced pawn

1 �f5 �h4 2 'it>f4 �h3 3 �3 'it> h 2 4 'it>e3 !

'it>g2
After 4 . . . �g3 5 l:tg 1 + �h4 6 Wf4 Wh3 7 'it>f3
bad is 7 . . . Wh2?? 8 l:.b1 , when Black loses
because of zugzwang.
5 Wd3 Wf3 6 Wc3 a2 7 'it>xb2 (or 7 .S.f1 + )
with a draw.

Intermediate check before the captu re

of a pawn
1 .S.g6 ! Wd7 2 .S.g4! g2! 3 .S.xg2 'lte6 4 l:!.g5 ! ,
winning thanks t o the fact that the black king
is cut off from the pawn along the 5th ra nk.
I n Maizel is's book he g ives a position by
Sozi n , wh ich d iffers only in that the wh ite
king is on a7. In this case after 1 l:.g6! Wd7
there is a second solution: 2 'it>b6 'it>e7 3
'it>c5 'it>f7 4 .S.g4 �f6 5 �d4 ! (5 .S.xf4+? 'it>g5 6
l:!.f8 'it>g4 7 �d4 g2) 5 . . . Wf5 6 .S.g8 and wins.
But with the king on a8 the analogous
variation no longer works: 1 l:.g6 ! Wd7 2
�b 7? rtie 7 3 rtic6 'it>f7 4 l:.g4 'it>f6 5 Wd5 'it>f5
6 .l:l.g8 f3 ! 7 Wd4 (7 l:.xg3 'it>f4 8 .l:tg8 f2 ; 7
 How to Study the Endgame ctJ 15

.l:!.f8+ Wg4 8 'it>e4 f2 9 'it>e3 �h3 with a d raw) Some of the ideas mentioned operate not
7 .. .f2 8 'it>e3 f1lt:l+ ! . only in endings of the g iven type. Thus, for
example, the rook should be placed to the
rea r of the more advanced pawn in nearly
Which pawn to advance?
every case, when it is fighting against two
con nected passed pawns.
Mar6czy - Tarrasch
San Sebastian 1 9 1 1
Alekh ine - Tartakower
Vienna 1 922

There was a n easy win by 1 l:t x h 2 'it;xh2 2

'it>a6! (the i m med iate 1 'it;a6! is also possi­
ble) 2 �g3 3 b5 �f4 4 b6 'it>e5 5 b7 .l:tb1 6
...
Alexa nder Alekh ine analyses the natural
Wa7 �d6 7 b8�+. Note the move 2 'it>a6 ! . conti nuations 36 'it>c2 , 36 �c4. 36 g5 and 36
Firstly, Wh ite advances the pawn behind l:th2, and shows that they a re sufficient for a
wh ich the rook is not standing. Secondly, his d raw at best. There is only one way to win.
remaining pawn is fu rther away from the
36 .ti.d5 ! !
enemy king, which does not manage to
attack it. Wh ite's fantastic move fi nds a precise
explanation , from the standpoint of typical
2 a6? is a mistake in view of 2 . . . 'it;g3 3 'it;b6
ideas for such endings.
Wf4 4 a7 'it;e5 5 �b7 'it>d5 6 b5 'it>c5 , when
the black king succeeds in ' locking on' to the The variations springing from this rather
b-pawn . Or 4 b5 'it>e5 5 'it>a7 'it>d6 6 b6 .l:!.b1 ! unlikely move (it attacks one solidly de­
fended pawn and allows the immediate
7 'itb7 (7 b7 'i;;c ?) 7 . . . 'i;;c 5.
advance of the other) are quite simple when
The game went 1 'i;; c 6? �c1 + 2 'it>b6 �c4!
we have descried the basic idea: The black
(threatening the interference 3 ... .l:!.h4) 3 pawns are inoffensive:
.l:!.xh2 l:txb4+ 4 'i;;c5 .l:!.a4 5 �b5 �xa5+ with a
1) When they occupy squares of the same
draw.
colour as their bishop, for in that case
It wou ld be possible to expa nd fu rther the White 's king can hold them back without
store of typical ideas, but for a start it is difficulty, by occupying the appropriate white
sufficient to l i m it ou rselves to these, the square, for example 36 . .f2 37 l:!.d1 e4 38
.

ones most used . 'it;c2 j)_f4 39 �f 1 and 40 'it>d1.

16 w How to Study the Endgame

2) When the rook can be posted behind

them, but without loss of time, for example
36...e4 37 'af5 Ji..g3 38 g5 e3 39 'ii.xf3 e2 40
.U.e3 (Aiekh ine).

It makes sense to also examine endi ngs

which are closely linked to those being
stud ied . I n the given instance - sharp rook
endings, transposing into endings with rook
against pawns. In them we encou nter ideas
with which we are already familiar.

Alekh ine - Bogolj u bow

World Championship Match , This position could have occu rred i n the
1 9th Game, 1 929 game Lapin - Utyatsky (Bryansk 1 965).
1 . . . 'it>c2
2 .l:i.c7+ 'it>b2 ! !
Only this paradoxical move , suggested by
Utyatsky, leads to a win. It i nvolves the idea
of i nterference. For example, if 3 'it>g4 B lack
decides matters with 3 . . . .l::t a 5! 4 .l::t c6 'it>a3! 5
.l:r.xg6 b2 6 :b6 l:ta4+ and 7 . . . l:tb4 .
3 .l::tc 6 .l:i.a4!
4 .U.xg6 'it>a3
4 . . . 'it>c3 or 4 . . . 'it>a2 is also not bad .
5 l:tb6
5 l:tf6 b2 6 l:tf1 l:Ic4 7 l:tb1 l:tc1 .
5. . . b2
I n the game there followed 70 ... 'it>g4? 71 b7 Threatening the interference 6 . . . .l::t b 4 .
f5 72 b8'ik l:Ixb8 73 l::t x b8 and White won
6 l:txb2 'it>xb2
easily by approach ing the pawn with his
7 g4 'it>c3
king . But Efim Bogoljubow could have saved
the draw by employing the 'shoulder-charge' . 8 'it>h4 'it>d4
70 . . . 'it> e4! 9 'it>g5 'it> e5
The black king must be placed in the path of 1 0 h4 'it> e6
the opponent's king. And B lack wins easily.
Of cou rse , we will also meet new ideas Thanks to the threat of i nterference, B lack
which operate in sharp rook endings. The forced his opponent into hurrying to give up
most important of them is inteference. We his rook for the pawn. In the event of the rou­
have already encountered it in the analysis tine 2 . . . 'it>b1 ? , interference no longer occurs
of the Ma r6czy-Ta rrasch ending. Now we and White can wait until the pawn reaches
will examine a far more complicated exam­ b 1 . From there, incidental ly, it takes longer
ple. for the king to reach the opposite wing.
 How to Study the Endgame

3 'it>g4 b2 (3 . . . �a5 4 �c6 b2 5 .l::tx g6 Wa2 Thus we should build u p our theory of the
6 J:(b6 with a d raw) 4 �g5 �a 1 (4 . . . .l::i. b 3 endgame in the most econom ical way, by
5 'it>xg6 or 5 g4) 5 �b7 b 1 'iV 6 �xb 1 + 'iti>xb 1 . singling out the most generally u sed tech­
n iques and the most i mportant exact posi­
tions. How best to assimilate and consoli­
date this material is a nother matter. Here
one cannot get by without a familiarity with
additional examples, i ncluding complicated
practical endings (such as the one we have
just been analysin g ) . It is useful to try and
solve a series of tra i n i ng exercises on the
given topic. And above all, I recommend that
you analyse i ndependently those endings
which you happen to encounter.
What does an i ndependent analysis of
endgame positions g ive us?
1 ) We learn new ideas and methods,
Wh ite can now ach ieve a d raw i n various expa ndir;�g our system of knowledge, and we
ways. It is useful to examine the resulting refine the i nformation we a l ready have.
variations, since i n this way we will repeat 2 ) After analysing a large amount of mate­
and consolidate our knowledge of endings rial , we have a better u nderstanding of what
with rook against pawns. I n the analysis featu res are typical and i mportant and
extreme care has to be taken - despite the should therefore be i ncluded i n the 'system' ,
apparent simplicity, here one can easily go and which are accidental i n character. As a
wrong . result we form o u r endgame impressions
1 ) 7 'it>xg6 'it>c2 8 g4 'iti>d3 9 h4 'iti>e4 1 0 h5 most clea rly, economical ly, at the same time
'it>f4 11 h6 �a6+ 12 'iti>h5! with a d raw without om itting anyth ing i mporta nt.
(shoulder-charge). It is amusing that Utyatsky 3) It improves our analytical mastery.
suggests 1 2 'iti>g7? 'it>g5 1 3 h 7 .l:f.a 7 + 1 4 'it>g8
4) At times some players gain the impres­
'.t>g6 1 5 h8tt:J+ 'it>f6 1 6 g5+ 'it>xg5 1 7 tt:Jf7 + ,
sion that they largely u nderstand the secrets
but we already know that, accord ing to
of chess and that to fi nd the best move in the
theory, after 1 7 . . . 'it>f6 1 8 tt:'Jd6 .l::t a 5 (or
majority of cases is no problem . They only
18 . . . We6) Black wins.
need not to blu nder, and to obta i n the
2) 7 Wxg6 'it>c2 8 h4 (in Utyatsky's opinion, open ings they want. Analysis helps to rid
this move loses) 8 . . . .l::t x g3+ 9 'it>f6 .l::i. h 3 1 0 themselves of such ill usions, and shows
'it>g5 Wd3 1 1 h5 'it>e4 1 2 h6 �e5 1 3 'iti>g6 what an enormous wealth of ideas is
'.t>e6 14 'it>g7! (but not 1 4 h 7? l:tg3+ 1 5 'it>h6 sometimes concealed in the seemingly
'it>f7 1 6 h8tt:J+ 'iti>f6 ) 14 . . . 'iti>e7 ( 1 4 . . . l:tg3+ 1 5 most modest position. It guards against
'it>f8!) 1 5 h7 l:tg3+ 1 6 'it>h8 ! , savi ng the game superficial ity, and aids the development of
thanks to stalemate. such important traits as precision, accu racy,
3) 7 g 4 Wc2 8 h4 .Ug3 9 'it>f4! .l::t h 3 1 0 'it>g5 industriousness, and so o n .
'.t>d3 11 h5 gxh5 1 2 gxh5 'it>e4 1 3 h6 'it>e5 1 4 5) A n analysis o f you r own games enables
'lt>g6 'iti>e6 1 5 'it>g7! with a d raw, a s i n the deficiencies i n you r play to be objectively
previous variation . diagnosed .
18 � How to Study the Endgame

6) Analysis sometimes leads to interesti ng U nfortunately, this entire variation is a

results, afford ing creative pleasure . comedy of errors, at the basis of wh ich is
Dreev's natura l , but in the g iven instance
Once when I was looking through Chess i ncorrect striving to defi n itely place his rook
lnformator, an ending annotated by one of behind the opponent's passed pawn .
my pupils, Alexey Dreev, caught my eye. After 9 g5 B lack saves the game by 9 . . . l:th 1 !
(the place for the black rook is to the rear of
Dreev - Moskalenko
the more advanced pawn ) 1 0 g6 llh5+!
USSR Young Masters Championsh i p ,
(remember the Fridstei n-Lutikov ending), or
Lvov 1 985
1 0 '1tg6 '1td7 1 1 '1th7 '1te6 1 2 g6 llg 1 ! (and
this we have already seen in one of the
variations of the Mar6czy-Tarrasch ending).
Wh ite should not g ive up his rook. I n stead of
5 h5? he wins by 5 .l:te 1 ! a2 6 l:r. a 1 '1tc7 7 h5
'1td7 8 h6.
But before this B l ack went wrong : he could
have d rawn by 4 . . . a2! (instead of 4 . . .l:.a4? ) 5
l:ta3 l:.b5+ 6 '1tg6 l::l.a 5 7 :Xa2 .l:!.xa2 8 h5
'1tc6 .
Even so, the ending is won . On ly, the rook
should not be p laced on e3.
2 l:te1 ! a2
3 .l:.a1 '1tc6
A draw resu lts from 1 l:!.e 1 ? a2 2 lla 1 '1td7 , 4 h5 '1td6
or 1 .l:l.e3? .l:.b4+ 2 '1tf5 l:.a4 3 .l:.e 1 a2 4 h5 (4 5 h6 lih2
.l:ta 1 'it>d7) 4 . . . a 1 'ii' 5 .l:txa 1 l:txa 1 . 5 .. .'.ti>e6 6 h7 .l:.b8 7 llxa2 '1tf6 8 l:.h2 llh8 9
1 l:te6+ ! l:Ih6+ '1tg7 1 0 '1tg5 .
The black king faces a choice . I n the game it 6 '1tg5(f5)
moved towards the kingside, but by placing Also possible is 6 h7 l:txh7 7 :Xa2 .l:th8!? 8
his rook beh ind the passed pawn Wh ite won l:.a4! or 8 l:ta6 ! followed by 9 'i.t>g5, but not 8
easily: 1 . . . '1td7 2 lla6 a2 3 g5 '1te7 4 '1tg4 g5? '1te6 and not 8 l:te2? lif8+ 9 '1tg3 l:tg8
'1tf7 5 '1th5 llh2 6 .Ua7+ '1te6 7 '1tg6 l:tb2 8 with a d raw.
h5 .l:tb8 9 h6 l:tg8+ 1 0 'it>h5 '1tf5 1 1 l:.a5+ 6 . . . '1te7
B lack resig ned .
7 '1tg6
1 . . . '1tb5
7 h7 is also strong .
In lnformatorthe following analysis is given:
7 . . . '1tf8
2lie5+ 1 ! '1tb6 (2 ... 'it>b4 3 l:te8 '1tb5 4 J:tb8+ !
8 h7
and 5 .U.a8) 3 .l:te3 ! (all the exclamation
marks are by the annotator) 3 ... .l:!.b4+ 4 c;tf5 And Wh ite wins.
l:.a4 5 h5 a2 6 l:Ie1 a 1 'ili' 7 .l:txa 1 l:.xa 1 8 h6 It remains to analyse the rook check at e5,
'1tc7 9 g51 '1td7 (9 . . . lih 1 1 0 g6! l:txh6 1 1 g7 which i n fact deserves not two exclamation
l:lh5+ 12 c;tf4 lih4+ 1 3 'it>f3 l:.h3+ 14 'it>g2 ) marks, but more probably one q uestion.
1 0 h7l::!. h 1 1 1 g 6 a n d wins. mark. Let us verify 2 :e5+?! '1tb4 ! ? .
 How to Study the Endgame l2:J 19

If 3 J:le8? Black plays not 3 . . . 'it>b5? , but

3 ... a2 ! . Then 4 l:ta8 .l:!.c2 ! with the threat of
5 . . . l:lc4+ , 6 .. Jk5(c3)+ and 7 . . J:ta5(a3) leads
to an immed iate draw. And if 4 l:tb8+, then
4...Wc4 (or 4 . . . 'it>c3) 5 l:ta8 l:tb4! (threaten­
ing interference: . . . '.tb3+ and . . . .l:ta4) 6
�a2 '.tb3+ 7 '.t>f5 'it>xa2 8 h5 .l:!.b5+! (the
king must be driven to the u nfortu nate h4-
square - this is not difficult to ach ieve, using
the rook's long range) 9 �g6 l:!.b6+ 1 0 �g5
l:lb5+ 11 Wh4 l:i.b 1 ! (and now the rook
moves to the rear of the pawns) 12 h6 ( 1 2 g5
'it>b3 13 g6 l:tg 1 ! ) 12 ... .l:!.h 1 + 1 3 'it>g5 'it>b3 1 4
'it>g6 'lt>c4 1 5 g5 �d5 1 6 'lt>h 7 'it>e6 1 7 g6 out that the answer is no. After a l l , the black
J:lg 1 ! with a draw. king too is not well placed - it is cut off on the
After 2 J:le5+?! 'it>b4!? all the same the rook h-file.
has to be retu rned to the first ra nk. But then 1 'lt>g1 ! h4 2 l:tg8 f3 3 llf8 'lt>g3 (or 3 . . . l:i.g2+
it is clear that the check was poi ntless - 4 'it>f1 'lt>g3 5 l:tg8+ 'lt>h2 6 l:If8 ) 4 l:tg8+ 'it>f4
White is forced to calculate the lengthy
5 l:tf8+ 'lt>e3 6 l:te8+ 'it>d3 7 l:td8+ 'it>e2 8
variation 3 .l:te 1 a2 4 .l:ta 1 'it>b3 5 h5 l:!.b1 6
l:te8+ 'it>d 1 9 l:tf8(e3) with a d raw.
l::!.x a2 '.txa2. Now it is incorrect to play 7 h6?
However, if it is Black to move he wins, by
J:lh 1 8 g5 '.tb3 with a d raw. B ut as yet the win
depriving the enemy king of the i m porta nt
has not been missed : 7 g5! .l:th 1 8 g6! or
g 1 -square.
7 .. J:tf1 + 8 'lt>g4! (8 'it>e5? l:th1 ! ) 8 . . . 'it>b3 9 g6.
1 . . . Wh2! 2 l:!.g8 ( i n the event of 2 l:!.f7 or 2
J:th7 Black wins by 2 . . . �g3 ) 2 . . . h4 3 l:!.g7 (3
In 1 976 the USSR Championship was held .l:tg4 h3 4 l:txf4 'lt>g3 5 .Uf8 .l:t b 1 + 6 �e2 h 2 )
in Moscow. I n the very fi rst round my friend
3 . . . h3 4 l:tg8 f 3 ( o r 4 . . . l:tg2 ) and wins.
Boris Gulko adjourned his game against
grandmaster Tai manov in a complicated
rook end ing. Before the resumption h e Taimanov - G u l ko
asked m e to join in the analysis. 44th U S S R C h a mpionsh ip, Moscow 1 976
In order to figure out precisely some very
intricate variations, we had to turn to the
theory of the rook endgame with f- and h­
pawns. The elementary information a bout
these endings, wh ich was given above, was
not sufficient for us. However, the necessary
positions could not even be found i n books
on the endgame, so that we had to
sup p lement 'official' theory with our own
analyses. Here is a very important key
position that we found (see diagram).

White's king is cut off on the back rank. Does

th is mean that he is bound to lose? It turns
20 � How to Study the Endgame

42 . . . 'it>e3 48 .Ud7?
The sealed move. We assu med that 48 .Uf5 also did not help in
43 l:ie2+ 'it>xf3 view of 48 . . . �b7 49 'it>f1 (49 'it>f2 'it>g4 50 l:tf6
44 gxhS gxhS h4) 49 . . . 'itog4 50 ltf2 .l:!.b 1 + ! 5 1 'it>g2 f5 . To
prevent the king from being pushed onto the
45 l:ie5 �g4
back rank, the wh ite rook must guard the
Weaker is 45 . . ..l:txb2 46 ltxh5! 'it>e4 47 �h8,
2nd rank, where it is too passively placed .
and Wh ite should gain a d raw.
Black wins easily, by advancing his pawns.
46 l:txd5 �xb2
Alas, a mistake crept i nto our a nalysis. By
47 'it>e1 conti n u ing 49 .Ua5! (instead of the losing
After 4 7 .Ud4+ 'lt>g3 48 'lt>e 1 f5 49 l:id5 f4 50 king moves) Wh ite exploits the long-range
�g5+ (50 l:ixh5 .Ub 1 + 5 1 'it>d2 f3) Black does power of his rook and d raws by d riving the
not play 50 . . . 'it>f3? 5 1 l:ixh5, but simply opponent's king to a less good position:
50 . . . Wxh4 ! 5 1 �g8 'lt>h3 52 'it>f1 �h2 ! , 49 ... 'it>g4 50 �a4+! 'it>g3 5 1 lta3+ 'it>g2 52
achieving a winning position, since h i s king �a2+ 'itg 1 53 'ite2 ! (now i n a n u mber of
succeeds i n reach ing h2. variations it becomes possible to shut the
But now we have reached the cu l m i n ation of king on the edge of the board after playing
the entire endgame. the rook to the g-file) 53 . . . '1t>g2 54 'ite 1 + ! .
48 . . . f6!
We thought that this subtle move was the
only correct one, since i n the event of 48 . . . f5
49 .Ug7 'itoh3 50 'it>f1 the wh ite king succeeds
i n reach ing g1 (50 . . . 'ith2 does not work i n
view o f 5 1 .Ug5). B u t here too w e were
wrong! After 50 . . . h4! 51 �g 1 Black has the
winning resou rce 51 . . . .Ub4 ! , which he did not
have i n our basic position - there the pawn
was a l ready sta n d i ng at f4 . 51 .Ug8 'it>h2! is
hopeless, as is 51 �g5 f4 52 'it>g 1 f3 52 l:if5
'it>g4 (the rook is placed too close to the king
and is u n able to give checks) . A good
illustration of how carefully and cautiously
The natural move 47 .. .f5? is a m istake. After one should use theoretical knowledge: a
48 Wf1 .Uh2 49 �g 1 .Uxh4 50 �g2 f4 51 l:id3 slight change in the position , and wel l­
a ' normal' (i.e. drawn ) position with f- and h­ known procedu res and eval uations may
pawns is reached , and with the black rook prove invalid .
bad ly placed . If 48 .. .f4 there follows 49 49 �1
�g5+ 'it>xh4 50 l:ig8 'it>h3 51 'it>g 1 with a 49 .Ug7 does not help in view of 49 . . . �b5! 50
draw, si nce the wh ite king has reached g 1 . �f2.l':lf5+ 5 1 'it>e3 ( 5 1 'it>g2 l:ig5+) 51 . . . 'it>h3
And if 49 . . . �3 (instead of 49 . . . 'it>xh4) , then with a n easy win. Black simply adva nces his
50 �g 1 l:ib1 + 5 1 'it>h2 'it>f2 52 �xh5 f3 53 king and his h-paw n , and then blocks the g­
lta5 'it>f1 54 �g3 f2 55 .Ua2 �b3+ 56 �g4 , fi le with his rook, and the wh ite king proves
and White gives up his rook for the f-pawn . to be too far away from the rook's pawn.
47 . . . 'itoxh4! 49 . . . 'it>g4
 How to Study the Endgame fLJ 21

White's position is hopeless, since his king of 46 llxd5). We examined a provisiona l

is cut off on the first rank, and he has been variation , which was fa r from forced , and
unable to shut in the opponent's king on the which i n add ition (as was revealed with the
h-file. help of a computer database of endings with
50 .Ug7+ <bf5 a small n u m ber of pieces , created more
51 .l:!.h7 'it>g6 than a q u a rter of a century later) contained
n u merous mistakes . However, at the time
52 .l:!.h8 f5
we were attracted not by the variation itself,
53 .l:!.g8+ �f6 but its concl uding position.
54 �g1 f4
46 . . J1b5 47 llg5+ 'it>xh4 48 .l:!.f5 .l:!.b7
55 'it'f1 �f5 (48 . . . �g4 49 .Uxf7 h4 comes into considera­
56 'lt>g1 h4 tion) 49 .U.xd5 �g4 50 .l:i.d4+ 'lt>g5 51 �d3
57 .Ug7 'it'e4 ( 5 1 l:!.d8 is also possible) 51 .. Jbb2 52 'it>e3
58 .Ua7 'it>f3 h4 53 �f3 h3 54 .l:!.g4+?
59 .Ua3+ �g4 The only way to d raw is 54 .l:!.d5+! f5 55 l:td8
We h ave already met this position , when we h2 56 .l:!.g8 + ! �f6 57 l:!.h8, or 54 . . . 'it>g6 55
were discussing the basic ideas of endings 'it'g4 ! h2 56 .l:!.g5+ �f6 57 l:!.h5 'it>e6 58 'it>f4 .
with f- and h-pawns. 54 . . . 'it>f5 55 llf4+ 'iii e 6
60 .Ua8 'lt>g3 55 . . . �g6! wins: 56 <bg3 h2 57 .Ug4+ �f5 58
61 .Ug8+ �f3 llh4 .U.c2 , or 56 l:!.g4+ 'it'h5 57 .Ug8 .l:!.b3+58
62 .l:!.h8 .Ub1 + 'it'f4 (58 'iiif2 l:!.b1 ! ) 58 . . . llb6! 59 .Ug5+ �h6
63 'it>h2 �f2 60 .U.g3 (60 .l:!.g8 .l:!.f6+ 61 �e5 �h7 ! )
64 .Uxh4 f3 60 . . . l:!.b4+! 6 1 'it>f5 .l:!.b5+.

65 .Ua4 �f1 56 .U.h4 h2? (Black should go back with his

king: 56 . . . �f5) 57 11h6+? (57 .U.h5! is
Wh ite resigned .
essenti a l , a i m i ng to provoke . . . f7-f5 in a
After 66 Wg3 f2 67 l::!.a 2 .l:!.b3+ 68 �h2 G u l ko
situation where the black king has not yet
wanted to win in the q u i ckest way -
occupied the e5-sq uare) 57 . . . 'it'e5 58 .Uh8
68 . . . .Uf3!. The 'scientific' 68 . . . .U.e3 69 .Ub2
.l:!.c2 59 .Uh4 (the threat was . . . 'it'e5-d4-c3-
.Ue8 70 .l:!.b1 + �e2 71 .l:!.b2+ �f3 72 .l:!.b3+
b2-c1 etc . ) 59 .. .f5 60 'iit g 3
.l:!.e3 73 .l:!.b1 .U.e 1 is also good , only not
69 . . . �e 1 ?? (instead of 69 . . . lle8 ! ) , as Jose
Raul Capablanca once played in a similar
position . After 70 llb1 + �e2 his opponent
Vera Menchik could have d rawn with the
obvious 71 'it>g2 ! . But there followed 7 1
.Ub2+?? <bf3 , and Mench ik resigned . This
happened at the 1 929/30 Hastings tou rna­
ment. This curious incident shows how
careful one has to be when playing even the
most simple endings.
Gulko and I reached another i nteresting
position, important for the theory of the
endgame, when analysing 46 �c2 (instead
22 � How to Study the Endgame

The move that suggests itself, 60 .. .f4+?, 69 �g4 �f1

does not win : 6 1 �f3 l:c3+ 62 �g2 �e4 63 70 �g3 �g1
�xh2 (63 l:th8 ! ) 63 . . . l:!.c2+ 64 �h3! (64
The king has a rrived just in time!
'it>g 1 ? �e3 65 l:.h8 l:.c1 + 66 c;t>h2 f3 67 l:.e8+
�2 68 l:ta8 c;t>f1 ) 64 . . . �f3 65 l:th8 .l:.c7 66 Th i rty years later I d iscovered the possibility
l1h6 (66 �h2? 'it>f2 ) 66 . . . l:te7 67 l:.h8 'it>f2 68 of a more tenacious defence. I n stead of 66
l:ta8! f3 69 l:.a2+ l::te 2 70 l:ta 1 (or 70 .Ua8 'it11 l:th8 it makes sense to play 66 l:td 1 !? .
7 1 �g3 f2 72 �f3 ! �g 1 73 lig8+) with a
draw.
Let us imagine that it is Wh ite's tu rn to move .
He will be forced to play 6 1 .l:!.h8 (6 1 �f3 is
not possible on account of 61 . . . .l:tc1 ! 62
l:!.xh2 l:tc3+ ), and the black king can ad­
vance, bypassing its rook, in order to
approach the h-pawn along the first rank.
Note that it is the c2-sq uare that the black
rook should occu py. If it is on d2 or e2 , the
white rook is no longer obl iged to leave the
4th rank (there is the move 'it>f3 ! ) . With the
rook on b2 the king's route via the q ueenside
around its rook becomes too long .
I n other words, the position is one of mutual To win, it is sufficient for Black to return with
zugzwang. Wh ite must be given the tu rn to his king to the f-pawn while the enemy rook
move . is tied to the 1 st rank. But how to ach ieve
60 . . . l:td2 this? If 66 . . . c;t>c4 the opponent repl ies 67
61 c;t>t3 l:ta2 .U.a 1 (67 l:tf1 .l:!.d2 ; 67 'it>f4 l:tf2+ 68 �g 3
61 . . . l:td 1 ? 62 l:.xh2 l:td3+ 63 �e2 . l:td2 ), after which it is pointless to play
62 c;t>g3 67 . . . 'it>d5 68 l:a5+ c;t>e6 (68 . . . �e4 69 %1a4+
l:tc2 ! !
�e5 70 .l:!.h4) 69 .l:.a6+ �e5 70 l:lh6
63 l:th8
(intending 71 .l:.h4) 70 . . . �e4 71 .l:.e6+ �d3
63 �3 l:tc 1 ! .
72 .l:td6+ �c3 73 .l:td 1 !, a n d so on .
63 . . . �e4
Before bringing the king back, it is important
64 l:te8+ �d3
to place the rook on d2. Then the ma noeu­
65l:ld8+ vre of the wh ite rook to h6 (by analogy with
65 l:th8 l:!.e2 ! , intending 66 . . . �d2 . the variations just considered ) loses its
65 . . . �c3 strength - Black again advances his king ,
66l:lh8 and the wh ite rook ca n no longer reach d 1 .
The most accu rate is 67 . . . l:ta21 (not i m medi­
66 l:tc8+ �d2 or 66 . . . 'it>b2
ately 67 . . . l:td2 68 l:ta4+ �d5?! 69 l:th4, and
66 . . . .l:te2!
it is necessary to sta rt all over aga i n ) 68 l:tb1
In the event of 66 . . . �b2? 67 �f4 �c 1 68 lld2 69 �3 (69 l:ta 1 �d5) 69 ... �d5
c;t>xf5 �d 1 69 c;t>g4 the black king is too late. (th reatening . . . 'it>e5-f6-g5 ) 70 l:tb5+ �e6
67 'it>f4 'it>d2 7 1 lib6+ �e5 72 .l:r.h6 c;t>d4 73 .l:ld6+ �c3 7 4
68 'it>xf5 �e1 lic6+ �b2 75 .l:!.h6 �c1 , and Black wins.
 How to Study the Endgame ttJ 23

Now let us again remember the ending with .Ud 1 + 1 0 'itc6 .Uh 1 ! . With the king on f6
which we beg a n : Gl igoric-Smyslov. In the Wh ite does not have the i mporta nt move 1 1
note to Wh ite's 3rd move the variation 3 h6 .Ue 7 , and i n the event of 1 1 l::td 7 (or 1 1 'it>b 7)
J:la1 ! was analysed . 1 1 . . .'itf5 the black king succeeds, after
elimi nating the f4-paw n , in retu rning to g6 in
ti me.

I I . I M P ROVI N G YO U R TEC H N I Q U E
I w i l l now talk about how you c a n improve
you r tec h n i cal mastery. For this you need to
study problems which a re common to a l l (or
to many) types of endings. Problems, such
as the enhanced role of the king in the
endgame, zugzwa ng (and a very importa nt
specific instance of it - mutual zugzwang,
and correspond ing sq uares), the advisabil­
ity of this or that exchange, and so on. It is
especially i mportant to sense the spirit of the
endgame, to develop the optimal mood for
However, we did not consider the attempt, by
playing it, and to understa nd the psychologi­
playing 4 l:!.g7+ �f6 (the retreat to the 8th
cal chess laws which apply here.
rank is hopeless, of cou rse) 5 l:!.c7 , to reach
the position which we have just been d is­ All this is best studied by analysing practical
cussing . Knowing of the impending danger, endings, played by g reat masters of the
Black can avoid it without great d ifficulty - endgame. As an example, let us look at a
the defensive resources are q uite sufficient. game by gra n d master Ulf Andersso n .

5 . . . 'it'g6
Andersson - Franco
5 .. J::tg 1 + 6 �f3 l::t h 1 is also good .
Buenos Aires 1 979
6 h7 .l:i.h 1 ! English Opening
But here it is wrong to interpose a check: 1 tDf3 tDf6
6 . . . l::tg 1 +? 7 'it>f3 l':Lh 1 8 'it>e4 .
2 c4 g6
7 'it'f3
3 lDc3 d5
Noth ing is g iven by 7 f5+ 'itf6 . I n reply to the 4 cxd5 tDxd5
waiting move 7 l':Lb7 Black can also wait:
5 e4 lDxc3
7 . . .l::th 2, not fearing 8 l::t b 5 'it>g7 ! 9 l:tg5+
6 dxc3
�h8! . Also good is 7 . . . l:tg 1 + 8 'it>f3 .Uh 1 9
'it'e4 l:i.e 1 + , since with the rook on b 7, as we Andersson likes and knows how to play the
know, the king's route to outflank the rook is endgame, and so already in the open ing he
too long : 1 0 �d5 .Ud 1 + 1 1 'it>c6 l:tc1 + ! 1 2 happily exchanges the queens.
�b6 l:i. h 1 with a d raw. 6 . . . 'i!t'xd 1 +
7 . . .�5! 7 'it>xd 1 f6
The simplest way of demonstrating that the 8 i£.e3 e5
position is d rawn . B ut Black also does not 9 tDd2
lose after 7 . . . 'it>f6 ! ? 8 'it>e4 l:te 1 + 9 'it>d5 9 i£.c4 ! ? .
24 � How to Study the Endgame

9. . . i.e6 1 3lt:Ja5! 0-0-0+

1 0 i.c4 i.xc4 14 'itc2 i.e7
1 0 . . . 'it>f7 looks more logica l , but here too Now if 1 4 . . . h5 Wh ite would have repl ied 1 5
after 1 1 '>t>c2 lt:Jd7 1 2 b4 Black experiences .l::i. h d 1 .l:ixd 1 ( 1 5 . . . i.e7) 1 6 .l:ixd 1 i.h6? 1 7
certa in difficu lties: 1 2 . . . lt:Jb6 1 3 i.b3 or i.xh6 .l::i.x h6 1 8 c4 .l::i. h 7 1 9 c5 lt:Jd7 20 c6 with
12 . . . h5 1 3 g3, intending 14 f4 . advantage. Even so, 1 4 . . . h5 was a useful
1 1 lt:Jxc4 lt:Jd7 move - the exchange of rooks would have
eased Black's defence.
1 5 a3!
Andersson prepares an offensive on the
queenside with c3-c4-c5 . This positional
th reat provokes the opponent into d u bious
activity.
15 . . . f5? !

1 2 b4!
I n the endgame one should carefu lly watch
for the opponent's ideas and if possible
frustrate his plans. Here Black wanted to
equalise the game completely with 1 2 . . . i.c5.
12 . . . lt:Jb6?!
An inaccuracy! The only defect of Black's
position is that his bishop is more passive 1 6 i.xb6! !
than the opponent's. He should have tried to
The 'automatic' 1 7 f3 wou ld have allowed
exchange it, by playing 1 2 . . . h5!? followed by
the opponent to gain cou nter-cha nces , by
. . . i.h6. The game Andersson-Mestel (Hast­
attacking the e4-pawn ( . . . lt:Jb6-d7-f6). An­
ings 1 978/79) continued 1 3 f3 i.h6 1 4 i.f2
dersson makes a timely correction to his
lt:Jb6 1 5 i.xb6! ( 1 5 lt:Ja5? 0-0-0+ ) 1 5 . . . axb6
plan. A move earlier the exchange of minor
1 6 b5 'it>e7! (in the endgame the king is best
pieces would not have g iven anyth i n g : 1 5
placed in the centre of the board - therefore
i.xb6?! axb6 1 6 lt:Jc4 b5, but now the e5-
Black avoids queenside castl ing) 1 7 a4
pawn comes under attack.
l:thd8+ 1 8 '>tc2 'it>e6 , and Jonathan Mestel
managed to retain the balance. Wh ite acted 16 . . . axb6
more d i rectly in the game Log inov-Sideif­ 1 7lt:Jc4 i.f6?
Zade (Aktyu binsk 1 985 ): 1 3 'it>c2 i.h6 1 4 Black defends too passively. He should
i.xh6 .Uxh6 1 5 .l:thd 1 0-0-0 1 6 lt:Ja5 l:thh8 1 7 have thought about 1 7 . . . .l::i. h f8 ! . If 1 8 l:tae 1 ,
l1d3, and retained slightly the better chances. then 1 8 . . . b5! 1 9 lt:Jxe5 fxe4 20 .l:hf1 i.g5.
1 2 . . . f5!? came into consideration . After 1 8 exf5 l:lxf5 1 9 f3 Black has a choice
 How to Study the Endgame lLJ 25

between the i nteresting, althoug h q uestion­

able piece sacrifice 1 9 . . . e4? ! 20 .l:.he1 exf3
21 I:!.xe7 fxg2 22 I:!.g 1 .l:tf2+ 23 'itb3 b5 24
ttJe5 h5! (with the idea of 25 . . . c6 and
26 . .l:i.dd2 ) and the q u ieter conti nuation
. .

19 i.g5!? (th reatening 20 . . . e4 or 20 . . . b5)

. . .

20 l:the 1 b5 21 tt:Je3 i.xe3 with a n inferior,

but tenable double-rook ending (2 1 . . . .l:.ff8 ,
intending 22 . . . e4 , also comes i nto consid­
eration ). Little is changed by 1 9 .l:.hf1 i.g5
(weaker is 1 9 . . . .l:.df8 20 f3 e4 21 I:!.fe 1 .l:i.g5
22 g4! .l:i.xf3 23 .l:.xe4 .ltf6 24 ki.d 1 ! with the
threats of 25 .l:!.e8+ and 25 h4) 20 .l:!.ae 1 b5.
As was pointed out by Maxim Notki n , a
similar double-rook ending a rises after With his active play on the a-fi le Wh ite has
1 7 . . .fxe4 ! ? 1 8 l1ae 1 .l:f.hf8 1 9 .Uhf1 .1L.h4 ! ? 20 tied down the opponent's forces and forced
g 3 .ig5 21 llxe4 (2 1 a4 ! ? ) 21 . . . b5! 22 tt:Je3 his rooks to move off the open file.
.be3 23 .l:!.xe3. 26 �b3!
18 a4! This u n h u rried manner of play is typical of
Wh ite not only consolidates the position of Andersson . J ust in case he improves the
his knight at c4 , but also beg ins a n offensive position of his king and awaits a conven ient
on the queenside. 1 8 b5 was less accu rate moment for the fu rther strengthen ing of his
on account of 1 8 . . . fxe4 and 1 9 . . . 11d5. position. This is the way to convert an
advantage in the endgame - by doing
18 . . . .1L.g7
everything possible to restrict the oppo­
19 .l:!.he1 .l:.he8 nent's possibil ities, and then, without hu rry­
20 b5! ing, look for new breaches in his defences.
20 a5? b5 was less good . The target (the b6- To many the rule 'do not hurry' may seem
pawn ) should be fixed , and only then paradoxical, but in fact it is seen in practi­
attacked . cally all the endings of games by the great
20 . . . f4 masters of the endgame. Look carefully at
the endings of Capablanca and Flohr, and
21 aS bxa5
you will see with what slowness, sometimes
22 .Uxa5 b6
bordering on tedium, they convert an advan­
23 .l:.a7 tage (Sergey Belavenets).
Threatening 24 tt:Jxb6+. 26 . . . .1L.d8?!
23 . . . .ltf6 It was this that Andersson was waiting for!
24 .l:!.ea1 .l:!.e6 27 I:l.a8+ 'it>d7
25 .l:!.1 a6! 28 .l:.a2 !
Creati ng the strong th reat of 26 tt:Ja5 and 27 A convenient moment has arrived for a
tt'lc6 . reg rouping of the forces : exploiting the poor
25 . . . .l:f.de8 position of the bishop on d8, Wh ite seizes
control of the d-file. But Alisa Galliamova's
(see diagram) suggestion 28 ll6a 7 ! , with the idea of 29
26 w How to Study the Endgame

l:tb8 and 30 tLlxb6+, was possibly even 39 . . . ii.g7

stronger. 40 f3 .l:tb8
28 . . . ii.f6 41 tLle6 ii.f6
29 .Ud2+ ci;e7 42.l:lc6
30 l:ta7 ! Black resigned , since 42 . . . l:!.b7 43 .l:td8+
Of cou rse , there is no point in Wh ite leads to mate, while if 42 . . . l:tc8 , then 43 b6 is
exchanging his active rook for the oppo­ decisive.
nent's passive rook. A classic example of virtuoso endgame play!
30 . . . llc8 The study of such endings assists the
31 l:.d5 development of taste for the endgame and
32 h3 improves tech n ical mastery.
In such positions Andersson loves to make
waiting moves. From the methodological point of view it is
32 . . . ci;e7 usefu l to see the same problems d isplayed
33 tLlb2! in a negative form - by exa m i n i ng exam ples
in wh ich typical endgame m i stakes a re
The knight has done an excellent job at c4 ,
made. The fol lowing game was played on
and now it moves to d3, from where it will
the women's board i n a com petition for
support the advance of the c-pawn, and
Moscow higher education establishments i n
from where it can itself advance fu rther via
1 972/73.
b4. Note that Wh ite did not play this a move
earlier, since he was afraid of the reply
32 . . . c6 - he waited u ntil the opponent's king Sicilian Defence
had gone to e7. 1 e4 c5 2 tLlf3 tLlc6 3 d4 cxd4 4 tLlxd4 tLlf6
33 . . . 'it>e8 5 tLlc3 e6 6 tLldb5 ii.b4 7 tLld6+?! ci;e7 ! 8
33 . . . .l:!.d6 was more tenacious, after wh ich it ii.f4? ! e5! 9 tLlf5+ 'it>f8 1 0 ii.g5 d5! 1 1 ii.xf6
was best to reply 34 J::[xd6 ! �xd6 35 c4 , gxf6?! ( 1 1 . . . 1ixf6! suggests itself) 1 2 exd5
intending 36 tLld3, 37 c5+ and 38 �c4 . i.xf5 1 3 dxc6 ii.xc3+ 14 bxc3 'ii'x d 1 + 1 5
3 4 tLld3 ii.g7 .l:.xd 1 bxc6 1 6 ..1d3 ( 1 6 l:.d6 ! ? )
34 . . . c6 35 l:.dd7! cxb5 36 tLlb4 .
35 c4 i.f6
36 c5 bxc5
37 tLlxc5 .l:.e7
37 . . . .l:r.b6 38 tLld7 ! .
38 I:r.a6 !
Systematic play! With gain of tempo Wh ite
gains control of the 6th ran k - he prevents
. . . c7-c6 and obtains the e6-square for his
knight.
38 . . . ii.h8
39 ci;c4!
Agai n , just in case , Andersson improves the
position of his king . 16 . . . e4?
 How to Study the Endgame QJ 27

After achieving the better endgame. Black 28 'it>xd2 rtle7

immediately makes a positional mistake - 29 'ite3 f5
she places a pawn on a sq uare of the same 30 'itd4 'itd6
colour as her bishop. 1 6 . . . .ie6 1 7 .ie4 �e7
31 c4 f6
18 i.xc6 .l:tac8 1 9 .ie4 l:txc3 suggests itself.
32 c5+ 'itc6
1 7 �c4 .l:!.g8?
33 .id5+ �c7
Again a fu ndamental mistake - the player
with Black does not pay attention to her 34 'itc4 h6
opponent's possibil ities. Of cou rse, 1 7 . . .'ite7 35 'itd4 �h5
followed by 1 8 . . . .ie6 was correct. 36 'it>e3 'itd7
1 8 l:td6! �e7 37 h 3 �e7
19 .l:txc6 .l:i.gc8 38 'itf4 .ig6
20 l:txc8 .l:i.xc8 39 g4
21 .ib3 .l:i.xc3 A slight inaccu racy. In accorda nce with the
22 rtld2 .Uc8 principle 'do not h u rry ' , before changing the
23 l:te1 .ti.g8 pattern of the play Wh ite should have
strengthened her position to the maxi m u m .
24 g3 l:td8+
I n t h e g iven position - b y movi ng t h e a2-
25 'itc3
pawn off a light square (a square of the
same colour as the bishop). Perhaps after
39 a3! Wh ite was concerned about the reply
39 . . . h5. But after this Black, with nearly all
her pawns on squares of the colour of her
bishop, would undoubted ly be lost.
39 . . . fxg4
40 hxg4 h5!
When defending an inferior endgame it is
usefu l to exchange as many pawns as
possible.
41 gxh5
With the pawn on a3 Wh ite would have
played 41 .ixe4 .if? 42 g5! , elimi nating the
The advantage is now with White, who is f6-pawn, wh ich hinders the wh ite king.
th reatening 26 f3 . 41 . . . �xh5
25 . . . �f8? 42 'it>xe4 �g4
Moving the king away from the centre in the 43 'it>f4 .ih3
endgame is nearly always a mistake . 44 .ie4 SL.e6
25 . . . �g6 26 f3 f5 was preferable. 45 a3 .id7?
26 Ite2 .ig6 After 45 . . . .ic8 it was stil l possible to put up a
27 �d2 l:txd2? ten acious defence . The move i n the game
An incorrect evaluation . The bishop end­ loses i m mediately.
game is lost. 27 . . .Itc8+ was more tenacious. 46 c6! �e8
28 � How to Study the Endgame

Otherwise 47 �f5 . c-pawn (after which the b-pawn could

47 c7 '>t>d7 become weak) and to control the entire
48 �c6+ ! board up to the fifth rank. This is achieved by
moving the king to e3 and by placing the
Black resig ned .
rook at c3, the knight at d4, and the pawns at
b4 and f4. A fter he has attained such a
A player's endgame tech nique is based on a position, White will be able to advance his
mastery of the whole arsenal of ideas he has queenside pawns.
accu mulated - from u nderstanding the spirit The following moves a re easy to understa nd
of the endgame and its most general laws, - Capablanca consistently carries out his
to minor tech niques which he encounters pla n .
when studying his own or other players'
33 ttJd4 .l:tb7
games. To illustrate this, I will acquaint you
with one ending of my own , broken up into 34 b4 �d7
elementary components . 35 f4 �e7
First we will examine fou r ' i ntermed iate 36 �f2 .l:ta7
prod ucts', which , incidentally, a re q u ite 37 .l:tc3 �d6
instructive in themselves . 38 .l:td3 '>t>e7
39 '>t>e3 .l:ta4
1 ) It is wel l known that in the endgame the 40 .l;tc3 'it>d6
role of logical thinking increases . One must
41 ti.d3 '>t>e7
be able to compile plans, outl ine a scheme
for arranging the pieces, and so on. A classic 42 ti.c3 �d6
example is provided by the following ending. The req u i red a rrangement of the pieces has
been ach ieved . Now Capablanca wants to
regroup his forces, by playing h i s knight to
Capablanca - Ragozin
c3 (or c5 ).
Moscow 1 936
43 ttJe2 g6
44 .l:td3+ �e6
45 �d4?
In win n i ng positions even highly experi­
enced players someti mes involu nta rily relax
and make tactical oversights, riski ng losing
the fru its of their correct strategy. That is the
case here : Wh ite's last move is a serious
inaccuracy (45 f5+ ! gxf5 46 ttJf4+ or 46 . . . 'it>e7
46 ttJc3 ! was correct), which cou l d have
been exploited by 45 . . . �b5 ! . I n the event of
46 .l:te3+ �d6 Black creates the th reat of
47 . . . c5+ , while after 46 ttJc3 �xd3 47 ttJxa4
�f1 he regains his pawn.
This is what Jose Raul Capablanca writes 45 . . . .l:ta6?
about this position : 46 .l:te3+ �d6
White 's plan is to prevent the advance of the 47 ttJc3
 How to Study the Endgame .Qj 29

The queenside pawns a re now ready to 54 .Ud6+ 'it>b7

advance. I n passing 48 4Je4+ is threatened . 55 fxg5 hxg5
47 . . . f5 56 .ll g 6 l:i.f8
48 b5 It aS 57 l::i. x g5 f4
48 . .ll x a3 49 4Je4+ fxe4 50 .ll x a3 SLxb5 5 1
. .
58 4Jd4 :l.c8
.l::i.g 3 . 59 .ll g 7+ �b6
49 'it>c4 i.e6+ 60 l:tg6+ 'itb7
50 'it>b4 c5+ 61 lLlb5 .U.f8
51 bxc6 i.g8 62 4Jd6+ 'it>b8
52 4Jb5+ 'itxc6 63 h4
Black resig ned .

2 ) You will have noticed, of cou rse , that

when playing the endgame Capablanca
twice repeated moves. Here is what Sergey
Belavenets writes about this:
The repetition of moves in the endgame
plays an important role. Disregarding the
fact that it gains time for thinking, it can be
mentioned that, by repeating moves, the
active side acquires certain psychological
gains. The defender, whose position is
inferior, often cannot withstand it, and he
53 .U.d3! creates a further weakening which eases his
opponent 's task. In addition, repeating moves
Take note : Wh ite does not advance his
enables the position to be clarified to the
passed pawn , but switches to a n attack on
maximum extent. We know that some
the enemy kingside pawns. This is fu lly in
upholders of 'pure ' chess will severely
the spirit of an importa nt principle in the
criticise us for this advice. But we cannot
conversion of a n adva ntage - the 'principle
refrain from advising players: sometimes
of two weaknesses' . After creating a second
repeat moves in the endgame. In the
weakness in the opponent's position , by
struggle every chance has to be exploited,
playing against it and then , in case of
and there is nothing ugly or unethical in
necessity, again switching the attack to the
repeating moves.
first weakness, you convert you r adva ntage
i n the most methodical way.
3) Let us exa m i n e an exa mple from one of
I n the broad sense of the word , a wea kness
my own games.
i n the opponent's position can be not only a
vulnerable pawn or a badly placed piece, but
(see diagram)
also our own passed pawn, which he is
forced to blockade, o r a n invasion square , Wh ite has a n obvious advantage, but for the
which he has t o cover. moment the invasion poi nts a re secu rely
53 . . • g5 defended .
30 � How to Study the Endgame

Dvoretsky - Kikiani noeuvring, the opponent does not know

Kiev 1 970 what he should fea r i n the fi rst instance .
Kikiani decided t o prevent f3-f4 , which in
fact was hardly a threat, since it wou ld have
weakened the e4-pawn .
41 . . . g5?
42 tt'lbc5!
There will no longer be a more convenient
moment for the plan ned i nvasion at c5: the
black rook is not defending the b7-pawn,
and the bishop is stuck at d 8 .
42 . . . l:tb8
43 tt'ld7 tt'lxd7
44 .l:!.xd7+ .l:!.e7
45 tt'lc5 'itoe8
46 l:!.xe7+ cJ;; x e7
35 . . . i.d8
47 l:Id7+ 'it>e8
I did not even beg in to examine 36 tt'lbc5
seriously, since I noticed an opportu n ity to 48 l:!.xh7 tt'lc7
gain a tempo by a simple triangulation 49 h4 gxh4
manoeuvre with the bishop. 50 gxh4
36 i.a7 ! .l:!.a8 Black resig ned .
37 i.e3
Th reatening 38 tt'lbc5 . 4) Let us examine a nother endgame by
37 . . . i.e7 Capablanca.
38 i.b6
If now 38 . . . i.d8 the move 39 tt'lbc5 gains in Capablanca - Yates
strength - the b7-pawn is not defended . New York 1 924
38 . . . .l:!.ab8
We have reached the position with which we
began, but with Wh ite to move.
39 g3
The 'do not h u rry' principle in action: while
the opponent is unable to do anyth ing, all
the even slightly usefu l moves should be
made. Why not, just in case, take away the
f4-sq uare from the knig ht?
39 . . . i.d8
40 i.a7 .l:!.a8
41 i.e3
Wh ite is not averse to repeati ng his ma-
noeuvre. Faced with such u n h u rried ma- Note the pretty knight circuit, than ks to
 How to Study the Endgame ctJ 31

which White won a pawn. The threat was 64 tt:Je8+ 'it>h7 65 lZ'lf6+ 'lt>g7
40 t'Llc3 �c5 66 g5 with mate.
41 t'Lle4 .l::t b 5 64 fxg5 hxg5
42 t'Lled6 .l::t c 5 65 hxg5 ..ltg2
43 t'Llb7 .l::t c 7 66 .l:!.e8 l::. c 7
44 t'Llbxa5 66 . . . .l:txe8 67 lt:Jxe8+ Wf8 68 g6 ! .
The rest is accu rate , Capablanca-style 6 7 :ds tt:Jc6
conversion of the advantage. Wh ite's first 68 tt:le8+ 'it>f8
objective is to improve the placing of his 69 tt:Jxc7 lt:Jxd8
pieces: first his knig hts, and then his rook. 70 'it>c3
44 . . . ..ltb5 Centralisation of the king .
45 tt:ld6 ..itd7 70 . . . ..ltb7
46 t'Llac4 .i:!a7 7 1 'it>d4 ii.c8
47 t'Lle4 h6 72 g6 tZ'lb7
48 f4 ii.e8 73 tt:Je8! lt:Jd8
49 t'Lle5 .l:!a8 74 b5 'it>g8
50 l:!.c1 ii.f7 75 g5 'it>f8
51 l:tc6 ..ltg8 76 g7+ 'it>g8
52 t'Llc5 J::. e 8 77 g6
After strengthening his position to the Black resig ned .
maximum and tyi ng down the opponent's
forces, White beg ins to prepare the advance
And now see how a l l this i nformation
of his passed pawn .
enabled the following ending to be won .
53 .l:ta6 l:te7
54 'it>a3 ii.f7 Dvoretsky - Privorotsky
55 b4 t'Llc7 Kiev 1 970
56 .l:!.c6 t'Llb5+
57 'it>b2 lt:Jd4
58 .l:ta6 ii.e8
59 g4!
Aga i n , as in the game against Ragozi n ,
Capablanca operates in accordance with
the principle of two weaknesses . For a time
he defers the advance of his passed pawn
and lau nches an attack on the kingside.
59 . . . 'it>f6
60 lt:Je4+ 'lt>g7
61 lt:Jd6 ..ltb5
62 .l:ta5 ..itf1
63 .l::t a 8 g5 29 .l:!.a5 b4
32 � How to Study the Endgame

29 . . .c4 30 i.xd4 was no good , but 29 . . . 4Je6 44 . . . 4Jc3?

was worth considering. After the move in the Belavenets was right - Black could not
game Black will no longer have any withsta nd the pressure, and he h i mself
counterplay. avoids the repetition of moves.
Now, following the example of the Capa­
blanca-Ragozi n ending, Wh ite outlined a
scheme for the deployment of his pieces.
Obviously, the knight must be played to e4 ,
the king brought up to f3 , the rook placed on
a6 and the bishop on the c1 -h6 diagonal,
and , final ly, the pawn advanced to f5 .
30 4Jd2 i.e7
31 4Je4 4Jd7
32 .l:ta6 'i;f7
33 'i;f3 4Jb8
34 l:ta8 4Jd7
35 i.c1 4Jb6
36 .l:!.a6 4Jd5
37 f5 gxf5 Now Wh ite carries out a curious ci rcu lar
38 gxf5 manoeuvre with his knight, resembling that
The plan has been successfu lly carried out. which Capablanca made against Yates.

38 . . . �d7 45 4Jd2 ! 4Jd5

Here I saw that the triangulation method for 46 ttJc4 �f6

gaining a tempo, found two rounds earlier i n 47 4Je5+ was th reatened .
the g a m e against Kikia n i , m i g h t a g a i n come 47 4Jd6+ �e7
in usefu l . The only d ifference is that here the 48 4Je4
tempo is won not by the bishop, but by the After making four successive moves, the
rook. knight has retu rned to where it beg a n . But
39 l:Ic6 �c7 Black's defences are now com pletely d i sor­
40 .l:!.h6! 'it>g7 ganised . The threat is 49 4Jxf6 4Jxf6 50
41 l:Ia6 �f7 �g5. If 48 . . . �f7. then 49 �d6 is decisive .
It is now White's turn to move . 48 . . . �h8
42 i.h6 .l:!.c8 49 �e6+ �d8
42 . . . l:td7 is now bad because of 43 l:Ia8 and 50 i.g5+
44 �h8, winning the h7-pawn . Black resig ned .
43 �a? �c7
44 �a6 Thus by reflecting on the games of g reat
If 44 �a8 there is the reply 44 . . . .l:i.c6 , and so players and the recommendations which
the rook returns to a6. I n the event of they give in their commenta ries, and by
44 . . . l:Ic8 Wh ite would probably have played considering your own competitive experi­
45 i.g5 ! ? , but it did not prove necessary to ence, you can sharply improve you r tech ni­
weigh up this move. cal mastery.
 How to Study the Endgame lb 33

In conclusion I offer several exercises, i n you tra i n i n g in the practical application of the
which a rook fig hts against opposing pawns. theory of this type of endgame.
The process of trying to solve them will offer

Exerc ises

1 . Wh ite to move 2. Wh ite to move

3 . Wh ite to move 4. Wh ite to move

34 How to Study the E ndgame

5. Wh ite to move 6. Wh ite to move

7. White to move 8. White to move

 How to Study the Endgame ctJ 35

Sol utions

1 . Yu . Averbakh ( 1 980). '>tg4 h 1 'if leads only to a d raw, since 5

1 �e6! e4 �g3?? 'ifh8 is not possi ble. The g7 -pawn
2 l:i.g5! ! must be kept on the board .
The only winning move, the point of which is 1 'it.>h7 ! ! h4
to place the rook beh ind the passed pawn 1 . . . g5!? 2 ..tg6 g4 is a nother try, hoping for 3
with gain of tempo, and the n , after ascertain­ Wxh5? g3 4 'it>g4 g2 5 ..th3 � h 1 ! with a
ing the position of the enemy king , to send d raw. To avoid stalemate, the h5-pawn must
the wh ite king i n the opposite d i rection, on be left al ive : 3 Wg5 ! ! .
an outflanking manoeuvre . 2 'it>g6 h3
2 . . .'lt>d2(d3) 3 l:i.d5+! �c2 4 l::t e 5! 'it> d 3 5 3 'it>g5 h2
Wf5!; 4 ..tg4 h 1 'if
2 . . . 'lt>f2(f3) 3 l:i.f5+! 'it>g2 4 I:te5! 'it>f3 5 'it>d5 ! ; 4 . . . g5!? also fa ils to save Black: 5 'it>g3
2 . . . 'lt>e2 3 �e5 e 3 4 �e4 . h 1 �+ 6 'it>f3 g4+ 7 'lt>xg4 tLlf2+ 8 'lt>f3 tLld3 9
The hasty 1 l:i.g5? leads to a d raw in view of J:ta4 (or 9 'lt>e3 tLle5 1 0 l:i.a4 ), and the knight
1 . . . �f4 ! 2 Wf6 e4 . Both 1 �d6? e4 2 J:tg5 will soon be caught.
�d3{d2 )! and 1 �f6? e4 2 :g5 Wf3(f2 )! a re 5 Wg3
also incorrect, since the outfl a n king ma­
noeuvre becomes u n realisable.
4. V. Bron ( 1 929).
2. E. Kolesni kov ( 1 989) . 1 l:!.c8+! 'it>e7!
1 'lt> f7 suggests itself, b u t after 1 . . . ..txd4! 1 . . . �d7 2 :f8 ; 1 . . .�f7 2 :c4 .
W hite cannot win: 2 ..te6 ..te3 3 l::tf5 d4 4 2 I:tc7+ �e6
l:\.e5+ Wf2 5 l:i.d5 We3 (6 �d5 ! is not 3 .l::i.c 6+ ..te5
possible), or 2 J:tf6 We3 3 l::t e 6+ Wf3 4 :d6 4 .Uc5+ 'it>e4!
We4 5 �e6 d4 (6 Wd6! is not possible). If 4 . . . 'it>d4 , then 5 J:tf5 ..te3 6 Wg5 g3 7 'it>g4
1 l:!.f8 ! �xd4 g2 8 l:!.xf3 + .
2 �f7 'it>e4 5 l:!.c4+ 'it>e3
3 l:!.e8+! Wf3 The checks a re at an end: if 6 l:!.c3+, then
4 �d8! 'it>e4 6 . Aid2 is decisive .
5 ..te6 d4 6 �xg4! f2
6 ..td6! d3 7 l:l.g3+ 'it>e4
7 'it.>c5 'it>e3 8 l:l.g4+ �e5
8 ..tc4 d2 9 �g5+ 'it>e6
9 'it.>c3 1 0 .l:i.g6+ ..te7
The king has a rrived just in time. 1 1 I:tg7+ ..tf8
1 2 l::t g 5 ! f1 'if
3. J . Moravec ( 1 9 1 3). 1 3 �f5+ 'ifxf5
The direct 1 Wxg7? h4 2 �g6 h3 3 'it.>g5 h2 4 Stalemate .
36 � How to Study the Endgame

5. V. Sokov ( 1 940). 8 c:Ji;d6 c:Ji;c8

The routine 1 c:Ji;e7? throws away the win in 8 . . . c:Ji;e8 9 l:te7+ and 1 0 l:.e 1 .
view of 1 ... c:Ji;b4! 2 l:.e 1 (otherwise 2 ... c:Ji;c3) 9 c:Ji;c6 c:J; b8
2 . . . a5 3 c:Ji;d6 a4, and the black king 1 0 l:lg8+ c:Jl;a7
'shoulder-charges' White's. The move . . . c:Ji;b4!
11 l:.g7+ c:Ji;a6
must be prevented .
12 l:.g8 c:J;as
1 .:.b1 1 ! c:Ji;a2
1 3 c:Ji;c5
2 .l:te1 ! aS
The black king is u nable to escape from the
3 c:Jl;e7 c:Ji;b3
pursuit.
3 . . . a4 is hopeless : 4 c:Ji;d6 a3 5 c:Ji;c5 c:Ji;b2 6
l:.e2+ (6 c:Ji;b4 a2 7 l:te2+ c:Ji;b1 8 c:Ji;b3 is also
good , or 7 . . . c:Ji;c 1 8 l:.xa2 c:Ji;d 1 9 c:Ji;c3) 7 . V. Pachman ( 1 960/6 1 ) .
6 . . . c:Ji;b1 (6 . . . c:Ji;b3 7 l:.xe3+ ) 7 c:Ji;b4 a2 8 c:Ji;b3. 1 .l:tf1 ! c3
4 c:Ji;d6! 2 lih1 + ! !
Only not 4 lbe3+? �b4 5 c:Ji;d6 a4 6 l:te4+ It i s very i mportant to place the rook on g 1
c:Ji;b5! with a d raw (again a 'shoulder­ with gain of tempo. White loses after 2 c:Ji;f??
charge' ). c:Ji;h6 3 �f6 c:Ji;h5 4 c:Ji;f5 c:Ji;h4 5 c:Ji;f4 �h3 6
4. . . a4 c:Ji;e3 c:Ji;h2 ! , when he ends u p i n zugzwan g : 7
4 . . . c:Ji;b4 5 �d5 a4 6 c:Ji;d4 a3 7 l:.b 1 + . l:ta 1 �g2 or 7 l:td 1 c2 8 c:Ji;xd2 cxd 1 'ii' + 9
c:Ji;xd 1 e3 (9 . . . c:Ji;g2) 1 0 c:Ji;c2 c:Ji;g2 1 1 c:Ji;c3
5 c:Ji;c5 a3
c:Ji;f1 ! 1 2 �d3 c:Ji;f2 .
6 l::txe 3+ c:Jl;a4
2. . . c:Ji;g6
6 . . . c:Ji;b2 7 c:Ji;b4 a2 8 l:te2+ c:Ji;b1 9 c:Ji;b3 .
2 . . . c:Jl;g7 3 .l::l. g 1 + �f8 4 l:tf1 + c:Ji;e8 5 l:.h 1 .
7 c:Jl;c4 a2
3 Itg 1 + �h5
8 l:te1 'it>a3
4 c:Ji;fS � h4
9 c:Ji;c3
5 'it>t4 c:Ji;h3
6 'it>e3 c:Ji;h2
6. J. Peckover ( 1 960). 7 .l:.f1 1
1 c:Ji;d5 ! ! The same zugzwang position has now
Everyth ing else loses : a risen with Black to move.
1 :b8+? c:Ji;a5 2 llg8 c:Ji;a6; 7 • • • c:Ji;g2
1 c:Ji;e3? c:Ji;c5 2 .l:tc8+ c:Ji;b6! 3 c:Ji;f2 g 1 'iW+ (or 8 l:la1 ! c:Ji;g3
3 . . . c:Ji;b7); 9 .:.g1 + c:Ji;h2
1 !tg7? c:Ji;b3! 2 c:Ji;e3 �c4 3 .l::l.c 7+ �b5 ! . 9 . . . c:Ji;h3 1 0 .U.h 1 + c:Ji;g2 1 1 .:.a 1 ! or 1 0 . . . c:Jl;g4
1 • . • c:Ji;b3 1 1 l:.g 1 + c:Ji;f5 1 2 .:.f1 + c:Ji;e5 1 3 l:td 1 .
2 l:r.g3+ c:Jl;a4 1 0 l:.f1 !
3 l:tg4+ c:J;as Black is u n able to win .
4 .l:.g8 �b5
5 l:tg7! c:Ji;b6 8. V. Hortov ( 1 982 ).
6 l:.g6+ �c7 Which pawn should be advanced? This
7 l:r.g7+ 'it>d8 q uestion ca n only be solved by a deep
 How to Study the Endgame ctJ 37

calculation of the variations. 1 8 'it> b 7 l::rg 8 1 9 a6, also does not work.
1 g7! l::r b 8 Black repl ies 1 4 . . . �c6! 1 5 a6 .l:te8(d8) 1 6 h7
2 'it>g 1 'it>g3 .l:te7(d7)+ with perpetual check.
3 'it>f1 'iii> f3 14 . . . ltb8+

4 'it>e 1 'it>e3 1 5 Wa6! 'it>c6

5 'it>d 1 'it>d3 1 6 Wa7 l:tg8

6 'it>c1 .l:.c8+ 1 7 a5 'it>d6

7 '>t>b2 .l:b8+ 1 7 . . . 'it>c7 1 8 h 7 .

8 'it>a3 1 8 'it>b7 ! 'it>e6

8 'it>a 1 ? 'it>c2 .
1 9 a6 'it>f6

8. . . 'it>c3 20 a7 '>t>g6

9 'it>a4 �c4 2 1 a8'if

1 0 'it>a5 'it>c5 The black king was only just too late.
11 'it>a6 '>t>c6 With the pawns on g6 and h7 the king is able
to attack them a move earlier.
1 2 'it>a7 .l:tg8!
1 h7? .l:tb8 (or 1 . . . 'it>g3 2 Wg 1 .l:Ib8) 2 'it>g 1
1 3 a4 'it>d6
'it>g3 3 'it>f1 Wf3 4 'it>e 1 'it>e3 5 Wd 1 'it>d3 6
14 'it>b6! Wc1 l:tc8+ 7 'it>b2 .l:i.b8+ 8 'it>a3 'it>c3 9 'it>a4
If 1 4 �b7?, then 14 . . . 'it>e6 1 5 a5 ( 1 5 '>t>b6 Wc4 1 0 Wa5 Wc5 1 1 'it>a6 'it>c6 1 2 �a 7 l:th8!
l:'tb8+ 1 6 We? l::rg 8; 1 5 Wc6 .l:!.c8+) 1 5 . . . 'it>f6 1 3 a4 Wd6 14 'lt>b6 l:ib8+ 15 'lt>a6 Wc6 1 6
1 6 a6 �g6 1 7 a7 Wxh6. 'it> a 7 llh8 1 7 a 5 'it>d6 1 8 'it>b 7 We6 1 9 a6 'it>f6
14 a5?, hoping for 14 . . .�e6? 1 5 'it>b6 l:tb8+ 20 a7 'it>xg6, or 1 8 'it>b6 I:Ib8+ 1 9 Wa6 'lt>c6
16 'it>c7! .l:tg8 1 7 'it>c6! lieS+ ( 1 7 .. .<iio>f6 1 8 h7) 20 'lt>a7 .l:.h8 2 1 a6 .l:.e8(d 8 ) ! .
38 cj{

Mark Dvoretsky, Artur Yusu pov

The Theory and P ractice of Rook E n d i n gs

M a rk Dvoretsky

Oendings which demand the most inten­

f all the types of endings, it is rook

sive study. Why is this?

Fi rstly, they occu r more often than other
types. A good half of all the endings that
occur in practice are rook endings.
Second ly, here there exists a fai rly deta i led
theory of positions with a small amount of
material (for example, rook and pawn
agai nst rook), which may also be repeated
in our games. This theory should be mas­
tered .
I n other types of endgame the situations
with a min imal number of pawns are either A d raw is inevitable. The only way to try and
qu ite simple, or not very importa nt. This free the king from i mprison ment is by
means that there a knowledge of exact playing the rook to b8. But then the black
positions is hardly ever req uired - it is l i kely king will stand guard in place of the rook.
that you will never need it. It is sufficient to 1 l::t h 2 'it>d7 2 l::t h 8 �c7 3 .l::i. b 8 �c1 (of
know the typical ideas and methods. But in cou rse, 3 . . l::t h 1 is also possi ble) 4 �b2 �c3,
.

the rook endgame you can not get by without and White ca n not strengthen his position.
studying a considerable n u mber of exact Let us move the king and rook one file to the
positions. rig ht.

I offer for you r attention one of the sections

of rook endgame theory - endings with a
pawn on the rook's fi le. As usual , we will
begin our analysis with the simplest cases .
And in genera l , we will not delve too deeply
into theory - we will merely pick out the most
important positions and the ideas involved
with them.

1) Stronger side's king i n front

of the pawn
 The Theory and Practice of Rook Endings ctJ 39

Now Wh ite wins, since the black king does shadow' of its wh ite opponent (say, at c3) ,
not succeed in reach ing c7. or, with t h e black rook on t h e 7th rank - 'in
1 .l:!.h2 <tle7 2 .l:!.h8 �d6 If 2 . . . �d7 , then 3 the shadow' of its own rook. We merely
l::!.b 8 .l:!.a 1 4 �b7 .Ub 1 + 5 �a6 .Ua 1 + 6 �b6 mention these ideas, but we will not study
l::!. b 1 + 7 'it>c5 . With the black king at d6 it is the m . Sometimes they a re sufficient for a
no longer possible to escape via c5 , and a d raw, sometimes not.
different route has to be fou n d . Let us add a wh ite pawn on h5. Noth ing has
3 .Mb8 .Ma 1 4 'it> b 7 .l:!.b1 + 5 'it>c8 l:!.c1 + 6 'it>d8 changed . B lack does not pay any attention
l::!. h 1 7 .U.b6+ �c5 to it. It is also a d raw with a wh ite pawn on
This is the only su btle moment. It is g5.
hopeless to play 8 .l:!.e6? .l:!.a 1 or 8 .l:!.a6? But with a pawn on f5 White wins. After 1 f6+
.Mh8+ 9 Wd7 l:th7+ 1 0 �e8 l:th8+ 1 1 �f7 �f7 ( 1 . . . '1t>xf6 2 :fa+ ; 1 . . . �h7 2 f7) 2 .l:!.h8
J::!.a 8 with a d raw. Black loses his rook.
8 l:!.c6+ ! �b5 (8 . . . �d 5 9 lia6 .Uh8+ 1 0 �c7
l:.h7+ 1 1 'it>b6) 9 l::!. c 8 lih8+ 1 0 'it>c7 lih7+ 1 1 It is no accident that I have 'chewed over'
'iii b8 these elementary cases in deta i l . You
should have a very clear impression of
2) Stronger side's rook in front of the them , and should always remember and
pawn ; pawn on the 7th ran k make use of them when considering more
complicated positions.

Kha u n i n - Fridman
Len ingrad 1 962

This is a standard defensive scheme:

Black's rook is beh ind the enemy pawn, and
his king is on g7 (or h7). The wh ite rook is
tied to the pawn and ca n not move from a8. If
1 'it>b6 , then 1 ... ltb1 + . The king has no In the game there followed 1 . . . hxg3 2
shelter from the vertical checks. After d riving hxg3? g4+ ! 3 fxg4, when a d raw was
it away, the rook retu rns to a 1 . inevitable, since Wh ite was left with a
I should mention that other, more com p l i­ knig ht's pawn (whether one or two is of no
cated and less rel iable systems of defence particu lar importa nce).
also exist: the black king may hide ' i n the He cou l d have won by 2 <tlxg3! 'it>h7 3 h4!
40 � The Theory and Practice of Rook Endings

gxh4+ 4 �h3 �g7 5 f4, when the f-pawn not to release the rook from a8. The
advances with decisive effect. resulting position is known in endgame
theory as the 'Vancura position' (from the
3) Stronger side's rook in front of the name of its d iscoverer - Joseph Va ncu ra ) .
pawn ; pawn on the 6th ran k What can Wh ite d o ? If a6-a7 there always
follows . . Jla6 (of cou rse, the black king will
not move from g7 and h7). If the pawn is
defended by the king, there fol lows a series
of checks , and then the rook returns to f6 .
For example: 3 'it>d5 l:tb6 4 �c5 l:tf6! (the
best square for the rook ! ) 5 �b5 .l:!.f5+ ! etc.

If in the d iagra m we move the wh ite king to

f4 , we obtain a position which was analysed
in 1 950 by Pyotr Romanovsky. 1 . . . .l:i.f1 + ? 2
'it>e5 .l:lf6 is now bad because of 3 .l:!.g8+ ! . But
all the same there is no other plan - only the
switching of the rook to the 6th rank.
Therefore let us play 1 . . . l:tc1 ! . If 2 'it>e5 there
follows 2 . . .lk6 - and we reach Va ncura's
The m a i n difference between this position d rawn position . Wh ite must take the oppor­
and the previous ones is that the wh ite king
tun ity to remove his rook from a8: 2 l:tb8
now has a shelter agai nst vertical checks -
.l:!.a 1 3 .l:r.b6 (weaker is 3 .Ub7+ 'it>f6 4 a7
at a7. It heads for there, in order to free the 'it>e6 ).
rook from having to defend the pawn .
With the rook on a8 Black's king was tied to
But the black king is u nable to run to the the kingside, but now it ca n head towards
queenside: 1 . . . ci;f7? 2 'it>e4 (2 a7? 'itg7 is the pawn . But this must be done cautiously:
prematu re) 2 . . . cJ;;e 7 3 a7! 'itd7(f7) 4 .l:!.h8. he loses after the hasty 3 . . . �f7? 4 . 'it>e5 �e7
2 . . . .l:!.a5 (instead of 2 . . . �e7) is also hope­ 5 . .l:!.b7+ 'it>d8 6 . a 7 . Correct is 3 . .. .l:r.a5! 4 We4
less: 3 �d4 'it>g7 4 'itc4 �f7 5 'it>b4 .Ua 1 6 'it>f7! 5 'it>d4 (if 5 .l:!.h6, then 5 . . . 'it>g7 ! , but not
'it>b5 l:tb 1 + 7 'itc6 .l:!.a 1 8 'it>b 7 .l:!.b 1 + 9 'it> a 7 5 . . . cl;e7? 6 a7 Wd7 7 l:th8 ! ) 5 ... 'it>e7 6 'it>c4
'ite7 1 0 l:tb8 �c1 1 1 'it>b7 (but not 1 1 .Ub6? �d7 7 ci;b4 Ua1 , and the d raw is obvious.
'itd7) 1 1 . . .l:tb 1 + 1 2 'it>a8 .l:!.a 1 1 3 a 7 , and a Note that the kings had a race to reach the
situation that is familiar to us arises: the queenside. If the wh ite king had been closer
black king does not reach c7 in ti me. to the pawn , Black's king might not have
I n view of this analysis, Siegbert Ta rrasch a rrived in time. This means that the attempt
considered this position to be won . But later to switch the rook to the 6th ra n k can not be
(in 1 924) a saving plan was fou n d . It is delayed - this plan must be carried out as
based on the fact that the a6-pawn provides soon as possible.
the king with a shelter against vertical
checks, but not against horizontal checks.
The system of defence exam ined by us is
The rook must be transferred to f6 .
very important. In particular, th is is how the
1 . . . .Uf1 +! 2 'it>e4 .Uf6 ! defence should be a rranged when the
It is important to attack the pawn , in order opponent has two extra pawns - 'a' and ' h ' .
 The Theory and Practice of Rook Endings ltJ 41

3 . . . .l:!.b6
Strangely enoug h , i n books on the endgame
this position is not a nalysed . It had to be
stud ied independently. White wins. The
main reason is that the black rook does not
have the i mportant f6-square , and the 6th
ran k proves too short.
4 �5
Again threatening 5 l:Ia7 + .
4. . . �b5+
5 �6 l:f.b6+
6 'it>e5
6 'it>f7 .:tb7+! is pointless.
The h-pawn does not help White - the d raw
is just as elementary as in the previous 6 . . . .l:!.c6
example. If 1 'it>b5 there follows 1 ... .l:.f5+. Of cou rse, B lack does not have time to
After driving away the king, the rook contin­ captu re the g5-pawn: 6 .. J:tb5+ 7 'it>d6 (7
ues its watch of the 6th rank. If the wh ite 'it>d4) 7 .. J:txg5 8 .l:!.e8 .l:!.a5 9 .Ue7 + 'it>g6 1 0
pawn is repositioned at a5, the black rook a7. He also loses q u ickly after 6 . . . 'it>g7 7
would be placed on the 5th rank, and so o n . 'it>f5! .l:!.b5+ 8 �g4 .l:tb6 9 'it>h5 and 1 0 .l:la7+.
7 'it>d5 l:f.b6
Now let us analyse a position with a- and g­ 8 'it>c5 l:te6
pawns. 8 . . Jig6 9 .l:!.a7+ 'it>g8 1 0 'it>d4 .

1 . . . 'it> h 7 ! 9 .l:.a7+! �g6

2 �h5 In the event of 9 . . . 'it>g8 the white king returns
Threatening 3 .l:!.a7+ �g8 4 g6 and 5 'it>h6. to the kingside.
2 . . . .l:!.h6+! 1 0 'it>b5 .l:!.e5+
3 'it>g4 1 1 'it>c6 .U.e6+
3 gxh6? - stalemate! 1 2 'it>c5!
42 � The Theory and Practice of Rook Endings

The decisive zugzwa ng! familiar to us, while after 1 1 . . . 'it>g7 the rook is
deprived of an impo rtant square, from where
I n stead of 9 l:!.a7+ Wh ite also wins by 9 'it>bS it could g ive a check. N ow the wh ite king
.l:!.eS+ 10 'it>c6 .l:!.e6+ 1 1 'it>cS! (but not 1 1 boldly advances: 1 2 'it>bS lieS+ 1 3 Wc6
'it>c7? .l:.g6 1 2 a 7 .l:!.g7+ ! with a d raw). Here lie6+ 1 4 �c7 'it>h7 (there is no longer the
too Black is in zugzwang! The variation reply 1 4 . . . .l:!.g6) 1 S a7! l:!.a6 ( 1 S . . . .l:!.e7+ 1 6
1 1 . . . lieS+ 1 2 �d6 .UxgS 1 3 .Ue8 is a l ready 'it>d6) 1 6 'it>b7 and wins.

Artu r Yus u pov

Aonfidently find his way in typical rook

practical player should be able to On prophylactic g rounds it makes sense to
remove the king beforehand from the first
endgame positions. See how, making use of rank: 4 1 'lt>g2 ! ? . Now after the exchange of
the ideas we have just exami ned , I was able q ueens the black rook does not come to the
to save a difficult ending against an ex-world rea r of the passed pawn . Even so, Black
champion. gains sufficient cou nterplay, by conti n u i ng
4 1 . . . cS 42 �c4 'ifxc4 43 .l:!.xc4 �c7 followed
Karpov - Yusupov by . . . �f7-e6-dS, or 42 a6 lia7 43 'i!VaS (43
l:!.a4 f4) 43 . . .'ir'c6(d6).
Linares 1 99 1
41 a6 'iia 2
The pawn has to be halted . In the g iven
insta nce it is not the rook that is placed
behind it, but the q ueen . I thought for a long
time about the possibil ity of conti n u i ng the
fight in the middlegame, but I did not fi nd
anyth ing convincing and I decided not to
avoid the exchange of q ueens.
42 'ii'c 4+
Anatoly Karpov did not th ink for long over
this move. He had to reckon with the threat of
a cou nterattack by 42 .. J�d 1 + and 43 . . . 'i!Vb 1 .
42 . . . 'iixc4
43 .l:.xc4 .Ud 1 +
It i s Karpov to move. What possibil ities does
Of cou rse, the rook i s switched to the rea r of
he have?
the passed pawn . This very i mportant
Black must seriously reckon with 'it'c3-c4 device is merely a pa rticular insta nce of the
(immediately, or after the preparatory 4 1 general principle of rook endings, which
a6). But, after exchang i ng queens, h e can says that the rook should be active .
give a check on d1 and place his rook
44 Wg2 lia1
beh ind the passed pawn - this is a very
important defensive resource, typical of rook 45 l:.c6
endings. When a pawn is attacked from the rea r, it is
 The Theory and Practice of Rook Endings 43

usually preferable to defend it with the rook

from the side, rather than by standing in
front of the pawn . The rook on c6 is very
active - it is controlling the 6th ran k and
attacking the c?-pawn .
45 . . . �f8
Sooner or later the wh ite king will try to
break through on the queenside. Black
begins a cou nter-plan - he plays his king to
d7, in order to activate his c7-pawn or
ach ieve the exchange of several pawns.
46 f4
If 46 'it>g3, then 46 . . J:i.a4 .
46 . . . .Ua3 ! 51 . . . 'it>c7 !
Subsequently every tempo may prove deci­ 52 .Uxg7+ 'it;b6
sive - therefore the wh ite king's passage to 53 l:tg6+ Wa7
the queen side must be h i ndered as much as 54 .l:i.c6
possible. Ka rpov aims to elimi nate as many black
47 �f1 l:ta2 pawns as possible. In the event of 54 l:txh6
48 �e1 'it;e8 l:txf2 followed by 55 . . . l:txf4 the d raw is
49 �d1 �dB! obvious.

An accu rate move . 49 . . . Wd7 suggests itself, 54 . . . .Uxf2

but I was concerned that after 50 .Ug6 the 55 .l:!.xc5 .Uxf4
g7-pawn would be captu red with check. Of A text-book position with a- and h-pawns is
cou rse , 49 . . . .Uxf2? was premature in view of reached . Of cou rse, 55 . . . Wxa6 was possi­
50 a7 l:ta2 51 l:txc? , and with his king cut off ble, but it was more method ical to play 'by
along the 7th rank, Black loses qu ickly. But the book' , especially since I was short of
now the captu re on f2 is th reatened . time on the clock.
50 .l:!.g6 c5 56 l:tc6 .l:!.g4
Not 50 . . . .Uxf2? 51 l:txg7 l:ta2 52 l:tg6 . 57 �d2 �g5
Therefore Black activates his passed pawn . 58 .Uxh6 f4
51 Wc1 59 �e2 f3+
In such situations one somewhere has to Black does not need this pawn . If you know
stop making common sense moves a n d , for certa in that a position is d rawn , you
after accu rately calculating a way t o d raw, should try to ensure that extraneous details
force events . Such a moment has now (like a 'non-theoretical' pawn ) do not acci­
arrived . dentally h inder you .
6 0 �xf3 .Uc5
61 l:th8
Here , just in case, I adjourned the game. To
my surprise, Karpov tu rned up for the
resumption and made a few more moves.
44 � The Theory and Practice of Rook Endings

61 . . . �g5 Therefore Black must first g ive some checks:

62 'it>e4 l:!.c5 68 . . . l:tc7+! 69 '1t>f6(e6) .l:lc6+ with a d raw. But
on the d-file the rook would be too close to
63 'it>f4 �c4+
the king , and after 68 . . . .l:ld7+ 69 'it>e6 Black
64 'it>e5 .Uc5+ would lose.
65 'it>e6 .Ug5 67 l:!.h7 �xa6
66 �f7 .Uc5! Only now, when the rook has gone to h 7,
In such positions the c-file is the best place ca n the a6-pawn be taken . But with the rook
for the rook. If now 67 h6 l:c6 68 h 7, then the on h8 it should be ignored .
rook should be placed beh ind the pawn , but 68 h6 l:!.c7+
the immediate 68 . . . l:!.h6 loses to 69 'it>g7 . Draw.

M a rk Dvorets ky

1\ lthough the ideas that we have been 7 1 l:!.b3 �c4 72 .l:lf3 �b4 73 l:!.f4+! etc. When
./"\d iscussing are elementary, by no means you know the plan of defence, the moves
all players are familiar with them . Even make themselves - here there is noth ing
grandmasters sometimes 'flounder' in stand­ cu nning.
ard theoretical endings. Here is a tragic­ However, the highly experienced grandmas­
comic example. ter Laszlo Szabo had no idea of how to play
these types of endings, and he lost a
Szabo - Tu kmakov completely d rawn position . Apparently Vladi­
Buenos Aires 1 970 mir Tukmakov also did not know them , since
he commented on the cou rse of the play as
follows: 'Theory considers this endgame to
be drawn , but I seemed to win q u ite
convincingly. '
6 6 'lt>g2?! Wd6
67 '.tf2?! l1a2+
68 'it>e1 ?
68 'it>g 1 ! would sti l l have led to a d raw.
68 . . . .l::t a 1 + !
6 9 'it>e2
69 Wd2 l:i.h 1 ! 70 .l:!.xa5 h3 7 1 l:!.h5 h2 and
72 . . . l:i.a 1 .
69 . . . a4
70 l:!.h6+
For Wh ite it is sufficient simply to wait,
70 l:!.xh4 a3 7 1 .l:!.a4 a2.
keeping the a5-pawn under fi re , in order not
to release the rook from the a-file. For 70 . . . 'it>e5
example, 66 l:!.b5 �d6 67 l:!.f5 .l:i.a 1 68 'it>h2! 7 1 .Uh5+ 'it>f6
a4 69 l:!.f4 ! a3 70 l:!.f3 ! 'it>c5 (70 . . . a2 71 l:!.a3) 72 �f2 a3
 The Theory and Practice of Rook Endings l2J 45

73 'it>g2 .Uc1 48 . . . 'it>g6

74 l1a5 l:Ic3 49 h5+ �f7
White resigned . A clear demonstration of Black's ignorance
of theory: G u revich , l i ke Szabo i n the
I found another, similar example i n the previous example, incorrectly moves his
magazine New in Chess, i n an article by king to the opposite wing.
Tony Miles about the 1 989 U SA Champion­ 50 Wf4 �c6
ship. He analyses the ending of a game by 51 'it>e5 'it>e7
the winner of the championship, Stua rt
52 'it>d5 .Uh6
Rachels, against g randmaster Dmitry G u re­
53 Wc4 �h8
vich . It would appear that none of the m ,
includ ing the commentator, w a s fam i l i a r with 54 l:te5+ 'it>f6
the ideas of the g iven endgame. 55 l:tc5 rt;e7
56 rt;b4 .l:!.h6
Rachels - G u revich 57 �a5 Wd6
U SA Championship, Long Beach 1 989 58 l:tg5 �c7
59 a4 Wd7
60 .l:!.g7+ Wc8
61 .Ug5
Here the game was adjourned . In home
a nalysis it is i mportant to look i n a book and
fam i l i a rise you rself with the theory of the
endings that may a rise d u ring the resump­
tion - i n the g iven instance, with the theory
of the endgame with two extra a- and h­
pawns. But G u revich d i d not do this.
61 . . . c.t>d7
62 .ti.c5 Wd8
63 .ti.d5+ rt;e7
Miles writes: 'Si nce rook and a- and h­ Black changes his plan of defence and
pawns vs rook is often d rawn , it is hard to returns his king to the kingside. I n the event
believe that Wh ite should win this. The of 63 . . . <t;c7 he was probably concerned
defensive task is not simple, though . ' about 64 rt;b4 with the threat of supporting
O f course, i t is not essential to give u p the the h-pawn with the king .
pawn , but from the practical point of view 64 'it>b4 We6
this is the best cou rse. Subsequently you no 65 �c5 �e7
longer have to think, since you will be acti ng 66 l:tg5 �f7
in accordance with theory. Remember: this 67 �d5 �f8
is h ow Yusu pov acted i n his game agai nst
68 .Ue5 Wf7
Karpov. Otherwise you will be forced to act
independently in a position which , although 69 We4 l:.c6
drawn , is unfamiliar, and it will be easy to 70 <t;t5 .l:!.c4
make a mistake . 71 .Ua5?
46 w The Theory and Practice of Rook Endings

71 a5 was stronger. 79 'it>eS .l:tc6

80 l:.d7

71 . . . l:tc6
71 . . . 'it>g7! 72 .l:.xa6 'it>h7 would have forced a Here M iles makes an amusing com ment: 'If
draw. I n principle, delaying this is now not 80 . . . 'it>f8 , then 8 1 l::i. d 6 l:!.c5+ 82 'it>f6 'it>g8 (or
without its dangers - after a l l , Black has to 82 . . . '1t>e8 83 h6} 83 �a6 wins.' But 83
reckon with the following plan: the wh ite .U.xa6?? 'it>h7! leads to an i m mediate d raw,
pawn goes to a5, the rook defends every­ whereas 83 'it>g6! wins.
thing along the 5th rank, and the king heads 80 . . . 'it>h8 could have been tried , i n the hope
for b7. of 81 .l:.d6 l:tc5+ 82 .l:!.d5 (82 'it>f6 'it?h7!)
72 'it>gS 'it>g7 82 ... �c6 83 'lt>f5 'it>g7 84 .l:.e5 (with the threat
73 .U.d5 �c4 of 85 l::i. e 7+ and 86 .l:!.e6) 84 . . . 'it>f7 ! . But the
74 aS l::i. c 6 subtle move 81 �e7 ! enables Wh ite to win:
75 �d7+ 'it>g8 after 8 1 . . . '1t>g8 82 'it>f5! �c5+ (82 . . . 'lt>f8 or
82 . . . �h6 - 83 �e6 ! and 84 'lt>g6) 83 .l:.e5!
76 l::i. a 7?
(only not 83 'it>f6? l::i.x h5 84 'it?g6 'it?f8 ! ) he
After 76 �e7! Wh ite's position is appa rently wins. This position occu rred later i n the
now won . For example: 76 . . . �c5+ 77 'it>g6 game .
.U.c6+ 78 'it>f5 l::i. c 5+ 79 �e5, or 76 .. .'it>f8 77
80 . . . .l:th6
.l:!.b7! (with the deadly th reat of 78 .l:l.b6)
77 . . .l:tc5+ 78 'it>g6 .l:!.c6+ 79 'it>h 7. 8 1 'it>fS
76 . . . �d6? The sealed move . Here the game was again
adjourned .
It was essential for Black to exploit his
opponent's mistake, by playing 76 . . . �c5+ 81 . . . �c6
77 'it>f6 �c6+ (neither 77 .. Jbh5? nor In the event of 81 . . . 'it>f8 ! ? (not allowing 82
77 . . .lba5? is possible, in view of 78 'lt>g6) .Ue7 ) 82 c.t>g5 l:tc6 Wh ite wou l d have won by
78 'it>e7 l:k5 (or - as recommended by M iles 83 �b7 ! .
- 78 . . . .Uh6) 79 l:txa6 .l:txh5 (or 79 . . . 'it>h7) with 8 2 .Ue7 ! l:tc5+
an obvious d raw. 83 l:i.e5 .Uc1
77 'it>fS .l:!.dS+ After 83 . . . l:i.c6 84 �e6 �c1 Wh ite has a
78 'it>f6 .l:.d6+ pleasant choice between 85 c.t>g6 and 85
 The Theory and Practice of Rook Endings 47

l::tx a6 �f7 (u nfortunately, 85 . . . 'it>h7 is not Baku l i n - Dvoretsky

possible because of 86 .l:.e6) 86 l:!.a7+ 'it>f8 Moscow Tea m C h a m pionship 1 974
87 .!::!. a 8+ 'it>f7 88 a6 �c5+ 89 <>t>e4 �c6 90 a?
l::!.a6 9 1 .l:!.h8. The fla n k attack on the rook's
pawn only works if the king is on g7 or h?.
84 'it>e6 !
The king heads towards the a6-pawn. Black
can no longer save the game.
84 Wg7 85 'it>d6 Wh7 86 l:!.c5 l:!.b1 87 Wc6
...

'it>h6 88 l:!.d5 .Ub2 89 l:!.d7 l:tb5 90 .Ud6+

'it>h7 91 .Ud5 l:!.b1 92 l:!.d7+ Wh6 93 l:!.b7
l::ta 1 94 'it>b6 <>t>xh5 95 'it>xa6 'it>g6 96 l:!.b5
00 97 'it>b7 .Ue1 98 a6 l:!.e7+ 99 Wb6 l:!.e6+
100 �a5 Black resig ned .
In these last two examples both the moves
and the comments made by g randmasters
make a comic i mpression , for one simple Reckoning that if 1 8 . . . �e6 the reply 1 9 tt:Jc5
reason - they were not sufficiently fam i l i a r was u n pleasant, I wa nted to play 1 8 . . . tt:Jc6.
with the basic theory o f rook endings. But my sense of da nger operated and I
began to have dou bts about the position
arising after 1 9 c4 d4 20 tt:Jc5 b6 21 tbd3.
Our next step should probably be an
Wh ite creates a pawn majority on the
analysis of positions that a re closely l i n ked
queenside, he secu rely blockades the d4-
with the type of endgame a l ready stud ied -
pawn , and he controls the e-file. The
namely, endings in which each side has two
adva ntage is on his side: perhaps not a
or th ree pawns on the kingside and one has
g reat advantage, but an enduring one.
an extra passed pawn on the q ueenside
When I showed this ending to Rafael
(usually a rook's pawn ). Such a situation
Vaga n i a n , an expert on the French Defence,
often occu rs in practice . But th is is a l ready
he assessed the position as unfavourable
another topic, and here we will merely
for Black.
mention it. I will restrict myself to one
example, in which the same idea was used By accu rately defending, one can probably
as we saw in the Karpov-Yusupov game, avoid defeat, but this is a d ifficult and
but did not see in the Rachels-Gu revich than kless task. Being an active player, I
end ing : the sacrifice of a pawn to transpose usually avoided this type of passive de­
into a theoretically d rawn position . fence , and endeavoured to fi nd a way of
sharply changing the cou rse of the play, of
forcing events , either with the aim of
clarifying the situation, or, on the contra ry, of
compl icati ng the play as much as possible.
Returning to the rook move to e6, I q u ickly
fou n d the forcing variation which occu rred in
the game.
18 . . . l:!.e6 ! ?
1 9 tt:Jc5 �xe1 +
48 � The Theory and Practice of Rook Endings

20 l:ixe1 �c8! Thirty years later I realised that endings of

20 . . . lt:Jg6 21 lt:Jxb7 .l::t b 8 22 lt:Jc5 �xb2 23 this type were by no means as harmless as I
l:te8+ lt::lf8 is not worth considering - the pin then thought. In 2003 in the theory of rook
on the knight is extremely dangerous. For endings with an extra pawn on the wing a
example, Wh ite ca n play 24 g3 f6 25 lt:Jd7 revolution occurred, and positions which
�f7 26 l:txf8+ 'it;e7 27 �b8 . had seemed completely drawn proved to be
won or at the least very dangerous. You can
21 l:l.xe7 'it>f8
find these new ideas in my articles or those
As you see, combinative vision sometimes of grandmaster Carsten Muller in the ar­
helps even when playing 'ted ious' end­ chives of the internet site Chesscafe. com,
games. and also in my book Dvoretsky's Endgame
22 .l:f.xb7 Manual from the same publisher. If I had
In the event of 22 l:ie2 l:l.xc5 the position is known about the future discoveries, I would
roughly eq ual. perhaps have rejected the combination (in
22 . . . .l:txc5 favour of 1 8. . . l:l.e8!? followed by 1 9 . . . 'it>f8),
and at the least I would have played the
23 c3 d4
endgame more carefully.
24 �f1
However, my decision to sacrifice a pawn
24 l:tb3?? d3 25 'it>f1 .l:te5 ! . was taken not only on purely chess g rounds
24 . . . dxc3 - it was i nfluenced , as wel l as the subse­
25 bxc3 �xc3 quent play, by certain extraneous factors . At
26 l:.xa7 Uc2 that time I was teaching in the chess
department of the I nstitute of Physical
Cu ltu re. We were planning to hold a tou rna­
ment on the Scheveningen system : stu­
dents (candidate masters) against masters,
and I had to fi nd some opponents for our
students. The Moscow Tea m Champion­
ship, held in the Centra l Chess Club,
attracted nearly all the Moscow masters ,
and this was the most convenient place to
hold discussions. After transposing i nto a
comparatively simple ending soon after the
start of play, I gai ned the opportun ity, by
making my moves without much thought, to
rush to the other boards and press players
who were strolling about to take pa rt in our
tou rnament.
I knew for sure that this was a d raw, and a
fairly simple one, and so without hesitation I Because my attention was d ivided , some of
went in for the exchanging combination. Of my moves were not the most accu rate , but I
cou rse , had I not made a previous study of nevertheless gai ned a d raw.
this type of endgame, I would hardly have 27 g3 g6 28 <it;g2 '.t>g7 29 'it;t3 h5 30 h4 'it>f6
decided to give up a pawn . Who knows how 3 1 'it>e3 �c3+
the game would have ended after 1 8 . . . lt:Jc6 , 31 . . . We6 or 31 . . . '>i>e5 ! ? was simpler, keep­
whereas as it was I easily made a d raw. ing the f-pawn under attack.
 The Theory and Practice of Rook Endings ctJ 49

32 �e4 �c2 33 f3 .l:I.e2+ view of 36 ... .l:!.b4+ 37 'it>e3 .l:I.b3+ 38 'it>e4

With the a-pawn stil l on its i n itial sq uare, .l:tb4+ 39 'it>d5 l:tb3, a lthough someti mes
Black could have permitted himself 33 . . .lk4+ (but, I th ink, not in the g iven instance) the
34 �d3 �c1 , intending to attack the g3- resulting complications favou r Wh ite .
pawn by 35 . . . �g 1 . 36 .l:!.a3 'it>f6 37 .l:ta6+ 'it>g7 38 .l:!.a4 'it>f6
34 �4 Ub2 35 .l:!.a6+ 'it>g7 (38 . . . f6 ! ? ; 38 . . . .l:l.f2 ! ? ) 39 g4 hxg4 40 fxg4
The confining of the king at g7 is rather �f2+ 41 'lt>g3 .l:I.c2 42 .l:tf4+ (42 g5+ 'it>e5 is
unpleasant for Black. However, i n such not dangerous for Black) 42 ... 'it>e6 43 a4 (43
situations he has ava ilable q u ite a good plan .l:I.f2 .l:I.c3+ 44 Wf4 f6 ) 43 ... f5 44 gxf5+ gxf5
of cou nterplay: .. .f7-f6 and at the first 45 .l:I.f2 .l:l.c4 46 .l:l.a2 .l:I.c3+ 47 'it>f4 lk4+ 48
conven ient opportu n ity . . . g6-g5. For exam­ 'l.t>g3 (48 'l.t>g5 .l:l.g4+ 49 Wh5 'it>f6 50 a5??
ple, 36 a3!? .l:I.b3 37 'it>e4 f6 ! ? . And it is not .l:l.g8 ) 48 ... .l:l.c3+ 49 'it>g2 .l:l.c4 50 h5 .l:!.h4
easy for the opponent to decide on 36 a4 in Draw.

Artu r Yus u pov

I extracts from my games, in which practical ,

shou ld now l ike to show you several fice - a procedure which has a l ready been
mentioned here severa l times. In the given
rathe r than purely theoretical rook endings instance the pawn is g iven up for the sake of
arose. However, initially the first example activating his own forces.
does not resemble an endgame at a l l . 2 1 .i.xf5 tt:Jxf5
22 tt::lxf5 'i!Vxf5
Lj u bojevic - Yus u pov 23 'ifxc5 .l:!.xd2
Linares 1 99 1 24 �xc7 .l:!.c8
25 1t'd6 .l:I.xc3
26 'i!Vxa6 .l:!.cc2
Of cou rse, the extra passed a-pawn is
da ngerous, but Black correctly calculated
that the pressu re of his rooks along the 2nd
ran k would enable him to mai ntain the
bala nce.
27 ifb6
If 27 �a7 , then 27 . . . d4 28 'iia 8+ 'it>h7 29 a6
(29 1i'f3 �xf3 30 gxf3 Ua2 is roughly the
same as that which occu rred in the game)
29 . . . .l:txf2 30 a7 �xg2+ 31 ifxg2 .l:I.xg2+ 32
'it>xg2 �g4+ with perpetual check.
20 . . . .i.f5! 27 . . . d4
Black carries out a tactical exchanging 28 iVd8+ 'it>h7
operation, involving a positional pawn sacri- 29 'ii' h 4 g5
50 � The Theory and Practice of Rook Endings

As Dvoretsky pointed out, 29 . . . iixe5 30 a6 The fol lowing example is also devoted to
l:ta2 was simpler, since if 31 a7?! Black has rook activity. Genera l ly speaking , the main
31 .. J:txa 1 32 l:txa 1 .l:i.a2 ! , and he remains principle in rook endings is that the rook
with an extra pawn. should be active!
30 'i¥h3
I n the event of 30 il'g3 Black would have Yusu pov - Barbero
conti nued 30 . . J:td3 31 f3 l:td2 with the Mendoza 1 985
dangerous th reat of 32 . . . '1Wf4 .
30 . . . iixh3
31 gxh3

Wh ite has a slight advantage thanks to the

fact that his rook is more active , and also as
a result of the rather un usual position of the
Things seem to be bad - after all, the wh ite black king at h6.
rook is positioned beh ind the passed a­ Now the most natural try seems to be the
pawn . However, thanks to a tactica l su btlety activation of the black rook by 28 . . . l:td8 29
Black nevertheless succeeds in stopping l:te7 b5 30 l:txa7 .U.d2 31 b3 c4 32 bxc4 bxc4 .
the pawn from beh ind. If the king were not at h6, Wh ite would
i mmediately have to agree to a d raw in view
31 . . . .tr.a2
of the unavoidable exchange of the queen­
32 a6 .l:.xf2!
side pawns. But here he can stil l play for a
33 l:txa2 win by 33 h4! c3 34 'lt>g2 c2 35 .l:!.c7 c 1 'ii
Forced . (35 . . . g5 is simpler, obtaining a theoretically
33 . . . .l:!.xa2 d rawn ending with h-pawn against f- and g­
34 .tr.xf7+ �g8 pawns ) 36 llxc1 .l:.xa2 37 .l:.c7 with the th reat
of 38 g4.
35 .l:.d7
Gera rdo Barbero carried out an operation
35 .Uf6 d3.
which also makes sense. Exploiting the fact
35 . . . l:txa6 that the pawn endgame is satisfactory for
36 .l:.xd4 'i;f7 Black, he decided to secure the 7th ran k for
37 h4 his rook.
Draw. 28 . . . l:tg8
 The Theory and Practice of Rook Endings lZJ 51

29 �1 l:.g7 36 hxg6+ hxg6

30 'it>e2 l:r.d7 37 f4
31 h4 Wh ite now has a serious advantage. B lack's
The immediate 31 l:.e5 came i nto considera­ king is tied to the g6-pawn, and his rook has
tion . to defend the 7th ran k - its activity will lead
31 . . . 'it>g7 to the loss of a pawn . And against passive
defence Wh ite is free to strengthen his
32 l::te 5! b6
position .
33 l:te6
37 . . .
38 b3
39 'it>e3 'it>g7
40 'i;e4 'it>t7
41 'i;f3
4 1 We5 is also good .
41 . . . .l:!.e7
42 'i;g4 l:r.d7
43 Wh4
Th reatening 44 g4 and 45 'it>g5 , when after a
check on the 5th ran k there follows f4-f5 or
'it>h6.
43 . . . l:.d2
Black decides to play actively, but it would
33 . . . 'it>f7?! have been better to do this a few moves
Black should have deprived the wh ite rook earlier.
of the i mportant c6-square. After 33 .. Jk7! 44 .l:!.c7+ Wf6
34 l::t d 6 Wf7 35 h5 'i;e7 the position would
45 .l:!.xa7 b5?
have remained roughly equ a l .
45 . . .Wf5 was more tenacious.
34 .l:lc6
46 .l:!.c7?!
The white rook is very wel l placed . It cuts off
After 46 .l:la5! 'itf5 47 '>t>h3 a second pawn
the enemy king along the sixth ran k and
would have been lost.
prevents the advance of Black's q ueenside
pawns. 46 . . . c4
34 . . . 'it>e7?! 47 I!.c6+ ! cJ;ts?
It was better to take active measu res to 47 . . . Wf7 was more tenacious, but this too
divert Wh ite from his plan ned offensive on would not have saved the game: 48 bxc4
the kingside: 34 . . .l:.e7+ 35 Wd 3 l::td 7+ 36 bxc4 49 a4 ! .l:!.d4 50 'it>g5 .l:!.d5+ 51 'it>g4 l:Id4
'it>c3 l::te 7, intending . . . .l::!. e 2 . 52 l:tc5! 'it>f6 (52 . . . c3 53 a5 .l:!.a4 54 'i;g5) 53
a5, and B lack is in zugzwang (53 .. J::t e4 54
35 h5! 'i;f7?
.l:!.c6+ 'it>f7 55 'it>g5).
Again passively played . 35 . . . gxh5! 36 .l:!.h6
48 l::tc 5+ 'i;e6
'it>d8 was essentia l . After playing his king to
c7, Black could then have advanced his b­ 49 l:!.xb5 c3
and c-pawns. 50 l::tc 5 c2
52 � The Theory and Practice of Rook Endings

51 b4 g5+ but he also loses after 48 'it>g4? 'ifi>d3 49 h5

52 fxg5 .l:.d4+ 'it>e4 50 'it>g5 'it>e5 5 1 'it>g6 'it>e6 52 h6 l:f.g 1 + .
53 'it>h5 l:txb4 From f4 the wh ite king 'shoulder-cha rges'
Black's, not al lowi ng it to approach the
54 l:txc2
pawn .
Black resigned .
48 . . . 'itd3 49 h5 .Uh1 50 'itg5 'it>e4 51 h6
I n itially the ending was al most eq ual . What 'Wte5 52 'itg6 'Wte6 53 'lt>g7 ! (but not 53 h7?
was the reason for Black's defeat? Fi rstly, l:tg 1 + 54 'ith6 cJilf7 55 h8lt:J+ 'it>f6 56 'ith 7
he defended very passively, and was too l:tg2 , and Black wins) 53 . . . 'it>e7 (53 . . . l:.g 1 +
late in activating his rook. And secondly, he 54 'it>f8 ) 54 h7 l:tg 1 + 55 'it>hB! with a d raw.
was caused a mass of problems by the
U nfortunately, my opponent fou n d a much
excellent position of the wh ite rook. Note the
stronger possibil ity.
intermed iate move 32 .Ue5 ! , wh ich secu red
the ideal square c6 for the rook. From here it 43 . . . .l:.f1 + !
tied down l iterally all the opponent's pieces 44 'it>g4 hxg3
and pawns. Now after 45 'it>xg3 'it>d3 46 :ta2 c3 4 7 h4 c2
48 l:1xc2 'it>xc2 Wh ite can no longer save the
Yusu pov - Tsesh kovsky game, since his king cannot go to f4 .
Moscow 1 98 1 45 .l:td2+ 'it>e3
46 l:tg2
46 l:tc2 would not have hel ped in view of
46 . . . l:r.f8 ! (46 . . . l:tf4+! 47 'it>xg3 l:td4 fol lowed
by 48 . . .<it>d3 is no less strong) 47 'it>xg3
l:!.g8 + , and the king on the h-file is extremely
badly placed . For example, 48 'it>h4 'it>d3 49
l:t.a2 c3 50 'it>h5 c2 51 lla 1 'it>d2 52 h4 c H i
5 3 .:txc 1 'it>xc1 5 4 'it>h6 'it>d2 5 5 h5 'it>e3 56
'it>h 7 l:tg 1 57 h6 'it>f4 58 'it>h8 'it>g5 59 h 7
'itg6. Or 48 c;i;>h2 'it>d3 49 .l:.a2 c3 50 h4 c2 5 1
l:ta 1 'it> d 2 5 2 W h 3 c 1 'ili' 53 l:txc1 'Wtxc 1 5 4 h5
'it>d2 55 'it>h4 'it>e3 56 h6 Wf4 57 'it>h5 .l:Ig5+ -
in both cases Wh ite loses.
46 . . . .Uf4+1
A typical situation : soon I will have to give u p 47 'iitx g3 c3
m y rook for the c-pawn and an endgame 48 h4 .l:Ic4
with rook against pawn will be reached .
49 l:.c2 $>d3
Every tempo may have a decisive infl uence
on the outcome. But now the fact that the wh ite king is cut off
along the 4th ra n k proves decisive.
The d irect 43 ... hxg3? (43 . . . 'it>d3? 44 l:.f2 ! or
44 l:tg2! comes to the same thing) 44 'it>xg3 50 l:lc1 c2
'ifi>d3 45 l:.a2 al lows White to save the game. 51 h5 'it>d2
The main variation is q u ite instructive. 52 l:th1 c 1 'ji'
45 . . . c3 46 h4 c2 47 l:txc2 Wxc2 48 'it>t4! 53 J:txc 1 'lt>xc1
Of cou rse , 48 h5?? l:td4 ! is bad for Wh ite , Wh ite resigned .
 The Theory and Practice of Rook Endings ltJ 53

Two ideas , typical of such endings, were zugzwang (for exa mple, 4 4 . . . .l::!. a 2 45 h4 ).
clearly seen d u ring the cou rse of the play: He is forced to excha nge pawns: 44. . . e2+
1) th e cutting off of the king along the 4th 45 'it>xe2 .Uxh3. N ow, after slightly improvi ng
rank - thanks to this, Black won the game; the position of his rook in a typical way: 46
2) the 'shoulder-charge' - Wh ite was hoping l:i.f6+! 'it>g7 47 l:i.c6 (threatening 48 l:i.c7+
to save hi mself by employing this device , 'it'f8 49 a7) 47 . . . 'it'f7 , Wh ite plays 48 'it'd2.
b u t with an i ntermed iate check Vitaly If 48 . . . l:i.a3 there follows 49 'it'c2 lla5 50
Tseshkovsky disru pted my plan. 'it>c3 .l:!.xg5 51 'it>b4 .U.g 1 52 'it'c5 g5 53 a7
l:i.a 1 54 'it'b6 g4 55 Wb7 and wins (the black
In the following, more complicated ending,
king is cut off from its passed pawn ).
similar motifs occu rred . And if 48 . . . .Ug3, then 49 a7 .Ua3 50 l1c7+
'it'e6 51 'it>c2 Wf5 52 '1t>b2 l1a6 53 'it'b3
'it>xg5 54 Wb4 (th reatening 55 .l:lc5+ and 56
Yus u pov - Timman
.l::!. a 5) 54 . . . 'it>h6 (the only defence) 55 'it'b5
Candidates Match , 5th Game, Ti lburg 1 986 .l::!.a 1 56 'it'b6 .

The eval uation of the position is not in Look in Mark Dvoretsky's book School of
question - Wh ite has a big adva ntage. The Chess Excellence 1: Endgame Analysis -
logical move was 38 a4! - it is i mportant to there in the chapter ' Rook against Pawns' a
advance the passed pawn as soon as very similar position is analysed . The best
possible. Black's passed pawn is not dan­ defence - 56 .. Jib1 + (if 56 . . . g5 the most
gerous - to . . . e4-e3 there is always the accu rate is 57 .l::!. c8 ! ) 57 'it>c6 .l::!. a 1 58 'it>b7
reply 'it>f1 .
·

IIb1 + 59 'it'c8 .l::!. a 1 - all the same does not

How could the game have developed? help: 60 'it'b8 'it>g5 61 a81i' I:Ixa8+ 62 Wxa8
38 l:i.d3 39 a5 c3 40 bxc3 e3 (th reatening
... 'it'f4 63 .Uf7+ ! (a typical i ntermed iate check
41 . . .l:i.d 1 + 42 'it>g2 e2 ) 41 'it>f1 .l:l.xc3 42 a6 to gain a tempo; the hasty 63 Wb 7? g5 leads
:t:l.a3 43 .l:l.b6 � (43 . . . g5 is bad : 44 Ug6+ to a d raw) 63 . . . 'it>e4 64 .l::!. g 7 ! 'it>f5 65 'it>b7 g5
rttf7 45 .l:i.xg5 l:i.xa6 46 .Ue5 l:i.a3 4 7 'it>e2 66 'it>c6 g4 67 �d5 ..t>f4 68 'it>d4 ..t>f3 69
:t:l.a2+ 48 'iin3 l:th2 49 'it>g3 l:.e2 50 'it>f4) 44 Wd3 g3 70 .l::!.f7 + and 71 'it'e2. Or 60 . . . g5 6 1
g5. a81i' l:txa8+ 6 2 'it'xa8 Wh5 (62 . . . g 4 6 3 .Uc5!
If Black stays passive, he ends u p i n - cutting off the king ! ) 63 'it'b7 'it'g4 64 'it>c6
54 � The Theory and Practice of Rook Endings

'it>f3 65 .l:if7+! We3 66 .l:ig7! 'it>f4 67 Wd5 etc. 40 . . . l:!.xc3?

Incidentally, now you will see why on the On the last move before the time control Jan
46th and 47th moves White replaced his Ti mman makes a decisive m i stake. He
rook at c6 - i n order to free the b6-square for thought that he wou l d always be able to
his king. Such 'trifles' sometimes decisively advance his pawn to e3, but he d i d not take
affect the result of the game, and on no account of the strong i m peding reply 41
account should they be disrega rded . .l:ie5 ! .
Unfortunately, in a time scramble I com mit­ If Ti m man h a d seen th is, the n , even without
ted a sign ificant i naccu racy. delving i nto variations, simply by the method
38 Wf2 .:d3 of com parison he would have preferred
40 . . . e3+ ! 41 'i.t>e2 .:xc3 . Here the black rook
Of course, the king must not be allowed to
is slig htly more active , and the wh ite king is
go to e3.
slig htly fu rther from its kingside pawns than
39 a4 c3 in the game. After 42 g5 l:!.a3 43 a5 '>t.>f7 44
39 . . . .l:ixh3? would have lost: 40 l:!.c5 l:!.b3 4 1 l:!.e5 '>t.>f8 the position wou l d apparently have
.l:ixc4 .l:ixb2+ 4 2 We3 . After the inevitable fal l been d rawn . For example: 45 l:!.e6 .l:i.xa5
o f t h e e4-pawn Wh ite ach ieves a n ideal (45 . . . l:!.a2+ followed by 46 . . J::t x a5 is even
construction - his rook defends both pawns more precise) 46 I!.xg6 .l:te5 4 7 h4 'it>f7 48
along the 4th rank, not al lowi ng any Itf6+ 'it>g? 49 .l:.f3 .l:!.e4 50 h5 .l:ih4 51 h6+
counterplay. The king can cal mly approach Wg6 .
the a-pawn. 41 l:.e5 !
There was also the i nteresting move 39 . . . Wf7 Here t h e g a m e w a s adjou rned . Analysis
with the idea after 40 .l:ic5?! 'i.t>e6! of showed that Wh ite wins without d ifficu lty.
supporting the passed e-pawn with the king .
41 . . . l:!.c4
40 a5 c3 4 1 b4 ! is stronger (but not 4 1 bxc3
.l:ixc3 with a draw). After 4 1 . . J::t d 4 42 \t>e3 ! The situation arising after 4 1 . . .l:!.xh3 42
l:.c4 4 3 l:.c5 l:.xb4 4 4 .l:ixc3 l:.a4 45 .:c5 the .l:ixe4 has a l ready been d iscussed . In the
same winning position, as occu rred later in event of 42 . . .'>t>f7 (with the idea of bri nging
the game, is reached . the king to g5) the strongest is 43 g 5 ! .
40 bxc3 42 a5 .l:i.a4
43 'it>e3 �g7
44 g5 'it>f7
45 h4 'it>g7
46 'it>f4
Also good is 46 .l:!.e 7 + 'it>f8 4 7 .l:!.e6 .l:!.a3+ 48
�f4 (or even 48 �xe4 .l:!.xa5 49 .l:.xg6 .l:!.a4+
50 '>t>f5 .l:.xh4 51 .l:!.a6).
46 . . . <tlt7
47 .l:.b5
If now 47 . . . �e6 , then 48 l::t b 6+ 'it>d5 49 .l:!.xg6
or 49 a6, while if 47 .. :.t>g7 there follows 48
l:!.b7+ 'it>f8 49 .l:i.b6 .l:i.xa5 (49 . . . '1t>f7 50 a6) 50
.l:i.xg6 with an easy wi n . I n the last variation
 The Theory and Practice of Rook Endings LtJ 55

we clearly see the d ifference i n the position Moreover, the alternative 38 a4 ?! would in
of the black pawn - with the pawn on e3 fact have thrown away the win. The variation
there would have been no w i n . considered by Yusupov - 38...'iJ.d3 39 a5 c3
47 . . . e3+ 40 bxc3 e3 4 1 �f1 l:f.xc3 42 a6 .l:!.a3 43 l:!.b6
48 Wxe3 �e6 Wf7 44 g5 e2+ 45 Wxe2 11xh3 46 .l:tf6+ 'it>g7
4 7 .l:i.c6 �f7 48 �d2 l:!.g3 49 a 7 'iJ.a3 50 .l:!.c 7 +
The only active chance - in reply to the rook
'it'e6 5 1 Wc2 Wf5 52 Wb2 etc. is unconvinc­
check the king can now go to f5.
ing. Instead of the incorrect 5 1 ...Wf5? Black
49 l:tb6+ �f5 should hold his ground: 5 1 ...�d6! 52 �g7
50 a6 'it>g4 Wc5 53 �b2 l:f.a6 54 Wb3 .:ta 1 , and White is
What would have happened after the cap­ unable to strengthen his position. He in turn
ture of the h4-pawn? Of cou rse, 51 .l:!.b5+ could have successfully forced events ear­
and 52 .l:!.a5 - how can one not exploit an lier: 4 1 a6! (instead of 4 1 �f 1) 4 1 ...'1J.d1 + 42
opportun ity to place the rook behind the Wg2 e2 43 a7 l:f.a 1 (43...'ii g 1 + 44 �f3) 44
passed pawn! l:f.e5 l:f.xa7 45 l:f.xe2. But this happened only
51 l:f.xg6 as a result of Black's mistake 40...e3?.
51 .Ud6? was less good in view of 51 .. .<it>h5 According to Kuzmin's analysis, by playing
and 52 ....ttx h4 . 40...'1J.xc3! 4 1 a6 'iJ.a3 42 .l:!.b6 Wf7 43 h4 e3
he would have gained a draw.
51 . . . �xh4
52 �d3 � h5
I n concl usio n , here is a very complicated ,
53 .l:!.c6 'it>xg5
purely a n alytical ending.
54 'it>c3 'it>f5
55 �b3 l::t a 1
Yusu pov - Mestel
56 'it>c4 � e5
Esbjerg 1 980
57 �c5!
The final touch - a 'shoulder-cha rge' .
57 . . . I1a2
58 �b6
Black resigned .

The 1998 No.5-6 issue of the magazine

Shakhmaty v Rossii published an extensive
article by Alexey Kuzmin, in which the
grandmaster pointed out a number of mis­
takes in the analysis of the Yusupov­
Timman ending. It turns out that 40... e3+
would not have saved Timman. White suc­
ceeds in breaking through decisively with his
king: 41 ..ti>f3! 1hc3 42 l:!.e5 l:ta3 43 �f4! This interesti ng position with a n u n usual
.!:!.xa4+ 44 �g5, and the resulting ending with wh ite pawn configuration on the kingside
two pawns against one is won. This means a rose immed iately after the adjou rnment.
that there is no reason to criticise his choice Only a win offered me chances of taking first
on the 38th move. place in the tou rnament and achieving the
56 <Jif The Theory and Practice of Rook Endings

grandmaster norm , and so I had to spend

the entire free day analysing the adjourned
position.
42 a4! bxa4
43 'it>c3
White has an extra piece in play - his king.
This factor will tell , for example, if Black
plays 'accord ing to the ru les' (but in fact
routi nely): 43 . . . .l:!.a8? - placing his rook
behind the passed pawn . In this case there
follows 44 l::t d 2, then .l:!.a2 and 'it>xc4 , and the
a-pawn will be lost. Black cannot defend in
this way - he is clearly too late with his
counterplay. 1 ) 4 7 ... .l:.c2+ 48 'i¥tb5 .l:i.b2+ 49 '.tc6 .l:!.c2+ 50
The best chance was 43 . . . e5! . Jonathan 'it>b7 .l:!.xf2 (note that Black has employed a
Mestel did not play this, because he was typical procedu re - he has fi rst d riven the
afraid of 44 .l:td2 'i¥te6 45 'it>xc4 . But after king onto the sq uare in front of its own pawn ,
45 . . . .l:i.c8+ 46 'it>b5 l:ta8! (46 . . . a3? loses to 47 and only then captu red a pawn) 5 1 'i¥ta8
'.ta4 .l:!.c3 48 b5) 47 'iil c6 (47 l:ta2 'iil d 5! 48 .l:!.xf3 52 b7 l:!.b3 53 llxa4 ! (all the same the
.l:i.xa4 .l:!.b8+) 47 . . . a3 48 .l:!.a2 .l:i.c8+ 49 '.tb7 pawn has to be captu red , so it is better to do
.l:!.c3 50 b5 'iild 5 51 b6 'it>c4 Black saves the this i m mediately, in order to halt the passed
game. 52 '.taB 'it>b3 53 .l:!.xa3+ 'iil x a3 54 b7 e5-pawn, even if only for a �oment) 53 . . . f5
.l:!.b3 leads to a d rawn pawn ending, while if 54 b8'ii' .l:.xb8+ 55 'i¥txb8 .
52 Wa6 , then 52 . . . 'it>b3 53 b7 Wxa2 54 b8'ik The resulting sharp position would appear to
.l:!.b3 55 'ii'xe5 'it> b 1 56 'iVe 1 + 'it>b2 57 �e2+ be won , for example: 55 . . . e4 56 .l::t a 5! (it is
'lt>b 1 58 Wa5 a2 59 'it>a4 .l:.xf3 ! (59 ... .l:.b7 ! ? i mporta nt to cut off the black king ) 56 . . . h6
6 0 'it'e4+ 'lt> c 1 6 1 'il¥c6+ 'it>d2 is also (56 . . . e3 57 .Ua3 f4 58 gxf4 'it>f5 59 llxe3) 59
possible) 60 'ii' d 1 + 'it>b2 6 1 'Yi'xf3 a 1 'ti'+ 62 .l:.a6+! '>td5 60 llxg6 e3 6 1 l:!.g8 '.te4 62 'iil c7
'it>b4 iVa? with a drawn queen ending. 'it>f3 63 'it>d6 e2 64 l:te8 'it>xg3 65 .l:!.xe2 f4 66
I was intending 44 .l:!.a7 ! , but then Black 'it>e5 or 66 h5.
activates his rook by 44 . . J!d8 ! . After 45 b5! 2 ) 47 ... .l:!.xf2 48 l:!.xa4 (48 b7 llb2 49 'i¥tc5
(weaker is 45 'iilxc4 .l::t d 2 46 b5 l:tc2 + ! ) does not work in view of 49 . . . a 3 ! ) 48 . . . Wd7?!
4 5 . . . 'iil e 6! (45 . . . .l:i.d3+? is bad : 4 6 Wxc4 l::txf3 (what happens after the strongest move
47 b6 .l:!.xf2 48 .l::t x a4 , or 47 . . . .l::t b 3 48 b7 Wf5 48 . . . '1¥td6! will be seen in the analysis of the
49 'it>c5 ) 46 'i¥txc4 (46 b6? .l:.b8) 46 . . . .l::td 2 47 followi ng variation) 49 .Ua7+ 'it> c6 50 .l:.xf7
b6! (47 .l:.xa4 .l:.c2 + ! ) Black would have had 'i¥txb6 5 1 'i¥td5, and Wh ite must win, since
to make a difficult choice between 47 . . . .Uc2+, the black king is too far away from the
47 . . . Uxf2 and 47 . . . '1¥td6. kingside pawns.
3 ) 47 . . . '1¥td6 48 llxa4 (48 :Xf7 a3 49 .l:.xh7
a2 50 l:!a7 .l::txf2 leads to a d raw) 48 . . . .l:.xf2
Black loses after 48 . . . 'it>c6? 49 .l:!.a7 f5 50
.l:!.xh7 'i¥txb6 51 .l:!.h6 .l:!.xf2 52 .l:!.xg6+ 'it> c7 53
'i¥td 5 .l:!.xf3 54 'iilx e5.
49 .l:!.b4 l:!.c2+ 50 'iil b 5
 The Theory and Practice of Rook Endings ltJ 57

43 . . . l:tb5?
44 'it>xc4 l1d5
45 lla7 l:td2
46 b5
In essence, Black has s i m ply lost a tempo.
In the analogous positio n , wh ich we have
a l ready ana lysed , the pawn stood at e5 and
the king cou l d be brought i nto play with
. . . 'it>e6. Now this resou rce is not available,
and therefore Black has no defence. If, for
example, 46 . . . l:!.xf2 , then 47 b6 l:.b2 48 b7
a3 49 'it>c5 a2 50 'it>c6 .
46 . . . .l:tc2+
Initially I thought that the pawn ending
arising after 50 ... 'it>d71 51 b7 'it>c7 52 l:.c4+ 47 'it>b4 e5
l:!.xc4 53 'it>xc4 'it>xb7 54 'it>d5 was won i n 47 . . . l:!.xf2 48 l:!.xa4 ltxf3 49 b6.
view o f 54 . . . 'it>c7 55 'it>xe5 'it>d7 5 6 'it>f6 'it>e8 48 b6 l:txf2
57 'it>g7 h5 58 g4. 49 b7 'it>f5
But Black can defend more strongly: 54 . . . f6 ! . 49 . . . l:.b2+ 50 'it>c5, threatening 5 1 l:ta6+ and
Now it does not help to play 5 5 'it>e6 'it>c6 S 6 52 l:.b6 .
�xf6 'it>d5 57 'it>g7 'it>d4 58 'it>xh7 'it>e3 59
50 g4+ !
'ittxg6 'iit xf3 60 h5 e4 or 57 g4 'it>d4 58 h5
gxh5 (58 . . .'it>e3 is also possible) 59 gxh5 e4 ! 51 l:txa4
60 fxe4 'it>xe4 6 1 'it>g 7 'it>f5 62 'it>xh 7 'it>f6 ! B l ack resig ned .
with a draw. If instead 55 g4, then 55 . . . h5!
(55 ... 'it>b6? 56 g5! or 55 . . . 'it>c7? 56 'it>e6 'it>c6 This game also g ives an opportun ity for
57 g5! ) 56 gxh5 gxh5 57 'it>e6 'it>c6 58 'it>xf6 d iscussion a bout the tech n ique of playing
�d5 59 'it>g5 e4! , and again it is a d raw. the endgame. Every tempo, even a seem­
As you ca n see , to find the n arrow path ingly insignificant one (such as . . . e6-e5! ),
enabling Black to hold on is extremely can have a sign ificant and possibly decisive
difficult, even i n home a n a lysis. In any i nfluence on the outcome of the game. You
event, Black was obl iged to play 43 . . . e5! . should always choose carefully the most
The conti nuation in the game loses without accu rate way of putting your plans i nto
a fight. practice .
58 � The Theory and Practice of Rook Endings

Exercises

1 . White to move 2. Black to move

3. Wh ite to move 4. Wh ite to move

 The Theory and Practice of Rook Endings CZJ 59

5. Wh ite to move 6 . Black to move

Is 59 . . . �g4 possible?

7 . Black to move
8 . Wh ite to move

9. Black to move 1 0 . B lack to move

How would you eval uate the position? Is 1 . . . a2 good or bad?
60 � The Theory and Practice of Rook Endings

Sol uti ons

1 . A . Rinck ( 1 906). 3 'it'd?

1 f6 .ll xe2 3 .Ua7+ �g8 4 l:!.f7 �a5 5 .l:ta7 l:th5 ! .
1 . . . 'it>b5 2 �h8 .l:i.d7 3 �e8 , or 1 . . . lld4 2 l:!.e7 3 . . . l:th6!
�e4 3 .ll e 8 and wins. 4 �c7 �f6!
2 .l:th5+! The d raw becomes obvious, for example: 5
2 .l:!.h8? .l:!.f2 3 �f8 'it>b6 4 f7 'it>b7 5 'it>c4 .l:tf5 . a7 .l:!.f7+ ! (5 . . . .l:!.a6? 6 'it>b7) 6 'it>d6 .l:tf6+ 7
2... �b6 �e5 l:ta6 .
3 .l:tf5 ! !
4 . Vaisser-Martinovic
(Vrnjacka Banja 1 984 ).
2. Trabattoni-Barlov (Valetta 1 979).
1 'it'd 1 ! ! gxh4
It is not enough to find a good idea ; you must
also choose the most accu rate way of 2 �xh4+ 'it>xf3
carrying it out. Wh ite has two moves: 1 �e6 3 .l:!.h5
and 1 �g5. Which of them is correct? 3 .l:!.h3+ �g2 4 l:i.h5! l:.f2 5 'it>e1 is also
1 :C.e6! �g2+ possible.
2 'it>h1 ! .Uxg3 3 . . . 'it>g4
3 .l:!.xg6! 4 l:Ib5 f5
The game went 1 :C.g5? :C.g2+ 2 � h 1 .l:tf2! 5 We1
(2 . . . .l:!.xg3? 3 :C.xg6!) 3 'lt>g 1 .l:tf6. Wh ite found And the game soon ended i n a d raw.
himself in zugzwang and resig ned a few All other king moves lead to a loss:
moves later. A) 1 'it>d3? gxh4 2 .l:!.xh4+ 'it>xf3 3 .l:!.h5 'it>g4
and 4 . . .f5. The wh ite king is stuck on the
3. P. Romanovsky ( 1 950). 'long side' of the pawn .
The rook must be switched to the 6th ra nk, B ) 1 'it>e1 ? 'it>e3! 2 'it>d 1 gxh4 3 .l:txh4 f5! 4 f4
but how can this be achieved? 1 . . . �f1 +? 2 .l:!.a 1 + 5 �c2 .l:!.f1 6 .l:!.h3+ .l:!.f3 7 l:th8 .l:!.xf4.
�e5 .l:!.f6 3 l:tg8+ is not possible, and C) 1 'it>f1 ? 'it>xf3 2 'lt>g 1 (2 'it>e 1 .Ua 1 + 3 'it>d2
1 .. J1b 1 ? 2 .l:ta7+ 'it>h6 3 .l:!.b7 .l:!.a 1 4 a7 also gxh4 4 .l:!.f5+ ..t>g3 5 .l:!.xf7 h3 6 .l:!.g7+ �f4 7
loses . .l:tf7+ �e5 8 llh7 h2) 2 . . . .Ug2 + !
1 . . . �a5+! Not 2 . . . gxh4? 3 .l:!.f5+! (3 .l::i.x h4? Wg3)
2 'it>e6 3 . . . �g3 4 .l:txf7 or 2 . . . g4? 3 .l:!.f5+ �g3 4 h5!
If 2 'it>e4 , then 2 . . . l:!.b5 3 .l:!.a7+ (3 .l:!.c8 lla5 4 .l:!.g2+ (4 . . . .l:!.a 1 + 5 .l:tf1 ) 5 �f1 l:th2 6 �g 1
:C.c6 'it>f7) 3 . . . 'it>g6 4 .l:!.b7 .l:!.a5 5 a7 'it>f6 6 with a d raw.
'it>d4 'it>e6 7 'it>c4 'it>d6 8 'it>b4 Ita 1 (or 3 �h 1 (3 'it>f1 tlh2 4 l:i.xg5 .l:!. h 1 + 5 .l:!.g 1
8 . . . 'it>c6) with a draw. l:::tx g 1 + 6 �xg 1 �g4 ) 3 . . . g4 4 l:!.f5+ 'lt>g3 5 h5
2 . . . .l:!.h5! ! (or 5 .l:!.xf7) 5 . . . .Uf2 ! , and Wh ite can resign.
The only way! 2 . . J::tb5? is bad in view of 3
.l:ta7+ and 4 l:!.b7 . If instead 2 . . . .l:!.g5?, then 3 5. Dorfman-Khol mov (Saratov 1 98 1 ) .
.l:ta7+ 'it>g8 4 �f6! .l:!.a5 5 .l:ta8+ �h7 6 'it>e7. 43 .l:!.d5! .Ue4
 The Theory and Practice of Rook Endings l2J 61

43 .. J1b8 44 .l:txd7.
44 .Uxb5
And White retains excellent winning chances,
for example: 44 . . . d5 45 .Ub7 ! , cutti ng off the
king on the 8th rank, or 44 .. :;itt? 45 �e2 !
and 46 Wd3 .
The game conti nuation 43 l1xd7? w a s a
mistake because of 43 .. J1e4 ! . The point is
that if 44 l:td4 Black can go i nto the pawn
ending: 44 . . . l:Ixd4! 45 exd4 )!;>f7 , and if 46 d5
rt:ie7 47 'it>e3 , then 47 . . . Wd 7 ! 48 d6 (48 Wd4
rt:id6) 48 . . . 'it>c6 ! 49 Wd3 c.t>d7 ! 50 'it>d4 'it>xd6.
White's comparatively best chance is to go
into a queen endgame by 46 )!;>g2 (instead
of 46 d5} 46 . . . )!;>e6 47 �h3 )!;>d5 48 )!;>h4 6. Portisch-Petrosian (Ca n d idates Match ,
rt:ixd4 49 'it>g5 'it>c4 50 'it>xg6 �xb4 51 'it>xf5 1 2th game, Palma de Mallorca 1 974) .
rt:ic4 52 'it>g5! (52 We5 b4 53 f5 b3 54 f6 b2 I n t h e g a m e Black preferred t h e cautious
55 f7 b 1 'iV 56 f8'ii' 'ife 1 +) 52 . . . b4 53 f5 �d5 ! 59 ... We6 and after 60 'it>c5 he made a
54 f6 'it>e6 5 5 �g6 b3 5 6 f7 b 2 57 f8'if b 1 'iV+ decisive mistake : 60 . . J1c2+? (60 . . . Wd7!
58 \t>g5 (58 Wg 7 it'b2+ 59 �g8 'ir'f6 ! ) - i n was necessary, with good d rawing chances).
this case t h e opponent is stil l req u i red to There followed 61 c.t>b5 Wd6 62 'it>a6 'itc6
defend accu rately, although objectively the 63 l1a1 l:Ic4 64 b7 l1b4 65 l1c1 + �d7 66
position is a d raw. lieS B lack resigned .
44 g4 fxg4 45 l1d4 g3+! 46 'itxg3 (46 Wf3 As was shown by Igor Zaitsev, the active
g2) 46 ... l1xe3+ 47 'itg4 l:r.b3 48 f5 � 49 king move wou l d have secu red a d raw, but
rt:Jf4 gxf5 50 'it>xf5 'ite7 51 )!;>e5 l:Ib1 52 )!;>d5 only if Black had fou n d a far from obvious
.!:!.c1 Draw. defensive idea .
An interesting attempt to play for a win was 59 . . . 'it>g4!
suggested by Viorel Bolog a n : 43 �f3?! I1e4 60 I1a4!
44 g4 ! .l:txb4 (44 . . . Wf7 45 .Uxd7+ We6 46
Threatening 6 1 Wc3+ .
.!:!.g7 �f6 47 l1b7 .l:txb4 48 g5+ We6 49 l:Ig7 )
45 gxf5 gxf5 4 6 .l:txd7. 60 . . . c.t>h3 ! !
6 0 . . . Wg3? i s hopeless: 6 1 Wc5 f5 6 2 l:!.b4
.l:tc2+ 63 c.t>d6 lk8 64 b7 .l:tb8 65 Wc7 .l:th8
(see diagram)
66 b8'iV .l:txb8 67 .l:txb8 �xh4 (67 .. f4 68 Wd6
f3 69 We5 f2 70 .l:tf8 Wxh4 71 We4) 68 )!;>d6
Black can apparently hope to save the game 'lt>g3 69 )!;>e5 h4 70 Wxf5 h3 71 l:tb3+ Wg2
after 46 . . . l:!.c4 47 .ti.d5 .l:Ic7 48 .ti.xf5 (48 .ti.xb5 (7 1 . . . 'it> h4 72 �f4 h2 73 .l:tb 1 ) 72 'it>g4 h2 73
.M.f7) 48 . . . .ti.b7, for example: 49 l:td 5 b4 50 lib2+ 'itg 1 74 'itg3 , or 68 . . . �g4 69 )!;>e5 h4
J::i.d2 b3 51 .l:tb2 Wf7 52 'it>e4 )!;>e6 53 70 �d4! (70 .l:tb4+? f4! 71 I1xf4+ Wg3)
'.t>d4(d3) 'it>f5 , or 49 'ite2 b4 50 )!;>d2 b3 51 70 . . . h 3 (70 . . . f4 7 1 'it> d 3 'it>f3 72 l1h8 ! ) 71
rt:ic1 .l:te7 52 .l:tg 5+ 'it>f7 53 l:tg3 Wf6 . )!;>e3 h2 72 .l:tg8+ �h3 73 �f2 h 1 4J+ 74 Wf3 .
61 'it>c5 f5
62 .l:tb4 .l:Ixb4!
62 � The Theory and Practice of Rook Endings

63 �xb4 f4 Wh ite needs to obta i n the same position , but

64 b7 f3 with Black to move .
65 bB'iV f2 1 l:ta2 ! ! Wh5
And Wh ite cannot wi n . 1 . . . 'it>g5 2 l:tg2+! 'it>h5 3 .i':lh2, or 2 ... Wf5 3
.llf2+ 'lt>e4 4 .l:!.f6 ! .
7. Petrosian-Karpov 2 .l:th2 .l:!.h6
(44th USSR Championsh ip, Moscow 1 976). 3 .l:l h 1 !
51 ... 'it>h6? 52 f7 �h7 (52 .. J1a 1 53 'it>g8) 53
h6 'lt>xh6 (53 . . . .l:!.a 1 54 .l::t xc2 also does not
help) 54 'lt>g8 leads to a position from
Emanuel Lasker's famous study. Wh ite wins
by grad ually pushing back the opponent's
king: 54 .. J:ig 1 + 55 �h8 .i':lf1 56 .i':lc6+ Wh5
57 'it>g7 l:r.g 1 + 58 'it>h7 .l:!.f1 59 l:lc5+ 'it>h4 60
'it>g7 .l:!.g 1 + 61 'it>h6 l:tf1 62 .l:!.c4+ 'it>h3 63
�g6 l:tg 1 + 64 Wh5 l:tf1 65 l:tc3+ �h2 66
l:txc2+.
51 . . . 'it>h8!
52 f7 tl.a1 !
53 'it>e7
I n the game there followed 53 l:i.xc2 .l:!.a8+ Now it is Black who is in zugzwan g . He is
54 'it>e7 .l:ta7+ 55 'it>f6 .l:ta6+ 56 'lt>g5 .l:ta5+ unable to win .
57 Wg4 .l:ta4+ 58 'it>g3 .l:ta3+ 59 Wg2 'lt>g7
3 . . . 'it>g4 4 .l:!.g 1 + �f3 5 .l:tf1 + �g2 6 .Uf6 .l:!.h8
60 .l:tf2 'it>fB 61 .l:!.f5 l:ta6! (6 1 . . . .l:la7? 62 h6
7 l:txe6 h3 8 l:tg6+ 'it>f2 9 l:lf6+ We2 1 0 .l:!.g6!
.l:!.xf7 63 h7 or 62 . . . l:ta6 63 .l:!.h5) 62 �g3 .l::t h 6
h2 11 .l:!.g2+ 'it>f3 1 2 l:lxh2 .l:txh2 1 3 e6.
63 'it>g4. Draw agreed in view of 63 . . . .l:lh7.
53 . . . l:te1 +
54 'it>f6 tl.f1 + 9. Larsen-Kavalek
(7th match game, Solingen 1 970).
55 'it>g6 .Ug 1 +
56 'it>h6 Wh ite wants to play .l:.c4 . If Black is forced
c1 "ii'+ !
totally onto the defensive with . . . .l:!.a7, then
57 .l:!.xc1 .l:lg6+ !
Wh ite's extra pawn together with the pas­
Either captu re leads to stalemate. sive black rook should ensure him a
straig htforward win.
8. Makarychev-Vasyu kov That is what happened i n the game:
(Vilnius 1 980/8 1 ; variation from the game). 1 . . .'it>g7? 2 .l:!.c4 .l:ta7 (2 ... .l:tb3 3 l:lxa4 l:txg3
Black will win if he can manage to advance 4 tl.g4+) 3 'it>c3 h5 4 Wb4 'it>g6 5 .Uc6+ 'it>g7
the h-pawn just one step more. After the 6 tl.c5 Wh6 7 'it>b5 l:le 7 (otherwise 8 .l:!.c4) 8
obvious 1 .l:la 1 ? �h5 (th reatening 2 . . . h3) 2 'it>xa4 l:te3 9 g4 hxg4 1 0 hxg4 .l:te4+ 1 1
l:th 1 l:th6! Wh ite ends up in zugzwan g : 3 �b5 .Uxg4 1 2 a4 .l::!.g 1 1 3 a5 l:lb1 + 1 4 Wc6
'it>e7 'it>g4 4 .l:lg 1 + �f4 with the th reats of .l:la 1 1 5 'it>b6 l:tb1 + 1 6 .l:tb5 .l:lf1 1 7 a6 tl.f6+
5 . . . 'it>xe5 and 5 . . . h3, or 3 tl.h2 'it>g4 4 l:tg2+ 1 8 'it>a5 :tf7 1 9 .l:lb6+ 'it>g5 20 tl.b7 .l:lf1 21 a7
'it>f3 5 .l:!.h2 �g3 and 6 . . . h3. 'lt>h6 22 .Ub6+ 'it>g7 23 .l:la6 Black resigned .
 The Theory and Practice of Rook Endings ctJ 63

1 . . .Wf7 , suggests itself, i n order to meet 2 3 �h2 'lt>h5

:!:ic4 with the counter-attack 2 . . . �b3 ! . But
White plays 2 g4! , intending 3 h4 and only
then 4 l'ic4. After 2 . . . �e6 3 h4 'it>d5 Wh ite's
threat is parried , but 4 g5! creates a new
threat: 5 �g3 followed by 6 �g4 or 6 h5 (rook
behind the passed pawn ) . Black's position
becomes critica l .
1 . . . h5! !
The only way to save the game! I n the event
of 2 �c4 .ti.b3! 3 .UXa4 .l:t.xg3 there is no check
o n g4 - the position is d rawn . If 2 h4, then
2 . .l:l.g7 and 3 . . . .ti.g4, if there is noth ing better.
. .

The rook on g4 is very active - it attacks the

white pawn , defends its own and restricts the
mobility of the wh ite king . Finally, if 2 g4 there
is the reply 2 . . . h4! , fixing the white h3-pawn Can a nyth ing be done agai nst the th reat­
as a target for a cou nter-attack along the 3rd ened . . . g6-g5 ?
rank (in the event of .l::t c4 ). 4 .Ua4! 'it>h6
5 l1a6! 'it>h5
1 0. Moiseev-Bagi rov (Moscow 1 956).
6 .Ua4! g5
Sooner or later Black will have to play . . . a3-
a2 (the march of the king to the a2-square is 7 g4+ ! !
obviously u n real). The only question is This is the whole point - Black can no longer
whether at the same time he can obta i n a obta i n a passed pawn on the f-file.
secon d passed pawn on the f-file. The game concl uded : 7 ... 'lt>xh4 8 gxf5+ g4
There was a stra ig htforward win by 1 . . . g5! 2 9 'lt>g2 .l:!.b1 1 0 llxa2 .l:!.b4 1 1 .i:tc2 g3 1 2 .Ua2
hx g 5 'it>g6 3 .l:Ia7 a2! 4 'it>h2 'it>h5! 5 'it>g2 g6 'it'g4 1 3 llc2 .l:!.f4 1 4 llc8
6 �a4 'it>xg5. Then Black moves his king
After 1 4 .Uc7 .Uf2+ 1 5 'lt>g 1 .Ua2 Wh ite
and plays . . . g6-g5 and . . . f5-f4 , achieving
ca nnot play 1 6 .Uxg7? 'it>f3 . It was for the
his a i m .
sake of such variations that Black left the f5-
I n the game he chose a different, far less pawn al ive .
successfu l move order.
1 4 . . . .Uf2+ 1 5 'it>g1 .l:!.e2 1 6 .Ua8 'it>f3 1 7 .l::!.a 3+
1 . . . a2? .Ue3 18 .Ua1 g2 ( 1 8 . . .'it>g4 1 9 .Ua8) 19 'lt>h2!
2 .Ua6! 'it>h6 ( 1 9 .l:!.b 1 ? 'it>g3 20 .l::!. a 1 .Uf3 2 1 .Ub 1 l'if1 +)
I n the event of 2 . . . g5 3 hxg5 the king can n ot 19 ... 'it>f2 20 .l:!.a2+ .l::!. e 2 21 .l::!.x e2+ 'it>xe2 22
approach the g5-pawn. 'it>xg2 Draw.
64 �
Mark Dvoretsky

F rom the S i m ple to the Com p l i cated

The Theory of E n d i ngs with O p pos ite -co l o u r B i s h o ps

Wendgame the most important thing is

hen trying to master any type of after a bishop check), then 'it>e5 and f5-f6 .
I n order to cou nter this pla n , Black must take
to lay a firm foundation: to pick out the most control of the e6-sq uare. But from where , d 7
importa nt theoretical positions, ideas and or b3? We will consider both possibil ities.
tech niques, which u nderlie our notions of After 1 . . . i.. b 3? the position is lost. First
the endgame in question . As a rule, the Wh ite gives a verifying check with his
necessary basic knowledge is made up of a bishop, i n order to determine the position of
small nu mber of rather simple positions, but the enemy king . I n so doing it is important
they must be understood i n every detail and that the bishop should prevent the king
firmly remembered . from wedging itself between the pawns
A successfu lly developed system of basic after White plays e5-e6. Hence , 2 i.. g 5+ ! .
endgame knowledge provides a rel iable Next the white king makes a by-pass
guide in the analysis of more complicated manoeuvre to the aid of the e-pawn, on
situations and helps them to be more the opposite side to where the oppo­
successfu lly understood . How this occu rs , I nent's king has moved. For exa mple:
will show using an example of endings with 2 . . . 'it>f7 3 'it>d4 i.. a 2 4 �c5 i.. b 3 (4 . . . i.. b 1 5
opposite-colou r bishops. e6+ and 6 f6 ) 5 �d6 and 6 e6+ . Or 2 . . . 'it>d7 3
�f4 i.. a 2 4 i.. h 4 i..f7 5 �g5 �e7 6 �h6+
Connected passed pawns �d 7 7 �g7 i.. c4 8 'it>f6 and 9 e6+ . After the
pawns have reached e6 and f6 , even if the
Let's discuss in detail the followi ng elemen­
th reat of f6-f7+ is parried , Wh ite repeats the
tary ending.
same proced u re : a verifying check with the
bishop and a by-pass by the king .
It is incorrect to play 2 il.b4 + ? �f7 ! (Black's
only hope is to provoke a prematu re e5-e6+
and wedge his king between the pawns) 3
�d4? i.. c2 ! 4 e6+ �f6 5 e7 i.. a 4 with a d raw.
As soon as the pawns are blockaded on
squares of the same colour as their
bishop, the draw becomes obvious.
Thus, with his bishop on b3 Black loses. On
the other hand , 1 i.. d 7! 2 i.. g 5+ �f7 1eads
. . .

to a n easy d raw. Su bseq uently Black waits,

movi ng his bishop between c8 and d7. I n
order t o prepare e5-e6 , Wh ite wou ld need
Wh ite is threatening to play e5-e6 (perhaps to make a by-pass with his king from the left,
 From the Simple to the Complicated l2J 65

but this is impossible, since the king is tied to

the defence of the f5-pawn.
The following ru le suggests itself: the
bishop should be positioned such that,
while preventing the advance of one
pawn, it simultaneously attacks the other.

We will use the ideas from the basic position

just examined for an analysis of other
positions. First some com paratively simple
ones.
Let us shift all the pieces one ra nk forward .
What has changed?
After a l l , with the king on f8 White's only plan
- a by-pass with the king from the right - is
impossible: the edge of the board prevents
it.
But it is Wh ite to move , and he shuts the
opponent's king in the corner by 2 .il.. c4 ! and
then carries out the standard manoeuvre -
the by-pass with the king from the left: 'it>h5-
g4-f5-e6-f7 .
After 1 if..f8 ! the by-pass is no longer
. . .

possible, but what about the threat of

zugzwang? I n order to put the opponent in
zugzwang, Wh ite must deprive the king of
the g8-square by playing 2 if.. c 4. But after
In the event of 1 . . . .i.b4 there is no d iffer- 2 .i.xh6! 3 �xh6 things end in stalemate .
. . .

ence. Wh ite wins in exactly the same way (a

check and a by-pass by the king); moreover,
here, as it is easy to see , both checks at b6
and b5 are equally good .
After 1 .1i.d8 2 if.. g 6+ (or 2 .i.b5+) 2 �f8 3
. . . . . .

Wf5 Black loses because of zugzwang - in

contrast to the previous position , he no
longer has a waiti ng move with his bishop.

(see diagram)

The proximity of the edge of the board

introduces new featu res i nto the evaluation
of the positio n . Let us verify 1 . . . if.. b 2. If Black
could also manage to play 2 . . . Wg8 and
3 . . . 'it>f8 , the d raw would become obvious. I n all the situations exam i ned earlier the
66 � From the Simple to the Complicated

weaker side was a1m1ng to g ive up his 'Accord ing to the ru les' Wh ite should give a
bishop for the two pawns . Here, of cou rse, check on h5, to control the g6-square . The
this familiar plan of defence will no longer black king should move to e7, forcing the
save him. Does this mean that Black is wh ite king to make a by-pass to the right,
doomed? It turns out that he is not - wing where there is little space for ma noeuvri ng.
pawns can sometimes be stopped without 1 i.h5+ �e7 ! (after 1 . . . 'it>g7? 2 �e4 there is
resorting to the bishop sacrifice . noth ing to prevent the by-pass by the king
1 . . i.d1 ! 2 �h4 (otherwise g4-g5 cannot be
. from the left) 2 'it>g4 .ib2 3 .ig6 (otherwise
played ) 2 . . . � 3 g5 �e6 ! 4 g6 'itf5 ! Wh ite the king can not advance, but now the
cannot advance either his king (the edge of i mporta nt g6-sq uare is i naccessible to the
the board prevents this), or his h-pawn . And king) 3 . . . .ic3 4 �h5 (th reatening 5 �h6, 6
if 5 g7, then 5 . . . .ib3 and 6 . . . .ig8 , with a .ih5 and so o n ) 4 . . . ..tg7 ! 5 .ih7 �f7 ! 6
secu re l ight-sq uare blockade of the enemy .ig6+ �e7, and Wh ite has been u nable to
pawns. ach ieve his a i m - to prepare f5-f6+ .
A s w e know from the basic position , the
The followi ng example is much more d iffi­ check from the other side also does not
cult. ach ieve a nyth i n g : 1 .ic4+ �g7 ! 2 �e4 .id2!
3 f6+ �g6 .
Even s o , t h e resources for playing for a win
M. Henneberger are not yet exhausted . The black king can
1916 fi rst be l u red to g7, and only then the bishop
switched to the e8-h5 d iagon a l , preparing a
by-pass by the king from the left.
1 �g4 .ib2
2 �h5 'it>g7!
The threat was 3 �h6; 2 . . . .ig7? is bad in
view of 3 .ic4+ and 4 'it>g6.
3 .ib5 i.c3
4 i.e8 .id4
4 . . . �f8 5 .ig6 �g7 is equally good .
5 .ig6
In the event of 5 'it>g4 (th reatening 6 .ih5, 7
'it>f3 , 8 �e4 and so o n ) the black king
succeeds i n switch ing to e7: 5 . . . �f8 ! 6 i.h5
�e7 , tra nsposi ng i nto the fi rst of the
The black bishop i s not i n its best position
va riations we exa m i ned .
(the place for it is at e7 or dB). In the basic
theoretical position , with which we beg a n , 5. . . .ic3
agai nst such a bishop Wh ite won easily. I f 6 �g4
we reason logical ly, o n l y o n e factor, d isti n­ Wh ite's plan appears to have tri u mphed:
guishing the given position from the basic 6 . . . �f8 7 f6 is bad for Black, and otherwise
one, can prevent the implementation of the Wh ite plays 7 i. h 5 . But at this moment the
standard winning plan - the proxim ity of the black bishop succeeds in switch ing to its
edge of the boa rd . Let us see! lawfu l place.
 From the Simple to the Complicated ctJ 67

6. . . i.a5! ! After 4 . . . d3 5 i.g6 <itd4 6 i.h7 the d raw is

Because of the position of the bishop on g6, obvious (6 . . . e3 7 i.xd 3 ) .
7 f6+ is not possible. 5 i. e 2 <it>f5
7 i.h5 i.d8 5 . . . <itg5 is pointless: 6 i.c4 (intending i.g8-
Black has set u p the d rawn position which is h 7 ) 6 . . . d3 7 i.xd3 exd3 8 <itxd2.
basic to this type of endgame. 6 i.c4! (but not 6 i.h5? d 3 ) 6 . . . d3 7 i.g8
(alas, the crude 7 �xd2 with the i rresistible
th reat of i.xd3 also leads to a d raw) 7 ... <it>f4
At one of our tra i n i ng sessions Sergey 8 i.h7 <ite5 9 i.g6 'iii> d 4 1 0 i.h7 e3 1 1
Dolmatov and Vad i m Zviagintsev tried to i.xd3 'iii> x d3 - stalemate!
solve a study by Jan Ti mman, composed in
1 989.
The stalemate defence is pretty, of cou rse,
but how necessary is it? Dolmatov and
Zviagintsev had doubts about the evaluation
of the position arising after the captu re of the
d2-pawn . Let's set it up with colours re­
versed , so that it will be easier to draw an
analogy with ideas that a re already known to
us.

1 'iite2 (1 i.e2? e4 or 1 . . . d4 is bad for White)

1 ...e4 2 <itd1 ! ?
In Timman's opinion, in the event o f 2 <itxd2
'iite5 3 i.b5 d4 Wh ite loses, because his
bishop does not manage to switch to c2 : 4
ia4? i.g5+ 5 <ite2 d3+ . Therefore Wh ite
leaves the d2-pawn al ive and plays for
stalemate! He al lows the black pawn to go to
The wh ite bishop occu pies an ideal position ,
d3 and is ready to meet . . . e4-e3 with the
whereas on the h3-c8 diagonal the black
bishop sacrifice i.xd 3 ! .
bishop is not it its best place. Without the e2-
2 ... 'iit e 5 3 i.e2 ! pawn the win would not be i n doubt, but here
The attempt to keep the bishop on the Wh ite consta ntly has to reckon with the
queenside is incorrect: 3 i.a6? d4 4 i.b5 d3 th reat of the d iverting sacrifice . . . e2-e 1 'ii' ,
5 .ia6 <itd4 6 i.b5 <itd5! 7 i.a6 'it>c5! after which for an i n stant the bishop loses
(zugzwang) 8 i.b7 e3, or 8 <itxd2 <itd4 and control of important sq uares in front of its
wins. pawns. The q uestion is whether or not Black
3 d4 4 i.h5 'iii>f6
... is able to make use of this resou rce .
68 w From the Simple to the Complicated

1 �e3 3 . . . il.. h 3 4 �f4 (4 ii.b4 �c7 ! ! ; 4 �xe2 .Jli.g2 5

Threatening both the captu re of the e2- e6+ �e7 ! 6 il.. b 4+ �f6) 4 . . . �e7 ! 5 il.. b4+
pawn, and the march of the king to f6 . After �fl comes to the same th i n g .
1 . . . il..g4 (1 . . . il.. h 3? 2 �xe2 ) 2 �f4 B lack is 4 �4 ii.h3
unsuccessful with 2 . . . e 1 'ik 3 il.. xe 1 il.. h 5 Here the idea of playing the bishop to f7 no
(with the threat of 4 . . . il..f7 ) 4 e6+ �d6 5 �e4, longer works: 4 . . . il.. h 5? 5 e6+ �d6 6 �e4.
while if 2 . . . il.. h 3 3 �g5 e 1 'ik 4 il.. xe 1 il.. g 2 , 5 �g5 <t>e7!
then 5 e6+ �d6 6 il.. b 4+ �xd5 7 e 7, and the
Otherwise �f6 can n ot be prevented (as we
pawn queens. But what else can he do?
a l ready know, 5 . . . Ji.g2? is bad : 6 e6+ �d6 7
1 .. . �c7 ! ! ii.b4+ �xd5 8 e7).
The key idea of Black's defence! It i s 6 il.. b4+ �f7
important that the pawn should not advance 6 . . . �d7? 7 �f6 .
to e6 with check (for example, after 2 �xe2
7 �4
Si.e4 ). Without the e2-pawn Wh ite would
reply 2 e6, but here this leads to a n White has managed to l u re the opponent's
immed iate draw: 2 e 6 il.. xe6 3 dxe6 e 1 'ik 4 king to f7 and now his king heads in the
il..xe 1 �d6 (Black's moves can also be opposite d i rection - a by-pass to the left.
interposed ).
I n the event of 2 �f4 the simplest is 2 . . . il.. d 3
(2 ... il.. h 3 also does not lose) 3 �g5 il.. c4 or 3
e6 e 1 'ik 4 ii.xe 1 �d6 . Finally, after 2 il.. a 5+
�d7 3 �f4 the main defensive idea i n such
endi ngs proves possible - the switching of
the bishop to f7 : 3 . . . Ji.g6! and 4 . . . il..f7 (4 e6+
ct>d6).
After moving the pieces arou nd for a short
while, we decided that the endgame was
drawn and hence that Timman's study was
incorrect, since it contains a second solu­
tion .
Later, when I was on my own , I again set u p
I phoned Zviagi ntsev and told h i m about the
the pieces a n d found another attem pt t o play
plan I had fou n d . Half an hour later Vadim
for a wi n , based on zugzwang.
phoned me back and reported that the
2 il.. e 1 'it>d7 position was nevertheless d rawn !
3 il..a 5! 7 . . . ct>g6 ! !
Now it is bad to play 3 . . .<t>e7? 4 il.. b 4+ �f7 The o n l y chance o f saving the game is the
(4 . . . �d7 5 �xe2) 5 <t>d4 , when there is no bold manoeuvre of the king to f5 . After the
defence against the march of the king to d6 incorrect 7 . . . ii.c8? 8 'it>e4 ! B lack unexpect­
(active counterplay is too late: 5 . . . il.. c8 6 edly ends u p i n zugzwa ng and loses:
'itc5 ct>g6 7 e6 'it>f5 8 �d6 il.. a 6 9 �c6 ! �f6 8 . . . 'it>g6 9 e6 or 8 . . . ii.h3 9 'it>d4 . It is curious
10 Si.c3+ �e7 1 1 il.. e 1 ! , or immed iately 1 0 that the zugzwang here is mutu a l ; if it is
i.e1 ). Wh ite to move he ca n not win - 9 �d4 �g6!
3... .Jli.g4 transposes i nto the m a i n variation , analysed
 F rom the Simple to the Complicated CZJ 69

below, wh ile in the event of 9 ii.e1 ii.h3 1 0 Berger - Kolterma n

'it>d4 the black king retu rns to the queenside: Arkhangelsk 1 948
1 0 'it>e7 1 1 ii.b4+ �d7 1 2 �e3 �c7 ! ! etc.
...

8 'iite4
Threatening 9 e6.
8. . . ii.f5+ !
9 'iit d4 .tc8 !
1 0 'i.t>c5 �f5
11 'i.t>d6 .ta6
12 e6 .tc4
Or 1 2 . . . ..ltb5 - Wh ite can n ot w i n .
A s you see, t h e a nalysis proved t o be rather
difficult, and fu l l of by no means obvious
manoeuvres by both sides. But even so, at
the basis of the analysis were ideas which
we derived from the basic theoretical posi­ 1 �e2 b3
tion. 2 'it>d1 'i.t>b4
3 .th7 'iit a 3
Separated pawns 4 ii.g6
If now 4 . . . b2 (with the threat of 5 . . . 'it>a2),
Generally speakin g , the fu rther apart the then 5 ii. b 1 ! �b3 6 'it>e2 .
pawns are, the more d ifficult the defence. 4. . . 'i.t>b2
When I was young I lea rned a h u morous 5 .tf7 !
rule for assessing such endings: if you can
The threat was 5 . . . � a 1 and 6 . . . b2 . By
reach both pawns with the fingers of one
attacki ng the b3-paw n , Wh ite forestalls the
hand , then the position is d rawn ; if you can't
opponent's pla n .
(the distance between the pawns is too
great) the position is won ! 5 . . . 'it>a2
Alas, such a g u ide is too imprecise to be 6 ii.e6 �a3
trusted . In fact, here there exist many With the th reat of 7 . . . b2 8 ii.f5 �a2 .
different situations which it is not at all 7 .tf5!
necessary to study and remember. The Draw.
outcome usually depends on the possibil ity
of a breakthrough by the stronger side's king
Let us consider a more complicated ending.
to the pawn being stopped by the bishop, i n
order to queen it.
But the fol lowing ending should defin itely be
included in our system of basic knowledge.
70 � From the Simple to the Complicated

Yu . Averbakh But if it is Black to move he saves the game

1 954 - he can prevent the opponent's king from
going to the kingside and set up a secure
defence on the queenside.
1 . . . 'it>d7!
2 'it>c3 �e6
3 �d4 .ltb7
4 'it>c5 'it>d7
5 �b6 i.f3
6 a6 �c8!
The threat was 7 'ita? 'lot>c8 8 d7+! �xd7 9
'it>b8
7 'it>a7
N ow the threat of 8 d7+ must be parried by
the bishop, but from which square , c6 or g4?
I n the first edition of the monograph on 7 . . . .ltg41
chess endings ed ited by Averbakh, the After 7 . . . i.c6? 8 i.b4 Black ends up in
analysis of this endgame conta i ned a seri­ zugzwan g : 8 . . . i.d7 9 'it>b6 .lli.f5 1 0 d7+!
ous mistake - it was discovered by Yusupov, 'it>xd7 1 1 'it>b7 or 1 0 .txd7 1 1 a7.
. . .

when I i nvited h i m to try and solve th is

8 'it>b6 i.f3 !
position . However, Averbakh himself cor­
rected the mistake i n a su bsequent ed ition . 8 . . . 'it>d7? 9 'it>b7.

1 'it>c3 i.f1 2 'it>d4 .lte2 3 �e5 'it>d7 9 'it>c5 'it>d7

Now it is wrong to play 4 'lot>f6 i.d3 5 a6? 1 0 Wd4 'it>e6!

i.xa6 6 �xg6 'it>e8 , when the familiar And Wh ite ca n not wi n .
Berger-Kolterman ending is reached (with
reversed colours). Endi ngs with many pawns
The correct plan is to play for zugzwang.
From d3 the bishop defends the g6-pawn After studying for many years the theory of
along one d iagonal, and along the other it endings with opposite-colour bishops, I
prevents the advance of the a-pawn ; there­ observed several rules which , as it tu rned
fore it has no moves. The wh ite king must out, apply in nearly all such endings and
not be allowed to reach e 7 - this means that, g reatly help in confidently fi nding you r way
apart from d7, the black king also has two in them.
other squares: e8 and d8. The fi rst can be
Before tu rn ing to a description of my theory
taken away from it by placing the white king
of endings with opposite-colour bishops, I
on f7, and the second by moving the bishop
will show an example i n which , despite its
to c7.
simplicity, nearly all the rules that we will be
4 .ltc5 i.f1 5 i.b6 .te2 6 .tc7 i.d3 7 'it>f6 talking about a re displayed .
'it>e8 8 'it>g7 'it>d7 9 'it>f7 , and Black is i n
zugzwang.
 From the Simple to the Complicated 71

Textbook ex ampl e tion of an impreg nable position , in which it is

sufficient to stick to waiting tactics si nce
everything is secu rely blockaded and de­
fended .
The main theme in endings with oppo­
site-colour bishops is the theme of the
fortress. The weaker side aims to con­
struct a fortress, while the stronger side
aims to prevent its construction or (if it
has already been constructed) find a way
of destroying the opponent's defences.
I n the textbook example the concl uding
position constitutes a fortress. Wh ite does
not seek any active cou nterplay, but simply
wa its , and the opponent is u nable to do
If it is Wh ite to move he saves hi mself by anyth ing.
1 c5! i.. x c 5 2 �b3 e5 3 �e6 'l,;c7 4 'it>e4. When playing endings, an abil ity to analyse
Later he s i m ply plays h i s bishop up and positions logical ly, by th i n king i n plans and
down the h3-c8 d iagonal. schemes, is very important. The role of
logical th i n king is especially great in endi ngs
1 . Drawi ng tendencies. This i s perhaps with opposite-colour bishops. In the majority
the best-known property of endings with of cases they should not be 'played ' , but
opposite-colour bishops. Here it is some­ 'constructed' - fi rst look for the a rrangement
times possible to save the game when you of pieces and pawns which makes the
are 2-3 pawns down (as, for instance , i n the position impreg nable, and only then verify
example just exa m i ned ). And remember the by calcu lating variations whether it is possi­
endings with two con nected passed pawns ble to achieve the plan ned set-up and
- in what other type of endgame may such whether it is i ndeed impregnable.
an enormous material and positional advan­ The following mech a n i s ms constitute either
tage prove insufficient for a win? the most i mportant general methods of
The consequences of this rule a re obvio u s : constructi ng and destroying a fortress , or
the stronger side should be extremely featu res of the most typical and freq uently­
careful both when transposing into an occu rring types of fortresses.
ending with opposite-colour bishops,
and when playing such an ending - here I l l . Arrangement of the pawns. There is a
it is easy to run into a drawing counter­ wel l-known principle which prescribes that
chance. And for the weaker side, trans­ pawns should be placed on sq uares of the
posing into an ending with opposite­ opposite colour to those on which you r own
colour bishops is sometimes a last bishop moves. In endings with opposite­
resort here the chances of a draw are
-
colou r bishops this principle remains valid
sharply improved. for the stronger side (it is especially i mpor­
tant with regard to con nected passed pawns).
II. Fortress. A fortress is a system of But, contrary to the general rule, the
passive defence, consisti ng in the construe- weaker side should keep his pawns on
 From the Simple to the Complicated

squares of the colour of his own bishop and stops the two enemy pawns on f6 and
- in this case it is usually possible to ensure g5.
that they are securely defended . I ndeed , a But i n the Averba kh position analysed
pawn defended by the bishop can be earlier the bishop defends the g6-pawn
attacked only by the enemy ki ng, wh ich along one diagonal and restrains the passed
means that it rema ins invul nerable. In other a5-pawn along another. Such a situation is
types of endings such a pawn may be u nfavourable for Black. In the solution and
attacked not only by the king, but also by the false trail you saw two typical ways of
another piece (knight or l i ke-colour bishop). exploiting the defects of a 'torn ' bishop:
I n the textbook example the weaker side's zugzwang and diversion.
pawn is on a l ight sq uare - the colour of its
own bishop, and this factor ensures the VI. Pawns ' u nder attack'. A typical
solid ity of the fortress constructed by Wh ite . defensive procedure is an attack on the
I n the i n itial position the stronger side, with opponent's pawns by the bishop. I n thi s
his da rk-sq uare bishop, has only the one way either they are forced t o move onto less
pawn on e6 correctly placed on a lig ht­ favou rable squares of the colour of their own
square. If Black were able to approach it bishop (as i n the textbook example), or the
with his king , he would then play . . . f6-f5 and opponent's king is tied to the defence of the
easily convert his material advantage. The pawns (as in the basic position with two
only way to d raw is to force the e-pawn to con nected passed pawns or the Berger­
move onto a sq uare of the colour of its own Kolterman ending).
bishop. Endings very often occur where the stronger
side has a passed pawn . It must be blocked
IV. Nuances i n the position are more by the king (fi rst system of defence) or the
important than material. I n endings with bishop (second system of defence).
opposite-colour bishops the nu mber of pawns
on the board is often of fa r less importa nce VII. First system of defence - the weaker
even than seemingly insign ificant cha nges side's king blocks the opponent's passed
in the placing of the pieces or pawns. pawn , and the bishop defends its own
Therefore in endings with opposite-col­ pawns. This is the basic and usually the
our bishops, positional pawn sacrifices most rel iable method of defence.
constantly occur. Thus in the textbook Attempts to destroy the first system of
example Wh ite happily sacrifices a third ( ! ) defence always involve creating a sec­
pawn in order t o achieve a 'trifle' - shift the ond passed pawn, often by means of a
black e-pawn one step forward . pawn breakthrough.
VI I I . Second system of defence - the
V. Principle of one diagonal . Both for the bishop stops the passed pawn (or some­
stronger, and the weaker side it is very times two , along the same d iagon a l ) , while
important that the bishop should defend the king , expressed i n footbal l language,
its own pawns and restrain the enemy engages i n 'zonal defence' - it protects its
pawns 'without being torn', along one pawns and restricts the activity of the
and the same diagonal. In the concl uding opponent's king.
position of the textbook exa mple the bishop Attempts to destroy the second system
on the h3-c8 diagonal defends the h3-pawn of defence always involve breaking
 From the Simple to the Complicated ttJ 73

through with the king to its passed pawn I n a book h e wrote on the endgame, N i kolai
(sometimes after a preparatory diver­ Krog ius considered th is outcome to be
sionary attack on the opposite wing). perfectly logica l . I n fact the position is, of
cou rse , d rawn - this is clear at first glance, it
We wil l now do some tra i n i n g i n the being sufficient only to remember the d raw­
employment of this theoretical foundation ing tendencies with opposite-colour bish­
for the analysis of specific endings. We wi l l ops.
try t o approach them in a logical way: w e will How can one explain such a bad mistake in
point out which system of defence has been eva luation , made by a player who at one
employed or should have been employed by time was q u ite a strong g randmaster? I n my
the weaker side and i n what way it may be view, by a change of professio n : one by no
possible to try and destroy this fortress, mea ns fi ne day Krog ius decided to ' re­
whether the pawns a re correctly placed , qual ify as a manager' , first in his native
whethe r it isn't possible to put the oppo­ Saratov, and then in Moscow - he became
nent's pawns ' u nder attack' , whether, in head (and, it should be mentioned , a very
order to carry out some idea, it is possible to nasty head ) of the Chess Ad ministration of
sacrifice a pawn or two , and so o n . the U S S R Sports Committee . Apparently
Caissa is a jealous woman who seeks
vengeance when she is betrayed .
F u c h s - Khol mov 43 . . . f6!
Dresden 1 956 44 'it>d2
Wh ite's objective is to defend the kingside
with his bishop and not a llow the opponent
to create a second passed pawn there. For
the moment the move i n the game does not
yet spoil anyth i n g , but it was simpler to play
44 d5! 'it>xd5 45 'it>d3(d2) followed by i.e3-
b6-d8 (the f6-pawn ' u nder attack' ). The
d raw would then be obvious - after moving
his king to f5 and playing . . . g6-g 5 , Black
would be u nable to make any fu rther
progress.
44 . . . 'it>f5
45 .if4?
Now the opponent inevitably obtains a
Black wil l probably obtain a passed pawn on passed pawn on the kingside. Meanwh ile
the queenside, but it will be blockaded by the 'pawns u nder attack' proced u re could
the opponent's king (fi rst system of de­ also have operated successfu l l y here: 45
fence). The only wi n n i ng chance is to create .ih6! g5 (45 . . .'1t>g4 46 .ig7 f5 47 .ih6 or 47
a second passed pawn . For this Black d5) 46 i.g7 ! , preventi ng 46 ... '1t>g4 . 45 d5!
needs to play . . . f7-f6 , . . . 'it>f5 and g6-g 5 , i.xd5 (45 . . . g5 46 d6 i.c6 47 .id4) 46 i.d4
then exchange on h 4 and win t h e h-pawn. I n or 46 i.b6 g5 47 .id8 was also good . It is
t h e game Ratmi r Khol mov successfu lly evident that proced u res i n the playing of
carried out this plan and won . endings with opposite-colour bishops, such
74 � F rom the Simple to the Complicated

as the sacrifice of one's own pawns or Bogolj ubow - Ed. Lasker

attacks on the opponent's, were u n known New York 1 924
both to the national master Fuchs, and to
grand master Krog ius.
45 . . . g5
46 i.c7 'it>g4
47 i.d8 gxh4
48 gxh4 'it>xh4
49 i.xf6+ 'it>g4
50 'it>e3 i.d5
51 i..e 7 b5
Wh ite resigned - in the opin ion of Krogius,
because of the variation 52 i.. d 8 h4 53 f3+
i.xf3 54 'it>f2 h3 followed by the switching of
the king to the queenside (on the h3-c8
diagonal the black bishop defends its own Wh ite should win tha n ks to his powerfu l pair
pawn and restrains the wh ite d-pawn). I n of con nected passed pawns. It was simplest
fact 53 . . . i..xf3? i s a mistake , of cou rse, i n to bring the king to the centre : 36 �f2 .
view of 54 i.xh4 ! ; Black should play Apparently Efim Bogoljubow was striving to
53 . . . 'it>g3! . play as safely as possible - he wanted to
Meanwh ile, even after 4 5 i..f4 a d raw is stil l prevent . . . a7-a5 and with this a i m he
possible. A s was shown b y grandmaster decided to exchange the rooks. I n the game
Sergey Shipov, by conti n u i ng 47 'it>e3! his plan proved justified .
(instead of 47 i.d8) 47 . . . gxh4 48 gxh4 'it>xh4 36 '/J.c7 �f7 37 '/J.xe7+ �xe7 38 i.d2!
49 'it>f4 'it>h3 50 i.d8 Wh ite would have (foresta l l i ng Black's cou nterplay on the
saved the game. q ueenside) 38 . . . We6 39 'it>f2 Wd6 40 We3
And also later, just two moves before �c5 41 i.. a 5 , and Wh ite won easily.
capitu lation, it was possible to gain a d raw Remember the need to exercise caution
by choosing 50 �c3 ! (instead of 50 'it>e3?) when transposing into an ending with oppo­
50 . . . h4 51 'it>c4 h3 52 i.. e 5 �3 53 d5 'it>xf2 site-colour bishops , in view of the i n herent
54 'it>c5 i..f3 55 d6 i.. c6 56 'it>b6 �g2 57 d7 d rawing tendencies. As was shown by
(Carsten Muller, Frank Lamprecht). Alexander Alekh ine, Black could have saved
And yet Wh ite's 45th move was a fu nda­ the game.
mental mistake: instead of fi nding and 36 'fJ.c7? .Uxc7 !
erecting a secu re fortress, he allowed the 37 i.. x c7 b4!
opponent to complicate the play advanta­
The sacrifice of a pawn is a com mon
geously.
phenomenon in endings with opposite­
colour bishops.
38 axb4 i.a6!
39 d4 i.d3!
This is also a standard defensive procedure
- an attack on the enemy pawns. They are
 From the Simple to the Complicated ttJ 75

forced to move onto sq uares of the colour of The game concl uded : 44 ... a6? 45 'ifa7+
their bishop, where they completely lose 'it>h6 46 'i!Ve3+ 'lt>g7 47 �g5 ! (Black u nder­
their strength , since they can easily be estimated the strength of this move) 47 . . . 'ikd4
blockaded . 48 c7! ii.xg3+ 49 �xg3 Black resigned .
40 e5 ii.c4 It is not my i ntention to g ive a detailed
41 �2 a6 analysis of the ending. I will merely show
When defending, pawns should be kept on one way (I would not assert that it is the only
squares of the colour of the bishop. one, but i n my view it is the simplest) of
gaining a d raw. Why not i m mediately elimi­
42 'ite3 'itf7
n ate the main enemy - the c6-pawn ?
43 �4 h5
44 . . . ii'xc6!
The position is d rawn .
45 "it'xa7+
Taking into account the principle 'nua nces i n
Noth ing is given by 45 'i!t'f7+ 'it>h6 .
the position are more important t h a n mate­
rial', we should a lso check 38 'it>f2 ! ? (instead 45 . . . "iVc7 !
of 38 axb4 ), in order not to al low a blockade The bishop and the g3-pawn are attacked ,
ofthe central pawns. However, after 38 . . . bxa3 and therefore the exchange of q ueens is
there is no win for Wh ite - the a-pawn practically forced .
diverts the bishop from its control of the 46 'ii'x c7+ ii.xc7
squares in front of the con nected passed Transposing i nto an ending with opposite­
pawns. Here is an approximate variation, colour bishops is an important defensive
suggested by Igor Bondarevsky: 39 <>t>e3 a2 procedu re, with the help of which one can
40 .te5+ �f7 41 ii.b2 �e6 42 d4 ( 42 �f4 sometimes save a d ifficu lt position , and
h6) 42 . . . 'it>d6 43 d5 h6 44 'it>d4 ii.a8 45 e5+ therefore, of cou rse, the suggested plan
�d7 46 'it>c5 (46 e6+ 'it>d6) 46 . . . ii.b7 47 e6+ deserved serious consideration . Grand mas­
�e7, and Wh ite is not able to strengthen his ter Igor Khenkin was afra id that the end­
position . game was lost, since Wh ite has two extra
pawns. I n fact it is a simple d raw, and i n
Kharlov - Khenkin establishing t h i s w e are helped , apart from
Copen hagen 1 993 the general g u ide ('drawing tendencies' ) by
a qu ite specific one. If Wh ite g ives up his g3-
pawn, we obta i n the wel l-known d rawn
situation from the Berger-Kolterma n game.
But if he advances it to g4, Black replies
. . . g6-g5, and blocks all the enemy king's
approaches to the upper half of the board .
Here is an approximate variation:
47 �xh3 �6
48 �g4 ii.d6
While there is time, it is useful to force the
opponent's pawn to stand on a square of the
colour of its bishop.
49 b5 i.. c 7
50 �d5 ri;e7
76 � From the Simple to the Complicated

51 �c6 <t>f6 It appears that White is out of danger - his

52 �f3 �e7 bishop has defended his pawns, and his
53 g4 g5 king is blocking the opponent's passed
pawn . But the bishop is really very passive ­
54 'it>e3 �b6+
soon it will not have a single waiting move. If
55 �d3 'it>d6 the king can also be deprived of its mobility,
56 �c4 �e5 a zugzwang situation could result. This aim,
The draw is obvious - there is nowhere for strangely enoug h , is q u ite ach ievable: the
the wh ite king to break throug h . wh ite king is g radually pushed to b3 and the
black king will occupy the d 3-square , from
where it conti nues to tied down the oppo­
Vakhidov - Timoshchenko nent's bishop and at the same time th reat­
Tashkent 1 982 ens to support its passed pawn . But if the
pawn were slig htly fu rther away - on the a­
file - a d raw would become inevitable.
5 ..td1 'it>g3 6 �f3 'it>f2 7 'it>c2 b4 8 �b3
..tc3 9 �c4 h6 1 0 �d3 ..te1 1 1 �c4 ..td2 12
�d3 �c3 (zugzwan g ! ) 1 3 �c4 �e3 (again
zugzwan g ! ) .
If 1 4 � b 3 , t h e n 1 4 . . . �d3 (the decisive
zugzwang ! ) 1 5 �a2(a4) �c2 and wins.
The game went 14 'it>d5 b3 and Wh ite
resigned .
Now let us try to set up the second system of
defence - use the king for the defence of the
kingside. But this plan too is not altogether
rel iable - after a l l , the bishop will have to
For the moment Black is not threatening to perform two tasks: not only conta i n the
play 1 . . . <t>g3 in view of 2 g5! and 3 �g8 - he enemy passed pawn , but also defend its
is plan ning 1 . . . g5! and only then 2 . . . �g3. own e4-pawn , and along a d ifferent d iago­
Now White has to decide how he will defend n a l . This means that here too a zugzwang
his kingside pawns and which piece will hold
position is q u ite l i kely.
back the enemy passed pawn on the
1 ..td5 g5! 2 'it>e2 h6 3 ..tb7 ..tc5 4 ..td5 a6 5
queenside.
..tc4 (5 ..tb7 b5 6 a5 b4 7 ..txa6 �xe4 is no
In the game he chose the fi rst system of
better) 5 . . . b5! 6 axb5 a5 (Black happily
defence: he switched his bishop to the
sacrifices a pawn for the sake of creating a
defence of his pawns, and kept his king on
passed pawn ) 7 ..td5 a4 8 ..tc6 a3 9 ..td5
the queenside.
..tb6. Zugzwa ng! The wh ite bishop has no
1 .i.b3 g5! 2 .i.d1 moves, si nce it is 'torn ' between two
2 h4 gxh4 3 g5 does not work: 3 . . . �g3! (but d iagonals. I n the event of 1 0 Wf1 (e 1 ) 'it>e3
not 3 . . . 'it>xg5? 4 .i.e6 <t>f4 5 .i.f5 'it>g3 6 .i.h3 the black king breaks through to its passed
with a draw) 4 ..tg8 'it>xg2 5 ..txh7 ..tc5 ! , and pawn , while if 1 0 'it>d 3 , then 1 0 . . . Wg3 1 1
Black wins. '>t>e2 (as i t i s easy to see , 1 1 'i!i>c3 'it>xg2 1 2
2 ... a6 3 ..tf3 b5 4 axb5 axb5 <t>b3 'i!i>xh3 1 3 .i.e6 also does not help)
 F rom the Simple to the Complicated ltJ 77

1 1 . . .'it>xg2 1 2 ii.e6 �xh3 1 3 �f3 'it>h4 1 4 ii.f7 ing it let us be prepared to sacrifice a pawn !
ic? (again zugzwang because of the 1 g5! ! �xg5
bishop being 'torn ' ) 1 5 ii.e6 h5 1 6 gxh5 2 ii.g8! h5
�xh5 17 'it>g3 'it>g6 1 8 �g4 'it>f6 1 9 ii.d5
2 . . . h6 comes to the same th i n g . 2 . .<>t>f6 3
�e7! 20 'it>xg5 �d6, and Black fi nally carries
.

�c4 also does not achieve anyth ing.

out the main idea for destroying the second
system of defence - the breakthrough with 3 ii.f7 h4
his king to the passed pawn . 4 'it>c4
Is the initial position really lost for Wh ite? The d raw is obvious, since now the bishop
Let's use our knowledge of opposite-colour easily copes with the defence of the king side.
bishops to guess where a saving line might White's moves can be transposed : 1 �g8 h6
nevertheless be concealed . 2 g5!! �xg5 3 iLf7 .
First of all one should usually check the
basic system of defence - the fi rst. But how
to securely defend the kingside with the I n conclusion I offer a few exercises, in the
bishop, and prevent there the creation of a solving of which you will train you rself in the
second passed pawn? The ma noeuvre of practical appl ication of your theoretical
the bishop to f3 solves this problem , but it knowledge. I advise you a lso to look at the
inevitably leads to zugzwa ng. I s there no instructive endings with opposite-colour bish­
other way? Remember the procedu re ' pawns ops, analysed i n my book School of Chess
under attack' and for the sake of implement- Excellence 1 - Endgame Analysis.
78 w F rom the Simple to the Complicated

Exercises

1 . Black to move 2. White to move

3. White to move 4. B lack to move

 F rom the Simple to the Complicated LtJ 79

Sol uti ons

1 . S . Tarrasch ( 1 92 1 ). 1 . . . 'it>c6
It is not possible to prevent the advance of 2 'it>e2 !
the pawns to the 5th ra nk (for this the bishop 2 3i.f7? d5.
would have to be switched to c6) . But how 2. . . �c1
should the black pieces be deployed agai nst
While Wh ite is tied down , the black bishop is
pawns on the 5th ra nk? Obviously, bishop
switched to a better positi o n .
on f7(g8) and king on d7. It is this set-up that
must be prepa red . 3 �d1 3i.b2

1 .. . Si.c4! 4 'i!.>e2 .il.d4

1 . . ..ib5? is incorrect: 2 Si.b4+ ! (but not 2 5 �d1 'it>d6

ig3+? 'it>e7 ! 3 d5 Si.e8 4 e5 Si.f7 ) 2 . . . �c7 3 5 . . . �c7 is answered in the same way.
d5 .ie8 4 e5 jLf7 5 e6 - Black is one tempo 6 .il.f7!
short. Or 2 . . . 'it>e6 3 d5+ 'it>e5 4 Si.c3+ �d6 5 Again attacki ng a pawn!
litd4 .ie8 6 e5+ , and the bishop has not 6. . . b2
managed to reach f7 .
7 i.. g 6 �cs
2 i.. g 3+ �c6!
8 'it>e2 dS
Of cou rse , not 2 . . . �e6? 3 'i!.>d2 and 4 'it>c3 .
9 i..f5 �b4
3 �f4 .i.g8
1 0 3i.g6 �a3
4 �es 'it>d7
11 3i.b1 ! 'it>b3
5 d5 ii.h7!
12 'it>d1 'it>c3
'Pawns under attack' - Black does not al low
1 3 'i!.>e2 ii.c5
the o pponent's king to go to f6 . However, the
less accurate 5 . . . Si.f7 6 'it>f6 We8 ! 7 Si.f4 3i.g8 14 'it>d1 d4
was also sufficient for a d raw. 1 5 �e2 'it>b3
6 �f4 ii.g6 The last hope : 1 6 �d 1 ? d 3 ! is bad for Wh ite .
7 e5 .il.f7! 1 6 �d3 !
A basic drawn position has been reached . Black ca nnot make a n y progress .

2. V. C hekhover ( 1 950). 3. A. Norl i n ( 1 922 ).

If W h ite should succeed in winning the d7- The typical plan is to march the king to the
pawn, this will lead to a familiar position from pawn which is being stopped by the bishop,
the Berger-Kolterman game. But if he i.e. to f8 . But then Black will advance his a­
doesn't? Then he must at least force the b­ pawn , d iverti ng the bishop from the defence
pawn to take a step forwa rd , onto a square of the c7 -pawn.
of the colour of its bishop, so that the black The only winning chance is to switch the
king will be unable to break through via b2 . bishop to a5, from where on the same a5-d8
1 i.. e 8! diagonal it will defend its own pawn and stop
'Pawn under attack!' the opponent's. But fi rst the c7-pawn must
80 � F rom the Simple to the Complicated

be defended with the king, without allowing positional idea (imagine that Wh ite were to
. . . a7-a5-a4 . If the black pawn should reach play c2-c4 , b2-b3 and a2-a4 - then the b6·
a4, the position will become d rawn , for pawn would be transformed into a serious
example: 1 �c5? a5! 2 �b5 a4 3 �b4 �c8 . weakness). The move made by Black is not
1 �c3 ! Jl.. f7 bad in itself, but for the reason that it does
2 �b4 Jl.. e 6 not help to solve the main problem of the
position - the defence of the kingside
3 Jl.. e 5!
pawns.
It is important to vacate the d6-square for
40 'it>g3 �c8 41 'it>f4 'it>d7 42 j;_b4 �e6 43
the king beforehand. 3 �c5?! is inaccu rate
Jl.. c 3 Ji.. d 7
in view of 3 . . . Ji.. b 3! with the threat of 4 . . . a5.
If Black keeps his bishop at g6 and uses his
3... �c8 ! ?
king to stop the futu re passed pawn on the
I f 3 . . . JI..f7 , then 4 �c5 Ji.. b 3 ( 4 . . . a5 5 'it> b 5 ) 5
q ueenside (first system of defence), at an
�d6 (th reatening 6 �d7) 5 . . .'it>c8 6 Jl.. c3 ! , or
appropriate moment Wh ite will attack the
4 ... �c8 5 �c6! (th reatening 6 Jl.. c 3) 5 ... JI.. e 8+
bishop by h3-h4-h5 and obta i n a second
(5 . . . a5 6 <it'b5) 6 'it>d6 Jl.. f7 7 Jl.. c 3! and 8
passed pawn. For example, 43 . . . ii.g6 44
Jl.. a 5.
<lt>g5 �d5 45 g3 b5 46 h4 �c6 4 7 b3 cxb3 48
4 �b5! cxb3 'it>b6 49 a4 bxa4 50 bxa4 'it>a6 51 a5
The variation given by the author is slig htly �b5 52 h5 ii.e8 53 �xf5 ii.xh5 54 �xe4 with
longer: 4 �c5 Ji.. b 3! 5 'it>b5! 'it>b7 6 <l;b4! and an easy wi n . Therefore Black leaves his
7 'it>c5 . bishop on the queenside. U nfortunately for
4. . . 'it>b7 h i m , his king can not simultaneously defend
5 'it>a6 was threatened . the h7- and f5-pawns, and therefore his
5 'it>c5 bishop will be 'torn' between the defence of
Ji.. b 3
the f5-pawn and the struggle agai nst the
6 �d6 �c8
opponent's passed pawn .
7 Jl.. c3
44 g3 b5 45 �g5 �f7 46 h4 Jl.. c 8 47 �h6
The next move will be 8 .Jta5, after which the �g8 48 b3 cxb3 49 cxb3 f4
king will fi nally be able to win the bishop for
This is already desperation in a hopeless
the g-pawn .
position. If 49 . . . j;_d7 Aaron N i mzowitsch
gave the fol lowing variation : 50 ii.b2 ii.c8
(50 . . . JI.. e 8 51 �g5 j;_d7 52 'it>f6 , and the
4. Nimzowitsch-Tarrasch wh ite king breaks through on the queen side)
(Kissingen 1 928). 51 a4 bxa4 52 bxa4 j;_d7 53 a5 ii.c8 54 ii.a 1 ,
Black must decide how to combat the and Black i s i n zugzwa ng (54 . . . ii.a6 55 �g5
threatened attack by the king on his kingside ii.c8 56 'it>f6).
pawns. The 'active' 39 . . .f4? is hopeless: 40 50 gxf4 j;_d7 51 �g5 'it>f7 52 f5 j;_c6 53 �f4
Jl.. g 5 e3 (40 .. .f3 41 g4 ), and Wh ite has a (the standard plan: the king heads towards
pleasant choice between 4 1 fxe3 and 4 1 f3 the passed pawn wh ich is being combated
e2 42 Ji.. h4 followed by �g 1 -f2 . Fi rst let us by the bishop) 53 . . . <.t>e7 54 'it>e5 ii.e8 55
see what happened in the game. <l;xe4 i.c6+ 56 �e5 ii.e8 57 <l;d5 ii.f7+ 58
39 ... c4? <lt>c5 Jl.. e 8 59 Jl.. e 5 ii.d7 60 �b6 <.t>f7 61 f6
Moving the pawn onto a sq uare of the colour Jl.. e 8 62 f4 <.t>e6 63 'it>a6! Wf7 64 b4 �e6
of its own bishop is, in genera l , a sound 65 a4 bxa4 66 b5 Black resigned .
 From the Simple to the Complicated 4J 81

As usual, we should fi rst look for a possibil ity 40 g4 fxg4 41 hxg4 i.e2 42 'it>g3 i.f3 .
of setting up the fi rst system of defence - 40 . . . .if1 !
leave the king on the q ueenside and ensure 41 h4 h5!
the defence of the pawns by the bishop. If
42 �4
the principle 'pawns under attack' is remem­
Otherwise it is not possible to strengthen the
bered , the correct solution (poi nted out by
position .
Averbakh ) does not seem at all d ifficult.
42 . . . .ixg2
39 . . . .ib5!
The black bishop easily copes with the
40 'it>g3 defence of the kingside pawns.
82 �
Mark Dvoretsky

The Arith meti c of Pawn E n d i ngs

A rapid eval uation rule

Ppawns and an outside passed pawn for

ositions with a pair of blocked rooks'

one of the sides occu r qu ite often in practice.

Therefore it is usefu l to be able to evaluate
them quickly and accu rately. The winning
plan is obvious: march the king over to the
rook's pawn . The opponent has to elimi nate
the pawn on the other wing and then rush
with his king to the corner, in order to stop
the rook's pawn . I n which cases does he
manage to do this?
It is easy to see that the position has
become d rawn : 1 '>t>d5 'iM6 2 'it>c5 'it>xf5 3
'it>b5 '>t>e6 4 'it>xa5 'it>d7 5 'lt>b6 '>t>c8.
If, say, the kings and the f-pawn are sh ifted
one rank down or to the left, Black again
loses . But what happens if the queenside
pawns a re also sh ifted down?
Of course, if you have the position in front of
you , it is easy to g ive an answer to any such
q uestion . But i n practice such situations
often a rise at the end of lengthy variations,
which you have to calculate , and to lengthen
the calculation by several more moves may
prove d ifficult. It would be desirable to learn
to determ ine the evaluation of the position
Here White wins: 1 'it>d5 'it>f6 2 'it>c6 'it>xf5 3
i mmed iately, on first looking at it.
'it>b6 'it>e6 4 'it>xa6 'it>d7 5 'it>b 7.
A simple method of rapid evaluation was
Now let us shift the queenside pawns back
suggested by Walther Bahr in 1 936. To me
by one rank.
this ru le seems not altogether conven ient,
and besides it does not extend to cases
where the king is not to the side of the
passed pawn , but in front of it. I n con nection
with th is I should l i ke to offer a somewhat
 The Arithmetic of Pawn Endings lLJ 83

different method of rapidly evaluating such

positions.
1 ) The first rule coincides with Bahr's
analogous ru le: if the rook's pawn of the
stronger side has crossed the middle of
the board, the position is always won. It
follows that from a single glance at the first
diagram it may be concluded that the
position is won .
2 ) We will call the position i n the second
diagram 'norma l ' . This is what makes it
such :
a) between the queenside pawns there
passes the i nvisible demarcation line, sepa­ With Wh ite to move , it is a w i n : 1 a5! (the
rating the u pper and lower halves of the pawn has crossed the middle of the board ) .
board ; With Black t o move , i t is a d raw: 1 . a5! , a n d
. .

b) the black king , which is aiming for the c8- a 'normal' position a rises.
square, reaches there without loss of ti me.
This happens because the passed pawn
has already crossed the key h3-c8 d iago­
nal , or is on this diagona l .
A 'normal' position is drawn.
3) Each sh ifting of the kingside pawn one
square down from the h 3-c8 d iagonal is
equivalent to a tem po i n favou r of White. For
example, the pawn on f4 g ives one tempo in
favour of Wh ite , and the pawn on e4 g ives
two. One fu rther tempo for the stronger side
may be given by having his king not to the
side of the passed pawn, but in front of it.
But each shifting of the q ueenside pawns
one square down compared with the 'nor­ Wh ite wins: he has two tempi (the g2-pawn
mal' position g ives the defending side a is two squares lower than the g4-sq uare),
tempo. With the pawns on a3/a4 Black has while Black has only one. But if the
one tempo in his favou r, and with the pawns queenside pawns a re sh ifted down one
on a2/a3 he has two . ran k , the score becomes 2-2 and the
position is now d rawn .
White wins only if the sum of tempi,
calculated in this way, is in his favour.
The formulation suggested by me looks
rather complicated and cu m bersome, but if
you learn it thoroughly you will find it very
easy to use.
84 \t> The Arithmetic of Pawn Endings

Accord ing to the afore-mentioned rule, this

should be a d raw. And i ndeed : after 1 . . �5 .

2 'ith5 (2 'itg3 'itg5 - a 'normal' position)

B lack does not play 2 . . . 'iii>f6? 3 'itg4 , when
Wh ite acq u i res an extra tempo, since his
king is in front of the pawn , but 2 . . .'itf4! 3 h4
(3 'itg6 'itg3) 3 . . . 'itf5 4 'ith6 'itf6 5 �h7
'itm(f5) 6 h5 'itt6 ! etc.

N ow let us examine some more complicated

endings, i n which a mastery of the rule
suggested by me sign ificantly eases the
calculation of variations.
Here, of cou rse , it is White to move (with
Black to move he has to concede the P rivorotsky - Peterson
opposition and Wh ite can queen his f-pawn ) . Riga 1 967
Wh ite wins, since t h e score is 3-2 in his
favour (two tempi a re g iven by the pawn on
f3 and another one by the position of the
king in front of the pawn ). He wins by 1 'ite4!
'ite6 2 'itd4(d3). It would be a blu nder to
play 1 'ite3? 'ite5(f5), since then a position
with a tempo ratio 2-2 is reached (the wh ite
king is no longer in front of the pawn , but to
the side of it), and this means a d raw.
One more useful deta i l . Let us suppose that
White's passed pawn is a rook's pawn , and
his king is in front of it, but the enemy king is
confining its opponent to the h-file. This
situation is equivalent to the one in
which the king is to the side of the pawn.
Black's positional adva ntage is obvious. His
plan is clear: . . . 'itg6-f5-e4 and then an
attack by his bishop or king on the q ueenside
pawns. This plan can be prevented by
offering an exchange of bishops, but this
demands precise calculation .
1 ..td4! ..txd4+
1 . . . ..th6 2 'iii>f2 ..tc1 3 'itxf3 ..txb2 4 a4 with
equal ity.
2 cxd4 'iii> f5
3 'itf2 'ite4
4 d51
Otherwise 4 . . . 'itxd4 5 'itxf3 'itd3 .
 The Arithmetic of Pawn Endings l2J 85

4. . . 'it>xd5 Matanovic - Botvin n i k

5 'it>xf3 'it>d4 Belg rade 1 969
6 'it>e2 c3
If 6 . h4 7 'it>d2 a5 there follows 8 'it>c2 or 8
. .

a4, but not 8 'it>e2?? c3 9 bxc3+ 'it>xc3 , and

Black wins.
7 bxc3+ 'it>xc3
8 h4! !
The only way! Otherwise Black h imself
would have played 8 . . . h4! , then picked up
the a3-pawn and won , since his pawn on the
other wing has crossed the m iddle of the
board . But after the move in the game a
'normal' position a rises, and this means a
draw.
8. . . 'it>b3 In his notes Mikhail Botvinnik analyses two
9 'it>d3 'it>xa3 cou rses of action for Wh ite : 43 .l:td5 and 43
10 'it>c3 l::td 6+ 'it>e7 44 l:f.a6. In fact there is also a
We have reached the last of the textbook third : 43 'it>f2 ! , for example, 43 . . . exd3 44
examples that we analysed . 'it>e3 l::.a 1 (44 . . . .l::!. g 1 45 'it>f2 ) 45 l::.x d3 .l::i. x a4
10 . . . a5 46 l:td6+ followed by 47 l::. a 6, and Wh ite
11 'it>c4! 'it>a4 should gain a d raw.
'it>b5 But let's forget about this possibil ity and try
12 'it>c3
to choose the more accu rate of the two
13 'it>b3
possible rook moves.
Draw.
First, after analysing some short variations,
In the calculation of this ending, d ifferent we must try to disclose the d ifference
versions of this type of position a rose. If between them, and compa re their virtues
White does not evaluate them ' mechani­ and d rawbacks .
cally' , by using the rule g iven above , but
I n the event of 43 .l:td6+ 'it>e7 44 .l::i. a 6 a clear
tries to work out the variations to the end, he
d raw resu lts from 44 .. J1xd3 45 .l::i. x a5 or
each time has to calculate some ten more
44 . . . .l:!.d2+ 45 lt:Jf2 e3 46 'it>f3 ! (46 l:txa5
moves, and this is not at all easy.
.l:!.xf2+ 47 'it>g 1 is also possible) 46 . . . e2
(46 . . . exf2 47 'it>g2) 47 l:1a7+. However, the
The players in the fol lowing ending faced captu re on d3 with the pawn is un pleasant:
even more complicated problems. 44 . . . exd3! 45 .l:!.xa5 'it>d6. N ow 46 'it>f2? is
bad in view of 46 .. J1g 1 !; Wh ite is forced to
play 46 .l:!.a8 , allowing the black king to
approach its passed d-pawn . I s this rook
ending lost or d rawn? You can't say i m medi­
ately, and this means it is time to cut short
the calculation and switch to a verification of
the alternative possibil ity.
86 � The Arithmetic of Pawn Endings

43 l:td5! .l::t d 2+! .l:!.a6 (I again repeat, that we have ag reed to

Now the rook ending arising after 43 . . . exd3 forget about the existence of 43 �f2 ).
44 .UXa5 is not dangerous for White : 44 . . . d2 In the pawn ending Black has two plans of
45 l:Id5, or 44 .. .'it>e6 45 .l::t e 5+ �d6 46 �f2 action: bring the king to the centre in the
d2 (46 . . .l:tg 1 47 .l::t e 3) 47 <t;e2 .l::tg 1 48 �xd2 hope of putting the opponent in zugzwang,
.UXg3 49 .l:!.e3. On the other hand, thanks to or break through on the kingside with . . . g6-
the pin on the knight along the d-file, Black g5.
can interpose a check with his rook, ena­ The fi rst plan, as it is not d ifficult to see, is
bling him to tra nspose into a favou rable completely harmless: 46 . . .<t;e6 47 �e3 Wd6
pawn ending. (47 . . . �d5 48 �xd3 h6 49 g4) 48 �xd3 Wd5
44 �f1 .l::t x d3 49 g4 h6 (if 49 . . . fxg4 50 hxg4 h5 there is 51
45 .l::t x d3 f5 ! , although 52 gxh5 gxh5 53 �e3 also
A forced exchange, since otherwise the g3- does not lose) 50 g5! hxg5 (50 . . . h5 5 1 h4) 5 1
pawn is lost. fxg5 f4 5 2 h 4 �e5 53 �e2 with equal ity.
45 . . . exd3 46 . . . g5!
46 �f2 47 fxg5+!
Not 47 �e3? gxf4+ 48 gxf4 �e6 49 h4 '.td5
50 �xd3 h5.
47 . . . 'it'xg5
48 �e3 h5
49 �xd3 h4
The situation a rising after 49 .. .f4 50 gxf4+
Wxf4 has already been seen in the previous
example. The wh ite king does not manage
to attack the a-pawn, but it does not need to
- it is sufficient to shut in the enemy king on
the h-file, for example: 51 h4 (or 51 '.td4)
51 . . . �g4 52 �e4 �xh4 53 'it>f4 with a draw.
50 gxh4+ �xh4

Alas, here too it is not immed iately clear

whether White can save the game. But
nevertheless, pawn endings are usually
more forcing in character compared with
rook endings, and here, as a ru le, it proves
possible, by calculating a variation to the
end , to give an exact evaluation of the
position. Therefore we should concentrate
our efforts on the calcu lation of th is pawn
ending. If it should tu rn out to be d rawn , we
will make the move 43 .l::td 5, but if it is lost
then we will reluctantly have to go into the
unpleasant rook ending by 43 l:td6+ �e7 44
 The Arith metic of Pawn Endings lLJ 87

Which of the two natura l moves, 5 1 'it>e3 or 46 l:ta8 �c7 (Black repeats moves, to gain
51 '.t>e2 , should be made? Let us refer to the time for thought) 47 l:ta5 'lt>c6 48 l:ta8 �c5
rule g iven above . After Black wins the h3- 49 'it>f2 l:ta 1 ! 50 l:td8
pawn , accord ing to our a rithmetic he will The only saving chances were offered by 50
have one extra tempo, since the f-pawn is �e3 !? .l:!.g 1 ! 51 g4 fxg4 52 hxg4 l:txg4, and
one square higher than the key c1 -h6 now, probably, 53 f5 .
diagonal. Wh ite will d raw only if he can force
50 . . . '1t>c4 51 'it>e3 I:.e1 + ( 5 1 . . . .l:!.g 1 ? 52
the pawn to adva nce to f4 . bpk
l:.d4+) 52 'it>f2 �e2+ 53 'it>f3 .l:!.e6 ! 54 a5 'it>c3
It becomes clea r that 5 1 �e2? loses : 55 �c8+ 'it>d2 ! (only not 55 . . . �b3? 54 a6! d2
51 . . .'it>g3 (zugzwa ng) 52 h4 (52 �1 'lt>xh3 55 .l:!.d8 'lt>c2 56 a? or 54 . . . lixa6 55 �e3 .l:!.d6
53 'it>f2 'it>g4 , and Black has even two extra 56 �d2 ) 56 h4
tempi) 52 . . . �xh4 53 �f3 �g5 54 'lt>g3 'it>f6
Accord ing to analysis by Botvinnik, Wh ite
55 '.t>f4 'it'e6 56 �f3 'it>d5 and so o n .
would also have failed to save the game by
51 'it>e3 ! 'lt>g3 56 Uc7 h5 (56 . . . Ue1 ? 57 a6 l:!.a 1 58 a?) 57
52 'it>e2 'it>f2 �d 1 58 �f3 d2 59 �f2 l:te2+! 60 'i!tf1
Now it is Black who is in zugzwang and he is lie3 61 a6 (61 �f2 l:ta3 followed by . . . Ua 1 -
forced to advance his pawn . c1 ) 6 1 . . . �xg3 62 a ? Ua3 6 3 'it>f2 h 4 64 'it>f1
52 . . . f4 l:ta4 65 'it>g2 'i!te2 66 lie?+ '1t>d3 67 l:!.d7+
53 �1 �xh3 'it>e3.
54 �2 'lt>g4 56 ... .l:!.e1 ! 57 a6 lia1
55 �g2 Now if 58 ltc6 Black decides matters with
A 'normal' d rawn position has arise n . 58 . . . '1t>e 1 59 l:.e6+ 'it>f1 60 l:td6 (60 We3
l:te 1 + ) 60 . . . d2 61 l:txd2 �a3+ , and Wh ite is
Alexander Matanovic d i d n o t m a n a g e to mated ! 58 l:ta8 'i!te 1 59 a? d2 60 l:!.e8+ 'it>f1
calculate the pawn ending exactly and he 61 .l:!.d8 .l:!.a3+ leads to the same fi nish.
preferred to retai n the rooks. Let us see 58 lic7 �e1 59 'it>g2 l:txa6 60 l:te7+ 'it>d1 61
what this led to . �xh7 l:ta2+ 62 Wf1 d2 63 .l::!. c 7 l:ta1 64 'it>f2
43 .l::!. d 6+? 'it>e7 44 l:!.a6 exd 3 ! 45 �xa5 'it>d6 lic1 Wh ite resigned .
88 �

PART I I

Endgame Analys i s
Vlad i m ir Vu lfson

Typica l E n d i ngs with

Con nected Passed Pawns

I complicated analysis of an ending of one

should l i ke to show you the rather Black wants to tie the enemy rook to the
defence of the g3-pawn . 2 l::t h 4 is obviously
of my games. After studying it you will have hopeless, and therefore I stud ied , in the
a better understanding of the theory of rook m a i n , 2 hxg4 .l:i.xg4 . Things are difficult for
endings with connected passed pawns. Wh ite : noth ing is g iven by 3 .l:i.h6+ l:i.g6, and
so he is forced to play 3 l:th3, but here the
Zlotni k - Vu lfson rook is exceptionally passive .

Moscow 1 983 But here Boris Zlotnik unexpectedly made a

move which I had overlooked in my analysis.
2 'it>b2
The idea is clear - to avoid the capture of the
g3-pawn with check.
The move is an interesting one, and during
the game it seemed to me to be very stro n g .
B u t after a thorough analysis I began to
have doubts about its strength . The point is
that when Black elimi nates the g-pawn and
obtains a position with con nected passed
pawns, the basic method of defence is to try
and wedge the king between the pawns a n d
blockade the m . B u t here the king , in solvi n g
a pa rtial problem (i nvolving t h e g3-pawn ),
volu ntarily moves away from the queenside
The adjourned position ; it is my move . The
pawns.
national master Anatoly Donchenko sug­
gested an excellent idea for Black. Usually 2. . . gxh3
with an advantage you are recommended to 3 �xh3 .l::t g 4
avoid pawn exchanges, but this is an 4 l::t h 8
exception to the rule. With his rook on h3 Wh ite , naturally, can not
1 .. . g4! hope for success, and so he activates it.
 Typical Endings with Connected Passed Pawns CZJ 89

Black's reply is forced , since if 4 . . . .Uxg3? My opponent was the national master
there follows 5 .l::!. a 8 with an i m med iate d raw. Zlotn ik, a chess teacher in the I nstitute of
4 . . . aS Physical Cultu re. I g reatly respected h i m ,
Passed pawns must be pushed . and t o me he w a s a n expert. When you play
such a person , a defi n ite complex appears ,
S .l::!.c 8
you beg in t o fea r everyth i n g , and therefore i t
Aga i n it is not possible to captu re on g3 can b e d ifficult t o make an active move.
because of 6 .l:!.c6+ followed by 7 l:tc5 . I also
Besides, I did not consider that the position
reckoned with 5 .l:!.a8 , to force one of the
was one where every tempo cou nted , I
pawns to advance and allow the king more
thought my king would always be able to go
quickly to wedge itself between them. But I
over and captu re the g4-pawn , and for the
think that in this case too Wh ite would not
moment it would not be bad to help the
have been able to save the game.
queenside pawns .
5... .l:f.g5
What would have happened after 6 . . . rJ;e5 ?
The roo k defends the pawns from the side. Let us try to provoke the advance of one of
I n such situations this is the best place for
the pawns: 7 l:!.a8 . Black replies 7 . . . a4 , and if
the roo k . Now the black king is free to go 8 �a3(c3), then simply 8 . . . .l:txg4. The king
where it wants. has not managed to reach b4 and after 9
6 g4!? l1b8 l1g3+ 1 0 �a2 .l:!.b3 Black wins.
If 8 �b8 (instead of 8 �a3 ) , then 8 . . . �d4 9
�a3 �c5 1 0 lk8+ �b6 1 1 llb8+ �c6 1 2
lk8+ �b7 and 1 3 . . . I:txg4 with a wi n . The
fact that his king is cut off along the 6th rank
does not concern Black - his rook will free
the king by . . . l:tc4-c6 .
Thus, 6 . . . 'it>e5 was a very good move, but I
played differently.
7 l:tc1
Wh ite wants to place his rook behind his
passed pawn .
7 . . . �d 6 ?
7 . . . �d5 was far stronger.
Now the captu re of the pawn leads to a 8 .l:f.g 1
typical drawn position with connected passed I should l i ke t o dwell on this position in more
pawns, one which occu rs q u ite often : deta i l .
6 . . . .!::!.xg4? 7 .l:!.c6+ rJ; e7 8 .l:!.c5 l:tb4+ 9 'itta 3
(see diagram)
<ttd 6 1 0 .l:!.h5. If 1 O . . . l:tb1 there follows 1 1
lt>a2 . Black cannot strengthen his position ,
Wh ite has succeeded in sign ificantly activat­
since his king has n o shelter from the side
ing his rook. If his pawn were on g 5 , he
checks.
would u ndoubtedly be able to d raw. B ut with
6. . . � e6 ? the pawn on g4 his rook has not so many
6 . . .cj;>e5! suggests itself. Why did I reject this squares for manoeuvri n g . Black now has
move? The reason was a psychological one. two plans for playing for a w i n :
90 � Typical Endings with Connected Passed Pawns

1 ) play his king to the help of the queenside I n lessons devoted to the tech nique of
pawns; converti ng an advantage, an important
2) first captu re the g4-pawn with the king, principle has been mentioned : to make use
and only then return to the queenside. of any opportun ity to improve even slightly
We will first examine the simpler plan, you r own position and weaken the oppo­
involving 8 ... 'it>c5. It is obvious that if Black nent's. Here Black can move his king
can place his pawns on a4 and b4 he will win forward , but in this case the wh ite pawn
advances and there is no longer a win. The
easily. Therefore White's objective is to
hinder the advance of the pawns, l u re the only way to the goal is to interpose the check
rook away from g5 as soon as possible and 1 2 . . . .l:!.c5+ ! . If 1 3 Wb2, then 1 3 . . . .l:!.g5, and
begin advancing his passed pawn . the b-pawn advances to the 4th rank. I n the
event of 1 3 'it>d4 Black can either advance
First let us analyse 9 Wb3 . If 9 . . . b4 (with the
his pawn i m med iately, or fi rst play 1 3 . . . .l:!.g5.
threat of 1 O . 'i!i>b5) there follows 1 0 'it>a4
. .

There only remains 13 Wb4 l:tc4+ 1 4 'it>a3 .

'it>b6 1 1 .l:!.f1 .l:!.xg4 1 2 .l:!.f5 ! with an i m mediate
draw. The d i rect 1 4 . . .<it>a5? does not achieve
anyth ing: 1 5 'i!i>a2 b4 1 6 g5 b3+ 1 7 'it>a 1 ! a3
Let us verify 9 . . . a4+ 1 0 'it>a3 'it>b6! 1 1 'it>b4
1 8 g6 b2+ (for 1 8 . . . a2 and . . . 'i!i>b4-a3 Black
(preventing 1 1 . . . 'it>a5) 1 1 .. Jie5 ! . Against the
is just one tem po short) 1 9 'i!i>b 1 Wb4 20 g7
threat of 1 2 . . . l:te3 Wh ite has two defences:
'it>b3 2 1 .l:!.g3+ 'i!i>b4 22 .l:!.g 1 .
12 l:tg3 and 12 'it>c3 ( 12 'it>a3 .l:te2 is
unsu itable, since the king remains in a The correct move is 1 4 . . . .l:!.c2 ! (with the
mating net). th reat of 1 5 . . . 'it>a5) 1 5 'it>b4 .l:!.b2+ ! (nothing
is g iven by 1 5 . . . l:tf2 1 6 'it>c3 ; fi rst the position
After 1 2 .l:!.g3 .l:!.e4+ 1 3 'i!i>a3 'i!i>a5 (with the
of the wh ite king must be clarified ), and 1 6
th reat of 14 . . . .l:!.e2 ) 14 'i!i>b2(a2) b4 the black
Wa3 l:tf2 1 7 Wb4 .l:!.f3 is bad for Wh ite , while
pawns reach their goal sooner than the g­
after 1 6 Wc3 there follows 1 6 . . . a3 1 7 g5 b4+
pawn . No better is 1 3 'i!i>c3 b4+ 14 'it>d3 a3!
1 8 Wc4 a2 1 9 lla1 ( 1 9 g6 .l:!. b 1 ) 1 9 . . . b3 20
1 5 'it>c2 .l:!.e2+ etc.
g6 l::t b 1 2 1 g7.
Let us examine 1 2 'i!i>c3. Here the win for
Black is not obvious.
 Typical Endings with Connected Passed Pawns lZJ 91

"ilixc6+ "Y/Vc3 32 "Y/Va4+ 'it;b2 3 3 ikb5+ 'it;c 1 34

"YIVf 1 + 'lt;c2.
26. . . 'lt;c5 27 "ilia3+ (27 'flc2+ 'it;b4 28 "iie 4+
'lt;a3 29 'ii'e 7+ �b3 30 'ilif7+ 1Ic4) 27.. .'ild4!
28 "iib 4+ 'it>e3 29 ii'e7+ 'it;f3, and Black
blocks the next check with his queen or rook
- Dvoretsky.]
As you see , the win is very compl icated .
Besides, Wh ite's defence can be improved
at the very start of the variation . I nstead of 9
Wb3?! we can play more cu n n i ngly - 9
Wa3 ! , so that the pawn should move to a4
without check.
21 .l:i.xa1 22 g8 'ii' .l:tc1 + 23 'it>xb3 a 1 "i'i.
. . .
9 a4 (no better is 9 .. .<.tfb6 1 0 .l:!.g3 ! ) 1 0
...

[Nowadays, for the analysis of both opening .l:!.g3 ! . Now 1 0 . . . 'it>b6 1 1 Wb4 leads to a
and endgame positions, increasing use is position of m utual zugzwan g , and with Black
made of computers. John Nunn and Graham to move. 1 1 . . . .l:te5 is pointless, since 1 2 . . .l:Ie3
Burgess checked the concluding position of is not a th reat and White ca n simply
this variation on a computer, and it tran­ advance his paw n . After 1 1 . . . Wa6(c6 ) 1 2
spired that after 24 "YIVb8+ Black cannot .l:!.g 1 .l:te5 the move 1 3 . . . l:te3 i s n o longer
avoid perpetual check. For example, with deadly and again 1 3 g5 ca n be played .
the king on h7 there follows 1 "Y/Ve4+ 'lt;g7 2 Let us try 1 0 . Wc6 ! ? 1 1 Wb4 Wb6 .
. .

'le 7+, and with the king on c8 - 1 'ii'g 4+

rtt b 8(b7) 2 'ii'b 4+.
Nevertheless, Black has a way to win - but
instead of 2 1 . . . I:.xa 1 ? he should play 21 . . . b2!.
Here is the analysis by Nunn and Burgess:
22 gB'fi (22 l:lxa2 .l:i.c 1 + 23 'lt;d4 b 1 "YIV 24
ga"fk 'ikb4+ with a quick mate or win of the
white queen) 22 .. .1J.c1+! 23 Wd5 (23 J:!xc 1
bxc 1"fk+ 24 'lt;b3 'ilb 1 + 25 Wc4 'fib5+)
23 .l:!.c5+ 24 'it;d6 .l:!.c6+!
...

24 bxa1"ik? is premature: 25 "fibB+ 'lt;a6 26

. . .

'laB+ 'it;b5 27 "YIVb7+ 'lt;c4 28 "ilif7+ 'it;b4 29

'lb 7+ 1:.b5 30 "ike4+ 'it;b3 31 "ilie6+! with
perpetual check.
Now it is Wh ite who is in zugzwang. We
25 cJ,;d7
already know that he loses after 1 2 .l:!.g 1
After 25 cj;d5 bxa 1"YIV 26 "YIVb8+ 'lt;a5 27 .l:!.e5; let us see whether 1 2 .l:!.g2 l:I e 5 1 3
'fa 7+ 'it;b5 28 "ilib8+ I:.b6 29 'ii'e 8+ 'it>b4 30 Wa3 ! helps h i m . The d ifference compared
'ffB+ 'it;b3 the checks come to an end. with the position of the rook on g 1 is
25 bxa 1"YIV 26 "YIVb3+
...
i m mediately apparent: after 1 3 . . . Wa5 14 g5
No better is 26 "ikb8+ 'it;a5 27 "ik a8+ 'it>b4 28 the black rook can not invade at e2 . B lack is
'lb 7+ 'it;a3 29 "ilia7+ 'it;b3 30 "ilib7+ 'it;c2! 3 1 forced to advance his pawn : 14 ... b4+ .
92 � Typical Endings with Connected Passed Pawns

To where should the king move? The 10 . . . a4+?

outcome depends on this. Of course, 1 5 1 1 'it>b4 'it>xg4
'it>a2 ! , so that the a-pawn should advance 1 2 .l:!a3?!
without check. All the same Black plays
My opponent embarks on a ru inous course.
15 . . . a3 (but now without gain of tempo ! ) 1 6
He probably thought that he would be able
g6 'it>a4.
to restrict my king along the 3rd rank, but in
Threatening 1 7 . . . b3+ with mate . Wh ite loses fact his rook is badly placed here .
after 1 7 .l:. g 1 b3+ 1 8 'it>a 1 a2, but he finds the
12 . . . 'it>f4
defence 17 Zig4! l:te2+ 18 'it>a 1 ! . And now
18 . . . a2 is dangerous only for Black: 1 9 g7 1 3 .Uc3??
'it>a3 20 .l:!.g3+ b3 21 .l:!.xb3+! 'it>xb3 22 g8'it'+. Any move along the 3rd rank loses - the
Thus we have established that in the event rook should have moved off it.
of 8 . . . '1t>c5 Wh ite gains a d raw. 13 . . . .Ug3
In the game I moved my king the other way. My rook goes to b3, after which the pawns
8. . . 'it>e5 queen of the i r own accord . Black won easily.
You see , the king has nevertheless reached But how should my opponent have de­
e5, but instead of going there immed iately it fended? Let us assume that we do not know
has wasted time, by wandering about on the the theory of endings with such a pawn
e6- and d6-squares. a rrangement - let us try acting simply by
9 'it>b3 using common sense .
White intends by Ua 1 to force the advance Let us ask the q uestio n : 'What does Black
of one of the pawns, and then to establish want?' U ndoubtedly, to take his king to b2,
his king between the m . after which it will be possible to give up the
b5-pawn and queen the a-pawn . Let's try
9. . . 'it>f4
and hinder the movement of the king, by
1 0 Ita 1 cutting it off along a file.
12 .l:!.f1 .l:!.f5 1 3 .l::!. e 1 'it>f4 1 4 1:te2 .l:!.e5 1 5 1:td2
�e3 1 6 ltd 1 .l:t.g5 1 7 .l:!.d8 �e4 1 8 .:td 1 1:tf5. It
is important not to place the rook on e5 -
then a check on e 1 will d rive the king away.
1 8 . . . l:td5 is prematu re in view of 1 9 Ite 1 + .
Black must play for zugzwang. The white
king on b4 is ideally placed , so the position
of rook must be improved .
1 9 �d2 Itd5 20 .Uc2 (now 20 11e2+ 'it>d3 is
hopeless for White) 20 . . . 'it>d3 2 1 .l:tc8 (21
l1h2 is also not bad ) 2 1 ... 'it>d2 22 l:tc7 'it>d 1 .

(see next diagram)

Here I did not bother to ponder over which U p to this point Wh ite has not been in any
pawn to advance , and this was a mistake - particu lar danger, and he could have de­
one pawn move leads to a win , and the other fended in various ways . But here he must
to a draw. Let us first see what happened in make an accu rate move (23 .:tc3 or 23 l:f.h8),
the game. since Black has created the concrete threat
 Typical Endings with Connected Passed Pawns ttJ 93

(it is important to control the a3-square)

3 . . . �b2 . At first sight Wh ite is i n trouble ­
Black intends 4 . . . .l:1d4+ and 5 . . . a3. But let us
verify: 4 .l:.g3 l:td4+ 5 �c5 ! (5 �xb5? a3 6
.l:.g2+ �c3 7 .l:.g3+ .l:ld 3 ) . If the rook goes to
e4 , it is now possible to captu re the b5-pawn
and after 6 . . . a3 to begi n side checks. The
king has to step onto the d-file, but then the
rook attacks the a-pawn and this leads to a
d raw. This was Kasparian's conclusion.
I n this position I d iscovered another curious
subtlety: B lack can try 5 . . . :1d 1 ! . Again it is
not possible to captu re on b5, and 6 �b4 is
23 l:.d2 . For example, 23 'R.c7? l:td2 24
. . .
necessary, but then there follows 6 . . . l:. b 1 .
l:.c3 (24 �xb5 a3 25 �b4 a2 26 l:ta7 �c1 ) Now it is essential to take the pawn: 7 �xb5!
24 . . .U.b2+ 25 �a3 .l:. b 1 . After 26 l:th3 there
.
a3 8 �a4 a2 9 l:tg2+ , and the king is
follows 26 . . . �c2 with the threat of 27 . . . .l:.b3+, deprived of the important b 1 -square - d raw!
while if 26 l:tc8 , then 26 . . . �d2 , and the king [ There is another way to draw: 6 .l:t g2 + �b 1
approaches the pawns. It has acq u i red an 7 1:tg3 or 6 . . . � c3 7 �xb5 a3 8 � a4 -
excellent shelter from the side checks at a5. Dvoretsky.)
This is one of the i mportant winning posi­
tions. Let us retu rn to the position after Wh ite's
And now - the main d rawn position , which it 1 Oth move. We have seen that 1 O . . . a4+?
is also essential to know. leads to a d raw. Let us now analyse 1 O . . . b4!
1 1 �a4 ( 1 1 'R.g 1 .l:.x:g4 is hopeless) 1 1 . . .
H. Kasparian � x g4.
This pawn configuration is obviously stronger
than a4-b5, since after the sacrifice of the
a5-pawn the remaining b-pawn is more
da ngerous than the a-pawn , and affords
more wi n n i ng possibil ities. For the moment
the black king is free to approach the
q ueenside ( 1 2 .l:.f1 1:tf5 etc. ) . Let us see what
methods of defence Wh ite has agai nst
pawns on b4 and a5.

The first: playi ng for stalemate.

(see diagram)

Could we have reached this position? Quite If the black king goes to c3 , there follows
possibly. Wh ite could always have placed llc2 + ! . But this mechanism is easily de­
his rook on the 3rd rank. The only plan to stroyed - playing the black rook to the 2nd
play for a win is 1 . . . �c2 2 l:th2+ l:td2 3 l:th3! rank proves decisive.
94 � Typical Endings with Connected Passed Pawns

What is the simplest way to win here? Let's

The second: the attempt by Wh ite to improve the position of the rook: 1 . . . l:!.e5.
place his own king in the way. The threat is 2 . . . .l:.e3+ , lifting the blockade of
the pawns . Both 2 l:.d 1 + 'it>c5 3 'it>a4 .l:!.e3
and 2 'it>a4 'it>d3 (not i m mediately 2 . . J:te3? 3
'it>xa5 b3 4 'it>b4) 3 .l:tc8 (or 3 'it>b3) 3 . . . 1:!.e3 4
l:.c7 'it>d2 a re hopeless for Wh ite. You see
that the key sq uare for the king in this type of
ending is d4 - it is very i mportant to occupy
it! And after this - accord ing to ci rcum·
stances : if the wh ite king is at a4, the route
. . . 'it>d4-c3-b2(c2 ) becomes possible. With
the king on b3 it comes under check and the
black king can then go to c5 and b5.
It all seems to be very simple, but look at the
following position:

The drawback to the position of the king on

b3 is that it comes u nder check along the 3rd
rank.
1 . . . 'it>d4 2 1:!.a2 l:te3+ 3 'it>a4 (forced) 3 .. .'�c3,
or 2 l:!.a4 l:!.e3+ 3 'it>b2 'it>c4! 4 .l:.xa5 l:!.e2+,
and a won position, well known in theory, is
reached .
Black wins in roughly the same way with the
enemy rook on a8 (instead of a 1 ) : 1 . . . 'it>d4
with the idea of . . . .Ue3+ .

The third : t o cut off the black k i n g along

the file. What is the eval uation? Draw! The king
 Typical Endings with Connected Passed Pawns ltJ 95

cannot break th rough anywhere. The nu m ber of mistakes that the two players
We arrive at a general ru le for this type of made in this endgame! The reason was an
ending: inadequate knowledge of the theory of rook
- I f the black king is cut off i n its own half of
endings. They had no ' beacons' by which
the board , the position is d rawn ; they could be g u ided .
- But if it breaks th rough into the opponent's
ha l f of the board , the position becomes won . Thus in the game the moving of the king to
the kingside (alas, rather belatedly) should
have led to a d raw. Then I again retu rned to
A general conclusion for this type o f ending
the position after 8 . . . 'it>c5 9 'lt>a3.
also suggests itself:
Black's plan of moving his king towards the
white pawn and captu ri ng it leads to a wi n ,
whereas the plan o f playing the king to the
help of the queenside pawns only d raws.

However, later analysis sometimes intro­

duces serious corrections into seemingly
establ ished conclusions. On one occasion I
looked more carefully at the position after
8.. .�e5.

We have seriously stud ied only 9 . . . a4 , but

there is also another idea : 9 ... b4+ ! 10 �a4
l':!.d5! . The th reat is 1 1 . . .�d3. There is no
point in retu rn ing with the king : 1 1 �b3 �b5
(with the th reat of 1 2 .. J:td2 ), and Wh ite
loses. In the event of 1 1 .U.c1 + �b6 1 2 l:tf1
l:td3 1 3 l:tf6+ �c5 1 4 l:tf5+ �c4 1 5 �xa5 ( 1 5
l:tf4+ l:td4 and 1 6 . . . l:txg4) 1 5 . . . b3 Black
wins, since his passed pawn advances more
q u ickly than the opponent's, and also it is
supported by the king .
Let us verify 1 1 �xa5 l':!.d3 1 2 �a6
Why did White play 9 �b3 ? The im mediate (otherwise mate ; 1 2 l:!.c1 + l::tc 3 is bad for
9 J:ia1 ! was far more logica l . N ow the reply
White ) 1 2 . . . b3 1 3 g5 .
. . . b5-b4 is no longer possi ble, and after
9 . a4 1 0 'lt>a3 (not 1 0 �c3 .l:!.xg4 1 1 .l:!.b 1
..

l:ic4+ ) Wh ite need not fear 1 O . . . l:!.xg4 1 1 .l:!.b1

(see diagram)
with an immediate d raw. Black replies
1 0. . .'it>d5 , but 11 'it>b4 �c6 1 2 l:Lg1 leads to
an already familiar d rawn position .
96 � Typical Endings with Connected Passed Pawns

The direct 1 3 . . . b2? (with the idea of 1 4 . . . .l:!.a3+ In the event of 1 5. . . .l:!.g8 White saves himself
and 1 5 . . . .l:!.a 1 ) leads only to a d raw in view of by continuing 1 6 .l:!.g4+ (or 1 6 .l:!.g2) 1 6 . /ila3
.

1 4 .l:!.b 1 .l:!.a3+ ( 1 4 . . . .l:!.d2 1 5 g6) 1 5 �b7 .l:!.b3+ 17 .l:!.g3 1;; a2 18 .l:!.g5! b2 19 .l:!.a5+ �b3
16 cJi;c? 'it>b4 1 7 'it'd?! 'it>a3 1 8 'it>e6(e7) , and ( 1 9 . . . cJi;b 1 20 .l:!.g5) 20 .l:!.b5+ cbc3 21 .l:!.c5+
the white king, paradoxically, succeeds in 'it>d4 22 1lb5. If instead 15 . . . .l:!.a8+, then 1 6
uniting with its pawn . 'it>b7 (or 1 6 'it>b6 b 2 1 7 'it>c6) 1 6. . . 'J:.g8 1 7
1 3 . . . 'it>b4? is hopeless: 1 4 g6 b2 1 5 g7 lld8 'it>c6 b 2 1 8 'it>d5 ( a very important tempo;
1 6 g8'ii' l:txg8 1 7 l:.xg8, and if 1 7 . . . b 1 �? 1 8 Black cannot reply 1 8. . . 'ID<g7) 18 . . �c3 .

ltb8+. After 1 3 . . . 'it>c4? both 1 4 g6 b2 1 5 g7 (threatening 1 9 . . . rlxg7) 1 9 l:tg3+ 'it>c2 20

l:td8 16 l:tg4+ and 14 'it>a5 b2 1 5 g6 'it>c3 1 6 .l:.g2+ cJi;b3 21 1:!g 1 ! 1;; a2 22 1;; e6, and the
'it>a4 are possible [this last move is a mistake king succeeds in joining up with the g l­
in view of 1 6.. .'i:J.d8 1 7 g7 :aB+ 1 8 cJi;b5 l:tg8; pawn.
a draw is given by 16 g7 .l:!.d8 17 .l:!.g3+ 'it>c4 B) 13 . . . �c4 14 g6 (in the event of 14 'it>a5 b2
18 .l:!.g4+ 'it>c5 19 .l:!.g5+ 'it>c6 20 l:tg6+ 'it>c 7 1 5 g6 the simplest win is by 1 5 . . . .l:!.a3+!, but
2 1 .l:!.g1 - Dvoretsky.] 15 . . . .l:!.g3 16 .l:!.xg3 b 1"ik is also possible in -

However, Black finds a subtle solution: view of the unfortunate position of the white
13 . . . �d7 ! ! 1 4 g6 �g7 . I n this way the king) 1 4. ..1:l..d8! 1 5 g7 l:!a8+! (we already
mobil ity of the wh ite king is restricted - now know how a draw is gained after 1 5. . . l:!gB?)
it can neither approach the b-pawn , nor 1 6 'it>b7 ( 1 6 'it>b6 b2 is no better) 1 6. . . 11g8
move to the kingside. After 1 5 .l:!.g5+ 'it>b4 1 6 17 'it>c6 b2 18 'it>d6 'fJ.xg7! (with the king on
'it>b6 ( 1 6 .l:!.g4+ 'it>a3 1 7 'it>b5 b2) 1 6 . . . b2 1 7 b4 this move would not be possible), and
.l:!.b5+ 'it>c3 1 8 .l:!. c5+ (in the hope of d riving Black wins.
the king to b 1 and retu rn ing to g5) B lack C) 1 3. . . b2 1 4 .l:!.b 1 .l:!.d2! 1 5 g6 1:!g2 16 gl
replies 18 . . . 'it>d4! 19 l:tb5 .l:!.xg6+ and wins. r!g6+!, and on the next move Black will
[In fact this way to win from the last diagram capture on g7 either with check, or with a
is not the only one: in all the alternative threat of mate - Dvoretsky.]
variations Black's play can be improved.
A) 13 . . . 'it>b4 ? 14 g6 :dB! (instead of 1 4. . . b2) Thus our i n itial concl usion has been re­
15 g7 versed : the plan of moving to the help of our
pawns proves to be stronger than the march
(see diagram) of the king to the g4-pawn .
 tb 97

Mark Dvoretsky

Adventu res o n Adj o u rn ment Day

I always l ucky. From the examples demon­

t is said that winner of the first prize is his position. After three hours of exception­
ally i ntensive work my head was l iterally
strated below, you will see that i n the U S S R spl ittin g , but on the other hand it appeared
C u p (the cl ub tea m championship o f the that we had found a way to save the game.
country) in 1 976 in Tbilisi our Burevestn ik
team was indeed l ucky. But when you have Tal - Smyslov
by no means the most impressive l i ne-u p ,
Tbilisi 1 976
competitive good fortu ne alone is not enough
to win by an enormous marg i n (before the
last round we were a l ready T'h points ahead
of our nea rest rivals). Our success was
largely secu red by the friendly atmosphere
reigning in our team , the benevolence and
mutual help. An important role was also
played by our su periority i n the a nalysis of
adjourned positions (although from the
exam ples g iven below you would probably
not say this) - when they were resumed the
results of a good dozen games came as a
pleasant surprise to us.
On the even ing before the adjournment day
our leader Vasily S myslov adjou rned his
Of cou rse, it is u nfavou rable to play 42 g5?
game against Mikhail Tal in what was a
dangerous position for h i m . At a tea m .l::i.x h5 43 tt::l f6+ �g6 , and therefore Tal's
meeting he said that the fol lowing day he sealed move was obvious.
would need help with the analysis. 42 tt::l x d6 cxd6
' O f course, of cou rse, let's look at the Now the captu re of the b6-pawn allows
position together, ' grandmaster Ta i m anov Black to activate his king : 43 .l::i.x b6 �h6! 44
offered his services. .l:!.xd6+ (44 f4 ! ? exf4 45 .l:!.xd6+ also does not
'Thank you , Mark Evgenevich , but I would wi n ) 44 . . . �g5 45 f3 e4 ! with counterplay
like to work with Mark lzrai levich , ' S myslov sufficient for a d raw, for exa mple: 46 .l:!.g6+
replied . 'it>f4 47 fxe4 a4.
Of course , it is flattering to have the Wh ite must play more sharply.
reputation of being a good ana lyst, a lthough 43 .l::i.x d6!
at times it is slig htly onerous - after all, my We have a choice between 43 . . . b5 44 .l::!. a 6
game was also adjourned . But the fol lowing a4 and 43 . . . .l::i. b 1 . In both cases the opponent
morning Smyslov and I sat down to look at sends his king forward . The queenside
98 � Adventures on Adjourn ment Day

pawns do not advance very qu ickly - d u ring 44 �a6 a4

this time danger impends over the black 45 'it>e3
king. The following variation is an instructive We considered 45 'it>f3 to be more accurate.
one, illustrating the typical ideas in the Tal was concerned about the reply 45 a3, . . .

position and the difficulties facing Black. but in this case Black loses: 46 c.t>g2 .Ua 1 47
43 .. J�tb1 44 'it>f3! (44 g5? .l:tg 1 is prematu re) g5 b4 48 g6+! (not 48 �a7+? Wg8 49 g6 Iic1
44 . . . a4 (44 . . . .l:tb4? 45 'it>g3 and 46 g5) 45 50 h6 l:!.c8) 48 . . . Wh6 49 .l:!.a7 with the
'it>e4 a3 46 .l:td7+ 'it>h6 (46 . . . 'it>g8 47 .l:ta7 decisive th reats of 50 l:th7+ or 50 g7 'it>h7 51
.l:tb3 48 'it>f5 or 48 g5 is completely bad for h6 and 52 l'la8 . Therefore Smyslov would
Black) 47 \t>f5! (th reatening 48 g5+ \t>xh5 49 have played as in the main variation of the
.l:!.h7 mate) 47 . . . .l:!.g 1 48 .l:!.a7 .l:tg2 49 f4 ! exf4 analysis - 45 . . Jlc1 ! . But after the move in
50 .l:!.xa3 .l:!.g3 5 1 .l:!.a 1 ! f3 (not 5 1 . . .llc3 52 the game Black acq u i res an add itional
g5+) 52 'lt>f4 �g2 (52 . . . �h3 53 �b1 f2 54 possibil ity.
IDb6+ 'it>h7 55 �b1 �b3 56 �f1 'it>h6 57 45 . . . lle1 +
.Uxf2 also does not help) 53 'it>xf3 �c2 54
4 6 'it>f3
.Ub 1 , and Black is short of the one tempo
which wou ld enable him to block the pawns
securely and set up a familiar d rawn position
with king on g5 and rook on c5.
Even in such sharp endings, where every­
thing can depend on a single tempo,
someti mes it is not worth immed iately
delving into the mass of variations. You
should first logically weigh up the situation
and look for a plan, a general idea , which
should be carried out. What, do you think, is
this idea?
It turns out that Black should switch his rook
to the 8th ra nk. Fi rstly, from here it covers
the king - to checkmate it the opponent will
have to bring forward his king and both 46 . . . �c1
pawns, and this demands time. Secondly, it
46 . . . e4+ 4 7 'it>f4 �e2 was also q u ite possi­
may be possible to place the rook behind its ble. We analysed sharp variations such as
own pawn and advance it, while giving u p 48 'it>g5 �xf2 49 �a7+ 'it>g8 50 h6 e3 51 'it>g6
the other. �f8 and did not see how Wh ite could win.
So, the general plan has been fou n d . It is But on the other hand the main plan of
merely necessary to firid the most accu rate defence a lso seemed sufficient for a d raw,
way of implementing it. so that it was not easy for Smyslov to make
43 . . . b5 a choice . He real ised perfectly wel l that, in
As Tal commented after the game, 43 . . . .Ub1 view of the lack of time for analysis, in any
was nevertheless possible, but only in branch a mistake could creep i n . The only
connection with the plan indicated above : q uestion was, where was this more prob­
after 44 'it>f3 a4 45 'it>e4 Black should play able?
45 . . . .l:tb4+ ! 46 'it>f5 .l:!.f4+ 4 7 'it>g5 .Uf8 . 47 c.t>e4 �c4+?!
 Adventures on Adjournment Day C2J 99

This was how we intended to switch the rook is 9 Wd5! (but not 9 �c5 ? .U.g8!) 9...l:!.b4 10
to the 8th rank. True, in this way the position �e5 (because of zugzwang Black is forced
of the wh ite king is improved , although the to allow the king into the lower half of the
e5-pawn remains invul nerable. We rejected board) 10 ... l:!.b5+ 1 1 �f4 l:!.b4+ 12 �g3 f!.b5
47 . J:!.c8 ! , because we considered that
. . 13 '>t>g4 .i:i.b3 14 �h4 with a decisive
Black was lost in the position arising after 48 zugzwang.]
'it>xe5 .l:!.b8 49 g5 b4 50 .U.a7+ 'it>g8 51 �a4 To the grandmaster's question, which plan
b3 52 l1a 1 b2 53 l:!. b 1 . of defence it would be better to choose, in
Not long before the resumption Vasily reply I merely shrugged my shoulders.
Vasilievich came up to me. Without waiting for advice, he said that he
'You know, ' he said , 'it would appear that would th i n k about it once more at the boa rd .
White's th ree pawns do not win . ' And he made h i s choice in favour of the
main variation, which we had plan ned from
'That can't b e s o ! ' I said in su rprise , a n d I
the very start. Alas, it was here that a
tried to refute his conclusion , but without
mistake had crept i n .
success. Here is the key position .
4 8 'it>f5 .l:!f4+
49 'lt>g5 l:i.f8

1 l:tb3+ 2 'it>d4 ( 2 'it>d2 .U.b4 3 f5 .i:i.b5)

. . .

2 J:.b4+ 3 Wc3 l:txf4 4 l:txb2 .U.h4! with a

...

draw. 50 h6!
The discovery made by S myslov is instruc­ Here it all became clear to S myslov. I n o u r
tive and pretty, and I th ink that it is of analysis w e h a d somewhere given a check
considerable importance for the theory of on a?, after which there is no win . In sharp
rook endings. But we no longer had time to endings such as this, every tempo is
verify it thoroughly. precious - Wh ite leaves the king on h7, in
order to advance his pawn to g6 with check.
{Many years later grandmaster Carsten
Muller nevertheless found a winning plan for 50 . . . b4
White. He suggested 2 '>t>e4 l:!.b4+ 3 �f5 51 l:txa4?
l:!.b5+ 4 '>t>e6 (4 �g4 l:i.b4 is hopeless) An unexpected amnesty at the very last
4 1:!b6+ 5 �d5 l:!.b5+ 6 �c6 l:!.b4 7 f5 tlg4 8
. . . moment. Wh ite could have won by 51 �h5!
h6+ rtlh7 9 l:!.xb2 l:i.xg5 10 l:!.f2 etc. If Black b3 52 g5 l:tb8 53 g6+ 'it>h8 54 h7 'it>g7
waits: 7 . .U.b8 8 h6+ �h 7, the most accurate
. . (54 . . . b2 55 'it>h6) 55 l:ta7+ Wf6 56 g7.
1 00 � Adventures on Adjournment Day

51 . . . .l::tb8 43 tt:lb3
5 2.l:.a7+ If 43 g6 I was intending 43 . . . hxg5 44 hxg6
It is now pointless to play 52 �h5 b3 53 g5 .l::txa 1 45 .ih6 .l::t a 7! 46 f7 .l:.xf7 47 gxf7 b3. 1t
b2 54 g6+ 'it>h8!. later transpired that after 48 tt:lxe5 b2 49
52 .. . 'it>h8 tt:lc4! b 1 1i' 50 tt:lxd6 Wh ite does not lose, for
exa m ple, 50 . . .1i'a2+ 5 1 'it>g3 ( 5 1 'it>h3?
53l:ta2 b3
'iie 6+ and 52 . . .'iVxh6) 5 1 . . . 'iVf2+ 52 'it> h 3 (52
54.l:.b2 e4
'it>g4? tt:le3+) 52 . . .'ii' f3+ 53 'it>h4 'ii'f6+ 54
55 'it>f4 'it>h7 'it>h5. Apart from 45 .ih6, also possible is 45
Draw. tt:lh6 l:ta7 46 g7+ .Uxg7 47 fxg7+ 'it>xg7 48
tt:lf5+ 'it>g6 (48 . . . 'it>f6 49 tt:lxd6 b3 50 tt:le4+
That same day I too resumed by game (also 'it>f5 5 1 .ib4 b2 52 tt:ld2) 49 tt:lxd6 b3 50 lt:Jc4
a sharp endgame with passed pawns for b2 5 1 tt:lxb2 tt:lxb2 52 'it>f3 tt:lxd3 53 'it>e4 or
both sides). It was adjourned before Smys­ 52 . . . 'it>f5 53 'it>e2 .
lov's game, and so I had managed to look at
43 ... l:ta3
it, although , obviously, I no longer had time
to check the variations. 44 g6 hx g6
45 hx g6 .itS
V . Kozlov- D voretsky 46 tt:lx e5!
Tbilisi 1 976
The simplest way to d raw. 46 tt:lc1 b3 47
tt:lxb3 :txb3 48 .ih6 tt:le3+ 49 .ixe3 dxe3 50
'it>f3 was also possible. I merely wanted to
check whether my opponent might mix up
the move order by choosing 46 .ih6?. In this
case after 46 . . . tt:le3+ 47 .ixe3 (47 tt:lxe3
.ixh6 48 tt:lf5 .if8 49 g7+ .ixg7 50 fxg7+
�h7 5 1 tt:lc5 b3) 47 . . . dxe3 48 tt:lc 1 Black
does not play 48 . . . b3? , but 48 . . . .U.c3!.
46. . . .l:tx b3

Analysis showed that, amazi ngly enoug h ,

the position was a forced d raw.
41l:tx d1!
The sealed move .
41 . . . tt:lx d1
4 2 f6 .l:.a8
After 42 . . . tt:le3+ 43 .ixe3 dxe3 44 'it>f3 lla8
45 tt:lb3 .Ua3 46 tt:lc 1 b3 47 tt:lxb3 llxb3 48
g6 .if8 49 tt:lxe5 Black is u nable to convert
his extra rook. 47 tt:lf7+
 Adventures on Adjournment Day ltJ 101

I was expecting 4 7 i.. h 6 �b2+ 4 8 'it h 3 (after look at the position, to avoid any bad
48 'it>g3 there is the u npleasant reply oversig ht.
4 8 .. �d6!, and if 49 lt::l g 4, then 49 . . . �g2+! 50
.
47 . . . 'it>g8
�xg2 i.xf4) 48 . . . I:i.f2 49 i.. xf8 .l:.xf6 50 .iL.xb4 48 lt::lh6+ i..xh6??
with a drawn endgame. I n my analysis the
48 . . . <i£th8 was essential, with a d raw. The
move made by Wh ite i n the game was not
move i n the game should have lost after the
even made on the board , since I thought that
i nterposition of 49 f7+!.
after 47 . . .<i£tg8 48 lt::l h 6+ the exchange on h6
49 .iL.xh6?? �b2+
followed by playing the rook to f2 would
retain Black's extra paw n . And when my 50 � g3 l:tf2
opponent nevertheless went i n for th is Now it is Black who wins.
variation , I instantly (a typical mistake!) 51 f7+ Ilxf7
made the moves that I had planned before­ 52 gxf7+ <i£txf7
hand.
53 i..c 1 <i£te6!
You always have to reckon with the possibil­
54 'it f3 lt::lc3!
ity of 'holes' i n you r p reparatory a nalysis -
55 'it12 b3
after all , not all its details will have been
worked out with identical thorough ness. Wh ite resigned .
Perhaps there was no point i n again
checki ng all the previously prepa red varia­ As you ca n see , tou rnament fortune was
tions, but at least I should have taken a fresh i ndeed on our side!
1 02 �
Artur Yusupov

Solo for a Kn ight

o r what one horse power is capable of

There are many horses trained to help their riders so as

to run upon any one that appears with a drawn sword...
M ichel Montaigne

I\ t the end of the 1 6th centu ry the French 46 . . . lb d6!

F"'\> h ilosopher was probably taken on As Florian Gheorg h i u informed me after the
trust. I, on my own experience, have been game, he overlooked this knight move in his
fortunate enough to satisfy myself of the adjourn ment analysis. 46 . . . 'it>f5 would not
u nusual capabil ities of a knight. have g iven anything because of 4 7 'it>d3,
while if 46 . . .lbf6 there would have followed
Gheorghiu - Yusupov 47 f3!.
Luzern 1 985 47 'it>d3
Black's task would have been more compli­
cated after 47 f3 . It would a ppear that
47 . . . gxf3+ 48 '>t>xf3 lLlf5 49 'it>xf4 lbxd4
th rows away the wi n , since Wh ite activates
hi s bishop: 50 .tf2 lbxb3 51 .tb6 lbd2 52
.txa5 b3 53 i.c3 lbc4 54 g4 b2 55 i.xb2
lbxb2 56 a5, or 51 . . . lbc1 52 'it>e3! lba2 53
.txa5 b3 54 Wd2 h5 55 .tc3!. I nteresting
play results if i n stead of 54 . . . h5 Black plays
54 . . . d4!? 55 g4 'it>d5. In reply 56 'it>d3 !?
lbc 1 + 57 'it>d2 comes i nto consideration, but
Wh ite can also go in for a sharp va riation
suggested later by Mark Dvoretsky: 56 g5
'it>e4 57 h5 d3 58 i.c3! ttJxc3 59 'it>xc3 'it>e3
60 g6 d2 6 1 g7 d 1 'if 62 g8iV 'it'c2+ 63 'it>b4
The Romanian player went i n for this
b2 64 'ifg3+, and the king will hardly be able
position , erroneously assuming that he
to avoid perpetual check.
would be able to construct an impregnable
The correct conti n uation is 47 . . . lbf5! 48 fxg4
fortress.
ttJxd4+ 49 'it>d3 lLlf3 50 .tf2 lbe5+ 5 1 'it>e2 (if
45... f4!
51 'it>d2 there follows 5 1 . . . ttJxg4 52 .tb6
46 'it>e2 fxg3 ) 51 . . . f3+ 52 'it>f1 (similar variations
If 46 gxf4 , then 46 . . .<�:'ld6! 47 f3 (or 47 i.d2 occur after 52 'it>d2 lbxg4 53 .tb6 'it>e5 54
lbf5 48 h5 'it>f6 49 .te1 'it>g7 ) 4 7 . . . gxf3 48 .txa5 'it>e4 55 .txb4 f2 56 'it>e2 d4)
'it>e3 lLlf5+ 49 '>t>xf3 lbxd4+ 50 'it>g4 lbxb3, 52 . . . lbxg4 53 .tb6 'it>e5 54 .txa5 'it>e4 55
and Black wins. i.b6 (or 55 .txb4 '>t>e3 56 i.e 1 f2 57 .txf2+
 Solo for a Knight ctJ 1 03

I'Llxf2 58 a5 d4 59 a6 d3 60 a? d2 6 1 a8'ii head start: it req u i res j u st th ree more

d1'i'+ 62 �g2 'i!Vh 1 mate) 55 . . . d4 56 a5 f2! moves, whereas the black knight can reach
57 'it>g2 (57 a6 'it>f3 58 ii.xd4 lDh2 mate) the aS-sq uare only i n fou r moves. D isap­
57 . d3 58 a6 d2 59 a? f 1 'iV+ , and Wh ite
. . poi ntment awaits Black if he tries to q ueen
loses. his own paw n: 53 . . ltJf5?
. 54 a6 g3 55 a? g2
47... lbts 56 a8'ilt' g 1 'ii 57 'ilt'g8+. H owever, as we
Now Wh ite is in zugzwan g , and he h i mself is know, a well-trained horse is capable of
forced to break u p h is fortress. u n usual feats . . .
48 hS 53 ... lDhS!!

48 �d2 would also not have saved Wh ite i n 54�e3

view of 48 . . .fxg3 49 fxg 3ltJxg3 50 �f4ltJf5! 54 a6 ltJf4+ 55 �e3 lbe6 56 a? CDc? and
51 !Ji.c7 g 3 . wins.
4 8... fx g3 54... lbf6
49 fx g3 �f6 55�4 �x h6
50 h6 56 a6 ltJd7
White's last hope is the vul nerable placing of 57 a7 lb b6
the black pawns on the queenside. Thus the 58�x g4 �g6
careless 50 . . ltJxh6??
. is answered by 5 1
59�4 �
ixb4!.
60�g4 ltJa8
50... <;;t>g6!
Wh ite resigned .
Not so convincing is 50 . . . �e6 5 1 �f2 lDxh6
52 �e3 with chances of a d raw. When he
Yusupov - Li Zunian
made this move, Black had to calculate the
Luzern 1 985
variation which occu rred i n the game.
51 �d2 ltJx g3
52 .tx b4 ax b4
52 . ltJe4? 53 �xa5 g3 54 �c7 .
. .

53 aS

In this position the game was adjourned for

the second time. Although d u ring the fi rst
adjourn ment session I managed to win a
pawn thanks to the enthusiasm of the wh ite
knight, which accompl ished an heroic raid i n
In the race to q u een , th e white a-pawn has a the enemy rear - ltJg4-f6-g8xh6-g8-e7-
1 04 � Solo for a Knight

c6-d4-e2--g3 - a d raw still seemed to me to d raw after 64 . . . 'it>d5 65 'it>f4 Wc4 66 "2lf5
be the most probable outcome. After a brief 'it>d3 .
analysis it transpired that apart from the 6 4... 'it>d6
exchange of the g-pawn there was no other The more accu rate 64 . . . �b7 will be ana­
realistic plan of playing for a win . After this lysed later.
White is left with a single target - the e4-
65 tt::l g6 'it>d5
pawn . The impression was that Black could
66 'it>f4
fairly easily solve the problem of its defence.
However, serious work on the position W h ite's plan takes shape. The winning idea
inspired hope, and I began to real ise that the is to occu py the key e5-square with the
last wh ite piece , standing modestly at g 3 , knight. F rom there it not only covers the d3-
was truly a ' Montaig nian' knig ht. and c4-squares, but also aims for d7 or f7.
59... 'it>e5 66... 'it>c5
60 'it>h5 f6 In the event of the natural 66 . . . Wc4 White
61 g5 fx g5 would have won by 67 tt::l e 5+! 'it>c3 68 "2ld7!.
62 'it>x g5
The threat is 69 tt::l c 5, after 68 . . . �c4 there is
the fork 69 tt::l b 6+ , while if 68 . . . 'it>b4, then
Black is at the crossroads , since the bishop
simply 69 tt::lf6 .
can defend the pawn from various sides. For
67 tt:Je5! �b7
a long time the plan chosen by the Chi nese
player also seemed the strongest to me. Or 67 . . . 'it> b4 68 tt::ld 7 , and Black loses.
62... �c6 68 tt::lf7!
Since there is no satisfactory defence
against the threat of 69 tt::l g 5 (if 68 . . . 'it>c4 69
tt::ld 6+ ) , Black resigned .

Let us return to the position after 64 ttJe7.

I nstead of 64 . . . 'it>d6 Black had the more
cunning 64 . . . �b7. If now 65 tt::l g 6+ 'it>d5 66
�f4 , then 66 . . . �c4 67 tt::le 5+ 'it>c3 .

Black keeps his bishop on the b7-a8

squares, and when his king is evicted from
e5 it aims for d 3 .
63 tt:Jts �as
Of cou rse, not 63 . . . 'it>d5 because of 64
tt::le 7+ .
64 tt::le7!
Wh ite must prevent the passage of the black This is a position of mutual zugzwang. With
king to d3. For example, 64 tt::l h 6? leads to a Black to move he would be lost: 68 . . . �d2 69
 Solo for a Knight lZJ 1 05

ti:ld7, 68 . . . i.. a 8 69 lZ'ld7 '01tc4 70 lZ'lb6+, or Here Wh ite wins irrespective of the tu rn to
68 .. .'01tb4 69 tt:J f7 '01tc3 70 lZ'ld6 . But it is move.
White to move and he is u nable to win : 68 1 lZ'lg 6+ '01td5 2 'it>f4 i.f1
ti:lf7 'it>d3 , or 68 lZ'ld7 '01tc4 69 'iti>e5 i.c6 . If 2 . . . i.. e 2, then 3 lZ'lf8! i.. f3 4 lZ'lh7 '01tc4
The natural 66 '01tf4? was a m i stake; Wh ite (lZ'lf6+ was th reatened ) 5 lZ'lg5 .
can win by 66 '01tf5!! '01tc4 (66 . . . i.. c 8+ 67 '01tf4 3 tt:Je7+ '01te6
ib 7 68 lZ'le5 i.a8 69 lZ'ld7, and Black has no 3 . . . 'it>d6 loses i m med iately in view of 4 lZ'lg8!
defence) 67 lZ'le5+ '01tc3 68 '01tf4 , and the i.. g 2 5 lZ'lf6 .
situation analysed by us is reached , but with 4lZ'lc8!!
Black to move.
This at fi rst sight ridiculous move becomes
It remains to clarify what would have u nderstandable, if the goal of the knig ht's
happened if Black had carried out hi s plan u n usual route is noticed - the c3-sq uare .
more accu rately, i.e. obta ined the position i n 4... i.d 3
the last b u t o n e diagram with his bishop on
Other continuations also fai l to save Black:
aB (with the bishop on b7 White wins by 1
a ) 4 . . . i.g2 5 lZ'lb6 '01td6 6 lZ'la4 '01td5 7 lZ'lc3 + ;
ti:lf5, and if 1 . . . '01td5 , then 2 '01tf4 with the
irresistible threat of 3 lZ'lg3 ) . In this case 1 b ) 4 . . . '01td7 5 lZ'lb6+ '01tc6 6 lZ'la4 and 7 lZ'lc3 ;
ti:lf1! leads to a wi n . After 1 . . . 'it>d5 there c) 4 . . . i.. b 5 5 '01txe4 '01td7 6 lZ'lb6+ 'it>c6 7 lZ'ld5;
follows 2 lZ'ld2 'it>e5 (3 '01tf4 was threatened ) d ) 4 . . . 'it>d5 5 lZ'lb6+ '01tc5 6 lZ'le7+ 'it>d6 7 lZ'lf6 .
3lLlc4+ '01te6 (3 . . . '01td5 4 lZ'lb6+) 4 'it>f4 and 5 5lZ'lb6 i.. c 2 (otherwise lZ'lb6-a4-c3) 6lZ'lc4
ti:ld2, while if 1 . . . i.. b 7 - 2 tZ'lh2 '01td5 3 lZ'lg4 Although Wh ite has not i n fact managed to
'it>c4 4 lZ'le5+ 'iti>c3 5 '01tf4 , and aga i n a transfer hi s knight to c3, his achievements
familiar position is reached (cf. the last a re very considerable: the black bishop has
diagram). been forced onto the b 1 -h 7 diagonal , where
Thus the system of defence with the bishop it is less well placed .
on b7-a8 ru ns i n to a far from obvious 6...'it>d5 7lZ'ld2 i.d 3
refutation .

Black could have stuck to another line of

defence with his bishop on g2-h 1 . We will
examine the fol lowing important positions.

8 'it>f5
Black is in zugzwang and is forced to allow
the knight to go to f1 (8 . . . i.e2 9 lZ'lxe4 i.d3
1 0 f3) .
1 06 � Solo for a Knight

8...�c2 9 tL'lf1 �d1 If 9 t2Jg4, then 9 . . . �e6 .

Or 9 . . . �c4 1 0 tL'lg3 �d5 1 1 tL'lh 5 and wins. 9 �c4 10 tL'lf1! �d5 ( 1 0 . . . 'it>d3 1 1 tL'lg3 ) 11
•.•

10 tL'lh2 �c2 1 1 t2Jg4 'it>c4 12 tL'lf6�d3 13 tL'lg3 followed b y tL'lh5, tra nsposing into a
t2Jx e4�e2 14�f4, and White wins. win n i ng position which is a l ready familiar to
Here I should make a slight digression and us.
refer the reader to the start of th is i nteresting
endgame, where the author praises the
'enthusiasm of the white knight' . In order to
eliminate the last bulwark of Black's defence
- the e4-pawn , the wh ite knight had to
complete a veritable round-the-world jour­
ney (f4--g6--e 7-c8-b!H;4-d2-f1 -h2--g4-f6-
e4).

I n the position from the last but one d iagra m

it could have been Black t o move.
1...Sl.f1 2 tL'lg6+ �d5 3 �4 Sl.g2 4 tL'lh4!
Sl.f1
If 4 . . . .th 1 , then 5 �g3 ! �c4 6 �h2 , forcing
Although the bishop is at g2, a l l the same
the exchange of bishop for knight. Wh ite is able to win .
5 tL'lf5 Sl.g2 6 t2Jg3
1 tL'lf5!
Noth ing is g iven by 1 tL'lh5 .tf3 ! (but not
1 . . . .t h 1 2 t2Jf4 .tf3 because of 3 tL'lg6+ '.t>d5
4 �f4 .th 1 5 tL'lh4 �c4 6 �g3 �d3 7 '.t>h2
�e2 8 �xh 1 �xf2 9 t2Jf5) 2 t2Jf4 .t h 1 ! .
1...�d5
Or 1 . . . .tf3 2 tL'lh4 .td 1 3 tL'lg6+ �d5 4 'ii'f5
Sl.f3 5 �f4 , similar to the main variation.
1 . . . Sl.h 1 2 tL'lh4 ! is bad for Black.
2 tL'lh4! .tf1 3 �f5! �e2 4 tL'lg6 Sl.f3 5 �f4
.tg2
Black a lso loses after 5 . . . .te2 6 tL'le7+ �e6
7 t2Jf5 �d5 8 tL'lg3 .tf3 (the position of
mutual zugzwang from the last but one
A very important position of mutual zug­ d iagra m has been reached ) 9 tL'lf1 etc.
zwang . If it is Black to move he is forced to 6 tL'lh4 Sl.f1 7 t2Jf5 .tg2 8 tL'lg3
occupy the f3-square with his bishop, Agai n a fam i l i a r mutual zugzwang position.
allowing the knight to go to f1 . Wh ite wins.
6...Sl.f3 7 tL'lf1 Sl.d1 It is more d ifficult to win when Black moves
7 . . . Sl.g2 is bad because of 8 tL'ld2 ! Sl.h 1 9 first:
�g3. 1...Sl.h3! (preventing the i mportant manoeu­
8 tL'lh2 Sl.c2 9�f5 vre t2Jg3-f5-h4) 2 �h4! (White tries to give
 Solo for a Knight C2J 1 07

his opponent the move) 2 .i.c8... 5 lt:Jh5 .i.f3

Or 2 . . . .i.g2 3 'it>g4 .i.f3+ 4 'it>g5 .i.g2, If 5 . . . � h 3 , then 6 lt:Jf4 is possible.
reaching the position from the last diagra m 6 lt:Jf4 .i. h 1 7 'it>g5 ..tf3
with Wh ite t o move. This position has a l ready occu rred i n the
3 'it>h5! .i.d7 notes. I will remind you of the win n i ng
If 3 . .i.h3, then 4 'it>g5 , and White has
. . method : 8 lt:Jg6+ 'it>d5 9 'it>f4 .i.h 1 1 0 lt:Jh4
succeeded i n giving his opponent the move . 'it>c4 1 1 'it>g3 'it>d3 1 2 'it>h2 'it>e2 1 3 'it>xh 1
4 i.c8 5 lt:Jh5 .i. h 3 6 lt:Jf4 followed by lt:Jg6+
. . . 'it>xf2 1 4 lt:Jf5 .
and r.t>f4 transposes i nto variations analysed With this the a uthor concludes his analysis
earlier. of this interesti ng endgame. I will be very
4 'it>g6! .i.g4 indebted to readers for any corrections,
Or 4 . �c8 5 lt:Jh5 Wd5 6 tt:Jf6 +.
. .
refi nements or refutations.
1 08 �
Mark Dvoretsky

More about the 'Montaign i a n' Kn ight

W Yusupov
hen I saw t h e analyses b y Artu r A knight can stop a rook's pawn, if it
in the previous chapter I succeeds in 'touching' any square in its
was reminded of several stud ies on the path (apart from the corner square h1). In
same theme , which i n their time made a the g iven i n stance it is clear that the knight
strong impression on me. I hope that they will aim for the h2-sq uare . H ow to reach it­
will also appeal to you . Their beauty l ies in seeing as the black king stands in its way?
the unusual amount of work carried out by Here are some logical considerations, which
the wh ite knight, the paradoxical manoeu­ will make it easier to fi nd the solutio n . The
vres of the wh ite pieces, and above all the knight can reach h2 via g4 or f1 . Each of
precision and depth of logic behind these these routes can be control led sepa rately by
manoeuvres. the black ki n g . It is necessary to create a
Before enjoying the solutions, try to fi nd the 'double attack' - by th reatening to go to h2
answers you rself. I should warn you before­ by both ways . The knight ca n reach f1 via
hand: the problems are very difficult, and d 2 , and g4 via e5. Do you see the
you will most probably have to move the i ntersection point of these two routes?
pieces on the board (perhaps only the 1 tt'lb4! h5
second example might be solved i n you r
2 tt'lc6!
head ). But even so, don't b e i n a h u rry to
make moves - first reason to you rself about Not 2 tt'ld5+? 'itof3 ! , and the pawn can not be
the fi nal and i ntermed iate aims of both stopped . Note that a king restricts a
knight most effectively when there is one
sides, the plans they will carry out, any
square between them along a diagonal,
important intermediate positions, and so o n .
or two squares between them along a
N. Grigoriev rank or a file.

1 932 2... 'it>e4!

Of cou rse , not 2 . . . h4? 3 tt'le5 , when the g4-
square can be covered only by 3 . . . Wf4 , but
then there follows 4 tt'lg6 + .
3 tt'la5!!
Only in th is way can the knight reach the key
c4-sq uare , from where it ca n ai m for both f1,
and g4 . 3 tt'ld8? would have lost after 3 . . . h4
4 tt'le6 'it>f5 ! 5 tt'ld4+ 'ltog4 .
3... h4
4 tt'lc4!
4 tt'lb3? �e3 .
4. . . 'itof3! ?
 More about the 'Montaignian' Knight lLJ 1 09

The last tra p . 4 . . . h3 5 lZ'ld2+ and 6 lZ'lf1 leads king: 1 ..t>b3(a3) 'it>f7 2 'it>b4 'it>g7 3 'it>xb5
to an immed iate d raw. 'it>xh7 4 'it>c4 , and the king enters the sq uare
5 lDe5+! of the h-paw n . What, then , is the problem?
5 liJd2+? (hoping for 5 . . . <;i;Jg2? 6 ltJc4 ! h3 7 It turns out that Black can save a very
ltJe3+) is a mistake because of 5 . . . 'ite2 ! 6 i mportant tempo by avoiding the attack on
1De4 h3 7 ltJg3+ 'it>f2 , and B lack wins. I n this the knight and satisfying h imself with merely
variatio n he is able to deflect the knight from restricting its mobil ity: 1 .. .<.t?e6 ! 2 'it>b4 'itf5 3
its route to h 2 , a n d , as you can see , 'it>xb5 h 5 , and the pawn queens. This means
'touchi n g ' the h 1 -sq u a re with the knight that i n reply to 1 . . . 'it>e6 Wh ite must i m medi­
does not help Wh ite . ately tackle the h-pawn with his knight.
5. . . 'it>g3 In a p ractical game without much hesitation
Forced: the g4-sq u a re m ust be guarded , but many players would play 1 'it>b3 'it>e6 2lZ'!f8+
if 5 . . 'it>f4 , then 6 ltJg6 + .
.
'it>f5 3 ltJd7 h 5 , and only now th i n k about
where next to d i rect the knight. The pa ra­
6lZ'lc4! h3
doxical featu re of the position is that here
7lZ'le3! such a generally-accepted way of acting
White has achieved h i s a i m : If 7 . . . h2 8 lZ'lf1 + , does not work - it is necessary to th i n k
while after 7 . . . 'itf3 there follows 8 lZ'lf1 <;i;Jf2 9 earlier!
1Dh2 cJi>g2 1 0 ltJg4 <;i;Jg3 1 1 ltJe3 etc. 1 'it>a3!! <;i;Je6!
In concl usion I should mention that 2 lZ'lc2+? 2lZ'lf8+! 'it>f5
(instead of 2 lZ'lc6+ ! ) would be j ustified after
3lZ'ld7 h5
2.. .'it>e4? 3 lZ'la3 ! 'itd3 (3 . . . h4 4 ltJc4 ) 4 'it>b5!
and lbc4 , but 2 .. . '�f2 ! leads to a wi n . 4 ltJc5 h4
5lZ'lb3!!

D . Gurgenidze
This is why it is important to calculate all the
variations as early as the fi rst move - the b3-
1 970
sq ua re m u st be left free for the wh ite knight.
5 lZ'ld3? woul d have lost after 5 . . . h 3 6 lZ'lf2
h2 7 <;i;Jb4 'it>f4 8 'it>xb5 <;i;Jf3 .
5... h3
6lZ'ld2 h2
6 . . . 'it>f4 7 ltJf1 leads to a fa miliar d rawn
position .
7 lZ'lf1 h1�
8lZ'lg3+

(see diagram)

1 ltJg1
I n contrast to the previous position , here the
fi rst move can be made without th i n ki n g . But
This study is a n a rtistic adaptation of a what n ext - how to d rive away the e nemy
position by N i kolai Dmitrievich Grigoriev. king? For a start we at least need some idea .
The threat of wi n n i ng the k n ight is easily Let's see where the k n ight should a i m for, i n
parried by approach ing the b5-pawn with the order to create d ifficu lties for the opponent.
1 10 � More about the 'Montaignian' Knight

R. Reti, A. Mandler The knight's circular journey has enabled

1 924 White to achieve h i s a i m - he has given the
opponent the move . The rest is al ready
familiar to us.
8 ... 'it>e3
9 tt'le1 'it>d2
10 tt'lc2! 'it>d1
11 tt'lb4! 'it>d2
12 tt'ld5

R. Reti, A. Mandler

1 924

We find the square c2 . From here the knight

takes away the important e3-square , and it
is itself invulnerable i n view of the reply 'i!te2 .
The black king has to be at d2 or d 1 . But with
the king on d1 Wh ite has the decisive tt'lb4!
followed by tt'ld5. It is clear that here we
beg in to have mutual zugzwang positions.
Black has to play accu rately from the very
start. 1 . . . 'it>d3? is bad in view of 2 tt'lf3 'i!te3 3
tt'le 1 ! Wd 2 4 tt'lc2 ! Wd 1 ( 4 . . . 'i!td 3 5 'it>e 1 !
'>t>xc2 6 �e2 ) 5 tt'lb4! Wd2 6 tt'ld5.
1... 'iit>d2! Here Wh ite's task is even more d ifficult than
2 tt'lf3+ �d3! in the previous study. The winning plan
Now it is pointless playing 3 tt'le1 + 'it>e3 ! 4 suggested there (wh ich in itself was not at all
tt'lc2+ '1t>d2 5 tt'lb4 'it>e3 6 tt'ld5+ 'it>e4 7 tt'lf6+ obvious) does not work here: the left edge of
'ite3. To wi n , the opponent must be given the board i nterferes. After a l l , the knight has
the move. But how can this be ach ieved? no sq uare equ ivalent to the i m portant b4-
3 'iit>e 1! 'it>e3 sq ua re in the previous example.
4 tt'le5 'it>e4 True, a new possibil ity has a ppea red - the
In the event of 4 . . . 'iit>d 4 Wh ite wins by 5 tt'lg4! activation of the king along the route d 1 -e 1 -
'it>d3 6 'it>d 1 , breaking forward with the king, f1 -g2-f3 . It is obvious that Wh ite h a s no
since 6 . . .f3 fails to 7 tt'le5 + . other win n i ng pla n . Black will try to prevent it
by attach ing h i s king to the e2-pawn from d2
5 tt'lc4! '.t>d3
or d 1 . It is easy to imagine that here too
5 . . .f3 6 tt'ld2+, or 5 . . . 'it>d4 6 'it>e2 ! . mutual zugzwa ng positions ca n n ot be
6 tt'ld2 'it>e3 avoided . Let's try to work th i ngs out!
7 tt'lf3 'it>d3 Let's suppose that Wh ite's knight is on d3,
8 'it>f1! his king on f1 , and the black king on c2.
 More about the 'Montaignian' Knight ttJ 111

Then Black loses after 1 . . . 'it>d2? 2 lt:Jf4 (2 A s you remember, 4 . . . �b3? 5 �c1 ! �xa2 6
tt'le5 �d 1 3 lt:Jf3 is also good ) 2 . . . 'it>d 1 3 'it>c2 is bad for Black. He is forced to move
Wg2 . He must play 1 . . . 'it>d 1 ! 2 lt:Jf4 (2 lt:Je5 his king along the 1 st ran k , away from the
lt>d2 3 lt:Jf3+ �c3) 2 . . . 'it>d2 ! , and if 3 'it>g2 , c3- a n d c4-squa res.
then 3 . . . d 3 ! with a d raw. 4... �b1
From this the fol lowing conclusions can be 5 lt:Jb4 'it>b2
drawn : if Wh ite plays We 1 , then with the 6lt:Jd5 'it>b3
knight on d3 Black must reply . . . 'it>c2 ! , while
6 . . . 'it>b 1 ? loses i m mediately to 7 lt:Jc7 �b2 8
with the knight o n f4 , e5 or c5 the correct
lt:Jb5.
reply is . . . 'it>c3 ! . The position with the knight
7lt:Jc7!
on c5, black king on c3 and wh ite king on e 1
is one of mutual zugzwan g . The shortest route to e4 is via f6 . H owever,
the d i rect 7 tt:Jf6? is refuted by 7 . . . �c4 ! ! 8
The followi ng step i n our logical a nalysis of
'it>c2 d 3 +! 9 exd3+ �d4 and 1 O . . . e2. There­
the position is to clarify the importance of the
fore the knight chooses a more i ntricate
e4-square for the knight. Let us su ppose
route : d 5-c7-b5-d6-e4 .
that the knight stands on e4 , the wh ite king
on d 1 , and the black king on b3. Then if it is 7... �c3
White to move 1 �c1 ! is decisive. But things Here 7 . . . 'it>c4 8 'it>c2 is now pointless.
are no easier for Black if it is him to move : if 8lLlb5+ �c4
1...Wb2 there follows 2 tt:Jc5 ! 'it>c3 (2 . . . � b 1 3 9lt:Jd6+ 'it>c3
tt'le6 ) 3 �e 1 ! , and the afore-mentioned
9 . . . �d5 1 0 lt:Jf7 .
position of mutual zugzwa ng is reached with
10 lt:Je4+ �b2
Black to move .
1 0 . . . <,tJb3 1 1 'it>c1 .
Thus the knight m ust be brought to the e4-
square. This is not at all easy to ach ieve , 11lt:Jc5! 'it>c3
seeing as Wh ite consta ntly has to watch out 12 'it>e1!
for . . . d4-d3 . Wh ite has ach ieved his a i m - he has set u p
1lLle1 �b2 t h e requ i red position o f mutual zugzwang
with his opponent to move .
2lt:Jd3+ �c3
12 . .. 'it>c2(c4)
Noth ing is changed by 2 . . . � b 1 3 tt:Jc1 �b2 4
tt'la2 , while 2 . . . �b3 shortens the solution : 3 13 lt:Jd3
tt'lf4 Wb2 (3 . . . 'lt>c3 4 �e 1 ! 'it>c2 5 lt:Jd3 , and Aga i n zugzwan g !
Black is i n zugzwa n g ) 4 tt:Jd5 �b3 5 lt:Jc7 ! 1 3. . . �c3
etc.- cf. the main variation . 14 �f1 'it>d2
3 lt:J c1! 15 lt:Jf4
But not immed iately 3 lt:Jf4 because of The fi n a l , decisive zugzwa n g . 1 5 lt:Je5 �d 1
3 . . Wb3 ! , and if 4 lt:Jd5 there is the reply
. 1 6 lt:Jf3 is equally good .
4 ...'.t'c4 ! . 15 .. . 'it>d1
3... � b2 16 � g2
4 lt:J a2! Thanks to the lengthy knight manoeuvre ,
A manoeuvre , found in the solving of the t he wh ite king has fi nally gai ned t h e opportu­
previous study, also comes in usefu l here . n ity to break free.
1 12 �

PART II

Technique

Mark Dvoretsky

Converti ng a n Advantage

C ments . One of the most common and

hess players suffer from many ail­ time and energy on solvi ng the problems
wh ich confront h i m i n the fi rst half of a game.
serious is poor tech n ique i n the conversion But for its later part he often l acks strength,
of an advantage. Even champions some­ and here he makes serious errors . It was
times suffer from th is ill ness - it is sufficient only for this reaso n , for example, that he did
to remember the 1990 match for the world not win his Candidates semi-final match in
championship between Garry Kasparov and 1989 agai nst Karpov. Yusu pov constantly
Anatoly Karpov. outplayed his form idable opponent, but was
How frequently after an unsuccessfu l game u nable to convert this i nto wins on account
do we state with vexation to o ur opponent, of tired ness towards the end of a game.
trainer, or a casual spectator: 'I had a Here is one of the most a n noying examples.
completely won position ! ' But it is poi ntless
complaining about fate - it is better to th i n k Karpov- Yusupov
about the causes o f t h e mistakes made, a n d Candidates Match , 6th Game,
endeavour t o understand what playing o r London 1989
personal deficiencies are behind you r er­
rors. I will now dwell briefly on the main
factors which h i nder the normal conversion
of an existi ng adva ntage .

I. Tiredness to wards the end o f a game

It is clear that after several hours of
intensive struggle a player becomes tired .
But some become more tired than others. It
is in the last few min utes that the fate of a
game is often decided , and therefore many
additional poi nts ca n be gai ned by a player
who retains a sufficient reserve of energy
towa rds the end of a rou nd .
Th e play of grandmaster Yusupov is always
deep and i nteresti n g , and he spends much Yusu pov saw that i n the variation 38 . . a3? .
 Converting an Advantage ttJ 1 13

39 .l:la6 .Uf2+ 40 'it>e1 a2 41 f5 he would not the board . But it is not everyone's nervous
have time to win the rook: 4 1 . . . .Uxh2 42 f6 system that is ready for such prolonged
l:ih 1 +? 43 'it>e2 a 1 'i!V 44 .l::!.x a 1 .l::!.x a 1 45 f7. tension . Often a player com poses h imself
Therefore he played 41 . 'it>d7 , and after 42
.. only at especially important moments of the
f6 'it>e6 43 .l:ta8 ! 'i.t>xd6 44 f7 .l::!.x f7 45 .U.xa2 strugg l e , but when the main problems seem
lt>e5 46 lla6 the players ag reed a d raw. to h i m to be resolved , he loses his vig ilance
Yet Black could simply have captu red the and beg i n s acting ca relessly. It is here that
pawn: m istakes usually occu r.
38 . . . l:txh2! Mestel - L. Popov
39 .l:.a6 Olympiad , Malta 1 980
39 f5 .Uf2+ and 40 . . . .l:txf5.
39 . . . l:.f2+
40 'it>e1 .l:txf4
41 'i.t>e2 .U.e4
With an easy wi n , for example: 42 .l:ta5
(otherwise . . . h6-h5-h4) 42 . . . '1t>d7 43 l:.d5
h5! 44 IIxh 5 'it>xd6.
Why didn't Artur play this? By h i s own
admissio n , at that moment his mind had
simply switched off, and he did not see any
other possibilities apart from 38 . . . a3?.
If you are let down by ti red ness, perhaps it
means that all is not well with you r physical
preparation? The prescri ption i n such cases
Wh ite is a sound pawn to the good , but he
is clear - you m ust do more physical
now has to resolve a d ifficult problem: what
exercise and devote more time to sport, i n
position to go i n for, so that the opponent will
particular exercises for stamina (for exam­
have the least i n the way of counter­
ple, slow but long-dista nce ru n n i n g ) . Con­
chances. The following possibil ities suggest
sider devising a rational daily reg ime d u ring
themselves:
a com petitio n , enabling you to relax p roperly
a ) 25 .l:te 1 .l:I.d2 (25 . . .'it'c5 is less accu rate i n
and regain you r energy before a new game.
view o f 26 'iWa6 ! .U d 2 27 .l:.e2 );
Finally, also d u ring a game you ca n husband
your strength , by using for relaxation those b ) 25 'i�Nxc6 .l:.xd 1 + 26 .i.xd 1 'iVxa2 (stronger
brief min utes of respite when it is you r than 26 . . . 'i�Nxe5 27 'ifa6! ) 27 'ifb5 'iVa 1 28
opponent's turn t o move . H owever, a l l these �e2 a 5 , and it is not easy for Wh ite to
are fai rly serious questions, demanding a strengthen his position ;
special d iscussion , and not just a brief c ) 25 .Uxd8+ 'i!Vxd8 26 'ii'xc6 'i�Nd2 ;
mention . d ) 25 .l:I.xd8+ 'i!Vxd8 2 6 �xa7 'iVd2 .
I n every case Black retains cou nterplay, and
II. Insufficiently stable nervous system the outcome remains unclear.
Throughout a game it is very important to Jonathan Mestel foun d an excellent solu­
maintain ful l concentration and u n broken tion .
attention to everything that is happening on 25 .l:.xd8+! 'i�Nxd8
1 14 � Converting an Advantage

26 i.c4! ! a8'6'+ 'it>g? 40 'ii'a 6 "ii'x c3 Black gains q uite

The bishop will secu rely defend the king good saving chances - at any event, a
from f1 . For the moment the queen remains lengthy struggle stil l l ies ahead.
on b?, from where it defends the b2-paw n . 34 'ii'b ?! was stro n g , ensuring the advance
The a?- and c6-pawns are weak, and soon of the pawn while reta i n i ng the option, in
White is sure to create a passed pawn on case of necessity, of defending the bishop
the queenside. from a6. 34 g 3 ! also came i nto con sidera­
26 . . . �d2 tio n , after which Black does not have a
27 i.f1 'lt>g7 single sensible move .
27 .. .'ti'c2 was more tenacious, preventing 34 'it'b6 h4
the following strong move by the opponent. 35 a7?!
28 a4! aS Was it worth al lowi ng the open ing up of
29 'ir'b6 h5 Wh ite's own king position? 35 g 3 ! would
have g iven a n elementary w i n .
Black's last fai nt hope is to break u p the
wh ite king's defences by the advance of his 35 . . . g3!
g- and h-pawns. He no longer has any other 36 ii'a6?
possibilities. Here it is - relaxation when only one step
30 'ii'x a5 i¥xb2 away from victory! 36 fxg3 hxg3 37 h3 would
31 'ir'b4 li'a1 have won , for example, 37 .. .'i!Va2 38 'lt>h1
tt'lh4 39 ii'a6 'ii'f2 40 a8'ii' tt'lf3 41 'i!V8a7 c5
32 a5 g5
42 'i!Vxf? + ! 'i£txf7 43 'ii' b ?+ and 44 'i!Vxf3.
33 a6 g4
H owever, such a variation is too compli­
cated for a player i n time-trouble. In any
case , with more accu rate preceding play the
game could have been concluded far more
simply.
36 . . . gxf2+
37 'i£txf2
I n the event of 37 Wh 1 'ii' c 1 38 a8'it'? White,
paradoxical ly, even loses - after 38 . . . tt'lg3+!
39 hxg3 hxg3 there is no defence against
mate by the queen on h6.
37 . . . i¥xc3
Aga i n mate is th reatened .
38 i.d3 'i!Vd2+
It is clear that Black's position is completely 39 i.e2 'iid 4+
hopeless. But it is very dangerous, in
40 We1 i¥c3+
believing th is, to weaken you r attention and
41 Wf2
stop checki ng variations. For example, if
Wh ite plays 34 'ii'b 8 'ii'c 1 35 a?? (35 'ii'b 6! is D raw.
correct), then after 35 .. .'!"i'le3 ! 36 fxe3 'ii'x e3+ How can the nervous system be tra i ned to
37 'it>h 1 "it'c1 (e1 ) 38 'it'h8+ (38 'ii'b 6 �xf1 + endure prolonged tension? Here too it is
39 �g 1 li'a6 with equal ity) 38 . . . 'it>xh8 39 probably not possible to get by without
 Converting an Advantage ttJ 1 15

physical preparation ('healthy in body, healthy An awful move, the fact that White was very
in mind ! ' ) ; self-tra in in g exercises or even short of time is, to my mind, as little to be
yoga are probably usefu l . Specific chess considered as an excuse, as for instance the
training is also possible. You can p ractice statement of the law-breaker that he was
playing specially selected exercises, i n drunk at the moment that he committed the
which you have t o find a long series o f the crime. The inability of an experienced
only correct moves. Try conducting indi­ master to deal with the clock should be
vidual games or even entire tournaments considered as grave a fault as a miscalcula­
with the aim of concentrating to the maxi­ tion.
mum throughout the entire game. Wh ite would h ave won by 47 i..x a4 .U.Xb 1 48
'.i?xb 1 '>i'xd5 , and now, if there is nothing
Ill. Time-trou ble better, 49 �e8 ! ? We5 50 h5 Wf4 (50 . . . gxh5
Nearly every player can remember depress­ 5 1 �xh5) 5 1 hxg6 hxg6 52 i..d 7 (Aiekhine).
ing instances of time-trouble adventu res, in 47 . . . 'it>xd5
which the fru its of all the p reced ing work 48 i.. c 4+ '>i'd6
were ru ined. But even so I will show you one
49 l:!.xb7 tbc5+
more example, together with a n i nstructive
50 We3 tbxb7
assessment of his own actions, which was
given in his notes by an ex-world champion . Soon the players ag reed a d raw.
Aga i n I will not speak in deta i l about how to
Alekh ine - Tylor combat time-trouble. I will merely mention
Notti ngham 1 936 two main methods: 1) 'anti-time-trouble'
games; 2) time-study of games, with a
subsequent analysis of the reasons for
getting into time-trouble.

Poi nts are lost not only in you r own time­

trouble, but also in the opponent's. This
happens because a player often disregard s
a well-known principle o f h o w t o act i n such
s ituation s . If you have the better position,
never play on the opponent's time­
trouble. A ct exactly as you normally
would, not even remembering about
your opponent's lack of time. Why? By
playing q u i ckly, so as not to allow the
Black's position is, of cou rse, completely
opponent to th i n k over his moves, you
hopeless. If 46 . . . '>i'xd5 there follows 47
essentially d rive you rself i nto the same time­
::l.d1 + '>i'e6 (47 . . . '.i?e5 48 i..c 6) 48 '>i'c3 . Tylor
trouble as h i m . But in a d ifficult situation the
tries his last ti me-trouble chance .
opponent is fu lly composed and mobilised ,
46 . . . tbxa4 whereas you , by contrast, l u l led by your
47 '>i'd3?? advantage i n time and positi o n , a re awa iting
Regarding this Aiekhine writes i n the tou rna­ the fal l of his flag and a re u nable to play at
ment book: fu ll i ntensity.
116 � Converting an Advantage

I n the hope of exploiti ng this psycholog ical 1 6 h3

effect, in d ifficult situations some players
Before placing his rook on c1 , Tseitl in wants
deliberately get themselves i nto time-trou­
to d rive away the knight. I n the event of the
ble, and there they often change the
i mmed iate 1 6 l:.ac1 h e was concerned
unfavourable cou rse of the play.
about the reply 1 6 . . .'ilt'f5 , when Black ac­
q u i res tactica l ideas associated with the
Mark Tseitlin - Makarychev wea kness of the f2-point. The fol lowing
Krasnoya rsk 1 98 1 combinative variation is i nteresti n g : 1 7 e4!
tt::lx e4 1 8 tt::lxe4 'ifxe4 1 9 tt::ld 4 (the bishop at
c4 is u nder attack) 1 9 . . .'iVe5 20 ..tf4 'ifh5 21
h 3 tt::lf6 ( 2 1 . . . tt::lxf2 is worse: 22 �xf2 e5 23
g4! 'iih 4+ 24 ..tg3 �f6+ 25 �g 1 ) 22 g4!
tt::lx g4 23 hxg4 'ifxg4 24 'it'g3 'ii'x g3 25 .bg3
..txa2 (25 . . . ..ta6 26 b4 .l:.ad8 27 tt::l b 3) 26
..txb7 .l:1ad8 27 tt::l c6 . The concl uding posi­
tion is undoubted ly i n W h ite's favou r, but the
outcome is stil l not clear - too many pawns
h ave been excha nged .
16 . . . tt::l xf2 ! ?
T h i s p iece sacrifice is Black's best practical
chance . If 1 6 . . . i.xc3 the simple 1 7 hxg4
..tg7 1 8 l:tac1 would h ave followed .
Only 1 4 moves h ave been made, but Black's 1 7 �xf2 i.xc3
position is difficult, and i n addition h e had
1 8 bxc3 tt::le4+
already spent nearly all the time on his clock
- he had just 6 ( ! ) min utes left for 26 moves.
Obviously it is important for W h ite to
develop h i s rook at c1 as soon as possible,
in order to create pressu re on the c-file.
However, after the immediate 1 5 l:tac1 h e
h a s to reckon with 1 5 . . . h 6 .
1 5 h 3 ! ? looks strong , and after t h e retreat of
the knight - 1 6 l:tac1 . Black would probably
h ave repl ied 1 5 . . . h6, i ntending to create
compl ications after 1 6 ..tf4 cxd4 ! . But W h ite
can very well waste a tempo: 1 6 ..ltc 1 ( 1 6
hxg4 hxg5 1 7 l:lac1 i s also not bad ) 1 6 . . . tt::lf6
1 7 ..lte3 cxd4 1 8 tt::l x d4 followed by .l:!.ac1 . An
unhurried method of playing, when you
simply strengthen your position without 1 9 �g1 ?
allowing any counter-chances, is the T h i s is what Sergey M a karychev had to say:
most unpleasant for an opponent who is Such a disdainful attitude to one's own
in time-trouble. material can be explained only by the
1 5 dxc5 tt::l x c5 opponent's time-trouble. In the event of 19
 Converting an Advantage lZJ 117

<3te1 l'Dxg3 (19. . . 'iic7 20 il..f4 e5 21 l'Dxe5!

�e5 22 Jl..xe4 favours White) White would
have had some difficulty in consolidating his
position, but Black would not have full
compensation for the piece. Perhaps Tseitlin
preferred- at any price- attack rather than
defence, reckoning that the only significant
factor was the difference in the clock times?!

19 . . . Jl..x e2
20 il..f4
If 20 .l:Id5, then 20 . . . e6 (20 . . . 'iic6! ? ) 2 1 .l:Ie5
ixf3 22 Jl..xf3 ti:'Jxg5 23 l:l.xg5 f5 , and the
white rook is out of play.
20 . . . il..x d1 24 'ili'd4??
21 l:l.xd 1 'ikxc3 Symptomatic: despite the opponent's time­
21 . 'iic 5+ ! ? 22 1Wxc5 ti:'Jxc5 was perhaps
. .
trouble (or more precisely - because of it), it
sounder, i ntending . . . Uac8 , . . . f7-f6 and is Wh ite who com mits a decisive oversig ht.
. . . e6-e5 with a n excellent endgame for It was essential to play 24 �h2l'Dc3 25 'it'd4
Black. (otherwise 25 . . . .l:Ixe5) 25 . . .'iit'x d4 26 .l:Ixd4.
Makarychev g ives the fol lowing variation:
22 'i!Va4 e5! ?
26 . . . g 5 27 Jl..x g5! (27l'Dg4 gxf4 28L'Df6+ �f8
22...'i'c5+ 23 �h2 l'Dc3 24 'it'c2 leads to a 29 ti:'Jxe8 fxg3+ 30 �xg3? L'De2+ is bad for
sharp middlegame, in which Black's chances Wh ite , but 27 il..d 2 ! ? .l:txe5 28 Jl..x c3 is safe
are not worse, but White gains the opportu­
for h i m ) 27 . . . l:1xe5 28 il..f6 .l:Ic5 29 �d7 with
nity to launch an attack on the king, which in q u ite good d rawi ng chances for Wh ite .
time-trouble seemed unpleasant to me. The H owever, he also has the rig ht to lay cla i m to
move in the game forces exchanges, and
more , by playing 29 .l:Ig4 + ! �f8 30 l:tg7 , for
maintains for a time the strong position of example, 30 . . . .l:Iac8 3 1 �xh7 �e8 32 h4!
the knight on e4 (Makarychev) . As you see,
ti:'Jd5 33 l:l.h8+ 'it'd? 34 il.. h 3+ 'it>d6 35 lixc8
in time-trouble a n experienced player some­ ti:'Jxf6 36 Uxc5 �xc5 37 g4 with a serious
times manages both to calculate variations, advantage . Black's play can be improved on
and to assess the position correctly. the 28th move , by placing his rook not on c5,
23l'Dxe5 but more actively on e3. And instead of
23 i.xe5? 'it'e3 + ! (or 2 3 . . . l:!.xe5 24 L'Dxe5 26 . . . g5 it probably makes sense for him to
'le3+!) is bad for Wh ite, while 2 3 'it'xe4 exf4 choose 26 . . .L'De2 ! ? .
24 'i'xf4 'ili'e3+ 25 1Wxe3 l:txe3 26 .l:Id7 l:l.c8 24 . . . Uad8!
27 l:l.xb 7 l:l.c1 + 28 �f2 .l:.a3 leads to a n 25 Jl..e 3 Uxd4
unclear endgame.
26 il..x d4 'ifc2
23 . . . Wc5+
27 Ua1 l:l.xe5!
Wh ite resigned.
(see diagram)
It is curious that on all the moves that we
have see n , Makarychev spent just th ree
m i n utes - one half of his reserve of time.
1 18 Converting an Advantage

IV. Inadequate knowledge of

endgame theory
In the majority of cases the conversion of a n
advantage h a s t o be carried o u t i n the
endgame. It is clear that, if you are not
familiar with theory, you are much more
likely to make a mistake.

Wolff - Browne
USA Championsh i p , Durango 1992

retreat to the edge of the board (and yet he

coul d have kept it at f6) , and the h-pawn for
ever remains backward .
52 'it>e4 �f6
53 l:ta7+ 'it>g8
54 g4!
At last!
54 . . . �c3
Why g ive up the e6-pawn without a fight?
54 . . . h6 55 .l:tb7 �f8 suggests itself.
It is qu ite possible (although not i nevitable)
55 l:!.e7 i..f6
that Black will lose his e-pawn , and therefore
it is useful to have some i nformation about 56 l:lxe6 'it>f7
endings with two pawns agai nst two on the 57 Ua6
same wing. The most important concl usion
is this: by placing his pawns on h5 and g6,
Black sets u p an impreg nable fortress - the
opponent's king is u n able to approach his
pawns.
(see diagram)

It obviously makes sense for Wh ite to

prevent the erection of th is defensive sys­
tem , by playing g3-g4 ! But if it is Black to
move, he should play . . . h7-h 5 ! . But, alas,
neither player knew this position .
50 'it>g2?! �d4? !
51 �f3?! g6?
A strange move, wh ich sign ificantly worsens 57 . . . �c3?
Black's position . His king is now forced to Black should have placed his bishop on h4
 Converting an Advantage lZJ 1 19

and played . . . h 7-h6 . If Wh ite's pawn were

at h3, there would be a ltogether noth ing that
he could do ( I once fou n d this fortress i n the
course of a joint analysis with Boris G u l ko of
one of his adjourned positions). With the
pawn on h2 it is possible to march the king to
h3 followed by 'it>g3 and h2-h4 . H owever,
this plan is not easy to carry out, and also i n
reply to h2-h4 Wh ite has t o reckon with the
strong reply . . . h6-h5!
58 .l:i.a7+ 'it>g8
59 l:Id7?
If during the game Patrick Wolff had known
about the system of defence with the bishop Another i m portant theoretical position! Black
on h4 , he would u ndoubted ly have played loses, if his king will be shut in the corner
59 h4! followed by 60 h 5 . (with a l ig ht-sq uare bishop, by contrast, it
59 . . . .if6 would be a d raw). As is not d ifficult to see,
60 'it>f4 .ib2? fleeing from the dangerous corner does not
work in view of the u nfortunate position of
60 . . h6! .
the bishop: 73 . . . �f8 74 �f6 'it>g8 (74 . . . �e8
.

61 �c7?! 75 'it>e6 ) 75 l:!.g7+ �h8 (75 . . . �f8 76 l:!.d7) 76

61 h4! . Wg6 and wins .
61 . . . .if6 69 h6 �
62 g5 .id4 70 .Uc4 i.e5
63 h4 .ib2 71 �3 i.d6
64 �g4 .ie5?! 72 .l::!.c 8 'it>e6
64 . . .i.a3 ! 65 h 5 gxh5+ 66 �xh 5 .ib4 was
.
73 .t!.h8! �5
more tenacious. For many years this posi­
74 l:!.xh7 �xg5
tion was considered d rawn , but i n 1 99 3 the
75 lid7
chess composer Noam El kies nevertheless
found a wi n n i ng plan . Black resigned .
65 l:tc6! .ii. b 2 I n the second issue of the American Chess
Journal grandmaster Wolff gave an excel­
66 l:!.a6 .ii. c 3
lent commenta ry on this endgame. By
67 .t!.a4! .ie5
studying his analyses, you , for example, will
68 h5! .ic3 learn how Wh ite wins if he remains with a
No better is 68 . . . gxh5+ 69 �xh5 i.d6 (the pawn on h5 or g5 against a black pawn on
threat was 70 .Ua8+ 'it>g7 71 l:!.a7+ 'it>g8 72 h 7 . All this is very i nteresting and usefu l , but
�h6) 70 .Ua8+ �g7 (70 . . . i.f8 71 g6) 7 1 nevertheless not essenti al . Whereas the
.!:!.a?+ Wg8 7 2 g 6 hxg6+ 7 3 'it>xg6. fortress with black pawns on g6 and h5
should defi n itely enter you r store of end­
(see diagram) game knowledge.
Why i n particular this position? F irstly, here
it is sufficient to learn the assessment of the
1 20 � Converting an Advantage

position and the basic idea of the defence extra pawns , which , of cou rse , is a lso good
(not to allow the king to approach the enough to w i n . Why then should his decision
pawns) - you don't have to remember any be criticised? Wel l , because after 78 'it>g5!
complicated variations. Second ly, this as­ the game would have concluded - theory,
sessment (draw ! ) is automatically trans­ wel l known to Larse n , would have begun.
ferred to positions with a wh ite h- or g-pawn Here it would no longer be possible to go
agai nst a pawn on g6 (after a l l , Wh ite ca n wrong. But after the move chosen by him,
play g3-g4 and captu re on g4 with the pawn play conti n ues i n a n u nfam i l i ar position ,
or a piece). And above a l l , this position is the which means that t h e probabil ity of a
most u n iversal and i nformative . Very often mistake remains.
the pawns of both sides have not advanced 78 . . . �d8
fu rther than the 2nd or 3rd ran k , and then it
79 .l:!.c6 c.t>d7
is clear that Black should aim to play
80 .l:!.d6+ 'it>e7
. . . h7(h6)-h 5 ! , and Wh ite - g2(g3)-g4 ! .
81 f6+?
One of the methods of converting a n And here is the decisive mistake , which
advantage is to transpose i nto a n endgame leads to a d raw. Wh ite should have played
position that is known to be theoretically either 81 lle6+ Wf7 82 c6 , or 81 .Ud5 .
won . 81 . . . �f7
82 c6 'l.t>g6
Larsen - Torre
83 'i.t>f3 l:.e1 !
I nterzonal Tou rnament, Leningrad 1 973
This is the whole point - the king cannot
break through to either of its pawns .
84 �f4 .Ue2 85 .l:t d5 .l:!. c 2 86 .l:!. d6 .l:!.e2 87 f7
'i.t>xf7 88 c.t>f5 c.t>e7 89 .l::!.d 7+ 'it>e8 90 'iiif6
.Ue1 91 .l:!.d5 .l:!.c1 92 .l:!.d6 .l:!.f1 + 93 �e6
l!e1 + 94 �d5 .l:!.d 1 + 95 c.t>c5 .l:!.xd6 96 �xd6
'i.t>d8 D raw.

V. Poor tec h n ique in the conversion

of an advantage
We will consider this problem in more detail.
G randmaster Igor Zaitsev once suggested a
deep and valid idea : Tech nique is the art of
the past' . If this is so, then a rel iable way of
The simplest way to win is to sacrifice the
improving you r tech nique is to study classic
c5-pawn. After 78 'i.t>g5 ! .Uxc5 79 'it>g6
models, a n d , above a l l , examples from the
followed by 80 l:!.h8+ we reach an elemen­
games of players who were famed for their
tary theoretical ending, which is completely
mastery in this field . Players, such as Akiba
hopeless for Black, since his king is on the
Rubinste i n , Jose Raul Capablanca , Alexan­
'long' side of the pawn .
der Alekh i n e , Tig ra n Petrosi a n , U lf Anders­
78 .U.c7?! son . . . When playing through their games,
Bent Larsen decided to keep both of his th i n k about why they were able so easily and
 Converting an Advantage t2J 121

naturally to exploit even a seemingly imper­ nent has a bad dark-sq uare bishop, and as a
ceptible advantage, and you will g rad ually conseq uence - weaknesses on the light
beg in to notice their approach to such squares. How can I strengthen my position?
situations, those principles of converting an The plan of playing the knight from d2 to d5
advantage which they consciously or sub­ suggests itself: f2-f3 , i.e3-f2 and tt:'ld2-f1 -
consciously followed , and the tech niq ues e3. It would seem that it ca n be beg u n with
wh ich they employed . We will now examine either 22 f3 , or 22 tt:'lf1 . It also makes sense
the most general of these principles and to play 22 g 3 , depriving the enemy knight of
methods. the f4-square . Wh ich of these th ree conti nu-
ations is the most accu rate?
Maximum restriction o f t h e opponent's Wh ite m ust carefu lly look to see what active
cou nterchances resou rces the opponent has, and how he is
It is very important for every player to intending to play. The move 22 . . . tt:'lf4 should
possess 'prophylactic thinking' - the not con cern us too much - after 23 g3 tt:'le6
ability to constantly ask yourself: 'What the knight does not create any th reats from
does my opponent want, and what would e6 and does not control the weak d5-sq uare,
he do now if it were his move?' But the for which Wh ite is a i m i n g .
role of prophylactic thin king i ncreases par­ The attem pt t o create cou nterplay on the
ticularly when converting an advantage, kingside with 22 . . . 'ife6 ! followed by 23 .. .f5
when the maxi m u m restriction of the oppo­ looks more serious. For example, 22 g3
nent's possibilities, the e l i m i nation of the 'ii'e 6 2 3 'it'b3 f5 ! , and there is no time for 24
slightest cou nterplay or any usefu l opera­ 'ilt'xb7? i n view of 24 .. .f4 . Or 22 tt:'lf1 'it'e6
tions to improve his own positi o n, becomes (with g ai n of tem po ! ) and 23 . . .f5 .
probably the main principle. I n the second variation White h a s an i nteres­
I will show two examples from my own ting tactical resou rce : 23 'il:Vb3 f5 24 exf5
games . gxf5 25 i.xh 6 ! jLxh6 26 .l:!.d6 �f7 27 .l:!.xh6,
althou g h after 27 . . . tt:Jdf6 his rook is stuck in
Dvoretsky - B utnoris enemy territory, and 28 tt:'lfe3! f4 29 tt:'lf5
Kiev 1 976 'ike6 30 tt:'lh4 e4 leads to a rather tense
situati on . And in genera l , when you have
such a solid positional advantage, why
calculate such complicated variations? After
22 f3 'YWe6 23 �b3 the captu re on b7 is now
seriously threatened , and Black is forced to
d ivert either his rook, or his queen, to the
defence of the pawn . It is th is move order,
therefore , that enables Wh ite to be fu lly
prepared for the opponent's cou nterplay.
22 f3 ! tt:'lf4
23 g3 tt:Je6
24 tt:'lf1 f6? !
25 'i!tg2!
Another accu rate move . 25 i.f2? is prema-
Of cou rse, Wh ite stands better. The oppo- ture in view of 25 . . . tt:Jg5 followed by . . . 'iie 6,
122 � Converting an Advantage

when 26 h4 does not work because of 33 tt:Je7 ! i.xe7

26 . . . tt:Jxf3+ 27 'it>g2 tbd4 28 cxd4 exd4. 34 ii'f7+ 'it>h8
25 . . . 'it>h7 35 'ifxe7 'i!Vd6
26 i.f2 tt:Jc7 36 'i!Vf7 g5
27 tt:Jfe3 37 i.xc5! 'iid 2+
38 'it>h3 tt:Jd7
39 il..e 7 .l:!.g8
40 tt:Jd5! .l:!.g7
40 . . . g4+ 41 fxg4 'iig 5 42 tt:Jxf6 .
41 'iie 8+
Black resigned .

Zakharov - Dvoretsky
Ordzho n ikidze 1 978

Wh ite has carried out his plan and i ncreased

his advantage.
Usually, when w e look a t a g a m e , w e do not
pay any attention to modest moves such as
22 f3 ! and 25 'it>g2 ! . And yet it is thanks to
them that events have developed in the
desirable, qu iet way for Wh ite , and the
opponent has not gai ned the slig htest
opportun ity to activate his game or compli­
cate the play. But few are happy to defend
passively without any counterchances, and Black clearly has a great positional advan­
in such cases new mistakes or inaccuracies tage. When I ask, how he should conti nue in
are l ikely, making it easier for the stronger this position , usually either 29 . . ..l:!.c2 or
side to convert his advantage. 29 .. .f6 and 30 . . ..l:!.c4 is suggested . And
27 . . . tDb8 i ndeed , why not - seeing as Wh ite has no
28 tt:Jb6 'iie 6 cou nterplay?
29 ii'a4! tt:Jb5 But nevertheless, th i n k how you would play
30 .l:!.d5 l::tx d5 if it were Wh ite to move. And then you will
31 tt:Jexd5 fi nd a n idea which promises q u ite good
cha nces of a successfu l defence - the
Threatening 32 c4 tt:Jd4 33 'i*'xe8 ! .
ma noeuvre tt:Je5-g4-e3 with the aim of
31 . . . .l:!.d8 exchanging the mighty knight on d 5 . This
32 'ii'c4 �c6? knight dominates the position , cementing
32 . . . 'ifd6 was more tenacious. Now Wh ite together Black's q ueenside and making the
lands a decisive blow. rook at b5 a passive, inoperative piece . But
 Converting an Advantage ttJ 1 23

in the event of the knight exchange, the rook endgame tech nique, some of which we will
wil l immediately be transformed - after all, it encou nter i n the fol lowi n g examples.
is attacking the b6- and g5-pawns.
It becomes clear what the best move is. 'Do not hurry!' does not imply that you
29 . . . h5! can carelessly squander tempi. On the
Black retains all the advantages o f h i s contrary, every opportunity to gain a
position and prevents the opponent's o n l y tempo should definitely to taken into
promising idea . account and exploited.

The game did not last long .

30 �d2 f6 Leonhardt - Spielmann
3 1 tt:'lf3 l::tc 4 San Sebastia n 19 12
32 b3 l:lc6
The wh ite rook h a s ended up in a tra p .
33 h 4 g4
34 tt:'le1 tt:'lc7
35 l:!.xh5 'it>g6
Wh ite resigned .

'Do not h u rry! '

The inherent a im wh ich a player should
follow when trying to convert a n advantage
is not to win as q u ickly as possible. As yet no
one has offered p rizes for the smallest
number of moves. You should endeavo ur to
play with the utmost safety, exploiting a l l the One must possess great presence of mind,
resou rces of you r position and completely to not seize the booty immediately, but do
restricting the active possibil ities for you r this after several strong preparatory moves.
opponent. I t is n o t a misfortune i f you h ave The anticipation of victory often hinders the
to make an extra dozen moves on the board , objection evaluation of a position.
if thanks to this you make your task easier It is to this factor that I prescribe the loss of a
and reach you r goal more su rely. In a sharp half point in what was a very important game
middlegame you may be tempted by the for me - the most annoying instance in my
i mage of tiger, swiftly h u nting down its chess career. This was at the finish of the
quarry and tea ring it a part , but when San Sebastian tournament of 1 91 2, at a
converting an advantage i n the endgame moment when I had excellent chances of
you should rath er imitate a pytho n , slowly winning first prize. For me it was sufficient to
suffocating its victi m . win against Leonhardt . (Rudolf Spielmann).
. .

The pri nciple ' d o not h u rry! ' was fi rst The game went:
formulated (but not explai ned to a sufficient
46 . . . c;t> xd4?
degree) i n instructional material o n the
endgame prepared by the Soviet master 47 g6
Sergey Belavenets . I n fact, beh i n d this brief It transpires that Black is i n zugzwan g . I
formula are concealed various aspects of should mention that he re the zugzwang is
1 24 � Converting an Advantage

mutual - any move by Wh ite worsens his Reti - Romanovsky

position . Moscow 1 925
47 . . . 'it>d3
48 l:td7 d4
49 g7
After advancing his passed pawn to the 7th
rank, Wh ite easily parries all his opponent's
attempts. For example: 49 . . . '1t>c3 50 .l:!.c7+
'1t>b3 5 1 ki.d7 , or 49 . . . .l:!.g 1 + 50 Wb2 'it>e3 5 1
'it>b3 ( 5 1 �e7+ 'it>d2 5 2 ki.d7 d3 5 3 �e7 'it'd 1
54 �d7 d2 55 :I.e? is also possible) 5 1 . . . d3
52 'it'xb4 d2 53 'iit>c3 .
49 . . . �g6
50 'it>b2 Itg1
51 'it>b3
Draw. 28 .l:!.c4!
Of cou rse, Black should have played Wh ite is planning 'it'f3 and e2-e 3 . It is very
46 ... 'it>c3 ! . If 4 7 'it>d 1 , then 4 7 . . . l::.g 1 + 48 importa nt that, after the black bishop moves,
'it>e2 b3 is decisive . After 47 .l:!.c7+ 'it'xd4 the c5-pawn remains en prise. It will have to
Black captu res the pawn with gain of tempo, be defended by the rook, but then the white
thanks to the attack on g5. Since 48 .l:!.b7? bishop will occupy a n active position on d5,
.l::!.x g5 49 .l:txb4+ 'it'c3 is bad for Wh ite , he the king will obta i n the excellent square e4,
has to reply 48 .l:i.g7 . Now another tempo and the rook may be switched along the 4th
can be won by 48 .. .'it>c3 ! 49 ki.c7+ 'it>d3. ran k to the kingside.
Look at the position after 50 ki.g7 : as yet
28 . . . 'iit>f8
Wh ite has not done anyth ing usefu l , whereas
Black has eliminated the pawn and placed 29 'it>f3 lieS
his king on d 3 . 30 e3 .i.c3
A nd yet, contrary to Spiel mann's o p i ni o n, It would have been better to play the bishop
even here he apparently did no t have a win . to b2.
For example: 50 ... d4 5 1 g 6 .l:tc2+! ( 5 1 . . . b3
52 'it>b 1 !, but not 52 �b7? b2+ 53 'it'b 1 .l:txg6
54 'it'xb2 'it'd2 ) 52 'it'b1 ! (52 'it'd 1 ? .l::!.c6 53
l:tg8 'it>c3 54 g7 Itc7) 52 ... .Uc6 53 'it>b2! 'it>c4
54 l:!.g8 (54 'it>c2) 54 . . . d3 55 llg7 ! (but not
55 g7? �c7 with zugzwang), and Black is
not able to strengthen his position .

Ifthe opponent is deprived of counterplay,

before changing the pattern of the play
and initiating decisive action you should
first strengthen your position to the
maximum, by making all moves that are
even slightly useful.
 Converting an Advantage ctJ 1 25

3 1 a4! Pyotr Romanovsky tries to solve the prob­

Method ically played ! This move does not lem by tactical means, but the combination
come d i rectly i nto Wh ite's pla n , but in itself it he has prepared meets with a spectacu lar
is useful - it is clear that the pawn sta nds refutatio n .
better at a4 than at a2. The opponent will no 36 g4
longer have cou nterplay i nvolving the switch­ Of cou rse, not 36 h4? g6 and the wh ite rook
ing of his rook to a6, and if Wh ite should is tra pped .
somehow be able to capture the a 7 -pawn he 36 . . . g6
will obta i n a passed a-pawn . I t is not clear
37 .Uxh6! Wg5
whether these considerations will play any
38 l1h7 �xg4
role, but this is not important. Any opportu­
nity to strengthen the position even slig htly
should be exploited .
31 . . . �e7
32 �d5 l1c7
33 l:i.h4!
The black king is ready to go to d6,
defending the c5-pawn , and therefore there
is no longer any point in keeping the rook at
c4 . It is switched to the kingside to support
the pawn offensive on that pa rt of the board .
33 . . . h6
34 'it>e4 �f6
35 l1h5
Now it is clear what Romanovsky had i n
Now Black must do someth i n g , since g3-g4,
m i n d . I n t h e event o f 39 fxg6?? f5 or 39 �e6
h2-h4 and g4-g5+ is threatened . He should
fxe6 40 .Uxd7?? gxf5 Wh ite is u nexpectedly
probably have played 35 . . . g6! 36 fxg6 (36
mated . If 39 f6 Black was i ntending 39 . . . �g5 .
l:lxh6? ;t>g5) 36 . . .'it>xg6 37 l1f5 a5! (prevent­
H owever, after 40 �xf7! �xf6 4 1 �xg6
ing a4-a5-a6 followed by l1f5-f2-a2-a5-
.l:lxh7 42 �xh7 he most probably is u nable to
b5). The outcome would have remai ned
save the ending with opposite-colour bish­
unclear, althoug h Wh ite could have devel­
ops. Wh ite attacks and captu res the a7-
oped his i n itiative by h2-h4-h 5+ and .Uf5-
pawn (perhaps after fi rst playing his bishop
f1 -h 1 -h4-g4.
to c4 ) , after which one of the two passed
I checked a pretty attempt to prevent the pawns , a- or h-, is bound to decide the
closing-up of the queen side by 37 a5? ! . The outcome. But with the pawn on a2 such a n
captu re of the rook leads to a n immediate e n d i n g would certainly be d rawn .
m ate: 37 . . . �xh5?? 38 Wf5 . Black has a
Richard Reti found a prettier and more
d ifficult position after 37 . . . .l:td7 38 g4 .Ud6 39
convincing solution .
.tc4 . U nfortunately, there is a spectacul a r
refutation : 37 . . . c4! ! , and 38 dxc4? is not 39 �e6 ! fxe6
possible i n view of 38 . . . 'it>xh5 39 �f5 e4 40 39 . . . .l:te7 40 .l:lxf7 .l:lxf7 4 1 fxg6+ .
.txe4 .Uc5+. 40 fxg6! l1d8
35 . . . l1d7?! 41 .l:lxa7 'itg5
1 26 Converting an Advantage

42 g7 'it>h6 'Why didn't you immediately try to convert

43 a5 your material advantage?' I ventured to ask
the great chess virtuoso. My companion
The passed a-pawn comes decisively i nto
condescendingly smiled: 'It was more practi­
play. Now we can assess the true worth of
cal to wait. '
the move 3 1 a4 ! . With his pawn on a2 Wh ite
would have been unable to w i n .
43 . . . 'it>h7 Dvoretsky - Cooper

44 a6 .l:td6 P h i ladelph i a 1 990

The threat was 45 llb7 and 46 a ? .

45 h4
White's last reserve joins the battle.
45 . . . ..te1
46 h5 ..th4
47 h6
Black resigned .

In cases when one side's advantage is

insufficient for a direct win, it makes
sense to manoeuvre, without changing
the pattern of the position, in order to set
the opponent various problems, even if they Wh ite u ndoubtedly has a n appreciable
are not too complicated . And only when , positional advantage. H e fi nds a convincing
after failing to withstand the prolonged plan to exploit it, i nvolvi ng the creation of
pressu re , he blu nders or makes some error, th reats on the dark sq uares on the kingside.
can you then turn to decisive action.
1 8 ..txg7
Such tactics of setting the opponent an
1 8 ..tg5 was also not bad . H owever, 1 8
endurance test ca n sometimes also make
lt:ih2? would have been a serious i naccu­
sense in positions with a big advantage. By
racy in view of the strong reply 1 8 . . . f5 ! .
provoking h im i nto making a mistake , you
can make it sign ificantly easier to convert 18 . . . 'it>xg7
your advantage. 1 9 tt:ih2 h5
Mikhail Botvin n i k remembers: I also had to reckon with cou nterplay on the
c-fi le. If 1 9 . . . .Uc2 there was the strong reply
In 1936 in Moscow during the 3rd Interna­
20 lt:ig4 �h4 21 .Uac1 ! .UacB (2 1 . . . .l:txb2 22
tional Tournament I witnessed the resump­
lie? ..te8 23 llec1 is equally cheerless) 22
tion of the Capablanca-Ragozin game. The
.l:.xc2 .Uxc2 23 g 3 ! (23 l:l.c1 .Uxc 1 + 24 �xc1
ex-world champion had an extra pawn and
is also not bad , when the wh ite queen
hence a won endgame. To my surprise,
breaks i nto the opponent's position along
however, Capablanca did not undertake any
the c-file) 23 .. ."it'xh3 24 "ii'f6+ 'it>gB 25 "ii'd 8+
positive action, but stuck to waiting tactics.
'it>g7 26 tt:if6 .
Finally his opponent made an inaccuracy,
and the Cuban won a second pawn and 20 .Uac 1 llxc1
soon the game. 2 1 .Uxc1 llc8
 Converting an Advantage ttJ 1 27

22 l::!.x c8 ..txc8 from the fact that my opponent's pieces

23 g4! were tied down .
This is the whole poi nt! Now 2 3 . . . hxg4 24 27 'ii'a 3 aS
tt:lxg4 , and if 24 .. .'it'h4 - 25 'ii'f6+ (of cou rse , 28 1i'c5 'it>g6
it is also possible to delay this check) 29 '.t>g2
2 5 . . 'it'xf6 26 exf6+ is completely bad for
.
A useful prophylactic move, which in some
Black. He does not want to allow the variations prevents the black queen from
exchange of pawns on h5, while if 2 3 . . . 'i!Vh4 giving a check on c1 .
there follows simply 24 '.t>g2 and 25 lLlf3 .
29 . . . b6
23 . . . g5 30 �c2+
24 '*'e3 h4 30 'it'd6? does not work in view of 30 . . . ifxd6
31 exd6 f6! . But if Black should play his
bishop to a6, then by placing his queen on
d6 Wh ite will i mmed iately decide the out­
come. It makes sense to check whether the
opponent will go wrong.
30 . . . '.t>h6
31 �c6 ..ta6?
H e does! Of cou rse , 31 . . . 'it>g6 was correct.
Then I would probably have nevertheless
ag reed to win a pawn by 32 l\Vc2+ 'it>h6 33
'i'c1 �g6 34 tt:Jxg5 "ikxg5 35 'i!kxc8 , although
fi rst I would certa i n ly have pondered over
whether I had extracted everything possible
from the type of position now on the board .
After 25 f4 ! ? the h4-pawn is, of cou rse,
3 2 'it'd6 ! 'ifcB
doomed . But i n this case the wh ite king
becomes somewhat exposed , which may 33 "ilie7 'ii'g 8
give the opponent some counter-cha nces . 34 'ilff6+
For the moment I p referred not t o change Black resigned .
the pattern of the position and I tried to See how more easily (not more q u ickly, but
ach ieve success in positional manoeuvri n g , more easily) I was able to win , thanks to the
b y tying t h e black pieces t o t h e defence of fact that I did not h u rry to force matters .
the weak g5-pawn . Especially, since the
(see diagram)
possibil ity of f2-f4 will never ru n away.
25 ltJf3 'it>g6 Wh ite has the advantage. But what is it
26 'ikd3+ 'it>h6 better to play: 38 ..lkxd4 or 38 f6 ?
Here I noticed that I could win a pawn by 27 When you have a choice between advan­
'id2 (with the threat of 28 ltJxh4) 27 . . . 'it>g6 tageous positions with different material
28 'i'c2+ Wh6 29 'ir'c 1 �g6 30 ltJxg5 'i!Vxg5 balances, all other things being equal
31 'ir'xc8 . The queen endgame is al most you should choose the one in which the
certainly won , but again I did not want to material balance is the most usual, the
force matters and I tried to obta i n benefit most standard. Here you will have more
1 28 � Converting an Advantage

Dvoretsky - Bai kov it, of cou rse, but I wondered whether it

Moscow Championsh i p 1 972 wouldn't be possible to ach ieve even more.
After notici ng that 40 b3? is refuted by
40 . . . ttlxe5 , I for some reason completely
forgot about the same possibil ity in reply to
40 .l:Ie4 and I considered only 40 . .'�xc2. .

After discovering 41 �xc7+! and calculating

its consequences, this is what I played .
40 �e4?? ttlxe5
Wh ite's incorrect move could also have
been refuted i n a d ifferent way: 40 . . . ttld2 !?
41 .l:Ixa4 ttlxf3 , and after the bishop moves
Black has the decisive 42 . . . h 3 .
41 l::tx e5 �xc2
42 nds �c1 +
43 Wh2 �xb2+
experience and hence there is less
chance of a mistake in the evaluation of 44 Wh3 l::tx d5
the position or the subsequent play. 45 .ltxd5 c6
I n the event of 38 f6? �e3 39 f7 'iVxe6 40 Here the game was adjou rned . The situation
fB'iV l::txf8 41 'il*'xf8+ tLleB or 41 �xf8 ttle4 has completely cha nged - Wh ite's position
Black, with a pawn for the exchange, is absol utely hopeless, not only because of
complicates the play. Moreover, on a more the opponent's extra pawn , but also in view
careful exami nation of the resulting position of the dangerous position of the wh ite king.
it becomes clea r that the advantage has However, on the resumption I managed to
now passed to the opponent. confuse matters and save the d raw.
The simple captu re on d4 is much safer. I regarded my blu nder on move 40 as
38 �xd4 iVa4? merely an i nexplicable 'eccentricity' . But
38 . . . ttlb5! was much stronger. In reply 39 when I showed the game to ex-world
�c8? does not work in view of 39 . . . 'iVc6 , champion Tig ra n Petrosia n , he took a quite
while after 39 �e5 ttld4 40 �xd4 .l:Ixd4 d ifferent view of things.
Black has sufficient compensation for the ' H ow do you explain why you avoided
pawn deficit. Possibly Wh ite should reply 39 tra nsposing into the endgame? You were in
�d5 ! ? , when i n the event of 39 . . . �d 7 40 no doubt that it was won . But if you see an
�e3 ! �xe3 41 l::tx e3 tt:ld4 42 iVe4 ttlxc2 43 elementa ry solution , why then calculate
l::te2 or 43 l::!.d 3 he retains the advantage. variations and get i nvolved in an excha nge
However, after 39 . . . tt:lxd4 40 l::tx e8 ttlxf3 41 of blows?'
l::tx d8+ �xd8 the ending with opposite­ The moral of this sad episode is obvious.
colour bishops is probably d rawn . Always give preference to the simplest
39 �e5 ttlc4 way of converting your advantage, in
Now the simple move 40 'iVe4! would have which case the probability of a making a
forced the transition into an absol utely won mistake will be minimal. Avoid unneces­
endgame (40 . . . ttlb6 41 'ii'x a4 ) - Black's sary complications, and never play 'for
position would have been resig nable. I saw brilliancy'.
 Converting an Advantage '2J 1 29

Any 'trifles', capable of facilitating the squandered his entire advantage.

conversion of an advantage, should 41 .l:l.xd6 lt:lxd6
without fail be taken into consideration.
42 .Jtc7?!
If, for example, you do not h ave much time
Wh ite avoids the obvious 42 .l:txa4 because
left to the time control , make use of every
of the reply 42 . . . h 5 , which of cou rse,
opportun ity to repeat moves. And after the
however, does not change the evaluation of
time control has been reached , defi n itely
the positi o n . In pri nciple, the desire to fi nd
adjou rn the game, if you h ave a wi n n ing
the most accu rate way of exploiti ng your
position. If you don't do this, because of
advantage is commendable, but i n so doing
tired ness you may make a m istake and spoil
you must accu rately check the variations,
your position .
which Karpov did not do. I n cidental ly, a
This last piece of advice has ceased to be similar mistake was made in h i s commen­
topical i n view of the change i n the rules of tary by M i khail Tal , who recommended 42
chess competitions - games a re no longer .l:ld4 lt:lc8 43 .Jtc5 . I n stead of 42 . . . lt:lc8?
adjourned . The fol lowing example is never­ Black plays 42 . . . .l:l.e 1 + 43 �c2 .U.e2+ 44 �c1
theless still instructive: it shows that over­ (44 �d3 llxb2 45 l:!.xd6 .l:lxa2) 44 . . . a3! 45
confidence in success, combined with a l:txd6 .l:lxb2 , casting doubts on whether
certain haste and i naccu racy, is capable of Wh ite can wi n . So that the simple captu re of
having a negative effect on the play of even the a4-pawn is the most rel iable way to win .
such a player, deserved ly famed for his fine
42 . . . lle1 +
tech n ique, as Anatoly Karpov.
43 �c2 lt:le8
Karpov - Korchnoi Karpov simply m issed this straightforward
World Championship M atch , reply. Now, to avoid fu rther mistakes , it was
22nd Game, Baguio 1 978 essential for h i m to ask the a rbiter for an
envelope and to seal his move. But the
world champion contin ued in the same vei n .
4 4 .ta5 a3
45 llb8 Ite7
45 . . . lle2+ 46 'it>d3 l:!.xb2 d id not work in view
of 47 llxe8+ 'it>h7 48 lle2 .

White's position is absol utely won . Karpov

should have sealed h i s next move , after
which his opponent would probably not have
bothered to resume the game. But for some
reason the world champion made a few
more moves at the board , and i n the end he
1 30 � Converting an Advantage

46 i.b4?? j u st been discussing - 'do not h u rry! ' . If the

Through inertia Karpov decided that now too opponent is condemned to passivity,
the check on e2 was not da ngerous for h i m . don't try to achieve success at one point
Of cou rse, 4 6 bxa3 ( o r 46 b4 ) would have alone - to hold it the defensive resources
given an elementary win . may prove quite adequate. Play more
46 . . . .l:i. e2+ widely, and try to exploit weaknesses
(and if possible - create new ones) on
47 'it>d3?
different parts of the board - then it will
Showi ng the same inertia. 4 7 i.d2 ! axb2 48
be much more difficult to defend.
a4 was essential, stil l reta i n i ng excellent
chances of success. It is hard even to
understand what exactly Karpov overlooked , Alekh ine - Samisch
since now both captu res on b2 enable Black Baden-Baden 1 925
to save the game. Sensing this, Victor
Korch noi decided to adjourn the game at
this precise moment, so that the opponent
would not know which choice he had made.
47 . . . axb2
I n the variation 47 . . . .l::!.x b2 48 .l:i.xe8+ 'it>h7 49
.ltxa3 (49 �c3 .l:i.xa2 50 l:!.f8 f6 or 50 'JJ.e 7
'it>g8) 49 . . . .l:i.xa2 Black then plays . . . f7-f6
and . . . h6-h5 , obta i n i ng a d rawn position ­
there are too few pawns left on the board .
48 i.d2
If 48 .ltc3 , then 48 . . . b 1 'ii'+ ! 49 l:I.xb 1 l:I.xa2 50
l:I.b8 l:I.g2 51 l:I.xe8+ 'it>h7, and roughly the
same drawn situation arises as in the
H ow to convert the extra pawn? Advance it
47 . . . l:I.xb2 variation . There can follow 48
to the queening square? But B lack will set
l:I.e4 h5 49 gxh5 l:I.h2 50 l:I.g4 f6 5 1 'it>e4
up a blockade on the b6-squa re , the white
l:I.xh5, and to avoid . . . g7-g6 the wh ite rook
king will be exposed and there will be a
has to stay on the g-file.
danger of perpetual check. I should remind
48 . . . l:I.e7 you that queen and knight form a rather
49 a4 l:I.d7+ 50 'it>c2 'it>h7 51 l:I.xb2 h5! 52 dangerous duo, if they are in the vicin ity of
gxh5 tt:ld6 53 l:I.a2 tt:lxf5 54 a5 tt:ld4+ 55 the enemy king . Only in the event of the
�c3 (55 'it>b1 tt:lb3, then giving u p the knight queens being excha nged will the wh ite king
for the a-pawn ) 55 . . .tt:lc6 56 a6 l:!.d5 57 ii.f4 be able to advance fearlessly to the help of
l:I.f5 (57 . . . .l:i.xh5? 58 l:I.h2 ! ) 58 i.d6 l:I.d5 59 its passed pawn .
ii.. g 3 l:I.g5 60 i.f2 l:I.xh5 61 'it>c4 tt:la5+ 62 34 l\Yd4!
�c3 tt:lc6 63 l:I.a4 'it>g8 64 'it>c4 tt:la5+ Draw.
With this move and the next one White finds
the correct winning plan, which is to ad­
Th e pr i n ci p l e of two wea knesses vance his kingside pawns. The passed b­
This principle is essentially one of the pawn must not advance until later, once the
conseq uences of the more general ru le of danger of perpetual check has been re­
converting an advantage, which we have moved by the exchange of queens. A
 Converting an Advantage t2J 1 31

concrete and clear eval uation of the posi­ 43 'iVd3 'ii'x d3

tion , typical of Alekh ine - i n his commenta r­ 44 ..txd3 tt:'!c8
ies one can fi nd n u merous i n structive 45 il.xh7
features such as this.
Black resigned .
34 . . . "ike7
35 il.d31
Kotov - Pachman
Perhaps the most difficult move in the game.
Ven ice 1 950
Its purpose is to prepare an attack on the
point h7. The winning method which follows
leaves Black powerless to resist. (Aiekhine)
35 . . . 'ii'c 7
36 g4! �f7
37 h4 tt:'!b6
38 h5 gxh5
39 gxh5
On the kingside Black has been saddled
with a second weakness, and a very serious
one (I should remind you : the fi rst 'weak­
ness' is the opponent's passed pawn , and
Black con stantly has to reckon with the
threat of its advance). If now 39 . . . �g7 , then
40 h6+ ! . Wh ite's positional adva ntage is determ ined
39 . . . by his better pawn structu re , by the weak­
39 h6 was more tenacious.
. . .
ness of the c6-pawn . But this factor alone
would have been insufficient for a win , if
40 Jl.e4!
Black had now played 42 . . . h 5 ! .
Of cou rse, not 40 ..txh7? "ikxf3 4 1 'ilt'xb6? �f6?
42 . . .
'ld 1 + with perpetual check. If Wh ite de­
43 g4!
sired , he could now h ave exchanged the
queens and after 40 'ii'e 4 'i!i'xe4 41 ii.xe4 h6 A typical move . Wh ite fixes a second
42 'it>c2 g radually won the m i nor piece weakness i n the opponent's position - the
ending. H owever, the move in the game is h7-paw n . This was why it should have been
far stronger, since it enables h i m to fix the adva nced to h5. I should mention that
weakness on h 7 . 42 .. .f5?! was much weaker i n view of 43 h3
followed by g3-g4 , and if Black replies
40 . . . "ikb5
43 . . . h 5 , then he again acq u i res a second
41 h6 'it'b3 weakness - this time on g6.
42 ..tc2 ! 43 . . . �e6
Now that the pawn on h7 has been 43 . . . �g5 44 h3 h5 45 f4+ �h4 46 'it>g2 .
blockaded, the next step is to force the 44 'it>g2
exchange of queens. (Aiekh ine)
In the endgame you should never forget
42 . . . 'itb5 about improving the position of your
42 . 'ii'e 6 43 'i!Ve4 .
. . king.
1 32 Converting an Advantage

44 . . . l:i.b7 52 . . . �xc5
45 .l:!.e8+ 53 dxc5 rJii d 7
Before attacking the h-pawn it is useful to 54 .Uh8 'it>e6
lure the black rook to a more passive N ow 54 . . . l:i.a5 55 l:txh7+ �e6 (the king
position . can not go to f8) 56 l:lg7 is bad for Black.
45 . .
. .Ue7 Such 'trifles' play a very i m po rtant role in the
46 .Uh8 f6 conversion of an advantage.
47 h4 l:tb7 55 .:td8
48 'it>f3 llf7
49 .Ue8+ .l:i.e7
50 l:!.d8!
White wants to place his knight on c5. It is
important that after the exchange of minor
pieces the black rook should be tied to the
defence of the c6-pawn . Passivity of the
rook is a very serious drawback in rook
endings.
50 . . . .Ua7
51 lt:Jc5+ rJii e7

Alexander Kotov has successfully carried

out his plan of tra nsposing i nto a favourable
rook ending. In h is commentary he judged
the resulting endgame to be won for White.
H owever, in Jonathan Speel man's book
Endgame Preparation· this evaluation was
called i nto q uestion .
The English grandmaster's conclusion ap­
pears not be to com patible with the logi c of
the preced ing play - after a l l , White has
consistently outplayed his opponent and, it
would appear, has the right to count on
52 .Uc8 ! success. But what can be done - the
A method ical move . 52 .Uh8 was incorrect i n defensive resou rces in chess a re g reat,
view o f 5 2 . . . �xc5 5 3 dxc5 .Ua5! 5 4 .Uxh7+ especially i n rook endings, wh ich according
Wf8 . Now the c6-pawn is under attack and to Tarrasch a re not usually won . The work
Black does not manage to activate his rook. done by Kotov was not in vai n - al most out
Thus if 52 . . . .Uc7 there now follows 53 .Uh8. of noth ing he has developed a dangerous
Do you sense how uncomfortable it is to i n itiative and posed serious problems for the
simu ltaneously defend two weaknesses - opponent, which at the board the latter was
c6 and h7, and how much easier it would be u nable to solve.
to defend the weak c6-pawn alone? 55 . . . W e7?!
 Converting an Advantage lZJ 1 33

Consideration should have been g iven to

55 . . J:tc7! 56 .Ud6+ 'it>e5 57 'it>e2 g5 58 hxg5
fxg5 , and if 59 'it>d3 , then not 59 . . . .Uc8? 60
l:td7 h6 61 l:td6 ( Kotov) , but 59 . . . .U.f7! 60
l:txc6 .Uxf2 61 .l:tc8 'it>e6 ! (Speelm a n ) . I n­
stead of 56 .Ud6+ Wh ite can try 56 'it>f4 ! ? , but
after 56 . . . .l:!.d7 ! 57 .Uc8 d4 58 .l:i.xc6+ 'it>e7
(58 . . . 'it>d5? 59 .l:!.d6+ ) 59 exd4 (59 .Ud6? d 3 )
59 . .l:!.xd4+ Black again retains real d rawi ng
. .

chances.
56 .l:td6 .l::i. a 6
57 g5!
Wh ite clears a way i nto the enemy position
for his king. 67 .l::i.c 7+ �dB?
57 . . . fxg5 In Speelman's opm1on , even now, two
moves before resignatio n , it was sti l l possi­
58 hxg5 Wf7
ble for Black to save the game, a n d ,
59 'it>g3
moreover, very prettily: 67 . . . 'it>e6! 68 Uxh7
Not immediately 59 Wf4 .Ua4+ 60 'it>e5?? iic4+ 69 Wf3 l:txc5 70 l:tg7 .Uc6 ! ! . Now 70
l:te4 mate. .UXg6+ 'it>f5 71 .Uxc6 leads to stalemate , and
59 . . . 'it>e7 70 'it>g4 'it>d5 to a straightforward d raw ( 7 1
60 f3 .l:!.a3 .Uf7 .U a 6 72 .l::i.f6 llxf6 73 gxf6 'it> e 6 74 Wg5
'it>f7 ) .
61 'it>f4 .Ua4+
And yet Wh ite's position would appear t o be
62 'it>e5
won . Having seen through the opponent's
It would have been a mistake to play 62 e4? stalemate tra p , he should 'take a move back'
dxe4 63 fxe4 .l:!.c4 64 .l:txc6 Wd7 65 .Ud6+ - 68 .l::i. c 6+ ! 'it>e7 , then play his king to the
We? 66 .l::i.d 5 .Uc3 ! 67 'it>e5 .l:tc4 . The rook on queenside: 69 'it>d5 Ud 1 + 70 'it>c4 .Uc1 +
d5 is too passive and therefore it is not (70 . . Jif1 7 1 .Uf6) 7 1 'it>b5 li b 1 + 72 'it>a6 (with
possible to convert the pawn advantage. the th reats of 73 .Uc7+ or 73 .Ub6) , and after
62 . . . .Ua3! 72 . . . Wd7 bring it back, exploiting the fact
63 J::!.x c6 ! ? that the i m porta nt f6-sq uare is now accessi­
Serious consideration should also have ble for i nvasion: 73 .Ub6 .Uc1 74 'it>b5 .Ub 1 +
been g iven to 63 .Ue6 + ! ? 'it>d7 (63 . . . 'it>f7 64 75 �c4 .Uf1 (75 . . . .Uxb6 76 cxb6+ �c6 77
Wd6) 64 'it>f6 , for example, 64 . . . d4 65 .Ud6+ �d4 ) 76 Wd5 etc. (suggested by Sergey
We? 66 .Uxd4 .Uxe3 67 f4 , and Black's Dolmatov).
position is very dangerous. 68 .Uxh7 .Uxc5
63 . . . .Uxe3+ 69 .Uf7
64 �xd5 .Ud3+ Black resigned .
64 . . . .Uxf3 65 .l:!.c7+ and 66 .l::i.x h7 is hopeless
for Black.
65 'it>e4 .l::i. c 3
66 f4 .l:!.c1
1 34 w Converting an Advantage

Spassky - Korchnoi outside passed pawn by . . . g7-g6 and . . .h6-

Candidates Match , 5th Game, Kiev 1 968 h5.
29 . . . h5?
30 'it>h2 hxg4
3 1 hxg4 g6?!
32 g5!
Now a draw becomes the most probable
result: the pawn position is fixed on both the
queenside, and the kingside. ( Korch noi).
The game ended in a d raw on the 5 1 st
move .

Exchanging
G rand master Kotov remembered for a long
time the advice given to him by the
experienced master Vlad i m i r M akogonov at
This is what grandmaster Korchnoi had to
the i nternational tournament in Venice in
say:
1 950.
Despite the occasional inaccuracies com­
Don't sharpen the play - what for? Ex­
mitted, I consider my play in the middle
change the queens, and arrive at a position
stage of this game to be my best achieve­
where each side has a rook and two or three
ment in the match. But I wasn't quite able to
minor pieces left. Which piece should you
complete the strategic picture - at the
exchange, and which should you keep ?
decisive moment I failed to display the
There are few modern players who can
necessary know-how. What was the prob­
solve this question correctly. They under­
lem facing Black? I will allow myself to quote
stand tactics, but in this you are superior to
Bondarevsky: 'White's pieces are tied to the
them.
weakness at c2, but a single weakness he is
able to defend. Korchnoi was faced with the When trying to convert an advantage you
problem of starting play on the kingside, so consta ntly h ave to th i n k about the advisabil­
as to create a new weakness in the enemy ity of this or that exchange. One of the most
position. ' general g u ides is given by the fol lowing rule:
I realised that the move of the h-pawn Having a material advantage, the stronger
appeared too routine to be the best. And 1 side should aim to exchange pieces,
rejected 29. . . g5 on account of the concrete whereas the weaker side should aim to

variation 30 "illd2 f6 31 "ille1 !, when White exchange pawns.

neutralises his opponent's advantage. But

the best move- 29. . . f5 (suggested by Flohr)
completely escaped my attention! The point
(see diagram)
of the move is not only that after the
exchange on g4 White's f- and g-pawns will
be further weakened; a/so of considerable
importance is the fact that, after the ex­
change of queens, Black can create an
 Converting an Advantage ltJ 1 35

Vidmar - Thomas Now it made sense to switch the knight to

Notti ngham 1 936 e5: 38 tbd7 + ! �e7 (38 . . . �f5 39 lbb6! is bad
for Black) 39 tbe5 . I n stea d , Wh ite for some
reason retu rns his knight to the rear.
38 tbe4+ We 7
39 lDf2 llg8 !
40 �f3
If 40 tbd3 , then 40 . . . l:i.f5 ! , preventing 4 1
tbe5 .
40 . . . lbb5
41 �c5 .l::!. c 8
After the exchange of the e5-pawn , the
passed c6-pawn has been sign ificantly
wea kened , since Black has acq u i red the
opportu n ity to attack it with his king from d6.
Wh ite has a decisive advantage. H e should 42 .l:txd5?! exd5
now move his knight from c5 and then play 43 .l:!.c5 tbd4+
J:l.c5 , aiming to exchange the active black 44 'it>e3 tbf5+!
rook . 32 tbe4 �ad8 33 .l:!.c5 (the prophylactic Much worse was 44 . . . t2Jxc6? 45 J:.xd5 with a
move 33 g3!? is also strong) 33 . . . �d3+ 34 sign ificant advantage for Wh ite .
'iite2 �3d4 35 l::t 1 c4 is possible. 32 tbd7 �d6
45 'it>d3
(with the threat of 33 lbb6) and 33 .l:!c5 is
46 ti.xa5 .l:.xc6
even simpler.
47 l:i.a7 .l::!. c4
Milan Vidmar tried to carry out the same
idea , but he did it i n a very i naccu rate way, 48 �xh7 .l:!xf4
overlooking the opponent's counterplay in­ Black has managed to exchange a fu rther
volving a n exchange of pawns. two pai rs of pawns, and all his remaining
32 lbb7? g5! pieces and pawns a re excellently placed . A
d raw is now the most probable outcome.
33 g3 gxf4+
49 'it>e2 .l:!.c4
34 gxf4 l::tg 8
50 'it>d2 l:i.d4+
The first u npleasant conseq uence of Wh ite's
mistake - the h itherto passive rook at a8 has 51 'it>e2 .l:!.c4
come i nto play. 52 �d1 d4
35 .l:!4c2 f6! 52 . . . l:td4+ 53 Wc2 .i:lc4+ was simpler, seeing
Another pawn exchange, and moreoever as 54 �b3? will not do i n view of 54 . . . tbd4+
the strong wh ite e-pawn is forced off the 55 �a4 b3+ 56 'it>a3 l::ta 4+! 57 �xa4 bxa2.
board . 53 �d2 b3!
36 exf6+ �xf6 George Thomas forces the exchange of
Itis obvious that the last exchanges have a nother pai r of pawns .
considerably increased Black's drawing 5 4 axb3 .l:!.b4
chances. (Aiekh ine) 55 tbd3 .l:!.xb3
37 tbc5 l:::. g 4 56 l:td7+?!
1 36 � Converting an Advantage

56 h4 was stronger, but even then Black 37 �e4? �c6

would have successfu lly defended by acti­ 38 �xc6 'it>xc6
vating his rook: 56 . . . .l:!.b8 followed by . . . .l::i. g 8 . The exchange of bishops was bad , since
56 . . . �xd7 now the wh ite rook is forced to take up a
57 lLlc5+ �d6 passive position i n front of the enemy pawn.
58 lLlxb3 lL:le3! 39 'it>f1 a4
There are too few pawns left on the board for 40 'it>e2 a3
Wh ite to hope for success in the knight 41 'it>d3 a2
ending. 42 .l:ta1 'it>xc5
59 h4 lL:lc4+ 60 'it>c2 'it>e5 61 lL:lxd4 'it>xd4 62 The d raw has become obvious (if 43 'it>c3,
b4 'it>e4 63 'it>c3 lL:lb6 64 b5 'it>f5 65 Wd4 then 43 . . . l:tg8 44 g3 .l::!. f8 ) .
'it>g4 66 'it>c5 lL:la4+ D raw.
I n stead o f t h e incorrect exchange, White
could have activated his kingside pawns: 36
It should be remembered , however, that the g4!? (threatening g 5-g6-g7). But it was
rule we have just formu lated is too general safer first to central ise the king: 36 f3 ! a4 37
to be trusted uncond itionally - in chess such 'it>f2, and only then play g2-g4 . Such
universal laws do not exist. This is merely strategy would have been in accordance
one of the gu ides; the concrete featu res of with a principle of endgame play, formulated
the position often dictate a completely by Aaron N imzowitsch : 'The advance must
different cou rse of action . be a collective one/ '.

Eh lvest - Andria nov VI. Lack of concrete action at

Tal l i n n 1 98 1 the decisive moment
Let us su ppose that you r opponent has no
real cou nterplay and that you , i n accordance
with the principle 'do not h u rry! ' , are accumu­
lati ng advantages l ittle-by-little . But against
tenacious resistance by the opponent you
will probably be unable to win the game by
tech nique alone - at some point you will
certainly have to switch from positional
manoeuvri ng to the precise calculation of
variations, and seek a concrete way to the
goa l . Many players stumble at this point, with
various factors playi ng their part. There is
carelessness, arising in a nticipation of a
q u ick w i n , about which we have a l ready
The passed a-pawn promises Black defin ite spoke n . There is the fu lly u nderstandable
counter-chances, but even so Wh ite's mate­ aim to act 'with every comfort' , not exces­
rial adva ntage should be sufficient for a w i n . sively exerting yourself, and not subjecting
However, not with t h e plan chosen b y Jan you rself to the risk of making a mista ke in
Ehlvest. forcing play. There is the d ifficulty of deter­
36 �g6? 'it>c7 ! m i n i ng that turn ing point, when you have
 Converting an Advantage ctJ 1 37

already extracted the maxim u m from play­ 1 0 . . . i.xf3 ! ? 1 1 i.xf3 lt'lc6 came i nto consid­
ing accord ing to the principle 'do not h u rry! ' , eration .
which mea n s that i t i s time to find a concrete 1 1 .l:tad 1 .l:i.c8?
variation, one which exploits the advantage A serious mistake , after which Black falls
gained and advantageously changes the sign ificantly beh ind i n development and
character of the play. ends u p i n a d ifficult positi o n . H e should
I have noticed that brill iant positional players have castled .
such as, for example, Salo Flohr or Anatoly 1 2 'iVa3 ! lt'la5
Karpov, would successfully convert a n ad­
1 2 . . . cxd4 really was better.
vantage against opponents i nferior to them
in class. They manoeuvred , suppressed all 1 3 b3 ii.e7
active possibilities by their opponents , and 1 4 dxc5 ffi
when the latter failed to withstand the It is a bad sig n , if moves such as this have to
pressu re , they made mistakes and them­ be made. But if 1 4 . . . bxc5 there follows 1 5
selves broke u p their positions. But agai nst tt:Je5 i.xg2 1 6 �xg2 d6 1 7 'ifa4+ 'it>f8 1 8
opponents of equal class they often did not lt'ld7+ �g8 1 9 lt'lxc5.
manage to convert even a big advantage . 1 5 ii.h3 'i;;f7
F o r t h e reason that, w h e n faced with 1 6 i.xe6 was th reatened , and if 1 5 . . .'iVc7 ,
tenacious resistance, you can n ot afford to then 1 6 cxb6.
miss an appropriate moment for concrete
and precise actio n , and this is by no mea ns
the strongest aspect of such positional
players .

Flohr - Keres
1 8th USSR Championsh i p , Moscow 1 950
Queen's Indian Defence
1 lt'lf3 c5
2 c4 lt'lf6
3 g3 b6
4 i.g2 i.b7
5 0-0 e6
Wh ite has a n und isputed adva ntage . H e is a
6 lt'lc3 i.e7
pawn u p , the black king is stuck in the
7 d4 lt'Je4?! centre , and the d7- and e6-poi nts a re
A dubious move , which could have been obviously weak. But note that all these
called into q uestion by the energetic 8 d5! factors are not constant, but temporary.
tt:lxc3 9 bxc3 , and if 9 . . . i.f6 , then 1 0 e4! I magine that Black plays . . . bxc5 and . . . d7-
.bc3 11 i.g5 (Udovcic-Kovacevic, Zagreb d6 - then h e will consolidate his position .
1 969). The usual conti n u ation is 7 . . . cxd4. This means that Wh ite m ust act swiftly and
8 'i!Vc2 lt'Jxc3 decisively.
9 'iWxc3 i.f6 1 6 li.d2?
1 0 i.e3 tt:Jc6 An i nstructive commenta ry on the move
1 38 � Converting an Advantage

made by Flohr was g iven by g rand master After the move made by White, the picture
Isaak Boleslavsky: changes amazingly rapidly.
In this position could White really not find I should also add that after 1 6 b4 ! li:Jxc4 the
anything better than the strictly positional move 1 7 'ir'xa 7 is the strongest - 1 7 'i'b3
doubling of rooks? If White really wanted to (hoping for 1 7 . . . b5? 1 8 ii.xe6 + ! 'it>xe6 19
play positionally, he should have continued li:Jd4+ an d 20 li:Jxb5) is much worse in view
1 6 li:Jd4 ii.. xc5 17 'ika4 ( 1 7 'ir'c1 ! ? - of 1 7 . . . li:Jxe3 1 8 fxe3 ii.xf3 . However, also
Dvoretsky), and to avoid the worst Black after 1 7 . . . i.d5 1 8 �xd5 exd5 1 9 'ii'd 3 lbxe3
must exchange on d4. But the position 2 1 fxe3 'lti>e8 compared with the analogous
demanded other measures, and after the variation with 1 7 'ikxa7 the a7-pawn would
energetic stroke 1 6 b4! White would have have remai ned al ive.
gained an irresistible attack. Here are some 16 . . . bxc5
sample variations:
1 7 I!fd 1 d6
1) 1 6. . . Chc6 17 cxb6 ( 1 7 l:td2 is also not bad
1 8 lLle1
- Dvoretsky) 17 ... axb6 18 'ikb3 li:Jxb4 (if
Another passive move . 1 8 ii.f4 suggests it­
1 8 . . . ii.. xb4 both 1 9 c5 and 1 9 a3 are strong,
self, forcing the uncomfortable reply 1 8 . . Jk6
and even 19 ii.xe6+ ! 'lti>xe6 20 ii.xb6! 'ii'e 8
( if 1 8 . . . .i.xf3 1 9 exf3 li:Jc6 , then either 20
21 c5+ cj;;e 7 22 a3 ii.. a 5 23 ii.xa5 li:Jxa5 24
.i.xd6 li:Jd4 2 1 .i.xe6+ ! , or 20 �xd6 �xd6
'ii'b4 - Dvoretsky) 19 ii.xe6+! cj;; xe6 20
2 1 �xd6 'file 7 22 .i.xe6+ ) .
ii.xb6 'i!lixb6 (20. . . 'fle8 21 c5+ li:Jd5 22 e4)
21 c5+ ii.d5 22 1J.xd5 (22 'ir'e3+ - Dvoretsky) 18 . . . 'i!lib6
22. . . ii.xc5 23 1J.fd1 ! (of cou rse, 23 1J.xc5+ is The d6-point is easily defended , and there is
also good enough to wi n ; generally speak­ noth ing more with wh ich to attack it - the f4-
ing, you should not conti nue calculating square will be ta ken away from the white
variations, if the evaluation of the conti nua­ bishop by . . . g7-g5.
tion being analysed has become obvious - 1 9 'ir'c1 h5!
Dvoretsky) 23. . . ii.xf2+ 24 �g2 li:Jxd5 25 Black has a l ready seized t h e i n itiative. I f 2 0
'iVxd5+ 'lto>e7 26 'ii'xd7+ cj;; f 8 27 't!Vxc8+, and li:Jd3 , t h e n 20 . . . g5 2 1 b 4 ii'c6 . Possibly
wins. Wh ite should have tried 20 ii.g2 h4 2 1 b4!?
2) {6. Jhxc4 17 'iVxa7 liJxe3? 1 8 fxe3 ii.xf3 - after 21 . . . 'ii'x b4 22 .Ub2 'it'a4 23 �xb7 (23
19 'fud7. ii.xb7? �b8) 23 . . . li:Jxb7 24 ii.xb7 l::t bB
3) 1 6. Jhxc4 17 fixa7 ii.. c 6 1 8 ii.xe6+! (or 1 8 followed by 25 . . . hxg3 26 hxg3 'it'xa2 the
cxb6 l:ta8 1 9 ii.xe6+ ! - Dvoretsky) 1 8. . . 'lto>xe6 resulting position is d ifficult to eval uate .
19 li:Jd4+ rtJfl 20 li:Jxc6 .:txc6 21 '!J.xd7 'ike8 20 f3? ! h4
22 cxb6 li:Jxe3 23 fxe3, and White, with four 21 g4 li:Jc6
pawns for the piece and an overwhelming
22 li:Jg2?
position, wins without difficulty.
22 li:Jc2 was better.
4) 1 6. . . li:Jxc4 17 'flixa7 ii.d5 1 8 �d5 exd5
22 . . . li:Jd4
19 'ikb7 cj;; e 8 20 'iixd5 li:Jxe3 21 fxe3 'Wic7 22
1J.d1 'i:J.d8 23 cxb6 'iic 6 (23 . . . 'ii'x b6 24 23 1J.xd4
ii.xd7 + ..tis 25 li:Jd4 ii.xb4 26 li:Je6+ cj;; e 7 27 The time for combi nations was earl ier. In the
li'e4! - Dvoretsky) 24 b7 'iixd5 25 1:.xd5 rtJfl subsequent play Black converted his ex­
26 b5, and White's powerful pawns decide change advantage , although the opponent
the game. did not exploit a l l his chances.
 Converting an Advantage CtJ 1 39

23 ... cxd4 24 i.xd4 'iii'a 6 25 g5 fxg5 26 f4 has gained a tempo - he has brought out his
g4! 27 i.xg4 h3 28 'ii'e 3 l:Ih6 29 tt:Je1 l:tg6 bishop to a more active position . If i n stead
30 'ii'x h3 'ifc6 3 1 ti:Jf3 'ii'e 4 32 'iVg3? (32 2 3 . . J1c7, then 24 i.e5 i.d6 25 i.. x d6 tt:Jxd6
<t;>f2 llh6 33 'ii'g 3 ) 32 ... 'it>g8 33 .l:r.d3 l':tf8 34 26 e4 , exploiti ng the fact that the rook has
.ie3 e5 ! 35 'ii'g 2 exf4 36 i.d2 i.d8 37 h3 remained on d 1 .
l:te8 38 'iti>f1 d5 39 l:td4 'ii' b 1 + 40 i.. e 1 dxc4 But it ca n also be exploited by Black! By
41 l:.xc4, and Wh ite resig ned in view of giving u p two minor pieces for a rook:
41 . . . i.h4! (but not 4 1 . . . i.a6? in view of 42 26 . . . tt:Jxe4 ! 27 fxe4 (27 g4 tt:Jg5) 27 . . . i.xd 1
tt'ld2). 28 l:.xd 1 dxe4 , he retai n s excellent chances
The entire game convincingly illustrates a of saving the game. There is no point in
well-known aspect of Stein itz's theory - the Wh ite going i n for such a n exchange, and
player with an advantage must attack, as the move made by Petrosian m ust be
otherwise he risks losing his advantage. deemed the strongest.
I n th is clea r formula the word 'attack' must 23 . . . l:.c7
be interpreted broadly - often it is necessa ry
24 i.. e 5 i.d6
to fi nd some precise variati o n, forci ng
combination etc. , i n short - a concrete and 25 i.. x d6 tt:Jxd6
e nergetic way to exploit you r advantage . 26 l:tfd 1
T hreatening both 27 l:.xd5 , and 27 e4 tt:Jxe4
Petrosian - Spassky 28 g4 .
World Championship Match , 1 2th Game, 26 . . . ti:Jb5
Moscow 1 969

Wh ite's advantage has crystall ised . H e now

23 l:.c1 has numerous tempti ng conti n uations, but it
A natu ral move, reta i n ing for Wh ite a serious is not so easy to choose the strongest. If 27
positional advantage. I ndeed , the c6-pawn tt:Jc5 (with the threat of 28 tt:Ja6) , then 27 . . . a5
is weak, the knight has an excellent sq uare 28 ti:Jd3 (th reatening 29 ti:Jf4 , then 30 a4 and
at c5 , and the bishop at h5 is out of play. 3 1 tt:Jxd5) 28 . . . a4 29 ti:Jf4 i.. g 6, and no d i rect
But couldn't Wh ite h ave played more accu­ win is apparent. To 27 l:.c5 Black repl ies
rately - 23 i.h3 ? After all , after 2 3 . . . l:.b7 24 27 .. .f5 (defending against 28 e4) 28 l:tdc1
l:tc1 .l:.c7 compared with the game Wh ite l:te7 ! ? (28 . . . ti:Jd4 29 'iii>f2 l:tac8 ) 29 'iii>f2 i.e8
1 40 � Converting an Advantage

or 29 .l:!.xc6 tt:'ld4 . sive move 33 . . . .tea (34 tt:'le6+ and 35 tt:'ld4)

The strongest was 27 g4! .tg6 2a f4 , relying and created the th reat of 34 tt:'ld7 + .
on a tactical subtlety: 2a . . .f6 (or 2a .. .f5) is 33 . . . .Uaa7
not possible because of 29 l:!.xc6 ! . I n the
event of 2a . . . .te4 29 .txe4 dxe4 30 �f2
Black's position is hopeless i n view of the
weakness of his c6- and e4-pawns. But
2a . . . .th7 is also no better: 29 f5 (29 tt:'lc3 ! ? )
2 9 . . . g 6 3 0 e 4 dxe4 3 1 .txe4 .Uea 32 tt:'lc5
with an overwhel ming advantage than ks to
the trag i-comic position of the black bishop
and the terrible threat of 33 a4 .
Wh ite also had a nother promising possibil­
ity: 27 tt:'lc3 ! ? , emphasising the vulnerabil ity
of the opponent's central pawns. The idea of
g3-g4 and f3-f4 could have been put i nto
effect slightly later.
27 '.t>f2 f6! ?
Black is only just holding o n . One senses
2 8 e3?! that it is time to fi nd a concrete way to break
Petrosian conti n ues strengthening his posi­ through the opponent's defences. And there
tion , but now his advantage is somewhat is such a way. After 34 tt:'ld3 ! .tea 35 .!Llf4
reduced , since the black bishop is i ncluded Wh ite is threatening both 36 tt:'le6+ followed
in the defence of the q ueen side pawns . But by 37 tt:'ld4 or 37 tt:'lda , and 36 l:!.xc6 .txc6 37
meanwh ile White still had a concrete way of tt:'lg6 + . I n the event of 35 . . . �f7 he decides
achieving a won position: 2a tt:'lc5 ! .l:!.e7 (or matters with 36 l:!.xc6! .txc6 37 .l:!.xc6 tt:'lb5
2a . . . a5 29 tt:'le6 l:!.cca 30 .th3 with the 3a .te6+ l:!.xe6 (forced) 39 tt:'lxe6 , and the
u navoidable 31 tt:'ld4) 29 tt:'la6 .tea 30 a4! conversion of the extra pawn is not too
tt:'ld6 3 1 e4, and Black loses a pawn . d ifficult ( if 39 . . . a4 there follows 40 b4).
28 . . . .tf7 H owever, we nevertheless do not have the
29 .tf1 tt:'ld6 right to say that Wh ite's position is defin itely
30 l:!.c3 won . Even in a seem ingly d ifficult situation
30 .ta6!? came i nto consideration. one ca n usually fi nd resou rces , enabling
30 . . . �f8? defeat to be avoided or at least the
opponent's task to be sign ificantly compli­
A far from obvious mistake . Black should
cated . That is also the case here . For
have taken control beforehand of the i m por­
example, there is a clever exchange sacri­
tant f4-square, by playi ng 30 . . . g5! . After 3 1
fice: 35 . . . l:!.a6 ! ? 36 tt:'le6+ l:!.xe6 37 .txe6 f5
tt:'lc5 a 5 3 2 l:!.dc1 .l:te7 followed b y . . . .tea it
3a g4 g6 39 gxf5 gxf5 40 .Ug 1 �e7, and the
would not be easy for Wh ite to strengthen
u nfortunate position of White's bishop mean s
his position .
that t h e conversion o f h i s material advan ­
31 tt:'lc5 a5 tage is problematic. I n stead o f 3 6 tt:'le6 + , i t is
32 l:!.dc1 .l:!.e7 p robable that 36 a4! is stronger - subse­
33 .th3 quently the knight may be switched to d4 not
Wh ite has prevented the important defen- only via e6, but also via e2.
 Converting an Advantage lZJ 141

Often t h e best defence is active defence. I 38 . . . g41

recommend checking 35 . . . a4 ! . Here is a n 39 .i.g2 gxf3
approximate variation : 36 .:.xc6 .i.xc6 3 7 40 .txf3 ttJe4+
t:Llg6+ (37 l:txc6 axb3 3a axb3 .l:. a 2 + 39 'i!tg 1
41 .i.xe4 fxe4
:xe3 ) 37 . . . 'i!tea 3a CiJxe7 'i!txe7 39 l:!.xc6
axb3 40 axb3 .l:ta2+ 4 1 'i!tg 1 d4!? 42 exd4 42 CiJd4 .l:r.f7+
t:Llb5. For the moment the outcome of the 43 'i!tg2 lif6
game remains unclear - B lack's counter­ Black now has some i n itiative in con nection
attacking resou rces should not be under­ with the strateg ic th reat, after the exchange
estimated . of a pai r of rooks, of playing h i s king to d6.
Even so, the manoeuvre of the knig ht to f4 On the resumption the game ended i n a
was the correct p l a n . Petrosian played a d raw.
weaker move .
34 a4? ! Transformation of an advantage
The position can be u n h u rriedly embell ished , The best way of exploiting an advantage
if during this time the opponent is not able to sometimes i nvolves a favourable change in
do anything to strengthen his defences. This the character of the positi o n , giving u p some
is not the case here. It is dangerous to abuse advantages that you already have i n favour
the principle 'do not h u rry! ' . o f others . S uch a method is cal led 'transfor­
Apparently Petrosian w a s i ntending 35 tiJd3 mation of an advantage' .
.iea 36 CiJf4 'i!tf7 37 tiJe2 followed by CiJd4
and he wanted to prevent the opponent from Tai manov - Ste i n
replying 37 . . . tiJb5. Generally speakin g , it is 34th U S S R Championsh i p , Tbilisi 1 966/67
usefu l to fix the black pawn on a5 and to
deprive the knight of the b5-square . But if
this is played , it should be after the switching
of the knight to f4 , and therefore now Boris
Spassky forestal l s the main danger.
34 . . . g51
T h e f4-sq uare is taken u nder contro l .
35 l:.d1
An exchange sacrifice came i nto considera­
tion - 35 tiJd3 .tea 36 .llxc6 .i.xc6 37 llxc6
tOea 3a tiJc5 . Petrosian wants to play h i s
knight t o d 4 v i a a lengthy route - d 3-c 1 -e2 ,
but d u ring this time Spassky is able to
activate his forces.
35 . . . 'i!tg7 Black has a great positional advantage. He
has securely blocked the opponent's central
36 tiJd3 .tea
pawns, the bishop at g2 is 'bad ' , and the
37 CiJc1 f5 wh ite knight also lacks mobil ity. The most
38 tiJe2?! natu ral plan , which Leonid Stei n undoubt­
3a .i.g2 g4 39 f4 was better, with a p robable edly had in m i n d , i nvolves the advance of
draw. the q ueenside pawns.
1 42 � Converting an Advantage

26 'it>f1 34 lLlxf2 lbf2+

Wh ite intends to reinforce his central pawns 35 '>t>c3 I:!.a3+
by taking his king to e3 and if necessary 36 .J::f. b 3 l:i.axa2
placing his bishop on f1 . H i s rooks would 37 .l:tb5 �g6
then be freed to take action against Black's
38 l::t d 5 l:i.f5
queenside pawn offensive. At this moment it
39 I:!.d6+ l:i.f6
probably seemed to Stei n that the conver­
sion of his adva ntage by normal methods 40 l:i.d7 l:i.g2
would not be so easy. 41 d4 l:i.xg3+
If the opponent makes an unexpected Wh ite resigned.
move, hindering the implementation of Stein's energetic actions were crowned by
your plans, it is useful to ask yourself: complete success. But it seems to me that
'What may be the drawback to the what mainly told here was the psychological
opponent's move?' But even after asking effect of Black's un expected operation -
you rself such a question , it is not easy to there a re nevertheless some doubts about
come to the decision found by Stei n - it is its objective strength .
very much not in keeping with the u n h u rried Serious consideration should have been
character of the preced ing play, and with g iven to the reply 27 � h 3 ! ? suggested by
Black's intended pla n . G rigory Kaidanov. After 27 . . . g4 28 i.g2
26 . . . f5 ! ? Black ca n not play 28 . . . lbxd3? 29 �xd3
B y opening the f-file, on which t h e wh ite king l:i.xd 3 30 lbxd3 fxe4+ 3 1 lLlf4 , and nothing
stands for a moment, Black creates a th reat particu lar is promised by 28 . . . fxe4 29 i.xe4
to the d3-pawn . Of cou rse , such a move , or 28 . . . f4 29 gxf4 l:i.xf4 30 '>t>e2 .
freeing Wh ite's bishop and knight, could I n stead of 27 . . . g4 combinations involvi ng a
only be made by a highly dynamic, non­ sacrifice on d3 look tempti n g . However, if
routine player. 27 . . . l::t x d3 there follows 28 lbxd3 fxe4+ 29
27 exf5 lbf2 �xh3+ 30 '>t>e2 . 27 . . . lbxd3 is stronger,
�xf5
hoping for 28 �xf5?! lbb4 ! ! 29 l:i.xd4 tt:lxc2
28 'lt>e2
30 �xe6 .Uxf2+ 31 ;t>xf2 lbxd4 with a won
I would have preferred to part with a pawn minor p iece ending (32 i.f5 is bad in view of
immediately, by retu rning with the king to g 1 . 32 . . . lbxf5 33 exf5 g4! 34 We3 ;t>f6 35 Wf4
28 . . . lbg4! h 5 ) . Wh ite defends by 28 exf5 ! tt:lxf2
Threatening 29 . . . lbxf2 30 ;t>xf2 �g4+ 3 1 (28 . . . i.xf5 29 i.xf5 l:i.xf5 30 'it>g2 or 28 . . . tt:lb4
;t>e3 i.xd 1 . 30 I:!.xd4 lbxc2 3 1 fxe6 l:i.xf2+ 32 'it>xf2 tt:lxd4
29 l:i.b2 lbxh2 33 'lt>e3) 29 I:!.xd4 lbxh3 30 g4 with a n
u nclear ending.
The conversion of the extra pawn is not
difficult. I th i n k that i n reply to 27 �h3 Black should
move his bishop: 27 . . . �d7 ! . H owever, here
30 'it>e3 l:i.a4
too Wh ite retains some saving chances in a
31 �e4 �xe4
position where he is the exchange down : 28
32 lbx e4 lbg4+ �xf5 �a4 29 I:!.dd2 �xc2 30 l:i.xc2 , or a
33 'it>d2 lLlf2 ! pawn down after 30 . . . lbxd3 3 1 l:i.d2 lLlxf2 32
Remember: with a material advantage it is .J::f.x d4 lbxe4 33 g4.
advisable to exchange pieces. The strongest response to 26 . . . f5 would
 Converting an Advantage ltJ 1 43

seem to be the cool-headed 27 'it>e2 ! . I n the 'for ' and 'against'. It looks illogical, since
event of 27 .. .f4 28 gxf4 gxf4 there is the White voluntarily exchanges his 'good' bishop
satisfactory reply 29 ..th3 ! , while after for the opponent 's 'bad ' bishop, instead of
27 .. .fxe4 28 ..txe4 ltJg4 , as shown by P h i l i p p exchanging it for the knight (1 8 ..txb6+) and
Schlosser, Wh ite has t h e s i m p l e move 29 consolidating his advantage. But on a
l:tf1 ! (29 . . . ltJxh2 30 l:th 1 ) . deeper investigation of the position it be­
comes clear that after the possible ex­
The tra nsformation of a n advantage - giving change of rooks on the d-file and the
up some benefits that you a l ready have for transfer of his king to e6, Black covers his
the sake of achieving other benefits - is a vulnerable points and sets up an impregna­
rather complicated tech nique, accessible ble position. In this case his 'bad ' bishop
only to players with a subtle u nderstanding would play an important role.
of the game. After a l l , you have to assess For my part I should comment that after 1 8
the situation correctly, and precisely weigh g4 .l:xd 1 + 1 9 l:txd 1 l:td8 20 l:txd8 'it>xd8 2 1
up the pluses and m i n uses of the decision it.xb6+ axb6 22 Wc2 Wh ite also retains
being take n , i n order not to 'buy a pig i n a excellent chances of success. He plays his
poke'. And psychologica l ly it is not easy in a king to e4 and his knight to d 3 , with the idea
favourable position to ta ke sharp decisions, of a pawn offensive on the queenside, and in
depriving you rself of some advantages some cases even e2-e3 and f2-f4 .
gai ned earlier. 18 . . . l:txd 1 +
Petrosian recommended the pawn sacrifice
Petrosian - Ban n i k 1 8 . . . ..txc5 1 9 ltJxc5 .l:Ihe8 20 I!.xd8 'it>xd8 2 1
25th U SSR Championsh i p , R i g a 1 958 ltJxb7+ 'it>c7 2 2 ltJc5 e 4 (with the threat of
23 . . . a5 and 24 . . J::t e 5), but it is i ncorrect in
view of 23 ltJa6+ 'it>b7 24 liJb4 followed by
ltJc2 .
1 9 l:txd 1 ..txc5
20 ltJxc5 l:te8
2 1 ltJe4 .l:.e6
2 1 . . . Itf8 was no better: 22 g4 .l:.f7 (22 . . . ltJc8
23 liJc5 .l:.f7 24 liJe6+) 23 .l:l.d6 .
22 g4 a5
23 l:td3 liJd7
24 'it>c2
24 'it>d2 ! ? .
24 . . . b6
Anatoly B a n n i k hopes to ease his defence
Wh ite unexpectedly offered a n excha nge of
by exchanging knig hts with 25 . . . ltJc5 . Wh ite
bishops.
prevents this.
1 8 it.c5!
25 l:tf3 ! 'it>d8
Why? Here is Petrosian's explanati o n:
Before deciding on this move, i t was (see diagram)
essential to thoroughly weigh up everything
1 44 � Converting an Advantage

enced players often resort to repeating

moves, not only to gain time on the
clock, but also in the hope that the
opponent will try to change the unfa·
vourable course of the play and, by
avoiding the repetition, worsen his own
position. However, this should be done
carefully, avoiding the th ree-fold repetition
of the position which occu rred in the present
game. It is stra nge that neither Petrosian,
nor his opponent, noticed that after 39 . 'it>e8
. .

the position woul d be repeated for the third

time and Black had the rig ht to cla i m a draw.
Most probably this a l l h appened in a severe
26 a3!
time scramble.
Again Petros ian takes measures agai nst the
40 a4 lDd8
threat of an exchange - if 26 . . . ..t>e7 (intend­
ing 27 . . .l2Jc5) he had prepared 27 b4 axb4 41 lDh6!
28 axb4 . Then there follows c4-c5, when N ot al lowi ng 41 . . . l2Jf7 .
the opponent is altogether unable to breath . 41 . . . l2J e6
26 . . . c5 42 l2Jg8 lDf8
27 �c3 �e7 42 . . . 'it>f7 43 �d7 + ! 'it>xg8 44 'it>d5 is an
28 Ild3 elegant variation .
After provoking . . . c6-c5, which has weak­
ened the d5-poi nt, Wh ite retu rns his rook to
the d-fi le.
28 . . . .Uc6
29 Ild5 tt:'lf8
30 l2Jg3 l2Je6
31 lDf5+ ..t>e8
32 e3 l2Jc7
32 . . . lDd8 and 33 . . .l2Jf7 was more tenacious.
33 l::!.d 1 l2Je6
34 �d3!
The time has come to activate the king .
34 . . . .Uc7
Now 4 3 'it>d5 rt;d? i s pointless, while i f 4 3
35 �e4 l::!. c 6
�f5 there follows 43 . . . 'it>f7 44 tt:'lh6+ rt;g? 4 5
36 l2Jd6+ rt;e7
�d8 l2Je6 46 l::!. e 8 tt:'lc7 , and Wh ite loses h i s
37 tt:'lf5+ 'it> e8 knight.
38 l2Jd6+ rt; e7 How then ca n he break through the enemy
39 l2Jf5+ 'it> e8 defences? When the opponent is con­
When converting an advantage, experi- demned to passivity, one is very often
 Converting an Advantage ctJ 1 45

aided by a very important endgame variation 48 . . . tt:Jf8 49 'it>f5 'it>e7 50 tt:Jc3 tt:Jd7
device - zugzwang. 51 tt:Jd5+ '.t>f7 52 e4 h6 53 f3 .
43 l:td2 ! �f7 49 h6 tt:Je8
In the event of 43 . . .'�Jd7 Wh ite wins by 44 50 'it>d5 f5
lttf5 Wd8 45 e4 '.t>e8 46 f3 �d8 47 l:.xd7+! 5 1 '.t>xe5 fxg4
lttx d7 48 tt:Jxf6+. Note that, before sacrific­ 52 tt:Jc3 �e7
ing the exchange, it makes sense, i n
53 tt:Je4 '.t>f7
accordance with t h e principle ' d o not h u rry! ' ,
to make two preparatory pawn moves, 54 �f5 g3
strengthening the position to the maxi m u m . 55 fxg3 g4
If 4 3 . . . ne6 there also follows 4 4 Wf5 �f7 45 56 tt:Jg5+ 'lt>g8
�d8 �c6 46 tt:Jh6+ �g7 4 7 �e4! tt:Je6 48 57 �e6 tt:Jc7+
�d7+! '.t>xh6 49 '.i?d5 . 58 '.i?d7 tt:Ja6
44 tt:Jh6+ �e8 59 e4 tt:Jb4
45 tt:Jf5 tt:Je6 60 e5 tt:Jd3
46 �d6! 61 e6
The exchange of rooks, strengthening the Black resigned .
threat of an i nvasion by the wh ite king, leads
An excellent ending - i n it Wh ite used many
to a won knight ending.
of the principles for converti ng a n advantage
46 . . . �xd6 that we have been d iscussing.
47 tt:Jxd6+ �d7
48 tt:Jb5 tt:Jg7 I n concl usion I offer a few exercises, i n each
This leads two moves later to zugzwa n g , but of which you have to choose the most
that is also how things conclude i n the methodical way of proceed i n g .
1 46 � Converting an Advantage

Exercises

1 . Black to move 2. Wh ite to move

3. White to move 4. Black to move

 Converting an Advantage tD 1 47

5. Black to move 6. Wh ite to move

7. Wh ite to move 8. Wh ite to move

1 48 � Converting an Advantage

Sol uti ons

1 . Koberl - Szabo (Budapest 1 95 1 ) 25 l:tg5! b6 26 .Uxh5 e5 27 �xeS! bxc5 28

23 . . . a5! i.. f6 �e8 29 .Uh8+ 'it>d7 30 l:txd8+ l:txd8 31
24 tt:'lc1 a4! �xd8 'it>xd8 32 �3 (Wh ite's outside passed
pawn ensures h i m an elementary win)
By advancing his a-pawn , Black has pre­
32 . . . 'it>e7 33 �e4 �e6 34 'it>f4 f5 35 h4 'ittf6
vented the equalising manoeuvre tt:'le2-c1 -
36 h5 We6 37 Wg5 Black resigned .
b3, prepared the development of his rook by
. . . l:ta8-a6-b6 or . . . a4-a3 followed by . . . l:!.a4,
and, finally, created the precond itions for an 3 . M i les - N i kolac (Wij k aan Zee 1 979)
attack on the opponent's queenside. If now Noth ing is given by 48 l:!.f5 l:tg5.
25 tt:'ld3, then 25 . . . l:!.d8 26 �f1 l:td4 ! . 48 a4!
The game concluded a s follows: 25 'it>f2 a3 With th is u n h u rried move Wh ite forestalls
26 'it>e2 �b2! 27 .Uc2 (27 tt:'ld3 tt:'la4 ! ) the opponent's only sensible plan of . . b6- .

2 7 . . . l:r.d8 2 8 �f1 tt:'la4 2 9 tt:'ld3 tt:'lc3+ 30 b5-b4 and puts h i m in zugzwa ng. After any
�e3 tt:'lxa2! 31 tt:'lxb2 tt:'lb4 32 �c1 axb2 33 move by the knight from e4 , 49 l:tf6+ is
l:!.b1 tt:'lc2+! 34 'it>f4 (34 'it>e2 tt:'la3 35 �xb2 decisive. 48 . . . l:!.g5 49 .Uh7 is bad for Black,
tt:'lxc4) 34 . . . g5+ 35 'it>e5 l:td6! 36 c5 l:!.e6+ 37 while if 48 . . . l:!.h4, then 49 tt:'lg6 ! and 50 tbe5,
'it>f5 tt:'le3 mate. but not 49 .Uf5? because of the pretty reply
49 . . . l:!.h 1 + ! .
2. Bastri kov - Kiselyov (Sverd lovsk 1 946) 48 . . . �c6
Noth ing is g iven by 22 l:tg 1 + 'it>h 7 2 3 l:tg7 + 49 l:tf5!
'it>h6 or 23 l:!.g5 f6 (23 . . . Wh6? 24 ..te3) 24 It transpires that 49 . . . .Ug5 no longer defends
Ir.xh5+ Wg6 . Black's importa nt defensive the pawn i n view of 50 tt:'lxd 5 ! l::i. xf5 5 1 tbe7+
move . . .f7-f6 must be prevented . and 52 tt:'lxf5.
22 �e1 ! l:tfe8 49 . . . tt:'ld6
If 22 . . . e6 or 22 . . . l:tae8 , then 23 tt:'lc5 is 50 l:!.f6 .Uh4
strong .
51 g3 Itg4
23 l:!.g1 +!
52 �g2
23 ... 'it>h7 24 l:!.g7+ 'it>h6 25 l::i. xf7 is now bad
Black's position is now completely hopeless.
for Black.
Tony M iles q u ickly converted his advantage.
There followed : 23 ... Wf8 24 tt:'lc5 1:1ed8. As
52 . . . h5 53 tt:'lxh 5 �d7 54 'it>f3 .Ug8 55 tLlf4!
was poi nted out by g randmaster Matthew
.Uxg3+ (55 . . . tt:'le4 56 tt:'lxd 5 ! ) 56 'it>xg3 tt:Je4+
Sadler, 24 . . . Ir.ec8 ! ? was more tenacious,
57 �g4 tt:'lxf6+ 58 'it>f5 tt:'le4 59 tt:'lxd5 tt:Jd6+
when Wh ite should conti nue 25 tt:'ld7+! (less
60 �e5 tt:'lf7+ 6 1 'it>f6 B lack resig ned .
good is 25 .l::t g 5 b6 26 .Uxh5 f6 or 25 tt:'lxb7
.l:l.ab8 26 ..tg7+ ! WeB 27 l:!.b 1 f6 28 il.h6
'it'd?) 25 . . .�e8 26 tt:'le5 'it>f8 (26 . . . 'it>d8 27 4. Skembris - Torre
tt:'lxf7+; 26 . . . e6 27 l:tg8+ �e7 28 l:tg7) 27 (Olympiad , Luzern 1 982)
..te3 e6 28 �c5+! .Uxc5 29 tt:'ld7+ 'it>e7 30 Wh ite's pieces have hardly any active
tt:'lxc5 . possibilities. H owever, he nevertheless has
 Converting an Advantage tZJ 1 49

one chance to become active : 'iie 2 ! , i ntend­ 6. Smirin - Vogt (Saltsjobaden 1 988/89)
ing 'ii'b 5! . For example, 30 . . . h6? (generally In the event of 33 .l:.xa5? .l:i.f3 34 l:t.h5 l:txg3
speaking, this pseudo-prophylactic move is 35 .l:lxh7+ �g8 the passed g-pawn ensures
usefu l , but it does not parry the opponent's Black sufficient counter-chances. The attack
concrete threat) 31 'iie 2 ! 'ikxb3 (in the on the g3-pawn must be forestalled , and the
endgame Wh ite gains a d raw without d iffi­ move 33 i... e4! ? , made by l l ya S m i ri n , looks
culty) 32 'ikb5 b6 33 lt:Jf3, a n d the weakness a sensible solution to the problem . There
of the f7-point ensures Wh ite sufficient followed 33 . . . i... x a4? 34 l:txa5 i... e 8 (34 . . . i... c6
counterplay. 35 i... x c6 bxc6 36 l:tg5) 35 i... x b7 ltf1 + 36
30 . . . a6! �d2 i... g 6 37 c4 l:tf2+ 38 �c3 �g7 39 l:tg5 ! ,
The opponent's only active idea is pa rried , a n d Black, finding no way out, lost on time.
and Black will soon create th reats on the Lotha r Vogt could have excha nged either
queenside by moving his knight across to the roo ks , or the bishops. Try to estimate (I
there . mean esti mate - to ca lculate everything is
31 g4 lt:J e7 32 lt:Je2 .i d 2 33 lt:Jg1 lt:Jc6 not possible and you have to trust you r
(threatening 34 . . . lt:Ja5) 34 i... c 7 lt:Jb4 35 i... a 5 i ntu ition ) whether o n e o f t h e excha nges (or
lLJc2 36 i... x d2 'ikxd2 37 �g3 lt:Jxe3 ! 38 'ii'a 3 both ) offers realistic chances of saving the
lLJd 1 39 lt:Jf3 'ii'xf2+ 40 �4 g5+ Wh ite game. If you r a n swer is positive, this gives
resigned . g rounds for seeking an a lternative move to
the one chosen by Wh ite in the game.
5. Gragger - Barcza (Olympiad , Varna F i rst let us examine the bishop ending:
1 962 , variation from the game) 33 . . J:te8 34 ltxe8+ i... x e8.
If a passed pawn is blocked by a bishop, the
winning plan usually i nvolves breaking
through with the king towards the passed
pawn . But doing this i mmed iately does not
work: 1 . . . �e4? 2 �e2 .ih5+ 3 �f2 �d3 4
i.e? ! a4 5 i... d 6 �c2 6 i... a 3 with a d raw.
Black must first tie the wh ite king to the
defence of the queenside pawns, and only
then break through with h i s king on the
opposite wi n g .
1 . . . �c4!
2 i... c 7 a4
3 i... e 5 � b3
4 �c1 i... c 2! In the magazi ne 64 - Shakhmatnoe oboz­
4 . .ih5 also wins.
. . renie ( 1 996 No. 1 2), grandmaster Igor Zaitsev
5 i... d 6 suggested a clever breakthrough pla n : 35
b4! ? b6 (Black loses immediately after
Or 5 c4 b6.
35 . . . axb4? 36 aS or 35 . . . i... x a4? 36 bxa5
5 . . . a3
followed by 37 i... x b7) 36 b5 i...f7 1 (it is
6 bxa3 �xc3 important to forestall Wh ite's main th reat
Then . . . .ia4, . . . b7-b5 and . . . �d3-e2-f3-g2 . c4-c5) 37 �d2 �g7 38 �d3 , restricting
1 50 � Converting an Advantage

hi mself to the variation 38 . . . h6 39 c4 'it>f6 40 43 . . . 'it>d6(d8)? 44 «t>b6 h4 45 c7 or 43 h4 . . .

c5 'iite 5 4 1 cxb6 'iit d 6 42 'iit e 3 �b3 43 �c6 44 gxh4 g3? 45 c7! (now it is clear why the
�xa4 44 b7 'it>c7 45 b6+ 'iit b 8 46 �xa4 . king avoided the d4- and c5-sq uares - so
Here I do not agree with h i m - Black's that after 45 . . . g2 the pawn should not queen
resou rces are not yet exhausted . To say with check), but Black continues 43 . . . h4! 44
nothing of the attempt, by sacrificing the h7- gxh4 «t>d6 ! , l u ring the king to b6 where it will
pawn with 38 . . . 'it>f6 , to bring the king to the be checked , and then 45 . . . g3 (analysis by
queenside as q u ickly as possible i n order to Zaitsev) .
hinder c4-c5, he can also play more I n t he event o f 38 �c8 ! Black h a s two
accu rately in the cou rse of the plan exam­ possibilities:
ined by Zaitsev. a ) 38 . . . h 5 39 c6 h4 40 gxh4 g3 41 �h3 �e6
38 . . . h5 39 c4 'iit f6 40 c5 'it>e7 ! (but on no 42 �g2 'it>g6 43 �d2 �h5 44 'it>e3 'it>xh4 45
account 40 . . . 'it>e5? - the place for the king is 'it>d4 'it>g5 46 Wc5 �f4 4 7 'it>b5 'it>e5 (after
not in the centre, but in front of the pawns) 47 . . . 'it>e3 48 'it>xa5 Wf2 49 �h 1 the king
41 cxb6 'iit d 8 ! . How can Wh ite win here? It does not manage to return to the q ueenside
is not possible to queen a pawn : 42 �f5 i n time) 48 'it>xa5 'iii>d 6 49 b4 , and White
�b3 43 'iit d 4 �xa4 44 'it>c5 �d 1 45 �d6 should appa rently w i n .
�f3 - the bishop has a rrived just in time. b ) 38 . . .<it>f6 39 i.xg4 'it> e 5 ( i n t h i s way i n ,

And the position arising after 42 'it>e3 �b3 Zaitsev's o p i n i o n , Black g a i n s a d raw) 4 0
(42 . . . 'it>c8 ) 43 �c6 'it>c8 44 b7+ 'it>b8 45 b6 'it> d 2 W d 5 4 1 'ito>c3 i. a 2 42 b 4 axb4+ (or
.ii. d 1 is drawn . 42 . . . «t>c6 43 i.f3+ 'ito>b5 44 �e2+ 'ito>c6 45
The main contin u ation is 35 �xb7 �xa4 36 bxa5 'ito>xc5 46 'it>d2 ) 43 'it>xb4 'ito>c6 44 .ii.f 3+
c4. Black avoids an i mmed iate loss by 'ito>c7 45 'it>c3 , and there is noth ing to be done
playing 36 . . . i.b3 37 c5 �g7 .
against the decisive breakth rough of the
king to the king side. And th is means that the
exchange of rooks most probably would not
have saved Black.
H ow can the defence be improved? The
best saving chances i n such situations are
usually prom ised by transposing i nto a rook
ending (everyone knows the sayi ng: ' rook
endings are never won'). And so, 33 ... �xe4!?
34 .l::txe4 . H owever, after 34 . . . .l:!.g8? (34 ... h5?
35 l::t e 5) 35 Wd2 ! ? with the idea of 'it>e3-f4
Black is condemned to complete passivity
and should certainly lose.
In rook endings you should aim to
activate the rook. After 34 . . . .l::t f 1 + 35 'it>d2
If 38 c6 , then 38 . . . i.d5! 39 �a8 �e6 ! . I n the .l:!.f2+ 36 Wd1 (36 �d3 .l::tf3+ 37 .l::te 3 .l::tf2 38
event of 38 'iit d 2 'iit f6 39 'iit c3 Black fi nds the b3 'it>g7) 36 . . . h 5 ! 37 .l::t e 5 .l::t h 2 38 .l::t x a5 'it>g7
excel lent manoeuvre 39 . . . �d 1 ! with the followed by . . . 'ito>f6(h6) and . . . h 5-h4 Black
idea of . . . i.f3 . It is not apparent how Wh ite gains cou nterplay, but is it sufficient to save
can wi n . For example, after 40 c6 �f3 4 1 the game?
'iitc4 h 5 4 2 �a8 �e7 4 3 'it>b5 i t i s bad to play There is also another way of transposing
 Converting an Advantage LtJ 1 51

into a rook ending: 33 . . .l:tf1 + 34 �d2 l:tg1 position by adva ncing his queenside pawns .
35 i.. xc6 bxc6 36 l:txa5 l:tg2+ 37 'it>d3 I n reply t o 37 . . . �c6 noth ing is g iven by 38 b4
l:'l.xg3+ 38 �e4, and now either 38 . . . l:tg 1 39 .i::tf2+ (but not 38 . . . .Uf3? 39 b5) 39 �e 1 �g2
l:'l.c5 .Uf1 ! (38 . . . g3? 40 �f3 g2 41 b3 ! ) 40 40 b5 �f3 41 .Uxh7+ 'it>g8, but 38 �h4! is not
l:'l.xc6 �g7 41 a5 h5, or 38 ... I:tg2 ! ? 39 �c5 bad .
l:'l.f2 ! 40 a5 g3 41 'it>e3 ! (4 1 a6 g2 42 a7? Such a plan for converting an advantage
g 1 'i' 43 a8'ii'+ �g7 44 'ii' b 7+ �f7 or 42 .Ug5 (domi nation and the absence of cou nterplay
l:'l.xc2 43 a7 .Uc4 + ) 41 ... 'it>g7 ! ? (4 1 . . .�f1 for the opponent) is fully i n the spirit of
would appear to be worse : 42 I:tg5 c5 43 Anatoly Ka rpov. From the viewpoint of the
l:'l.xg3 .Ua 1 44 '.te4 I:txa5 45 'it>d5 , and Wh ite practical player, it is very important that here
is threatening 46 c4 followed by 'it>c6-b6 ) 42 practically nothing needs to be calculated (in
a6 l:f.f1 , and the position is most probably contrast to the 33 �e4 variation , where one
drawn . has to delve both i nto the bishop, and the
I n the rook endgame Black would h ave rook endgame), and this means that the
retained good d rawing chances. This factor probabil ity of mistakes is reduced .
casts doubts on the plan beg i n n i n g with 33 I ncidental ly, the fi nal conclusion about there
..ie4 , and forces us to seek other ideas. being only one solution to the i n itial end­
Here is a suggestion by grandmaster Viorel game position (and also about there being
Bologan . only one winning method i n the bishop
3 3 .Ug5! ending) is fu lly i n accordance with Zaitsev's
I n the first insta nce , as we know, 33 .. J:U3 view: My many years' experience of analysis
must be prevented . have convinced me that in tense, balanced
33 . . . �xa4 positions there cannot be two ways to win.
The same thought was also expressed by
34 I:txa5!
a nother experienced analyst, i nternational
An u n expected change of d i rectio n ! I n his master Gavri i l Veresov: In positions on the
commentary Smiri n considered only 34 border between a draw and a loss, we
l:'l.xg4 �c6 with a probable d raw. normally find there is only one solution.
34 . . . �c6
34 . . �e8! ? would seem to be more tena­
.
7 . Smyslov - Botv i n n i k (World Champion­
cious, aiming at the fi rst conven ient opportu­ ship Match , 3rd Game, Moscow 1 954 )
n ity to play . . . h7-h5 .
Although Black has th ree pawns for a piece,
35 �g5! his position is difficult. Vasily Smyslov could
The rook resembles a n a n noying fly. have decided the outcome in the middlegame,
35 . . . �f3 by breaking u p the opponent's pawn chai n
35 .. ..!:!.g8 36 l:f.h5 is no better. and open ing lines for his pieces b y h2-h3.
36 l:f.h5 27 Wkg2 !
The immediate 37 'it>d2 ! ? �f6 38 b4 is also Threatening 28 .Ue5 .
good . 27 . . . .Ufe8
36 . . . I:tf7 28 h3!
37 �d2 I n t he ga m e there followed 27 �e6+? �xe6
After skilfully tying down the enemy pieces , 28 .Uxe6.
White now wants simply to strengthen his Usually piece exchanges are the easiest
1 52 � Converting an Advantage

way of converting a material advantage. But play and weaken you r q ueenside pawns by
here, fi rstly, material is nominally balanced , 27 a3 i..f8 . The normal contin uation is 27 g3
and secondly (and this is more important), Wf7 . We note that the black king prevents
the fewer the pieces remai n i ng on the board , o ur rook from becoming active on the e-file,
the greater the role played by the pawns. whereas the black rook on the c-file, a long
28 . . .'it>f7 29 Itfe1 . way from the wh ite king, is very active and it
If 29 l:te5, then 29 . . . Itfe8, when 30 Itfe 1 restricts the wh ite pieces .
i.. c7 1eads to roughly the same position as in I myself wanted t o seize t h e c-fi le, Therefore
the game. And if 30 Itxd5 there follows I began checki ng 26 l:td 1 .
30 . . . Ite3 31 i.. b 1 ( 3 1 �d 1 'it>e6) 31 . . . .l:!.e2 26 �d1 ! 'it>f7
with sufficient cou nterplay for Black. Black hardly has the right to sacrifice a
29 ... �fe8 30 .l:!.xe8 .l:!.xe8 3 1 l:txe8 (31 l:td 1 second pawn by 26 . . . I:tc8 27 ..ll. xf6 M.c2 28
l:!.e3 32 'it>f2 l:th3) 31 .. .'lt>xe8. a4 bxa4 29 bxa4 (29 . . . i.. c5 30 i.. d 4; 29 .l::ia 2
. . .

White is not able to convert his extra piece , 30 .l:i.a 1 ) . In reserve I also had the transition
since his k i n g h a s nowhere t o break through i nto a bishop e n di n g: 27 l:!.c1 M.xc1 + 28
- the black pawns prevent this. But what a i.. x c1 f5 (28 . . . 'it>f7 29 'it>e2 'it>e6 30 Wd3 '.iid 5
wonderfu l target they presented in the 31 g4) 29 'it>e2 'it>f7 30 'it>d3 'it>e6 31 '.ii d 4,
middlegame! and i n all probabil ity Wh ite should gradually
32 i.. c 3 'it'd? 33 a5 i.. d 8 34 i.. b4 b6 35 a6 wi n .
i..f6 36 i.. c 3 'it>e6 37 'it>g2 g5 38 �e2 g6 39 27 l:!.c1 !
i..d 1 i..e 7 40 ii.. d 2 i.. d 8 41 ii.. e 3 . Now 27 . . . i.. d 2 is poi ntless in view of 28
Here the game was adjourned , and the .l:!.c7+ and 29 g3. White wants to calmly
players ag reed a draw without resuming. strengthen his position by g2-g3, l:!.c2 , and
Wg2-f3 ; his rook is consta ntly th reatening to
8. Dvoretsky - Zil berstein break i nto the opponent's position along the
(Ordzhonikidze 1 978) c-file. The resulting situation is more com­
To where should the rook move , e2 or d 1 ? fortable for Wh ite than after 26 l:i.e2 .
But isn't it all the same - after a l l , i n both The fu rther course of the game confi rmed
cases Wh ite remains a sound pawn to the that my eval uation was correct - the
good? But you should not approach the conversion of the advantage proved to be an
conversion of an adva ntage so frivolously - altogether easy matter.
otherwise very often disappointment will 27 . . . .l:!.d8 28 .l:!.c2 .Ud 1 + 29 'it>e2 l:!.e1 + 30
await you . You should try to d iscover the Wf3 llb1 3 1 i.. d 4! .l:i.d1 32 <lt>e4 a5 33 g4
difference between moves and choose the i.. d 6 34 l:!.c6 �e5 35 i.. x e5 (35 �e3
one which is i n some way better, more followed by f2-f4 is also strong) 35 .Me1 + . . .

accu rate than the other. 36 'it>d3 .l:!.xe5 3 7 f4 l:i.d5+ 3 8 'it>e4 l:i.d 2 39
If 26 l:!.e2 there follows 26 . . . �c8 , when 27 h4 Itxa2 (39 . . . h 5 40 g5 fxg5 41 hxg5) 40
.l::!.e 6 .l::!. c2 is pointless . Having an obvious Wf5 Itf2 41 Itxf6+ 'it>g8 42 .l:!.a6 Black
advantage, you don't want to compl icate the resig ned .
 ltJ 1 53

Artur Yusupov

Tech n i ca l P roced u res i n a

G ra n d m aste r Battle

Twas played
he g a m e which I would l i ke t o show you 9 l2lc2 l:te8
in a g randmaster tou rna­ Black is not in a h u rry to disclose his plans.
ment in the Spa n ish town of Linares. In its
1 0 i.. b 2 a5
i n itial stage the two players engaged i n a
Realising that for the moment the adva nce
difficult manoeuvring battle in a roughly
. . . e6-e5 is unfavourable, I ca rry out another
equal position . Then a n ending, slig htly
idea that is typical i n such positions, trying to
better for Black, was reached . It is i n struc­
' latch on' to the opponent's queenside. If
tive to follow those typical endgame proce­
now 1 1 a3 �b6 , and Wh ite has problems
dures, tha n ks to wh ich I was able fi rst to
with the defence of his b3-pawn .
increase, and then successfu lly convert my
advantage. 1 1 l:tb1 !
A deep prophylactic move . I n reply to
Salov - Yusu pov 1 1 . . . a4 , a pa rt from 1 2 b4 Black also has to
Linares 1 99 1 reckon with 1 2 bxa4 ! ? J::t x a4 1 3 i.. xf6 and 1 4
Reti Opening l:txb7.
1 l2lf3 l2lf6 11 . . . i.. h 5
2 g3 d5 11 . . . e5 is prematu re i n view of 1 2 cxd5 cxd5
3 i.. g 2 c6 1 3 tt'le3 (attacking the bishop) 1 3 . . . i.. h 5 1 4
4 0-0 ii.g4 l2lh4 . So why not retreat the bishop i n good
time?
5 c4
1 2 l2le3
A normal position for the Reti Opening has
arisen. In my view, 5 l2le5 ! ? i.. f5 6 c4 is Aga i n Wh ite prevents . . . e6-e5. For both
interesting , since i n the game after Black's sides it is difficult now to do anyth ing active .
reply the active knight advance is no longer As is usual in such situations, ma noeuvri ng
possible. beg i ns without any clearly defi ned pla n . The
two players merely operate with 'short'
5 . . . l2lbd7
positional or tactical ideas.
6 d3 e6
12 . . . i.. c 5
7 b3 i.. d 6
1 3 �d2
8 tt'la3
After 1 3 d4 i..f8 the e4-point is wea kened .
An unusual pla n . Now i n the event of . . . e6-
e5 the manoeuvre l2la3-c2-e3 will hi g hl ight 13 . . . �b6
a certain weakness in Black's central pawns; 1 4 a3!? �a7
however, if he avoids occu pyi ng the centre , Black i ntends in some cases to play . . . a5-
Wh ite's idea does not present any danger. a4. For example, if 1 5 l:tfd 1 there can follow
8 . . . 0-0 1 5 . . . a4 1 6 b4 i.. x e3 1 7 fxe3 dxc4 .
1 54 � Technical Procedures in a Grandmaster Battle

1 5 lLlc2 ii.f8 20 tL:lh4 ..th5

The bishop moves away from the tempo­ 2 1 lLlf5 ii.g6
gaining b3-b4 . 22 i.h3?!
1 6 ii.d4 The 'grandmaster d raw' after 22 lLlh4 would
1 6 lLle5 with equal ity came i nto considera­ have been the logical outcome. In his desire
tion. to play o n , Valery Salov comm its a sign ifi­
16 . . . �b8 cant inaccu racy - he loses control of the e4-
1 7 i.a1 point. I was able to exploit the 'hanging'
position of the wh ite pieces .
Wh ite thought that he had slig htly improved
the position of his bishop and worsened the 22 . . . a4!
position of the opponent's queen . The long-awaited advance!
23 cxd5
I had reckoned with this possibil ity and I had
prepared a simple i ntermediate operation.
But White had no choice: it was bad to play
2 3 bxa4? dxc4 24 dxc4 tt:Je4 (with the threat
of 25 . . . lLlg5 ) 25 'ir'xd7 'ir'xd7 26 lLlh6+ gxh6
27 i.xd7 .:.e7 28 .:.xb7 tL:lc5 , or 23 b4? dxc4
24 dxc4 lLle4 25 'iYxd7 (25 'ii'e 3 tL:lg5)
25 .. .'ifxd7 26 lLlh6+ gxh6 27 ii.xd7 .:.e7 28
i.h3 (28 .:.bd 1 lLlf6) 28 . . . lLld2 (28 . . . tL:lg5).
23 . . . axb3 !
24 �xb3
24 dxc6 bxc2 25 �xb 7 did not work in view of
25 . . . tt:Jc5 26 c7 itd5 .
17 . . . e5!?
24 . . . tL:lc5
After all these clever ma noeuvres I decided
25 .U.bb1 tL:lxd5
it was time for activity in the centre, since the
move lLlc2-e3 does not have to be feared - Now Black's position is preferable - the
the a3-pawn demands constant concern . opponent has a weak pawn on a3 .
Even so, this adva nce also has defi n ite 26 lLlfe3
min uses - it weakens the d5-pawn and the If 26 J:ifd 1 Salov was apparently concerned
f5-square. about 26 . . . e4! 27 d4 e3 ! .
1 8 lLlh4 1i'd8 ! ? 26 . . . 'it'g5
The opposition o f the queens is advanta­ Black contin ues the idea of exploiting the
geous to Black - in some cases the slight superiority of his pawn structu re.
undefended state of the wh ite queen may 26 . . . tL:lc7 ! ? 27 tL:lc4 f6 , mainta i n i ng the
tel l . tensio n , was also possible.
1 9 lLlf5 27 ii.g2 tL:lxe3
The position is roughly equ a l . The slight 28 �xe3
pressu re of the wh ite pieces is neutralised 28 h4!? ca me i nto consideration . In the
by Black's superior pawn formatio n. event of 28 . . . 'ir'h6 29 tL:lxe3! .l:i.xa3 Wh ite's
19 . . . ..tg6 pawn deficit would have been compensated
 Technical Procedures in a Grandmaster Battle a 1 55

by the bad position of the black q uee n , while 37 �e1 tt::'l c 5

after a d ifferent retreat he would have All the same the knight ca n not be main­
captu red on e3 with his q uee n , avoid i ng a ta ined at b3, and Black switches it to a4,
further spoiling of his pawn structure . intending to advance h i s queenside pawns.
28 . . . 'i!Vxe3 38 .Ub4 tt::'l a 4
29 fxe3 39 �d2? !
A natu ral move , b u t n o t t h e best, since it
does not prevent Black from ca rrying out his
pla n . 39 lldb 1 ! ii.a2 40 .Ua 1 ii.e6 4 1 .U ab 1
w a s stronger, w h e n t h e pressure on b7 does
not al low . . . c6-c5 to be played .
39 . . . c5
40 Ubb1 ?
A time-trouble error. After 40 .Ub5! .i.e8 4 1
l:tbb 1 b 5 Wh ite could have i ncluded his
bishop i n the defence: 42 ..lli. d 5 ! .
40 . . . l:i.d7!
41 l:tdc1 c4
Here Salov spent a lot of time, appa rently
Of cou rse, the endgame is more pleasant weig h i ng u p wh ich was the lesser evi l - the
for Black, but it is not easy for him to loss of a pawn or passivity, - and he chose
increase his adva ntage. the latter. In the variation 42 Ub5 cxd3 43
29 . . . f6 tt::'l b 4 (or 43 exd3 ..lli. c4 44 .Uxb7 .Uxd3+)
Black rei nforces his e5-pawn and prepares 43 . . . dxe2+ 44 �xe2 Wh ite does not have
.. . i.f7. ful l compensation for the pawn .
30 ..lli. c 3! Ua7 42 tt::'l b4 tt::'l c 5
31 �b4 ..lli.f7! 43 .Uc3
32 �f2 Now 43 . . . tt::'l b 3+ 44 �e 1 Uxa3 is u nfavou r­
Salov defends in accordance with all the able because of 45 dxc4 .
rules of the endgame - he covers his
weaknesses and brings h i s king towa rds the
centre .
32 . . . tt::'l b 3
33 .i.xf8 �xf8
34 1:i.b2 Ue7
When short of time it is usefu l to over­
protect the important b7-pawn .
35 �fb1 tt::'l d 2 !
I n time-trouble Black gains t i m e on t he clock
by repeati ng moves.
36 Ud1 tt::'l b 3
1 56 � Technical Procedures in a Grandmaster Battle

43 . . . e4! 52 l:tf1 followed by l::tf4 . I n the event of 50

Formally the move made by me is against lZJc3 lZJxc3 5 1 l::tx c3 b5 (and then . . . l::td a?)
the ru les (pawns are supposed to be kept on Black is effectively a piece u p .
sq uares of the opposite colour to you r own 50 . . . �g7
bishop), but in fact it is very strong , since it 51 .l'::f.f1 .i.e6
shuts in the bishop on g2.
I decided not to defend the h4-pawn with the
44 d4 king , fearing that after 5 1 . . . �h6 52 .l:If4 �g5
Forced . 53 gxf5 gxf5 54 I::f. c 1 it would be attacked by
44 . . . h5! the rook from g 1 . I n stead of this Black
44 . . .'�Jb3+ 45 'it>e 1 �xa3 46 .lli. x e4 l:!.a4 also exploits the departure of the white rook from
looks good , d riving back the knight and then the q ueenside and beg i n s decisive action
adva ncing the pawns. But i n th is case the there . All fully in accordance with the
wh ite pieces would have become active . I principle of two weaknesses.
preferred to play for a second weakness in 52 l:tf4 ctJb6
the opponent's position - sh utti ng the 53 gxf5 gxf5
bishop out of play (the fi rst weakness is the
54 .l:!.xh4
a3-pawn ).
45 'it>e1
45 h3 was better, in order after 45 . . .f5 to
have the reply 46 g4 (or 46 .:f1 g6 47 g4).
45 . . . lZJa4
46 .l:i.c2 f5
The bishop on g2 has ended up on a square
from which it is unable to escape. After the
game Salov heated ly exclaimed that it
wou ld have been better if it hadn't existed at
all - then he could at least have tried to
obtain some cou nterplay on the kingside.
47 h3
Trying to activate the bishop.
47 . . . g6! 54 . . . c3+!

The last black pawn occu pies a square of The q u ickest way of converti ng the advan­
the same colour as its bishop. Rules a re tage. The pawn moves i nto a th ree-fold
rules, but concrete considerations come attack, but noth ing can captu re it. For
first! It is important to be able to answer g3- example, if 55 ttJxc3 , then 55 . . . lZJc4+ 56
g4 with . . . h5-h4! �c1 lZJxe3 is decisive .
48 lZJa2 l::. a 5 55 '.t>c1 .lli. b 3
49 g4 h4 56 lZJxc3 .lli. x c2
The trap has snapped shut! 57 '.t>xc2 .l:!.xa3
50 'it>d2 I was expecting my opponent to capitulate,
Probably the only chance . Wh ite i ntends but unexpectedly Wh ite sacrificed his knight.
play against the h4-pawn with 5 1 gxf5 gxf5 58 ttJxe4 fxe4
 Technical Procedures in a Grandmaster Battle lZJ 1 57

59 .i.xe4 l:txe3 the q uestion: why doesn't the opponent

60 .i.d3 l:tg3 resign? In the event of 67 . . .l:.xd4+?? White's
.

61 <iii' d 2 last trap would have worked - 68 <iii' c 3 l:ta4

(or 68 . . . .:Z.fd5) 69 l:tg4+! .
The time control was reached , and I
immed iately decided to seal my move , to 67 . . . l:tf1
avoid any chance accidents at the board . Of 68 d 5 .l:r.c7
cou rse, with an extra rook I could have Wh ite resig ned .
continued playing, but it is my conviction
that in a won position one should not do this. In my view, the conversion of the advantage
Tiredness after six hours' play sometimes i n this game was rather instructive . Black
leads to mista kes such as 61 . . lt:'Jd5?? 62
.
made use of severa l i m porta nt proced u res
:Ih7+ . in such endings:
61 . . . <iit g 8
- when short of time he repeated moves,
I should mention that, despite my extra rook, and later too did not h u rry to force events,
I analysed the adjourned position q u ite but g rad ually strengthened his position;
accurately, to make my l ife easier d u ring the - not restricting myself to a d i rect attack on
resumption . After a l l , a few pitfalls stil l one weakness (the a3-pawn ), I tried to
remained . operate broadly, breaking u p the opponent's
62 <iii' c 3 defences from d ifferent sides: play against
Not the most tenacious. Now Black forces the d3-paw n , restriction of the l ig ht-square
the exchange of the minor pieces. bishop - the point is that d u ring a game it is
62 . . . lt:'Jd5+ very d ifficult to switch from the defence of
one point to another (especially in time­
63 <iii' c4 lt:'Je3+
trouble);
64 <iii' c 5 l:tg5+
- after obtaining a won position , Black was
65 <iii' b4 lt:'Jf5
not in a h u rry to mentally chalk up a point in
66 .i.xf5 l:.xf5 the tou rnament table, but he contin ued
67 e4! ? playing carefully, maintain ing his vigilance to
I n such situations it is usefu l to ask you rself the end.
1 58 \i?

Mark Dvoretsky

Lessons from One Parti c u l a r E ndgame

I straight away, because fi rst I should l i ke to

wi ll not come to the promised endgame their attention to variations, d isregardi ng
with their eyes (or ears) the a rgu ments of
discuss certain aspects of studying chess i n the author. I am convinced that in this way
general. they lose much - after a l l , often the most
At the boa rd w e operate with moves and valuable i nformation is concentrated in the
variations, but they are based on our word s . Sometimes it is worth dwelling even
u nderstanding of the game, the develop­ on simple, seemingly banal th ings - by
ment of which depends considerably on repeati ng them, and di scovering new as­
study and training work carried out earlier. pects to them, you reinforce your u nder­
So that this work should be prod uctive , it is sta nd i ng of chess.
insufficient merely to memorise specific It sta nds to reason that i n fact everything is
information - it is importa nt that on the basis fa r more complicated than it appears on
of it chess images arise. The most vivid paper. Most commentaries in chess maga­
images, which engrave themselves for a zines and books are su perfici a l , and some­
long time in you r memory, a re orig inal and times no more than hack-work. Once an
deep general ideas, manifested in clear, experienced master explai ned to me how he
convincing variations. worked : If you place two fi ngers over a page
Many ideas that are valuable for our self­ of text and you see that u nder them there
improvement are scattered about in com­ a re only moves, it mea ns that it is time to
mentaries to games, written by g reat play­ g ive a com ment. You write someth ing l i ke
ers. When studying such commenta ries I 'The Ruy Lopez always leads to a tense and
often look at the words even more than at complicated struggle' , and you r fee in­
the moves. As soon as I see an idea which creases by a rouble.
seems orig inal and i nteresting , and in some The ability to d istinguish real feelings and
way new for me, I immediately record it thoug hts from such verbal rubbish will come
together with the position in which it oc­ in useful to you not only in chess.
curred . I also record examples which suc­ Often the reverse picture occu rs . The author
cessfully demonstrate rules and eval uations seemingly has some i nteresti ng ideas, but
which have long been familiar to me - after he is not capable of i l l ustrati ng them with
all, it is useful to repeat them from time to worthy examples. If a g rand master anno­
time, and if possible in a vivid and memora­ tates h i s own game, this problem normally
ble form . As a result I have been able to does not arise: his general thoug hts are
accu mulate a qu ite extensive collection of closely l i n ked with what happened on the
the most d iverse chess ideas, illustrated by board . But as soon as he decides to write an
excellent examples. article or book on some extraneous topic,
Young players , when read ing books or problems immediately arise, since he may
l istening to lectures, usually devote most of not have appropriate material to h a n d .
 Lessons from One Particular Endgame ctJ 1 59

I remember looki ng through a book by from even a comparatively small amount of

Alexey Suetin Put k masterstvu (The path to material you will be able to extract a great
mastery). The head ings of many chapters deal of interesting and val uable i nformation.
seemed i nteresting , for example: ' Play by
analogy', ' U npromising positions', ' Loss of
Capablanca - Alekh ine
consistency' , ' P roblems of time utilisation
New York 1 924
when choosing a move ' , and so on. Here
some gen u i nely i m portant q uestions of
chess mastery were ra ised . The book would
have been excellent, if it had also been
possible to obta i n answers to them. Alas, in
fact it was hack-work, l i ke, i n my view, nearly
all Sueti n's books. The author delved i nto
hardly any of the problems raised . Most of
the examples were either feeble, or superfi­
cially analysed , and were often very weakly
l i n ked to the topic i n question. And without
adequate analytical material it is i m possible
to draw informative conclusions. Besides,
where was he to find good material? - after
all, at some point Suetin practically gave u p
serious play a n d d i d not conduct any real It is Wh ite to move . H e is a pawn up, but this
training work. Of cou rse, something re­ advantage is not easy to convert (remember
mained in his memory - and thus to the the semi-joki n g, semi-serious aphorism of
chapter head ings he attached the fi rst Siegbert Ta rrasch : ' Rook endings a re never
episodes that came i nto his head , if they won ' ) . Let us decide what candidate moves
were even slig htly appropriate . You look at
(or more precisely, what candidate plans)
the head ings and you become curious about
we have available. It is useful to i mmed iately
how the author u nderstands the g iven gather as many ideas as possible - other­
problem . You read fu rther, and you see : he wise, by delving too early i nto calculation , it
doesn't understand it at all, and gets by will be easy to miss someth ing that is i ndeed
merely with general words. im porta nt.
Probably the correct order of work is not
39 c5 immediately suggests itself. The
from topic to example, but vice versa: from
threat is 40 l:td6+ , winning the c6-pawn .
an interesting, wel l-analysed example to the
general isations stemming from it. This is The second suggestion is 39 '.td4 , in the
how we will study the classic ending which I hope of breaking through with the king to c5 .
offer for you r attention - i ncidentally, one of There is also a nother pla n : 39 h4 with the
my favou rites. idea of 40 g4, 41 .l::!. h 2 and so on - the wh ite
The heritage of the famous masters from the rook will occu py an ideal position to the rear
past provides an i nvaluable sou rce for self­ of the passed h-pawn .
improvement. It is important only that you do As you see, Wh ite has n u merous tem pti ng
not restrict you rself to rapidly playing the possibilities. In order to make the correct
book variations on the board , but endeavo ur choice, it will be essential to take i nto
to check and understand t he m . And then acco unt the opponent's cou nterplay.
1 60 � Lessons from One Particular Endgame

Let's take them in order and beg in with 39 fi nd a way to prevent the opponent's main
c5. If 39 . . :lt>e5?! there follows 40 l:Id7. I n the d efensive idea.
event of 39 . . . l:.b4?! noth ing is given by 40 Alekh ine suggests the surprising move 39
l:.d6+ 'it>e5 41 l:.xc6 l:.xe4+ and 42 . . . l:.a4, h 3 ! ! . Now if 39 . . . l:Ih8 the h-pawn is not
but 40 �f4 ! followed by 4 1 l:.d6+ is far more hanging and Wh ite replies. 40 c5. After
dangerous. The best defence was sug­ 40 . . . I:th4 the reply 4 1 .l:!.d6+ 'it>e5 42 l:Ixc6
gested by Alexander Alekh ine: 39 . . . l:.b5! 40 .l:!.xe4+ and 43 . . . .l:!.a4 is unconvincing, but 4 1
.l:!.d6+ 'it>e5 41 l:!.xc6 (4 1 l:!.d7 l:!.a5 or l:Id8 ! is very strong. A t t h e s a m e time Black
4 1 . . . l:!.xc5) 4 1 ... l:!.a5 . With such an increase must now seriously reckon with 40 'it>d4 , for
in the activity of his pieces, Black should not example: 39 . . . .l:!.b 1 (b4) 40 '>t>d4 '>t>d6 41 e5+,
lose. or 39 .. .f6 40 'it>d4 .ll d 8+ (40 . . . '>t>d6 4 1 c5+
Now let's examine 39 Wd4. Obviously, the 'it>e6 42 Wc4 ) 41 �c3 .Ub8 42 c5 'it>e5 43
king cannot be allowed to go to c5 . l:Id6 with an obvious advantage. 39 . . . 'it>e5 is
39 . . . 'it>d6? 40 e5+ does not help, and d a ngerous because of 40 l:Id7 . There only
therefore the reply 39 . . . l:.d8+ is forced . After remains 39 . . . c5 40 l:Id5 ( if 40 h4, then
4 1 'it>c3 the th reat of c4-c5 has become 40 . . . .ll b4 ! , but not 40 . . . l:Ih8 41 g3 l:Ih5 42
more serious, since now the c5-pawn may .l:!.h2 , and the 5th ra n k is too short) 40 . . .l:!.b2
.

be defended by the king . However, it is not 41 g4 (4 1 .l:!.xc5 .l:!.xg2 42 l:Ia5 is also good )
hard to forestal l Wh ite's plan: 41 . . . .l:!.h8! 42 4 1 . . . .l:i.xa2 42 l:!.xc5 .l:!.a3+ 43 'it>d4 .l:!.xh3 44
h3 .l:!.h5 (42 . . . .l:!.h4 also comes i nto consid­ .l:!.a5 with excellent winning chances for
eration). The rook is wel l placed on the 5th Wh ite.
rank - it controls the c5-sq uare (if 43 'it>b4 It is significant that Jose Raul Capablanca ­
a5+) and is able to attack any of the enemy a player with bri l l iant i ntuition - was unable
pawns. It is evident that Wh ite has not to come to the correct decision , suggested
ach ieved much . by Alekh ine - a player with a totally different
It remains to verify 39 h4. The reply way of th i n k i n g. A move such as 39 h3!!
39 . . . .l:!.h8! suggests itself (39 . . . f5? 40 exf5+ can n ot be called i ntu itive, based on 'general
is bad for Black). Wh ite plays 40 g3, considerations' - it could be fou n d only after
preparing 4 1 l:!.h2 and 42 g4 . How can this a deep and very concrete penetration i nto
plan be countered? Black is saved by the the secrets of the position .
same rook manoeuvre: 40 . . . llh5! 41 llh2
Many years ago I was helping Botvinnik by
l1a5 ! . Now 42 g4? is u nfavourable because
giving some lessons at his schoo l . On one
of 42 . . . 'it>e5 43 h5? l:.a3+ and 44 . . . l:Ixa2+, occasion , at the request of Mikhail Moisee­
and if 42 �f4 there follows 42 .. .f6 ! , prepar­ vich , I prepared an extensive endgame
ing in the event of g 3-g4 to exchange the
lesson for the young Ga rry Kasparov, which
opponent's most dangerous pawn by . . . g6-
included in particular a n independent analy­
g5+ ! .
sis of the Capablanca-Aiekhine ending.
We have establ ished that Wh ite does not Garry fou n d a nother way of forestall ing the
achieve anyth ing with the d i rect i m plemen­ switch ing of the black rook to the 5th rank ­
tation of any of our i ntended plans. H ow can the move 39 g3! ! . It appeals to me perhaps
he nevertheless conti nue playing for a win? even more than Alekh ine's recommenda­
Note that everywhere Black was saved by tio n , since it contains an add itional active
the switch ing of his rook onto the 5th rank. idea : 40 h4 ! . And there do not appear to be
Let's remember about prophylaxis and try to any d rawbacks : for example, if 39 . . . g5 there
 Lessons from One Particular Endgame ctJ 161

i s a pleasant choice between 4 0 h 4 a n d 40 45 l:.c3

l:.f2 with the threats of 41 .l:.f5 or 4 1 <it?d4 . Threate n i n g an exchange of pawns advan­
tageous to Wh ite : 46 lta3 l::t x c5 47 l:.xa7.
Now let us see how the game went. 45 . . . aS!
39 h4? ! .l:.h8 46 .l:.c2 l:te5
40 g3 .l:.h 5 ! 47 l:lc3 l:1h5
41 l:.h 2 .tla5 48 <it?f3 ! <it?e7 !
42 <it?f4 Both 48 . . . <it?e5? 49 l:ta3 and 48 . . . .l:.e5? 49 g4
42 g4? <it?e5 ; 42 <it>d4? c5+ . would have been i n correct.
49 <it?g4!
42 . . . f6!

Wh ite wa nts to strengthen his position by

The main danger has been elimi nated - if
<it?h3 and g3-g4 . What can be done to
43 g4 there is the reply 43 . . . g 5+ ! . The play
oppose this plan?
takes on a manoeuvring character. Capa­
blanca skilfully sets his opponent one 49 . . . <it?f71
problem after another, so that Alekh ine is In reply to 50 <it?h3 Alekh ine had prepared
required to defend with exceptional care . 50 . . . g5! 51 �g4 <it?g6 . Then he excha nges
pawns on h4 and oscillates with his rook
43 l:tc2 l:f.e5
between e5 and h 5 .
Otherwise after 44 c5 the rook would have
5 0 I!c4! <it>g7 !
been cut off from the kings ide and would no
Wh ite's subtle ma noeuvres have forced the
longer have been preventi ng Wh ite from
playing g3-g4 and h4-h5 . black king (which is obl iged to control the
g6-square) to move away from the centre .
44 c5
Capablanca sees that the most appropriate
A double-edged move , but otherwise the moment for transforming his advantage has
position can not be strengthened . Wh ite arrived . He gives u p his extra pawn , but
restricts the mobil ity of the enemy rook , but activates his rook to the maxim u m and
his own rook will be tied to the defence of the d rives the opponent's king onto the back
c5-pawn. ran k .
44 . . . l:th5 51 .l:td4! I!xc5
1 62 Lessons from One Particular Endgame

52 l:td7+ 'it>f8 no reason to g ive the opponent a passed

52 . . . Wh6? would have been risky: 53 I:!.f7 pawn .
.ti.c4 54 'it>f3 �c1 55 'lt>e3 .l:i.f1 56 We2 l:ta 1 57 58 .i::t a 6 .l:!.c5
�f6 .l:!.xa2+ 58 �3 . 59 We3 'it>f7
53 'lt>f4 60 'lt>d4 .ti.g5
If 53 l:ta7 there is the reply 53 . . . l:tc4 ! ? , but 6 1 ltxc6 .l:!.xg4
after the move i n the game Black could have 62 .l:!.c5 .ti.g5!
played 53 . . . .l:i.c2 ! ? .
In this position the players ag reed a draw in
53 . . . 'it>g8 view of the variation 63 l:txg5 fxg5 64 'it>e5
54 l:ta7 'it>f8 <t;g6! 65 rt;d6 cj;;f7! (after 65 . . . g4 66 e5 Black
55 a4! 'it>g8 would sti l l have had to defend a q ueen
ending) 66 'lt>e5 (66 e5?! We8 ; 66 Wd7 'iii> f6)
66 . . .'it>g6! .

With what topic should the ending of the

Capablanca-Aiekhine game be linked? Af­
ter a l ittle thought you will see that there is
no clear a n swer - i n the process of studying
the ending various aspects have come to
lig ht, identically important for the practical
player. Let's remember what we have seen:
1 ) Excellent example of a practical rook
endgame. Among the n u merous evalua­
tions and methods typical of this type of
ending, with which the two players operated,
I will single out a comparatively less trivial
Wh ite has done everything possible to
idea , which is very clearly expressed here.
strengthen his position and now is the time
An open line, for wh ich a rook should aim,
for decisive action . The logical consequence
may be not only a file, as usual, but
of his preced ing strategy would have been
sometimes also a ra nk.
the variation 56 'it>e3 ! .ti.c3+ 57 'it>d4 .l:!.xg3 58
.ti.xa5 cj;;f7 (in the event of 58 .. J:tg4 noth ing 2) Model example of accu rate defence. It
is given by 59 .ti.a7 .l:!.xh4 60 a5 .ti.h5! , but the is i nstructive to follow how Alekh ine, without
reply 59 h5 is unpleasant) 59 .ti.a8 or 59 h5. losing his presence of mind in a difficult
I n Alekhine's opinion, Black is able to hold situation , move by move patiently resolved
the position , but at any event he would have the problems facing h i m .
had to defend accu rately. 3) Various aspects o f t h e problem of
converting an advantage. Here we can
Unfortu nately, Capablanca did not want to
sharpen the play and he chose a d ifferent mentio n : the i m porta nce of looking for a n d
plan, which leads by force to a d raw. forestalling t h e opponent's cou nter-chances
(at the very start of the ending); the
56 g4 g5+!
maxi m um strengthening of Wh ite's position
57 hxg5 l:txg5! before changing the pattern of the play; the
Of cou rse, not 57 . . . fxg5+? 58 We3 - there is timely tran sformation of an adva ntage (the
 Lessons from One Particular Endgame lZJ 1 63

5 1 st move) ; fi nally, the need at some point tainly been trying very hard in this game, so
(the 56th move) to abandon positional as to draw nearer to Dr. Lasker, who was in
manoeuvring and choose a concrete cou rse, the lead, and who had won against me the
involving precise calcu l ati o n. previous day. I was convinced that if I had
4 ) Demonstration o f t h e i mportance of been in Capablanca 's position I should
prophylactic th inking. Without it, of cou rse, certainly have won that game. I had finally
it is not possible to fi nd the brill iant solution detected a slight weakness in my future
to the position on the 39th move . And opponent: increasing uncertainty when con­
subsequently too Alekh ine's defence was fronted with stubborn resistance! Of course I
based on taking account of all the oppo­ had already noticed Capablanca committing
nent's active plans and forcefully opposing occasional slight inaccuracies, but I should
them. not have thought that he would be unable to
rid himself of this failing even when he tried
5) Grounds for reflection about chess
his utmost. This was an exceedingly impor­
players with an i ntu itive way of th inking.
tant lesson for the future!
We have seen which decisions a re d ifficult
for them or altogether i naccessible. The Later, in a fa mous a rticle 'The 1 927 New
conclusion suggests itself, that even if you York tou rnament as a prologue to the battle
possess splendid i ntuition, you should de­ i n Buenos Aires for the world championsh i p ' ,
velop in you rself the abil ity to consta ntly Alekh ine once again emphasised t h e role
delve into the concrete details of the position that the game with Capablanca had played
and if necessary to accu rately calculate for h i m :
variations. This game, incidentally, was the starting
For a chess player it is very importa nt to point for my understanding of Capablanca 's
evaluate objectively the strengths and weak­ chess individuality.
nesses of a forthcoming opponent. A sou rce I will also g ive some other assessments by
of such evalu ations is provided by an Alekh ine of the style of his historic oppo­
analysis of games played by him. Some of nent, wh ich a re confi rmed by the ending we
the m will prove especially i nformative. have examined . They may seem exces­
I n the 1 920s Alekh ine was preparing for his sively sharp, but to some extent th is is
duel for the world crown against Capablanca . explai ned by the very tense personal rela­
This is what he recorded for h i mself after the tions which developed between the two
New York tou rnament of 1 924: champions. But objectively these assess­
ments seem to me to be just (of course, only
I took home with me from this tournament
'on a g rand scale' - ta king i nto account the
one valuable moral victory, and that was the
very high sta ndard of play in questio n ) .
lesson I learned from my first game with
Capablanca, which had the effect of a . . . Capablanca is by no means an excep­
revelation on me. Having outplayed me in tional master of the endgame; his skill in this
the opening, having reached a won position stage of the game is mainly of a technical
in the middlegame and having carried over a character and other masters in certain fields
large part of his advantage into a rook of the endgame surpass or used to surpass
ending, the Cuban then allowed me to him (for example, Rubinstein in rook end­
neutralise his superiority in that ending and ings).
finally had to make do with a draw. That . . . In Capablanca's games with the years
made me think, for Capablanca had cer- one observes increasingly less delving into
1 64 � Lessons from One Particular Endgame

the details of the position, and the reason for helpless weapon. On account of this 'lack of
this is his unshakeable (I am talking all the punishment' in employing not the best
time about the period before Buenos Aires) moves, he, on the one hand, got out of the
confidence in the infallibility of his intuition. habit of that concentration of thought during
The saddest thing for Capablanca is that this a game, which alone can give a guarantee
system of his of operating with 'good' moves against possible elementary oversights, and
almost without exception proved sufficient, on the other hand- his self-confidence grew
since to a great extent he was opposed in immeasurably and turned almost into self­
the positional sense by a more or less worship . . .
 ltJ 1 65

Mark Dvoretsky

G ra n d m aster Tech n i q ue

I London Chess Centre and showed the

n August 2005 I gave a lectu re in the

fol lowing position, ta ken from the magazine

64 Shakhmatnoe obozrenie - the ending of
the game was publ ished there with notes by
the winner Evgeny N ajer.

Yandem i rov - Najer

Russian Club Champions h i p ,
Dagomys 2004
11 ... d 5 !
A clever blow i n t h e centre , fi rst employed , I
th i nk , by Vasily lvanch u k against Alexander
Delchev i n the 2003 European Champion­
s h i p . In this way Black solves his opening
problems - i n fact, it is now his opponent
who has to act ca refully, to avoid getting i nto
d ifficu lties.
12 exd5 ( 1 2 e5 is d u bious i n view of
1 2 . . . tt:leB 1 3 tt:lxd5 e6 1 4 tt:lc3 .1Lxe5)
1 2 ... tt:lxd5 1 3 tt:lxd5 e6 14 .i.h6 exd5 1 5
.1Lxg7 'it>xg7 1 6 cxd5 ( 1 6 tt:lb5 ! ? ) 1 6 . . . 'iVxd5
1 7 tt:le2
Here peace was concluded in the game
Grandmaster Jonathan Rowson , who was Gdanski-Kempi nski from the 2004 Polish
present at the lectu re , su rprised me by Championship, and a move (and a year)
remarking that this position is reached more earlier i n a game Yandemirov-Biryu kov. And
or less by force in one of the modern in genera l , looking i n a computer database, I
open ing variations, and that he h imself had saw that in most of the games where th is
once played it. variation occu rred there was effectively no
1 e4 c5 2 tt:lf3 d6 3 .i.b5+ .i.d7 4 .1Lxd7+ play - the contesta nts ag reed a d raw
'i'xd7 5 c4 tt:lf6 6 tt:lc3 g6 7 0-0 .i.g7 8 d4 somewhere between the 1 1 th and 23rd
cxd4 9 tt:lxd4 0-0 1 0 f3 IreS 1 1 b3 moves.
1 66 � Grandmaster Technique

In the summer of 2003 I publ ished an article After the game Valery Yandemirov sug­
in the Russian newspaper Shakhmatnaya gested that he should have played 21 f4. Of
nedelya and on the Chesscafe site (it was cou rse, advancing the f-pawn is not some­
also publ ished somewhere else on the thing that one wants to do - one ca n decide
I nternet) with a suggestion that a ru le should on this move only after employing prophy­
be introduced forbidding conversations be­ lactic thin king and clearly appreciating the
tween players d u ring play a nd , hence, d anger th reatening Wh ite.
prematu re d raw agreements . Two years 21 . . . .l:r.xc1
later my suggestion was successfully adopted 22 .l:r.xc1 .l:r.d8
at the super-tournament in Sofia. Its un iver­
sal adoption would, I am sure, not only
lengthen games, but also lower the percent­
age of draws. As we will now see , even i n
such a q u iet and seemingly l ifeless situa­
tion, where a draw is indeed the most
probable outcome, it is possible to seek
resou rces and pose problems for the oppo­
nent.
17 .. .'it'e5 1 8 'it'd4 tt:Jc6 1 9 'ifxeS+ tt:Jxe5 20
�ac1
We have reached the position in the fi rst
diagram. The at first sight mysterious move
found by grand master Najer is the strongest
- it was thanks to it that he won the game. Now after 23 �c7 .l:r.d2 the only way of
20 . . . b6! ! avoiding the loss of a pawn is 24 tt:Jc3 (bad is
Let us try and reprod uce Black's logic. Fi rst 24 tt:Jc1 ?? .l:r.d 1 + 25 �f2 �xc 1 ), wh ich allows
he probably checked 20 . . . tt:Jd 3 and real ised the u n pleasant pin 24 . . . .U.c2 . The situation
that after 21 .l:r.xc8 .l:r.xc8 22 .l:r.d 1 the position after 25 tt:Jb5 .t!.xa2 26 �xa 7 �b2 looks
was eq ual. Then his attention was d rawn to dangerous for Wh ite: the b3-pawn is u nder
the possibil ity of 20 . . . .l::.x c1 21 �xc1 �d8 attack, and he has to reckon with the
with the idea of 22 . . . �d2 . Najer is an ma noeuvre . . . tt:Jd3-f4(e 1 ) .
experienced player and he knows that i n the He sho ul d probably have restricted himself
first insta nce you must check active replies to the accu rate move 23 .l:r.c2 . But one does
by the opponent - in the given case 22 �c7 . not want to place the rook passively,
There appears to be noth ing better than especially since after 23 . . . .l:r.d 1 + 24 'it>f2
22 . . . .l:r.d2 23 �f2 .l:r.xa2 24 .l:r.xb7, but here tt:Jd3+ the king has to be moved to the side -
Black's advantage is of a purely academic 25 'iit g 3, since in the event of 25 'it>e3?! tt::le 1
natu re, and he has practically no chances of 26 .l:r.c7 tt:Jxg2+ 27 'it>f2 .U.d2 28 .t!.xa7 ctJf4 29
success. As was confi rmed by the game Li 'it>e3 .l:r.xe2+ 30 'it>xf4 l:!.xh2 Wh ite comes out
Ruofan-Rowson , 2004 . a pawn down .
But after . . . b7-b6 in this variation Black It is q u ite probable that after 23 �c2 , and
would remai n a pawn up, since from a2 his perha ps also after 23 l:!.c7 , the position
rook defends the a7-pawn . would objectively have remai ned d rawn . But
21 .l::tfd 1 ? ! it is one th ing to calmly analyse at home,
 Grandmaster Technique ttJ 1 67

and with the aid of a computer, a nd q u ite �e 1 tt::l x a2 32 �d2 tt::l b 4 (and 33 . . . tt::l d 5) with
different to take a decision at the board . A an appreciable advantage.
forced d raw is not apparent, defending is 24 . . . tt::l d 3
unpleasant - i n such a situation it is easy to
25 .Ue2 l:txe2
lose you r bea rings.
26 tt::l x e2 �6
23 tt::l c 3?
27 tt::lc 3 'it>e5
A serious mistake! Wh ite wants to exchange
28 �f1 a6
the rooks, but fails to take acco unt of the fact
that the black king will be the first to reach Fro m the previous note it follows that
the centre . In a knight ending a more active 28 . . . �d4 was also possible, but i n this
king is a very important factor. situation the move i n the game is, of cou rse,
more accu rate.
23 . . . �d2
29 �e2 tt::lf4+
30 �d2
If 30 �2 . then 30 . . . �d4 is strong, as is the
suggestion of Ca rsten M u l ler: 30 . . . f5 ! ? , and
if 31 g3? �d4 ! .
30 . . . tt::l x g2
3 1 tt::la 4 tt::l h 4
32 'it>e2 b5
33 tt::lc 5

24 .Ue1 ? !
Obviously, w h e n he m a d e h i s previous
move , Wh ite was intending the manoeuvre
.l::!.e 1 -e2 . Of cou rse , the enemy rook on the
2nd ran k can not be tolerated , but he should
have exchanged the rooks i n a slig htly more
favourable way: by 24 �f1 (with the same
idea : �e 1 -e2) 24 . . . tt::ld 3 (24 . . . f6? 25 tt::l e 4)
25 .l:td 1 .l:txd 1 + 26 tt::l x d 1 �f6 27 tt::lc 3 �e5 . I n
the game this same position was reached
with Black to move . 33 . . . a5!
White can make use of his extra tempo by N ajer was not satisfied with the variation
playing 28 g3 (28 'it>e2 is evidently weaker: 33 . . . �f4 34 tt::lx a6 tt::l xf3 35 tt::l c7 tt::l x h2 36
28 . . . tt::lf4+ 29 �f2 a6 30 g3 'it>d4! 31 tt::l e4 tt::lx b5, i n which the play becomes sharper.
l2Jd5). But here too Black has a pleasant I n advancing his a-pawn he undoubtedly
choice between 28 . . . a6 29 'it>e2 'it>d4 30 foresaw the tactical subtlety o n the fol lowing
Wd2 tt::le 5 31 tt::le 2+ �c5 32 �e3 �b4 a nd move , which sign ificantly facil itates the
28 ... �d4 ! ? 29 tt::l b 5+ �c5 30 tt::l x a7 tt::l c 1 3 1 conversion of his advantage .
168 Grandmaster Technique

34 a4 'it>d5! Stean - Hort

I n this way Black ensures the safety of his Biel 1 98 1
important a5-pawn .
35 tt:Jb7 'it>c6
36 tt:Jd8+
36 tt:Jxa5+ Wb6 37 b4 bxa4 was completely
hopeless for Wh ite.
36 . . . 'it>c5
37 tt:Jxf7 bxa4
38 bxa4 'it> b4
39 tt:Je5 '.txa4
Black is a sou nd pawn to the good . The
outcome is decided .
40 'it>d3 'it>b3 41 tt:Jc4 a4 42 tt:Ja5+ Wb4 43
tt:Jc6+ 'it>c5 44 tt:Je5 'it>d5 45 tt:Jc4 tt:Jxf3 46
tt:Jb6+ We5 47 'it>e3 tt:Jxh2 48 tt:Jxa4 'it>f5,
of the opponent's possib i l ities, and the
and Black won .
suppression of a ny cou nterplay, any useful
It was pleasant for m e to read the fol lowing operations a i med at improving his own
comment by Najer about the move 20 . . . b6! ! : position . To put this principle successfully
i nto effect, use m ust be made of 'prophylac­
A useful move, from which one can identify a
tic th inking ' .
pupil of Mark Dvoretsky or at least a careful
reader of his books. Let us a s k ourselves what Wh ite wa nts to
play. H i s choice is l i m ited . There is no point
Evgeny Najer twice pa rticipated in my
in attacking the knight: 29 �d2 lle1 + 30 'it>b2
training sessions and has stud ied my books
�e5 . Obviously the only sensible operation
- obviously, his work has not been in vai n . I
is to bring the king to the centre : �b1 -c1 -
th ink that the g randmaster's com ments
d2-d 3 . It is t h i s that s h o u l d be prevented .
were evoked by the close similarity of this
28 . . . l::t e 1 + !
ending with one of the examples g iven i n my
book School of Chess Excellence 3 - 29 'it>b2 h5
Strategic Play, which I should now l i ke to
show you .
(see diagram)

Black's positional advantage is determ i ned

by the active placing of his pieces. But it is
rather frag ile: the slig htest inaccu racy, and it
wil l evaporate. Which is what happened i n
the game: 2 8 . . . h 5 ? 29 '.tc1 ! l::tf5 3 0 l::t xf5
tt:Jxf5 31 tt:Je4 g4 32 tt:Jc5+ We5 33 tt:Jxb7
tt:Je3 Draw.
A very important principle in the conversion
of an advantage is the maxi mum restriction
 Grandmaster Technique ltJ 1 69

Having forestalled the opponent's i ntention, And now 36 . . . �f4 , 36 . . . ll'lf3 or 36 . . . h 5 . The
Black can now calmly strengthen h i s posi­ concl uding position of the variation is very
tion . It is not easy for Wh ite to defend. For p romising for Black. But is it won? Couldn't
example, if 30 �f8 there follows 30 . . . .l:!.g 1 3 1 Wh ite have played more accurately some­
.!:!.f2 �e5 32 ll'le2 � e 1 33 ll'lxd4 'i£txd4, and where earlier? It is clea r that here everything
the dominating position of Black's king hangs by a thread : the slig htest add itional
guarantees h i m a g reat advantage i n the improvement to the defence, and the game
rook ending . will end i n a d raw.
But a completely d ifferent interpretation of Black's play can be improved . One is struck
the position is also adm issible. The black by the fact that he delayed slig htly - his
king is far more active than its wh ite knight did not i mmediately reach the neces­
opponent, which can be especially percepti­ sary sq uare f5. Of cou rse, it is desirable to
ble in a pawn or knight ending (Mikhail captu re on f5 with the knight. It is this that
Botvin n i k once remarked : 'A knight end­ explains the recommendation by Vad i m
game is the same as a pawn endgame' ). Zviagintsev: 28 . . . h 6!? (a mysterious move
Artur Yusupov suggested 28 . . . .l:i.f5 ! ? . Wh ite at first sight, wouldn't you ag ree! ) 29 �c1
replies 29 �xf5 (29 �d2 .l:i.f1 + 30 �b2 '.te5 .l:i.f5.
is u nfavourable for h i m ) .
Now Black would l i ke t o captu re with the
knight, but after 29 . . . ll'lxf5 30 lbe4 ! g4 3 1
lt:Jc5+! (weaker i s 3 1 lbg5+ �e5 3 2 lbxh7
lt:Je3 33 �c 1 lbxg2 3 4 'it> d 2 �f4 35 �e2
tt:lh4) 31 . . . '.te5 32 ll'lxb7 he can hardly hope
to wi n .
This means that he must play 2 9 . . . �xf5 ! ,
intending 3 0 . . . 'it>e5 a n d 3 1 . . . ll'lf5. Events
can then develop roughly as follows : 30 '.tc1
We5! 3 1 'it>d2 ll'lf5 32 Wd3 (if 32 �e2 or 32
g3, then 32 . . . �d4 is strong ) 32 . . . lbh4 33 g3
tt:lf3 34 h3 ll'lg 1 35 h4 gxh4 36 gxh4.

After 30 llxf5 l2Jxf5 3 1 'it>d2 ( 3 1 l2Je4 is now

poi ntless - the g5-pawn is defended)
31 . . . ll'lh4 Black must wi n .
However, the exchange on f5 i s not essen­
tial - 30 lbd 1 ! (but not 30 lbe4 We5) is far
more accu rate for White . For example:
30 . . . .l:i.xf2 3 1 ll'lxf2 ll'lf5 32 ll'ld3, i ntending 33
lbc5+ or 33 �d2 ll'lh4 34 ll'le1 . Little is
changed by 30 . . . We5 31 �d2 �xf2+ 32
lbxf2 lbf5 33 �d3, since if 33 . . . lbh4 there is
34 lbg4+ or 34 g3 followed by lbg4+. We
see that the position of the pawn at h6 is fa r
from idea l .
1 70 � Grandmaster Techniq ue

And yet Zviagintsev's idea is logical - simply knight ending - is much more concrete and
it must be put into effect slightly d ifferently. I req u i res carefu l checkin g . The q uiet pawn
suggest another mysterious move - i nciden­ move on the queenside, preparing the
tally, the same one that Najer made. exchange of rooks in the most favourable
28 . . . b6 ! ! version , can be fou n d only as a result of
29 Wc1 llf5! delving thoroughly i nto the secrets of the
position .
There a re two aspects to the process of
converting an advantage. On the one h a n d ,
it demands accu ract and methodical play,
and on the other hand - an abil ity some­
where to cut short the manoeuvri ng , and
fi nd and calculate a concrete way to the
goal . It is not easy to sense which is more
correct at a particu lar moment. I n the g iven
example both approaches seem to me to be
equally good , but th is does not often occur.

II
I t can happen that a position looks (and
Wh ite's position is d ifficult: 30 llxf5 tt'lxf5 or i ndeed is) completely won , and yet a pl ayer
30 tt'ld 1 llxf2 31 lbxf2 tt'lf5 32 tt'le4 h6, does not manage to convert it i nto a win.
th reatening either to attack the g-pawn Very often (if of course, things do not occur
(33 . . . tt'lh4 or 33 . . . tt'le3 ), or to penetrate with in severe ti me-trouble) the cause is a l oss of
the king onto the 4th rank. concentration and insufficient attention to
As you see, the similarity with the Yande­ those few resou rces wh ich a re stil l available
mirov-Najer endgame is not restricted to the to the opponent or which suddenly occur. A
fact that in both cases the key to the position very i mportant ski l l , enabling this type of
was an impercepti ble pawn move. Here mistake to be avoided , is prophylactic
there was also the identical nature of the th inking, about which I write in l iterally every
material and the pawn structu re , and also book of mine. The essence of it is that you
the highly important role played by the need to put you rself in you r opponent's
activity of the king i n a knight ending - a place , consta ntly asking you rself the ques­
recu rrent theme of Black's play in both tion : what can he u ndertake , and what would
examples. you do i n his place if it were him to move?
However in the Stean-Hort game fu nda­ I n some of the examples g iven below the
mentally different approaches to exploiting employment of prophylactic th i n king ena­
the advantage were possible. Which one do bled a player to fi nd the opti mal ways of
you l ike more? The fi rst is purely tech n ical exploiti ng an advantage, which at first sight
(restriction of the opponent's pla n ; u n h u rried were fa r from obvious. I n others, by con­
improvement of the position ), but it does not trast, such possibilities were not exploited
demand deep calculation and therefore it and the adva ntage evaporated .
enables time and energy to be saved . The
second approach - the transition i nto the
 Grandmaster Technique ttJ 1 71

Xie J u n - Larsen method of playing is too slow. He should

Monaco 1 994 probably have preferred 1 3 . . . �b4 , i ntending
Pirc Defence to a n swer 14 tt:'lg5 with 14 . . . .!::i.f8 , and 14 tt:'ld2
with 1 4 . . . 1i.f8 .
1 e4 g6 2 d4 �g7 3 tt:Jc3 c6 4 tt:Jt3 d6 5 h3
tt:'lf6 6 a4 0-0 7 ..ie3 tt:'lbd7 8 1i.e2 e5 9 1 4 tt:'ld2
dxe5!? Wh ite i ntends to retreat the q u een to a2 and
In the event of 9 0-0 Wh ite has to reckon occu py the c4-squ a re with her knight.
with the reply 9 . . . d 5 ! ? . 14 .. .'ii b4 ! ? was now essential , but Bent
Larsen conti n ues manoeuvring on the
9 ... dxe5
kingside, underestimating the strateg ic dan­
ger facing h i m .
1 4 . .. tt:J h 7 ? ! 1 5 'i¥b3 tt:J g 5 1 6 tt:Jc4 tt:Jc5?
The last chance was probably 1 6 . . . tt:Je6 ,
with the idea of sacrificing the exchange i n
t h e variation 1 7 tt:'ld6 tt:'ld4 ! 1 8 1i.xd4 exd4 1 9
tt:Jxe8 iixe8 .
1 7 'ii'a 3 tt:Jce6 1 8 'ii'x e7 �xe7 1 9 tt:'lb6 .:b8
20 ..ig4! .U.e8 (not 20 . . . h5? 2 1 ..ixg5)

10 0-0
Instead of this simple-minded move, 1 0
tt:ld2 ! was stronger, and if 1 0 . . . it'e7 1 1 tt:Jc4
.!::i.d 8, then 1 2 'ii'd 6! with adva ntage to Wh ite
(suggested by grandmaster Vlad imir Potkin ) .
1 0 . . . 'ife7 1 1 'ii'd 3 a 5 ! ?
Black forestalls t h e bind on his q u een s i d e by
a4-a5. He could also have considered
1 1 . tt:'lh5 ! ?, a typical ma noeuvre in such
. .

positions, with the idea of i nvad ing with the The knight on b6 is cramping the opponent's
knight on f4 , or even fi nding a conven ient position , and the pin on the h 3-c8 d iagonal
opportu n ity to play . . . f7-f5 . is extremely u n pleasant. There is no doubt
1 2 'ii'c4 .!::i.e 8 1 3 .l:.fd 1 h6 about Wh ite's enormous positional advan­
Black's last two moves a re log ically l i n ked: tage; the only q uestion is how to method i­
first the rook vacates the f8-squ a re for the cally exploit it.
knight (which has j u st been prevented from The attem pt to force matters , by i nvad ing
going to c5), or perhaps a lso for the bishop, with the rook on d7, is prematu re : 21 tt:Jxc8?!
and then control is taken of the g5-point, i n l:tbxc8 22 .i:f. d 7? ( 22 1i.xg5 hxg5 23 1i.xe6
order t o safeg uard t h e f7 -pawn from a n .l:i.xe6 24 .l:i.d7 is stronger, sti l l reta i n i ng a
attack b y t h e wh ite knight. H owever, this slight advantage) 22 . . . tt:Jd4! 23 .l:i.xb7 .i:f.b8.
1 72 � Grandmaster Technique

Black has activated his pieces, and h i s 27 .Uad 1 .U.f8

position is not worse . Black has prepared . . . f7-f5, but he does not
Let us ask ou rselves what the opponent in fact manage to play this - White has
would do if it were h im to move . Obviously everything ready for the decisive i nvasion .
he would l i ke to get rid of the pin by . . . h6-h5. 28 lt:lxc8+ .l::i.fxc8
2 1 h4 suggests itself, and after 2 1 . . . lt:l h7 22 29 .l::i. d 7+ 'it>f6
lt:lxc8 !? Itbxc8 23 ii.b6 Black's position 29 .. .<it'f8 30 ii.xe6 fxe6 31 .l::i. h 7 is also
remains d ifficult. But Wh ite has to reckon com pletely hopeless.
with 2 1 . . . h 5 ! ? . For example: 22 hxg5 hxg4 ,
or 22 ii.e2 lt:lh7 followed by 23 . . . lt:ld4 - i n
both cases with good chances o f equalising .
Wh ite nevertheless retains a sol id advan­
tage, by conti n u i ng 22 ii.xh5! lt:lh3+ (22 . . .
gxh5 2 3 hxg5 with a sound extra pawn ) 23
gxh3 gxh5 24 lt:le2 ! and 25 lt:lg3 . However,
the decision taken by the C hi nese player is
even stronger.
21 ii.xg5! hxg5
In this way Wh ite has eliminated the
possibility of . . . h6-h5 and mainta i ned the
pin on the knight at e6 . But without the
followi ng excellent move it would not have
been possible to cramp completely the
opponent's position . 30 ii.xe6 fxe6
What does Black want? Obviously his best 31 g4!
chance is to disturb the powerful knight at b6 The next move will be 32 .l:.1 d3 with
by the manoeuvre . . . ii.f8-c5 . u navoidable mate (after the immediate 31
22 lt:lb1 ! ii.f8 .U. 1 d3 there was the reply 31 . . . g4 ). Black
resig ned .
22 . . . lt:ld4 does not work: 23 ii.xc8 lt:lxc2 24
.U.a2 .
23 lt:Jd2 ii.c5 Vad i m Zviagi ntsev once prepared a new
plan for Wh ite i n one of the variations of the
24 lt:ldc4
French Defence and he suggested to me
The knight has arrived at just the right time, that we should subject his analysis to a
and now there is simply noth ing that the
practical testin g . We played a couple of
opponent can move , whereas at some point tra i n i ng games with a short time control ( 1 5
Wh ite wi ll i nvade at d7 with decisive effect. If m i n utes each for t h e game). A n d soon i n the
24 . . . .Uf8 (with the idea of 25 .. .f5) the Russian Cup a new testing occu rred - in a
simplest is 25 lt:Jxc8 f5 26 lt:lcb6 fxg4 27 duel with a well-known expert on the French
lt:ld7 or 25 . . . l:.bxc8 26 ii.xe6 fxe6 27 .l:.d2 Defence.
with a won position .
24 . . . ii.xb6
25 lt:lxb6 'itf8
26 .l:.d2 'ite7
 Grandmaster Technique ltJ 1 73

Zviagi ntsev - Vol kov I played 1 0 . . 'it'c4 1 1 i.d2 axb4 1 2 axb4

.

Russian C u p, Samara 1 998 l:txa 1 1 3 'it'xa 1 , after which I stopped to

French Defence th i nk . I n my prel iminary calculations I had
been i ntending 1 3 . . .i.xb4 14 'ifa8 i.xc3 1 5
1e4 e6 2 d4 d5 3 e5 c5 4 c3 'ii' b 6 5 tLlf3
'ii'x b8+ 'it'd? 1 6 'ii'x b 7 + 'it>e8, but here I
id7 6 a3 i.b5
realised that Wh ite would easily bring his
After his fa i l u re agai nst Zviagintsev, Sergey rook i nto play by either 1 7 'iVa8+ 'it'd? 1 8
Vol kov switched to 6 . . . a 5 ! ? and in this 'ifa3 , o r 1 7 i.xc3 't!Vxc3+ 1 8 'it>e2 , when
variation (with alternating success) he played there would be noth ing with which to defend
a couple of notable games: against Evgeny my king.
Sveshnikov and Peter Svidler.
Searches for a n acceptable defence proved
7 b4! ? cxd4 8 i.xb5+ 'it'xb5 9 cxd4 unsuccessfu l , so Black chose 1 3 . . . ttJc6 1 4
'iWa8+ ttJd8 1 5 it'a4+ ttJc6 1 6 b 5 'it'xa4 1 7
ttJxa4 ttJa? (with the faint hope of putting u p
a resistance i n t h e blocked position after 1 8
b6 ttJc6) 1 8 'it>e2 ! , and Vad i m confidently
exploited his overwhelming lead in develop­
ment.
1 0 ... 'ii'c 6 11 i.d2 axb4 1 2 axb4 .l:!.xa1 1 3
'ii'x a1 'ii'a 6 1 4 'it'a4+ ! (the same idea a s i n
t h e tra i n i ng g a m e ) 1 4 . . . 'it'xa4 1 5 ttJxa4 b 5
1 5 . . . ttJc6 1 6 b5 would have transposed i nto
a positio n from the tra i n ing game.
16 ttJc3! (much stronger than 1 6 ttJc5 - as in
my game with him, Vad i m aims to open
l i nes) 16 . . . i.xb4 17 ttJxb5 i.e 7 18 'it>e2 lLlh6
regularly u sed to employ the French
Defence with Black and I was happy if I
m anaged to get rid of my 'bad' l ig ht-squa re
bishop. Nowadays a spatial adva ntage has
come to be val ued more highly than it used
to be - White happily goes in for such
positions. It is no accident that the popu larity
of the 3 e5 variation in the Caro-Kan n has
increased sharply - there too Black's light­
square bishop may be q u ickly exchanged.
The diagram position occu rred i n both of my
training games with Zviagintsev. Black has a
choice: whether to continue q uietly, 9 . . . ttJd7!?
10 ttJc3 'ifa6 for example, or go i n for
sharper play, beg i n n ing with . . . a7-a5. I n the Wh ite's solid positional advantage is deter­
first game the tempti ng u nderm i n i ng of the m ined above all by the poor placing of the
queenside was tested . enemy knig hts . It is important n ot to allow
9 a5?! 1 0 ttJc3
. . . them i nto play.
1 74 � Grandmaster Techniq ue

1 9 .l::tc 1 suggests itself, but after 1 9 . . . 0-0 no

progress is made by 20 l:tc7 SLd8 , and the n ,
depending on Wh ite's move , . . .tt:Jc6 , . . . �b6
or . . . ttJf5-e7 . If 20 ttJa7 Black plays
20 . . . .l:!.e8 (intending . . . t2ld7-b6) 21 l:tc7 �f8
followed by . . . ttJf5-e7 - he retains a
defensible position.
After some thought the grandmaster found a
subtle pla n , enabling h im to immobil ise both
knights.
1 9 ttJa7 ! !
Subsequently I examined another, also very
strong method of play for Wh ite : 1 9 .l:!.a 1 ! ?
0-0 20 .l:!.a7. N o w 20 . . . 4Jc6 2 1 l:tc7 .l:!.b8
But now Wh ite must reckon with 2 1 . . . tt'lf5,
22 ttJc3 .l:!.b6 23 ttJa4 l:ta6 24 .l::t c 8+ tiJdB
as wel l as 2 1 . . .f6 . For example: 2 1 h3?! f6!?
25 ttJc5 .l:!.a7 is completely hopeless, while i n
(the knight aims for f7 , but 21 . . . t2lf5 22 g4
the event o f 20 . . . � d 8 2 1 �b4 ttJc6 (if
ttJh4 23 ttJxh4 3Lxh4 24 .l:!.b 7 �e 7 is also
21 .. J1e8 the simplest is 22 .l:!.a8 ) 22 �xf8
possible) 22 exf6 �xf6 23 .l::t b 7 (23 �f4 tt'ld7
ttJxa7 23 �xg7 Wh ite emerges a pawn u p .
24 l:tb7 t2lf8 ) 23 . . . 4Jf5 24 �f4 ttJxd4+ 25
B u t t h e conti nuation i n t h e g a m e is appar­
ttJxd4 e5! .
ently even stronger.
Also u nconvincing i s 2 1 l:tb7 �f8 followed
19 . . . 0-0
by . . . t2lf5-e7 .
1 9 . . . 'i¥td7 loses to 20 l:tb 1 .
2 1 g4! !
20 l:i.b1 !
The crux of Wh ite's plan! The knight at h6
The rook is better placed on the b-file than
remains u nder arrest, as does the one at b8.
on the c-file.
21 .. .f6? is no longer possible on account of
20 . . . .l:!.e8
22 exf6 �xf6 23 g5. Black loses after
If Volkov had foreseen his opponent's reply, 21 . . . t2ld7 22 ttJc6 ! (22 l:tb7 tiJfB 23 ttJc6 tt'lg6
he wou ld probably have preferred 20 . . . 4Jf5 . is less accu rate) 22 . . . t2lf8 (22 . . . �f8 23 .l::!. b 7;
If 2 1 l:tb7, then 2 1 . . . l:te8? 22 g4 t2lh4 23 22 . . . �a3 23 l:tb7 tiJfB 24 l:ta7) 23 ttJxe7+
ttJxh4 �xh4 24 �b4 ! and 25 �d6 is l:txe7 24 l:tb8 .
hopeless for Black - his knight at b8 will be
Possi bly Black's best practical chance was
lost. However, it is possible to defend by
to g ive up a knight for two pawns: 2 1 . . . t2lxg4!?
21 . . . h5! 22 h3 h4. Vad i m would probably
22 l:tg 1 ttJxe5, but of cou rse, this too is
have chosen 21 g4! t2lh4 22 t2lxh4 �xh4 2 3
completely hopeless.
l:tb7 , intending both �b4-d6, and f4-f5.
21 . . . �f8
22 h3 f5?!
It would have been better to wait passively.
(see diagram) The attempt by Black to free h imself meets
with a clear-cut refutati o n .
23 �xh6 fxg4
24 t2lh2!
 Grandmaster Technique tZJ 1 75

Black evidently overlooked this move . The

rest is easy.
24 . . . gxh6 25 lbxg4 lDd7 26 l:I.b7 h5 27
lixd7 hxg4 28 hxg4 lib8 29 lDc6 lib2+ 30
'ii1f3 i.. a 3 31 g5 l:!.b6 32 .Uc7 i.. b 2 33 �g4
.Ua6 (33 . . . i.. x d4 34 lbxd4 l:ib4 35 lie?
l:!.xd4+ 36 f4) 34 g6 .l::t a 1 35 lDe7+ �h8 36
Wg5 .l::t c 1 37 �f6 Black resig ned .

Lutz - Dautov
Germany, B u ndesliga 1 997

If 46 i.. d 1 ? ! , then 46 . . . l:I.c4 is u n pleasant.

C h ristopher Lutz admitted that he panicked
and reconciled h i mself to a d raw. There
followed 46 i.f5?! <J;e7 (47 i.. x e6 i.. x e6 48
f5 was th reatened ) 47 .Ug7+ <J;d6 48 .Ug6
<J;e7 Draw.
In h i s i nformative an d i nteresting book
Endgame Secrets Lutz gave a rather com­
plicated analysis, demonstrati ng that with
accu rate play he would stil l have retai ned
good winning chances. I h ave dou bts about
his assessment, but even if my a rg u ments
may seem u nconvincing , this essentially
Things seem to be completely bad for Black does not change anyth i n g : it is clear that i n a
- bad bishop, lack of space , wea kness on position with a great positional adva ntage it
g6 and an absence of any cou nterplay. But did not make sense to allow such sharp
even i n such positions ( I will put it more cou nterplay.
defi nitely - precisely in such positions) you
46 �d3 l:i.c3
shou ld be particularly attentive , by fi nding
46 . . . <J;e7? 47 Itg7+ 'it>d6 48 g4 lic3 49 i.. e 2
and neutralising beforehand all active re­
.l:i.a3 50 g5 is bad for Black.
sources for the opponent.
47 i.e2 lic2
The natural move 43 .l:i.h6?, made i n the
game, was a imed only at the a utomatic Not 47 . . . l:I.a3? 48 f5 l:I.xa4 49 f6 ! , but
reply 43 . . . i.. e 8 - then Wh ite plays 44 'it>f3 47 . . . i.. e 8!? 48 l:I.g7 l:I.a3 is qu ite possible and
followed by �g4-g 5 . However, Black gains will probably lead to the same position which
an opportu n ity to activate his game at the we reach i n the main variation.
cost of a pawn , and Rustem Dautov ex­ 48 Wf3
ploited it. After 48 �f2 i.. e 8 49 l:I.g7 Ita2 Wh ite cannot
43 . liff8 ! 44 .Uxg6 (of course, Wh ite cannot
. . play 50 Itb7? i.. h 5, while if 50 <J;e3, then
allow the p i n : 44 i.. x g6? lif6 , and then 50 . . . .td7 51 g4 l:I.xa4 52 g5 l:I.a3+ 53 Wf2
. . . .Ug8 or . . . i.e8) 44 ... .Ubc8 45 l:I.xc8 l:I.xc8 .Uc3 with cou nterplay.
1 76 � Grandmaster Technique

48 . . . �e8 What then should Wh ite h ave done in the

In the event of 48 . . . l:!.a2? 49 f5 .l::t.x a4 50 f6 i n itial position? Employ prophylactic think­
l:ta3+ 5 1 'i!.i4 (threatening 52 .l:tg8 ) 5 1 . . . l:tc3 i n g , g uess the opponent's plan and take
52 l:!.g8 l:tc8 53 .l:txc8 �xc8 the bishop measures against the switching of the rook
ending is lost for Black: 54 �d 1 ..td7 55 �a4 to the c-file, by movi ng the bishop off there
�e8 56 g4 ..tf7 57 �c2 (but not 57 'it>g5? e5) beforehand.
57 . . . �e8 58 g5 �xb5 59 g6 ..te8 60 g7 ..tf7 43 ..td3 ! !
6 1 i.g6 . N ow 43 . . . l:tff8? 4 4 ..txg6 or 43 . . . g 5 ? 44 fxg5
4 9 .l:lg7 l:ta2 5 0 l:tb7 is totally bad .
43 . . . .U.g8
Lutz continues 44 .Uh6 .llff8 45 .Uxg6 l:tc8 46
l:!.xc8 l:!.xc8 47 f5 , which , of cou rse, is good
enough to win . Vad i m Zviagi ntsev found
another, more i nteresting way.
44 l:th7 ! ? .l:tff8

Up till now we have been following Lutz's

analysis. His main variation goes 50 . . . ..th5+
5 1 g4 i.xg4+ 52 'it>xg4 .Uxe2 53 .l::t.x b6+, and
Wh ite should win the rook ending.
It makes sense for Black to retai n the
bishops.
50 ... .Uxa4 (instead of 50 . . . �h5+?) 5 1 .l::t.x b6+ Now a spectacular stroke proves decisive.
(here Lutz terminates the variation ) 51 . . . �e7
45 l:!.c6+ ! ! ..txc6
The idea of Black's defence is to pursue the
46 bxc6
opponent's king and bishop with his rook. If
52 .Ua6 , then 52 . . . l:ta3+ 53 'it>f2 l:.a2 or 53 Mate is th reatened : 47 .i::t.d 7+ �xc6 48
�g4 .l::t.a 2 54 ..tf1 .Uf2 . If 52 'it>e3 there ..tb5#, and noth ing is changed by 46 . . r.t>xc6 .

follows 52 . . . .l::t. a 3+, and the king is unable to 47 ..tb5+ �d6 48 l:!.d7#. Both 46 . . . l:!.d8 47
approach the rook (the vulnerable position ..tb5 and 46 . . . e5 47 fxe5+ are also hope­
of the bishop prevents th is), while in the less.
event of 53 'it>f2 Wh ite has to reckon with Black's best chance is 43 ... l:!.bf8 ! ? with the
53 . . . .l::t. a 2 (53 . . . .l::t. a4! ? ; 53 . . . .l::t. b 3!?) 54 g4 a4 idea of 44 . . . g5.
(54 . . . �g6? 55 f5) 55 �e3 .l::t. b 2.
(see diagram)
 Grandmaster Technique 4:J 1 77

reply 50 'i!Vd 1 !, preparing not only a check on

h 5 , but also 51 'ii'g 4 .
I n the event of 49 . . . lt:Jxf5?! Wh ite finds the
tactical resou rce 50 ..id4 ! , when 50 . . . lt:Jxd4?
51 'ii'x d5+ is not possible, while if Black
supports his d5-pawn by 50 . . . c6 , this ex­
poses the king and the wh ite queen breaks
free: 51 �a 1 .
The same reply 50 't!Va 1 ! is also good after
the i mmed iate 49 . . . c6? ! .
4 9 . . . h5?! i s temptin g , since the pawn is
invul nerable: 50 i.xg5?? lt:Jf3 + . In the event
of 50 b5 'ii'xf5 51 'ti'd 1 Black does not play
44 .l:lc2 is unconvi nci n g : 44 . . . g5 45 fxg5 .l:lf3 51 . . . tt:lf3+?! 52 'it>g2 g4 53 hxg4 �xg4+ 54
46 i.h7 l:ta3 47 g6 .l:!.ff3 48 .l:lg 1 ! ? .l:lxg3+ 49 'it>f1 'ii'e 4 (54 . . . 'it>e6 55 it'd3 ) 55 Wg2 ! , but
'.t>h 1 i.e8 50 g7 i.f7. However, after 44 .l:lh6 simply 51 . . . 'it>g6 ! with a n overwhelming
g5 45 .l:lxf6 .l:lxf6 46 fxg5 l:tf8 47 g6 .Ug8 advantage . A possible variation is 52 b6
(47 . e5 48 dxe5+ 'it>xe5 49 .l:lc7 �e6 50
. .
l!Ve4 ! 53 'ii' h 1 'ti'e5+ 54 'it>g 1 cxb6 55 i.xb6
.!:l.b7) 48 ..t>f3 e5 49 dxe5+ 'it>xe5 50 .l::!. c 7 'ii'e 6! followed by 56 . . . g4 .
Wh ite retains a sign ificant advantage. Wh ite is saved by the fi ne reply suggested
by Kasparov: 50 c4! ! 'i!Vxc4 (50 . . . dxc4 5 1
Ti mman - Kasparov �c6) 5 1 ..ixg5 ( now Black does not have
5 1 . . . tt:lf3+) 5 1 . . . lt:Jxf5 52 'ii'f3 .
USSR v. Rest of the World, London 1 984
It appears that we h ave looked at all the
sensible conti n uations and have not found a
solution. Garry Kaspa rov also did not find
one.
49 ... tt:lxf5?! 50 i.d4! c6 51 'i!Va 1 lt:Jxd4
(5 1 . . .11if3 52 'it'a8 ! ) 52 cxd4

Black has an obvious positional advantage ,

but i t is not easy t o exploit. O n close
examination it is d iscovered that every move
has one d rawback or another.
Of cou rse, he would l i ke to captu re the
pawn. But if 49 . . . 'i!Vxf5 ?! there is the strong Wh ite's pawns a re weak, but the open
1 78 \t> Grandmaster Technique

position of the enemy king g ives h i m 'ilkh 1 'ife5+ (or, as given b y Kasparov,
adequate counter-cha nces. 5 1 . . . ltJf3+ 52 'it>g3 'it'e5+ 53 �g2 ltJh4+ 54
52 . . .'i&'d2 �f1 'ii'xf5) 52 �g 1 'ii'xf5 - the f5-pawn has
If 52 . . . 'ii'f3 Kasparov g ives the variation 53 been captu red , and the opponent's pieces
"YWa7+ 'it>g6 54 'it>g 1 'ii'x h3 55 'it'b6 'iVe6 56 a re stil l fastened down in the corner.
b5 with equal ity. I n stead of 53 . . . Wg6 it Let us consider 50 �c5 .
makes sense to try 53 . . . 'it>f6 ! ? , when 54
'it>g 1 ?! 'i¥xh3 is now i neffective : after 55 'ii'b 6
"YWe6 there is no pin along the 6th ran k - the
queen is defended by the king . Wh ite must
play 54 'it'a2 , and if 54 . . . 'ii'f4+ 55 'it>g 1 'it'xd4 ,
then 5 6 'i!Va8 .
5 3 'it'a7+ 'it>g6 5 4 �g2 �xb4 55 'ii'd 7 !
"YWxd4 5 6 �e8+
Jan Timman does not h u rry to captu re the
pawn (56 'ii'x c6+ 'it>h5), hoping fi rst to
worsen the placing of the black pieces .
56 . . . Wf5?!
As was pointed out by Kasparov, the only
way to continue to play for a win was by
56 . . . �h7 57 'ii'd 7+ 'it'g7 - however, the Kasparov g ives 50 . . . h5 (with an exclamation
position arising in the variation 58 'ii'f5+ 'i!Vg6 mark) 51 Si.e3 'iVxf5 , overlooking a success­
59 'it'd?+ 'it>g8 60 'it'd8+ 'it>g7 61 'it'd?+ 'ii'f7 ful defence: 51 'ikc1 ! 'it>f7 (or 51 .. .'i&'xf5) 52
(6 1 . . . �f8 62 'i!Vd8+ 'i!Ve8 63 �f6+ �g8 64 'ii'e 3.
�xh6) 62 �xc6 it'f5 is objectively d rawn. Black retains a great advantage by continu­
57 'ii'd 7+ 'it>f4?! (57 . . . 'it>g6 ) 58 'ii'f7+ D raw. ing 50 ... ltJf3+! 51 �g3 ( if 5 1 'it>g2 both
5 1 . . . 'it>f7 52 �e3 h5 an d 5 1 . . . 'it'xc3 52 'i'd 1
Let us return to the i n itial position . Take note: 'i!Ve5 a re satisfactory) 51 . . . ltJd2+ 52 'it>h2
for the moment Wh ite is completely tied (52 Si.e3 ltJf1 + 53 'it>g2 ltJxe3+ 54 fxe3 'ii'e2+
down , and he not only cannot improve his 55 'it>g3 'ilt'xe3+, a n d the resu lting pawn
position , but it is hard even to suggest a ending is easily won ) 52 . . 'it>f7 .
.

move which would not worsen it. However, I n reply to 50 Si.d4 the tem pting 50 . . . 'ii'xf5 5 1
Black faces the same problem, since his 'ii'd 1 'iff4+ 5 2 'it>g 1 ltJf3+ 53 W g 2 g 4 5 4 hxg4
forces are now optimally placed . It is 'i!Vxg4+ 55 �f1 'i!ke4 56 ii.e3 h 5 is not
possible to play for zugzwang , only by altogether convi n ci n g .
decid ing on a far from obvious king move .
49 . . . �g8 ! !
Let u s look a t the opponent's replies.
(see diagram)
I n the event of 50 f6 'it>f7 things have not
become any easier for White - it is again
hard to offer h im any good advice.
50 'it'a1 concedes an important central
square to the enemy queen : 50 . . . 'i!ke4! 5 1
 Grandmaster Techniq ue ctJ 1 79

reta i ned an enormous advantage, most

p robably sufficient for a wi n .
O f cou rse , to find such a solution to the
position was extremely d ifficult, even with
the classical time control which was used at
that time in all serious competitions. And it
would be a ltogether i m possible with the
id iotically shortened control which is being
p ropagated now by F I D E . If, God forbid, the
officials a re successfu l , deep and subtle
ideas will completely d isappear from tou rna­
ment chess (with the exception of cou rse, of
opening fi nds, prepared beforehand with the
Wh ite pa rries the th reat of the h-pawn's help of com puter prog rams). And what then
fu rther advance with the unexpected move will there be for chess enthusiasts to admire
57 'it>g2 ! . - standard plans and tech n iques, and
simple combi nations which h ave occu rred a
50 . . .tt:'lf3+ 5 1 Wg2 "iie 4 is fa r stronger, and
thousand times? You don't have to be an
after 52 i.. e 3 Black ca n play as i n the 50
oracle to predict severe conseq uences in
'i'a 1 variation: 52 . . . tt:'l h4+ 53 'it>h2 'iVe5+ 54
the futu re both for chess l iterature , and for
Wg1 'Wxf5.
the popu larity of chess in genera l .
It only remains to analyse the desperate
attempt 50 b5 ! ? .

Ill
When a fru it i s ripe, i t should be gathered -
otherwise it will over-ripen and become
inedible. It is the same with the conversion
of a n advantage . It is i mportant not to miss
the appropriate moment for the favou rable
transformation of a n advantage, lead ing to
an i m med iate win or to a situation which can
be played al most a utomatical ly. If you delay,
su bsequently such a conven ient i n stance
may not present itself agai n .
Alas, i n overwhelming positions even very
strong players sometimes enjoy l ife, stop
acti ng concretely, and refuse to exert them­
It is justified only after Kasparov's sugges­ selves or calculate variations that a re even
tion of 50 . . . 'it>f7 (again his excla mation mark the slig htest bit complicated , expecting that
should be replaced by a question mark) 5 1 the fru it will itself fal l from the tree . The
b6! cxb6 5 2 ifa 1 with counterplay. Black possible consequences of th is a re very wel l
should be satisfied with a sou n d extra pawn illustrated b y t h e following example.
after 50 .. ."Yi'xb5 51 'iWd 1 tt:'lxf5.
Thus after 49 . . . Wg8 ! ! Black would have
1 80 � Grandmaster Technique

Anand - Kamsky the game by Viswanath a n Anand himself:

Candidates Match , 5th Game, 35 lLlc7 ! Ita7 36 lLlb5. Now 36 . . . lld7 37 'i'c6
Sanghi Nagar 1 994 leads to the loss of a paw n , while if 36 . . . .l:i:a6,
then 37 'it'd 5 .i.e? 38 l:!.c1 - the opponent
has practically no usefu l moves .
I should mention that in such situations the
knight on d5 looks fi nely placed , but that is
all - it is not itself attacki ng the weak d6-
paw n, and is covering it from pressure on
the d-file. Therefore the switch ing of the
knight to c4 or b5 is a typical pla n .
Anand decided not t o change t h e pattern of
the position, but conti n ued his unhu rried
manoeuvri n g .
35 l:!.d3? 'ifb7
35 . . . 'i!Nc8? was completely bad : 36 t2:lb6
'iVxc4 37 tt::lxc4 .i.e? 38 lld5 ( Ftacnik). In
Wh ite's position is strateg ically won , tha n ks lnformator Anand awards the move played
to the overwhel ming superiority of his an exclamation mark and a ;!; sign (slight
mig hty knight over the 'bad' black bishop. advantage for White). Of cou rse, this evalu­
It is also important to note that at the g iven ation is too pessimistic and was undoubt­
moment the white pieces are active, whereas edly infl uenced by the result of the game.
the opponent's are disunited . If he is g iven 36 llc3 llb8
time, Black will improve somewhat the 37 'i!Nd3 'it>g7
placing of his pieces (for exa mple, . . . �d8,
Anand considers 37 . . . .i.d8 ! ? to be a poor
. . . 'it>g7, . . . 'ii'b 7, . . . llb8), which , of course,
reply in view of 38 l:!.c4 followed by 39 b4
will not change the eval uation of the
(38 . . . 'i!Nxb3? is not possible on account of 39
situation , but will improve his defensive
'it'xb3 l::Ix b3 40 Itc8) . But I am not convinced
resou rces . This means that now, before this
that Wh ite is guara nteed a win after 38 . . . Wg7
occu rs , a concrete way of exploiting the
39 b4 axb4 40 llxb4 �a7 4 1 l::Ix b8 'i¥xb8.
advantages of Wh ite's position should be
We saw a similar situation i n Ftacnik's
sought.
variation (35 llb1 ), but there the bishop was
Ljubomir Ftacn ik exami ned the plan of u nable to return to the queenside.
creating a passed pawn on the queenside:
The next moment seems to me to be
35 llb1 !? 'i!Nd8 36 b4 axb4 37 'ifxb4 l:!.a5 - in
exceptionally important a n d i nstructive.
his opinion Wh ite has only a m i n i mal
advantage. I n my view, the adva ntage here
is very great - for example, Wh ite ca n play
38 g3!? followed by 39 h4, practically
depriving the bishop of any hopes of taking (see diagram)
part in the defence of the queenside.
However, Black's moves in this variation are
not obligatory.
A more convincing way was poi nted out after
 Grandmaster Technique lZJ 181

40 . . . ir'a5 ! - the q ueen ties the knight to the

d efence of the rook.

3 8 g3?
I n accordance with the principle 'do not h urry'
White gradually strengthens his position .
Alas, the useful move made by h i m is in fact Noth ing is g iven by 41 ltJb6 llf8 42 llc6
a serious mistake, which puts the win i n h 5 ! ? , and 43 ltJc4? 'i!Va8 44 .l:!.xd6 even
jeopardy. loses: 44 . . . 'iWxe4 45 I:.d 1 I:.d8. The correct
scheme of attack was suggested by Igor
What is the poi nt? How can this be
Za itsev: 41 g3! with the idea of 42 h4. The
explained a n d , what is far more i m portant,
black bishop turns out to be at the cross­
recogn ised d u ring a tou rnament game?
roads: a move along either d iagonal will
The point is that, apart from the long-term i nvolve serious concessions.
advantages of his position , here Wh ite also
I n the event of 4 1 . . . �d2? a swift mati ng
has one tem pora ry plus: the possibility of
attack proves decisive : 42 'ii'f3 .Uf8 43 'i!Vf6+
occu pyi ng the 7th ran k with his rook (the
'.t>h6 44 ltJe 7 ! . If 4 1 . . . ..lli.. d 8 there follows 42
immed iate 38 llc7 does not work because of
.Ud7 , and there is no satisfactory defence
38 . . 'ii'x b3, but he can fi rst sacrifice a pawn :
against a move by the knight: after a l l , the
.

38 b4 ). I n combination with the su bseq uent

rook's path to f8 is blocked by its own
attack of the queen on the f7-point, the idea
bishop.
looks tempting and therefore it should be
If 4 1 . . . llf8 ! ? is played immediately, then
thoroughly checked . After all, if Wh ite delays
after 42 I:.b7 Wh ite captu res the b4-pawn
slig htly, Black will cover the c7 -point with
and ach ieves a decisive su periority on the
38 . . . ..lli.. d 8, and such an opportu n ity will not
queenside - the enemy rook is not able to
occu r agai n . All the other advantages will
help, since it is tied to the f7-point. 42 . . . ..lli.. d 2
rema i n , of cou rse, but who knows whether
43 'i!Vf3 ..lli.. g 5 (43 . . . 'iWd8 44 a5) 44 h4 ..lli.. d 8 45
they will suffice for a win?
�b3 etc. is also hopeless.
S o , 38 b4! axb4 39 llc7 'it'a8 . I n lnformator
Possibly the best practical chance is a
Anand suggests only 40 'ii'f3? ! llf8 (after 4 1
bishop sacrifice: 41 . . . lla8 ! ? 42 h4 ..lli.. d 8 43
'i'b3 �dB Wh ite has noth ing special). The
.l:td7 'iWxa4 44 'ifxa4 I:.xa4 45 .Uxd8 b3 46
immed iate 40 �b3 ! is more logica l , blocking
ltb8 .l:txe4 47 .l:txb3.
the opponent's passed pawn and a i m i ng at
the f7-point. The only sensible reply is (see diagram)
1 82 � Grandmaster Technique

(4 1 .. .'�xf7 42 li:Jb6+) 42 .l:I.xh 7+! Wxh7 43

'iVc7+ Wh6 44 'it'xb8.
After 40 . . . ..td8 there follows 41 .l:d7, and i f
41 . . . b3, then 42 li:Jb4 ..t>h6 (42 . . . d 5 43 'i'xb3
- Black is tied hand and foot) 43 'it'xf7 'tixe4
44 'ii'x h 7 + 'it>g5 45 .U.xd6 with the th reat of 46
h4+ - Wh ite's attack is i rresisti ble. And in the
event of 4 1 . . . 1Wxa4 Wh ite can decide mat­
ters with both 42 ti.xf7 + ! ? Wh8 43 .U.f8+ 'lt>g7
44 ti.g8+ Wh6 45 lt:Je3 ! and 42 li:Jf6 ! ? 'ii'd 1 +
43 Wh2 Wxf6 (43 . . . d5 44 ti.xd 5) 44 'ii'x f7+
..t>g5 45 'ii'f8 ! (or 45 f4+ ! exf4 46 h4+ Wg4 47
'ii'x h7).
For the knight Black has two pawns, which The comparatively best cha n ces of a de­
u nder more favourable circu mstances could fence a re g iven by 40 . . . 'i'a5 ! ? 4 1 li:Jf4 (4 1
have promised reasonable saving chances. lt:Jxb4? d 5 ! 42 lt:Jxd5 'ike 1 + ) 4 1 . . . d5, al-
But here, tha n ks to the mai ntai ned blockade though after 42 lt:Jxd 5 b3 (or 42 . . . .U.f8 43
in the centre , it seems to me that Wh ite 1Wb3 ! ) 43 lt:Jc3 .l:tf8 44 .l:I.b7 Wh ite remains a
should be able to convert his material sou nd pawn to the good .
advantage. Fasci nating complications a rise in the varia­
I should mention that Wh ite also has tion 40 . . . b3!? 41 li:Jb6 d5! 42 .l:txf7+! 'lt>xf7
available another, sharper way of attacki n g : 43 �c7+ il.e7 44 lt:Jxa8 l:tb4 ! .
instead o f 40 'i&'b3 he c a n try 4 0 'i&'c4 ! ? .

Wh ite faces a choice: he can either block the

Now in t h e event o f 40 . . . 'ika5 t h e rook is pawn with his quee n , or d isrega rd it and play
defended and the knight can go to any for a n attack.
square. On the other hand, the a4-pawn is A) 45 'it'c1 b2 46 'it'b 1 Ji.c5! (th reatening
en prise, and the way is open for Black's 47 . . . ..td4 and 48 . . . ti.c4 ) 47 a5! Ji.. d 4 48 tL:lb6
pawn to the queening squa re . Let us dxe4 ! (the position arising after 48 . . . l:tb5 49
examine some variations. lt:Ja4 .l:txa5 50 lt:Jxb2 .l:b5 is most probably
40 ... 'it'xa4? is totally bad : 4 1 ti.xf7 + ! Wh6 lost) 49 ..t>f1 .i.xb6 ! ? 50 axb6 .U.xb6, and it is
 Grandmaster Technique i:tJ 1 83

not clea r how Wh ite can make any p rog ress convert hi s advantage. For the moment
- it would appear that the rook and the b2- Gata Kamsky prefers to retai n the rooks,
pawn neutral ise the queen . hoping to tie down the wh ite pieces by the
B) 45 lt:Jb6! b2 46 lt:Jc8 b 1 fi+ 47 'it>h2 .l:!.b7 pressu re on the b3-paw n .
(it is dou btful whether Black has anyth ing 40 �g2
better - the th reats to h is king created by the If 40 'i¥g4 there is the good reply 40 . . . 'i¥e6 ! ;
queen + knight duo a re just too da ngerous ) i n add ition Wh ite h a s t o reckon with 40 . . . f5! ?
48 lt:Jd6+ �f6 ( 4 8 . . . 'it>e6 49 exd5+ 'it>xd5 5 0 4 1 exf5 ir'xf5 .
tt:Jxb7) 4 9 'ii'x b7 'ii'c 1 ! ? ( 4 9 . . .'it'xb7 50 lt:Jxb7 40 . . . h5!?
d4 51 �g3 �e6 52 �3 1eads to a hopeless
It is useful t o deprive Wh ite o f t h e g4-square,
minor piece ending).
and i n some cases the pawn may also
advance to h4.
41 .l:!.c4 fib7

Black is th reatening 50 . . . fif4+ with per­

petual check. I n the event of 50 'ii'b 6 d4!
Wh ite again has to defend agai nst the same
42 .l:!.c3
threat - stra ngely enoug h , he does not h ave
a powerful enough d iscovered check. H ow­
It transpires that it is not so easy to create a
ever, he fi nds 50 fic7 ! ! 'i¥xc7 (50 . . .'�'f4+ 5 1 passed pawn on the queenside. Anand
<ttg 1 dxe4 5 2 'ii'c6 ! is no better) 51 lt:Je8+ rejected 42 b4 axb4 43 .l:!.xb4 because of
<ttf7 52 lt:Jxc7, when he should w i n, for 43 . . .'tl¥xb4! 44 lt:Jxb4 l:l.xb4 . If 45 'iWd 1 there
example: 52 . . . dxe4 53 a5 �c5 54 lt:Jb5 �xf2 follows 45 . . . .l:!.d4 46 fic2 (46 'iWb3 .l:!.xe4 )
55 g3! . 46 . . . h4 ! ? or 46 . . . �b6 ! ? , when Black should
not lose. H e also retains a tenable position
Thus 38 b4 ! axb4 39 �c7 fia8 40 �b3! or 40
after 43 . . . fia8 44 �xb8 'il&'xb8, for example,
'i'c4 ! ? would have promised Wh ite enor­
45 fic3 fib 1 46 �b4 ir'd3 47 a5 f5 ( Ftacnik)
mous winning chances . But now let us see
48 h4 f4 49 gxf4 'ife2 .
how the game developed .
42 h4 ! ? had some point. Black ca n wait, but
38 . . . �d8!
i n my view he can also take the pawn ,
Of cou rse, Black covers the c7-squa re . although this is risky. After 42 . . .'i!Vxb3 43
39 ir'f3 'iVd7 "i'Nxb3 l:l.xb3 44 .l:!.c8 �f6 l ossif Dorfman
Also after 39 . . . .l:!.c8 40 .l:!.xc8 'ifxc8 I am not awards 45 .l:!.c6 a n exclamation mark, but
sure that Wh ite would have been able to this is hardly justified . On ly, Black should not
1 84 � Grandmaster Technique

reply 45 . . . .l::!.d 3 (in the hope of 46 �xd6? (45 . . . 'ife6 is better, and if 46 I!f3, then
�e7 ! , when the bishop breaks free) in view 46 . . . .tb6 47 .l::!.f6 'it'd? ) 46 4Je7 ! 'it>f8 47
of 46 l:ta6 .l::!. d 4 47 f3 , when 47 . . . .l::!. x a4 48 4Jxg6+ fxg6 48 'ir'xg6 and wins (Anand ).
I!xd6 leads to the loss of the bishop. 45 'ifc2 'iVb 7
45 .. Jlb8 46 .Uxd6 �d8 followed by . . . 'it>f8- 46 'iVd3 �b6
e8 is correct, when White's advantage is not
47 �f3 �d8
too great. It looks more natural to play 45
.l::!.a 8 g5! 46 hxg5 .txg5 47 .Uxa5 .Ua3
(47 . . . .lld 3) 48 .Ua6 Ji.d2 , but here too Black
can fight on .
42 . . . 'iid 7
43 'ife2 �b6
44 �d2

How can Wh ite make progress? It would be

good to exchange the h 5-pawn for h i s g­
pawn by playing g3-g4 and taking on g4
with a piece , and then battering the enemy
king's defences by h 3-h4-h5 . However,
there appears to be no easy way of putting
this plan i nto effect. And i n addition, he
44 . . . .idS! constantly has to reckon with the u ndermin­
David Bronstein was once asked how he ing of the centre by . . . f7-f5 (th is can
had managed to save a game, where for a apparently be played in reply to 48 h4 ).
long time he had stood badly. 48 g4? ! hxg4
'Very simple , ' replied the g randmaster, ' I 49 hxg4
merely endeavoured not to worsen my It would be desirable to captu re with the
position, and, what is even more important, I quee n , but after 49 'ifxg4 there is the strong
didn't try to improve it. ' reply 49 . . . f5! , lead ing to an u nclear game.
That i s also the case here : Black should 49 . . . .l::!. c 8!
keep patient and wait - any activity will After the opponent has weakened his king's
merely make things easier for the opponent. defences, Kamsky happily exchanges rooks.
For example, in the event of 44 .. .f5? 45 exf5 He eliminates the potential th reat of the
�xf5 Wh ite organises a decisive attack by wh ite rook switch ing to the h-file, while the
46 4Jxb6 .l:txb6 47 .l::!.c 7+ c.t>f6 48 'i!Vd5 'ii'e 6 49 opponent now has to key an eye on
'ii'f3+ �f5 50 'iVa8 . The attempt to activate attempts by the black queen to break
the bishop with 44 . . . �c5?! is also u nsuc­ through on the kingside, and this factor ties
cessful - because of 45 'ii'g 5! 'ifd8? his hands, not al lowi ng h i m the freedom to
 Grandmaster Technique ctJ 1 85

play actively on the queenside. The main d ifference com pa red with the
However, 49 . . . �d7 was also possible: 50 previous example is the queenside pawn
'ilt'g3 �g5 5 1 l:tc7 'i!Ve6 , or 50 llc1 'ifb7 51 b4 structure. There the wh ite rook was able to
axb4 52 'ii' h 3 �g5 ! . operate on the c-file, an d one of the
promising plans was the creation of a
5 0 "ii'e 3 llxc3
passed a-paw n . Here the queenside is
51 'ifxc3 'it'a6 p ractically closed and therefore play has to
52 'it'c2 be created on the opposite side of the board .
The queen covers the e2-squa re . 52 b4 H owever, the blocking of the queenside also
axb4 53 �xb4 'ii'e 2 54 �g3 �h4+ 55 �xh4 has its pluses for Wh ite : she does not have
'ilr'xf2+ 56 �h3 �f3+ would have led to a to concern herself about the backward b3-
draw ( Ftacn ik). pawn , and the opponent is u nable to
excha nge rooks, as Kamsky did
52 . . . 'ifa7
I n itially J udit Polgar employs a well-known
But now the black queen is th reatening to
tech nique: she correctly deploys her heavy
invade at d4.
pieces on the half-open d-file - with the rook
53 'ii'd 2 'it'b7 i n front of the q uee n .
In the opinion of Yasser Seirawa n , Wh ite 32 lld3 �d8
would still have retained chances of success
33 'ii'd 2 .l:!c6
by playing 54 b4 ! ? axb4 55 'it'xb4 . I don't
34 'it'd1 �g8
th ink so - 55 . . . �c8 ! ensures sufficient
cou nterplay. 35 h4!
54 �d3
Draw.

Four years later a similar position again

arose i n one of Anand's games, but this time
he was representi ng the weaker side.

J . Polgar - Anand
Wij k aan Zee 1 998

J u d it realises that, by operati ng only on the

central file, it will not be possible to wi n .
Therefore she opens a 'second front' on the
kingside - i n full accordance with the well­
known method of converti ng a n advantage:
'the principle of two weaknesses' .
Black should have prevented the fu rther
advance of the h-paw n , by playing 35 . . . h 5 ! .
Polgar w a s i ntend ing t o have her way - by
1 86 � Grandmaster Technique

continu ing 36 f3 Wg7 37 �e2 followed by i nvasion along the d iagona l , but at a high
ltd 1 -h 1 and g3-g4 . But this plan is not easy price - Wh ite has acq u i red a new possibil ity:
to carry out - the wh ite king's defences a re c4-c5!
weakened , and Wh ite has to reckon with the Before playing th is, Polgar excha nged pawns
manoeuvre . . . i.b6-d4 and with . . . f7-f5 . on g6. A sensible decision , wh ich would not
35 . . . �g7? requ i re any commentary, had it not been for
36 h5 i.g5 the variation 43 c5?! dxc5 44 ifxc5 'ii'a 6,
37 'ii'f3 .l:i.c8 g iven by the grandmaster as the j ustification
for the move she made.
38 .l:i.d1
White prepares to switch her heavy pieces
to the h-file, in order to create threats to the
enemy king .
38 . . . .l:i.c6
39 ife2 .U.c8
40 .l:!.h1 'l!tg8
41 f3 !
Excellently played . Wh ite frees not only the
2nd rank (for the manoeuvre Wf1 and 'ii' h 2),
but also the f2-sq uare , from where the
queen will create the threat of i nvad ing
along the g 1 -a7 d iagona l . Another man ifes­
tation of the 'principle of two weaknesses':
to make the opponent's defence as d ifficu lt There is no longer time for 45 hxg6? in view
as possible, he must be given problems of 45 . . . li'e2+ 46 'l!th3 'ii'xf3 . H owever, White
over the entire boa rd . fi nds the spectacu lar stroke 45 tt'lf6+ ! ! . Now
45 . . . i.xf6? is not possible because of mate:
41 . . . .l:i.b8
46 "i!Vc8+ Wg7 47 h6#, and in the even t of
42 iff2 .l:.b7 45 . . .'ii' xf6 46 ifc8+ li'd8 47 'ii'x b7 'ii'd 2+ 48
\t>h3 Black does not gain sufficient compen­
sation for the lost exchange.
There only remains 45 . . . Wg7 46 tt'le8+ �g8,
but after 4 7 ifxe5 Black's position is difficult.
Here are some sample variations: 47 . . . �f8
48 ir'xg5 ife2+ (48 . . . Wxe8 49 'ii'e 5+ Wf8 50
hxg6) 49 �h3 l:!.b5 50 'ii'f6 ! .l:!.xh5+ 5 1 'it>g4
.l:l.xh 1 ( 5 1 . . . Wxe8 52 l:!.c 1 ) 52 tt'ld6 .l:i.h4+ 53
'it>xh4 ifh2+ 54 'it>g4 ifh5+ 55 'it>f4 , and the
checks soon come to an end , or 47 .. .f6 48
'iVd5+ 'it>f8 49 hxg6 'ii'e 2+ 50 �h3 'iVxf3 5 1
'i!Vc5+ 'lt> g 8 ( 5 1 . . . l:!.e7 52 g 7 + 'it>g8 5 3 'i'c4+
l:!.f7 54 .l:!.f1 �h5+ 55 �g2 ) 52 �c4+ followed
by 53 .l:l.f1 (there is also 52 tt'lxf6+!?).
Black has defended against the queen's 43 hxg6 fxg6
 Grandmaster Technique t2J 1 87

Now the shelter of Black's king is sign ifi­ After 43 . . . fxg6 things a re easier for Wh ite ­
cantly weakened (true, in retu rn he can hope there is no longer any reason for hesitatio n .
to exploit the open ing of the f-file for a 4 4 c5! dxc5
cou nterattack - but things never come to
45 'ilxc5 i.. d 8
th is).
Here 45 . . . l:.b8? is no longer possible be­
Anand rejected 43 . . . hxg6 because of 44
cause of the double attack 46 'iVa7(c7 ) .
'it'g 1 ! ? with an attack on the h-file. Perhaps
he should have taken a risk. After 44 .. .f5 45 46 lic1 !
'i'h2 �f6 the offensive is not so easy to
organ ise. I n the event of 46 'i¥h6 i.. g 7 47
'ii'g 5 (47 'ii' h 7+ 'iii f7 48 :a 1 l::!. a 7 with the
th reat of . . J la8-h8 ) 47 . . . l:l.f7 no mate is
apparent, and it is not possible to switch the
queen to the opposite wing, because of
counter-th reats to the king : 48 'i¥d8+ i..f8 49
'i!i'xa5? fxe4 50 fxe4? 'ii'g 4 , and it is Black
who wins.
Artu r Yusupov found the correct idea: 46
'iVh3! followed by the i nclusion of the g-pawn
in the storming of the enemy king's de­
fences. For example: 46 . . . .l:.f7 47 g4! fxe4
(47 . . .f4 48 'ilfh6 is u n promising for Black) 48
fxe4 l:tb7 49 'iit g 3 l:.f7 50 'ii' h 6 �g7 51 'ii' h 7+ On the h-file there is no longer anyth ing for
'itf8 52 g5 (Black is i n zugzwang ! ) 52 . . . .Ua7 the rook to do, and Wh ite switches it to the
(52 . . . l:.d7 53 l:la 1 ) 53 l2Jf6 Vf7 54 .l:[f1 i.. xf6 newly-opened c-file. Polgar wants to carry
55 'ir'h6+ 'ifg7 56 l:txf6+ l:.f7 57 1:.xg6 or 57 out the same reg ro u p i ng of the heavy pieces
l:.xd6 with a n easily won rook ending. as in the position with which we beg a n :
However, Polgar could also have played as place t h e q u e e n b e h i n d t h e rook (Itc4 ,
i n the game: 44 c5 dxc5 45 Wxc5 , althoug h 'ile3-c1 ), and then i nvade with the rook at
here this move is sign ificantly less strong. c8 .
Black creates sufficient counterplay, by 46 . . . �?!
conti n u ing either 45 . . . f5 ! ? 46 'i!fxa5 iic8 ! , or
A loss of a tempo. 46 .. .<it>g7 was more
as recommended by the I ndian player
accu rate .
Sundararajan Kidambi: 45 . . . .l:tb8 ! ? 46 'i!fxa5
l:.c8 47 l:.f1 (47 l2Jxb4 'ilfd6 or 47 . . . i.. e 7) 47 'ilfe3 'iii g 7
47 . . . .l:tc2+ 48 .l:tf2 l:.xf2+ 49 'iiixf2 'iVh3. (This Otherwise Black would have to reckon with
mea ns that i n the 43 c5? ! dxc5 44 'ii'x c5 48 'ifh6 .
variation Black should play not 44 . . . 'ilfa6? ! , 48 .l:1c4 l:td7
but 44 . . Jlb8 ! ) . 49 'ilfc1 h5
I t usually makes sense t o pose t h e opponent 50 l:r.c6 .Ud6
such a choice . At the board it ca n be hard to
50 . . . 'ilff7 51 "ifc5 was no better.
decide which contin u ation is objectively
stronger, or which at least is the more 51 .Uc8 'ili'd7
promising from the practical point of view. 52 'ii'c 5 'iii h 6
1 88 � Grandmaster Technique

58 lt:Jd5 i.d8 59 'i!Vc2 .l:td4 - Karolyi)

56 . . . .l:!.d7 (56 . . . 'it>g8 57 lt:Jxf6+ .l:!.xf6 58 'i'h6
'ii'e 2+ 59 'it>h3 iff1 + 60 ..t?h4) 57 .l:!.xd7+
'ii'x d7 58 'it'b6 .

53 llb8
There was probably no point in White
avoiding the natural move 53 lla8 ! ? (with
the idea of 54 l::ta 7), since if 53 . . . i.b6, then
54 'it'c1 + �g7 55 'it'g5 is strong ( Polgar).
Black's position is d ifficult: 5 8 . . . i.d8 (58 . . .
However, this possibil ity will never ru n away.
i.e? 5 9 'i!Vxa5 ; 5 8 . . . 'i!Vd8 5 9 'i!Vc6 ) 5 9 'i'b8
53 . . . i.f6 (the e5-pawn is attacked ) 59 . . . 'ife8? 60 {jjc?
The counter-attacking attem pt 53 . . . g5?! was and 6 1 'ti'xd 8 ! .
du bious in view of the weakening of the f5-
sq uare . Wh ite would have repl ied 54 .l:.a8!
'ii'e 6 (54 . . . g4 55 fxg4 hxg4 56 'iff2 ! Wg6 57
'ii'f8) 55 lt:Je3 l::t d 2+ 56 �f1 and wins (but, of
cou rse , not 56 ..t?g 1 ?? i.b6).
I n the opinion of Tibor Karolyi, who has
written an i nteresting book about Judit
Polgar, 53 . . . ..t?g7 was more tenacious. Then
Wh ite would have had a choice between
simpl ifying the position : 54 .l:!.b7 'ti'xb7 55
'i!Vxd6 and the more energetic 54 .l:!.a8 ! ? , and
then by analogy with Polgar's variation
considered earlier: 54 . . . i.b6 55 'ti'c1 i.d8
56 'i!Ve3 i.b6 57 'iig 5 .
54 'ife3+ i. g 5? Now comes t h e fi nal combinati o n .
The move i n the game al lows the H u ngarian 55 f4! exf4
player to bring the game to a spectacular 56 Ith8+!
concl usion . 54 . . . g5? 55 lt:Jxf6 .l:.xf6 56 l:!.h8+ Black resigned , not allowing h i s opponent to
would also have lost immediately. demonstrate the following pretty fi nish:
54 . . . �g7 was essentia l . Polgar g ives the 56 . . . �g7 57 'it'd4 + i.f6 58 �xf6 + ! ! .l::t xf6 59
following variation: 55 .l:!.a8 'ii'b 5 56 l::t a 7+ l::t h 7 + ! �xh7 60 lt:Jxf6 + .
(56 lt:Jc7 is less good : 56 . . . l::td 3 57 'ti'c1 "it'd? Comparing t h e two games, i t c a n be
 Grandmaster Techniq ue 4J 1 89

mentioned that, i n contrast to Anan d , Polgar of his c5-pawn (if he ca n q u ickly capture on
acted fa r more pu rposefully and consistently c6) .
with Wh ite. However, she encou ntered a It makes sense for B lack t o check carefully
sig n ificantly less stu bborn resistance on the the d i rect plan of advancing h i s pawn . And
part of the I ndian g randmaster than that put only if it transpires that this plan does not
up by Kamsky i n the fi rst example. work should he seek a more s ubtle way of
Wel l , all players have 'black' days. There is converti ng h i s advantage.
also another legitimate explanatio n : no one 44 . . . e3!
is perfect, and even outstanding g rand mas­ 45 l\i'c4+
ters have situations (each have their own ) in
45 'ili'f7+ does not help: 45 . . . "it'f2 46 �xg6
wh ich they feel less confident. And th is
"it'f4+ 47 'it>h 1 e2 48 'ili'd3 'it>f2 49 'ii'c2 �g5 .
means that there is still scope for fu rther
creative improvement. 45 . . . e2
46 'iif4+ 'iff2

IV 47 'iVc1 +

We will now analyse several more examples

(not so large-scale), where one of the
players faced the same problem : fi nding the
opti mal way of converting his advantage.
This can also be rega rded as a collection of
tests - a l l the positions are taken from my
card i ndex of exercises .

Szily - Ostvath
H u ngarian Championsh i p 1 954

Andras Ostvath undoubtedly calculated this

variation and decided that things would end
i n perpetual check: 47 . . . e 1 'iV 48 'iic4+ 'iife2
49 lli'f4+ 'ii' 1 f2 50 'ii'c 1 + etc.
But the perpetual check can be avoided by
promoting the pawn to a knight!
47 . . . e1 tt::l ! !
4 8 'i!Vc4+ 'it'e2
49 'ii'f4+ tt::lf3+ !
5 0 'iixf3+ ifxf3
51 gxf3 'it>f2
Wh ite is a pawn down , and the opponent's It is possible that the game can also be won
passed pawn is very da ngerous. All he can d ifferently. But no d i rect way is apparent
hope for is perpetual check, or the adva nce (apart from the one shown above) . B lack
1 90 � Grandmaster Technique

failed to cope with the problem and allowed 'it>g1 'it>g3 60 �h1 �2 Wh ite resigned.
his opponent to escape. Why did Black g ive u p h i s h-pawn? Why
44 . . 'ilfc1 ? 45 'ilff7+ �e1 (in the variation
. d i d n 't he advance it (38 . . . h6 or 38 . . h5)?
.

45 . . . '1t>e2 46 'i!Vxg6 'ii'f4+ 47 'it>g 1 'it'e5 a Obviously J a n Timman was concerned

draw is g iven by 48 'ii'd 6! , but not 48 �xc6? about a combination .
�a 1 + 49 �h2 'it>f2 with i nevitable mate) 46 38 . . . h6!
'ii'x g6 'i!Vf4+ 47 'it>g 1 'ii'f1 + 48 'it>h2 'ii'c 4 49 39 gxh 6 ! ? �xf6
'i\fxc6 e3 50 'ii'd 6, and the players agreed a
40 �h3
draw.
However, Igor Zaitsev showed that the
passed pawn could be stopped .
Browne - Ti mman
40 . . . l:tg4+!
Stockholm 1 972
41 'it>f2 �f4+
42 'it>g1
42 �g3 .l::i.f3+ , or 42 �e 1 �e4+ and
43 . . . .l::i.e 8 .

In the game 38 ... .l::i.c4?! was played , and if

White had repl ied 39 ttJxh7! he would have
created serious problems for the opponent
with the conversion of his adva ntage, in
view of the small amount of material 42 . . . .1Ld7 ! !
remaining on the board . For example: 43 h7 l:tf1 +!
39 . . . b4 40 ttJf6 .l::i. c3 4 1 �f2 ! . 44 'it>xf1 �xh3+
Walter Browne preferred 3 9 .l::i. h 3?! , but after 45 �f2 'it>g7
39 . . . b4 40 l::tx h7+ �e6 the enemy king
came into play and his position became
hopeless. There followed 41 �h3 �5 42
.l::tg 3 �c3 43 .l:!.g4 b3 44 �b4 �xg5 45 lt::lg 4
(see diagram)
.1Le4 (45 . . . �d5 would have won more
qu ickly, not fearing 46 h4+ �xh4) 46 �xe4
b2 47 .l:!.b4 .l::i.c 1 + 48 �g2 b1 'ii' 49 .l::i. x b1
�xb1 50 ttJf2 .l::i. b 2 51 'it>g3 �b3+ 52 'it>g2
�5 53 ttJh3 g5 54 lt::l g 1 'it>g4 55 h3+ 'it>h4
56 ttJf3+ �xf3 57 �xf3 �xh3 58 'it>f2 g4 59
 Grandmaster Technique ctJ 191

Kunitz - Dvoretsky S h i razi - Vasyukov

Bad Wiessee 1 997 Tirucch i rappalli 1 978

Any position with a n extra pawn prom ises Without resorting to d rastic measures, it is
Black good chances of success, but every­ hardly possible to convert the advantage. A
where the struggle still continues: for exam­ combination comes to Black's a i d .
ple, after 34 . . . l:tb5 35 'it>e2 or 34 . . . d 3 35 80 . . . g5!
l:11 xc5 .l:i.xb3 36 .l:i.c1 ! . H owever, he has
This move suggests itself: the th reat is
available a combination leading to a forced
81 . . . g4 and 82 . . . �e4 . However, this could
win .
not be played merely 'on general grou nds' ,
34 . . . d3! without precise calculation - Black had to
35 .U.1 xc5 reckon with the attack on h i s rook.
35 l:!.7xc5? d2 is totally bad for Wh ite. 81 'it>f3 h2!
35 . . . .l:i.h8 ! ! The exchange sacrifice 8 1 . . . l:!.xe5? 82 fxe5
3 6 'lt>g1 'it>h4 is sufficient only for a d raw: 83 l::i. a 1 !
g4+ (83 . . . �e6 84 .l:i.a6 �d5+ 85 'it>f2 is not
The rook is taboo: 36 .U.xd5 .l:i.h 1 mate . If 36
da ngerous for Wh ite) 84 'it>f4 �g6 (84 . . . h2
c.t>e 1 , then 36 . . . d2+ is decisive, while if 36 g3
85 'it>xf5) 85 e6 g3 86 e7 h2 (86 . . . g2? even
- 36 . . . .l:i.xc5 37 .l:i.xc5 .l:i.h 1 + 38 'it>g2 d 2 .
loses after 87 'it>f3 ! ) 87 .l:i.a8! 'it>h3! . Paradoxi­
36 . . . .l:!.dd8! cally, the goal is ach ieved by a sacrifice of
37 l:!.c1 d2 far more material - a rook!
38 .l:!.d1 .U.de8! 82 .l:i.a1
Of cou rse, not 38 . . . l:!.he8? 39 Wf1 . The Bad is 82 .l:i.h 1 ? �e4+ 83 'it>xe2 �xh 1 84
captu re of the d2-pawn is not possible fxg5 �f3 + .
because of mate . 82 . . . g4+ ! !
39 l:!.f1 .l:i.e1 Strictly speakin g , there was also a second
40 .l:!.d7 l:th 1 +! soluti o n: 82 . . . l:!.a2 ! ? 83 l:!.h 1 g4+ (but not
Wh ite resig ned . 83 . . . �h3? 84 'it>e3! g4 85 f5 with equal ity) 84
'it>g3 .U.a3+ 85 'it>xh2 g3+ (85 . . . �e4 is
1 92 � Grandmaster Technique

insufficient in view of 86 l::te 1 ! l::t h 3+ 87 'itr>g 1 Kra ntz - Yudovich

.l'.:th 1 + 88 'it>f2 g3+ 89 'it>e2 followed by i.d4) World Correspondence
86 'it>g2+ 'it>g4 87 .U.e 1 (87 l:ta1 l::t d 3 ) Tea m Championship 1 979-80
87 . . . l:ta2+ 88 'it>g 1 .ti.c2 ! ? and 89 . . . 'it>h3 - the
opponent can not hold this position. But if
Wh ite had played 82 .ti.c1 ! ? (instead of 82
l::ta 1 ), the rook sacrifice 82 . . . g4+ ! ! would
have been forced - it is now poi ntless to play
82 . . . l::t c 2? 83 l:t h 1 g4+ 84 'it>g3 , since there
is no check on the 3rd rank: the c3-sq uare is
controlled by the bishop.
83 'it>xe2
83 'it>g3 i.e4 .
83 . . . g3

Black can easily obta i n a queen end ing with

fou r pawns agai nst th ree on one wing:
39 . . . a2?! 40 i.xe6 'ifxe6 41 "ili'a4 'ikxc4 42
'ii'xc4 a 1 'ii' . But is it possible to win it? Is
there not a stronger a lternative?
After the exchange of m i nor pieces on e6
the white queen will occupy the a-file, and it
will be possible to defend the a3-pawn only
along the diagonal - then White will play
'ii'a 5 and c4-c5. The idea of breaking
through to the aid of the passed pawn via the
The two far-advanced passed pawns prove cu rrently barricaded e-file is not something
to be stronger than the rook. If 84 l::t a 8, then that i m mediately comes to m i n d .
both 84 . . . i.h3 and 84 . . . g2 are possible. 39 . . . e4! !
84 'it>f3 'it>h4 4 0 c5
85 i.d4 'it>h3 40 i.xe4? loses i m mediately to 40 . . . a2 41
86 i.f2 'ii'a 4 'ii'd 4 or 4 1 . . . 'it'b2 .
I n the event of 86 l:th 1 the switch i ng of the I n the event of 40 i.xe6 'it'xe6 4 1 'i'a4
bishop to the a8-h 1 d iagonal would have noth ing is g iven by 4 1 . . . 'ii'd 6 42 'ii'a 5 with
been decisive: 86 . . . i.d7 . the threat of 43 c5, while if 41 . . . 'ii'e 7 there
86 . . . g2 follows not 42 'ii'a 5 e3! , but 42 'ii' b 3. Black
Wh ite resigned in view of 87 .l:la8 i.g4+ 88 decides matters with 4 1 . . . e3! 42 'iix a3 (42
'it>e3 g 1 'it'. fxe3 'iix e3 is completely hopeless) 42 . . . e2
43 'ii'c 3+ 'it>f8 (43 . . . 'it>g8 ) 44 'it'e 1 'it'e4+! (but
not 44 . . . 'ikxc4? 45 f3 and 46 'it>f2 with
equal ity) 45 f3 'ike3 , and Wh ite's position is
 Grandmaster Techniq ue tLJ 1 93

lost: his q ueen has no moves, his king is not equally strong conti n uations: 46 . . . 'ili'a8 47
able to come to its aid, and the c-pawn will 'ii'a 2 'ii'a 6+ 48 'it>g 1 (48 '1t>e 1 'it'e6+ ) 48 . . . �a4
be stopped by the black king. followed by . . . 'ti'b4-b2 , or 46 . . . 'i!kh 1 + 47
Roughly the same picture results from 4 1 'it>e2 'ii'c 1 48 'itd3 'ii'f 1 + 49 'it>e3 'ii'e 1 + 50
�a? e3! 4 2 fxe3 (42 'ii'x a3 e 2 tra nsposes <t>d3 �xf2 .
into the previous variation) 42 . . . 'ii'e 4+ 43 45 . . . 'ii'f2+
'it>h3 'ii'f5+ 44 'lt>g2 (44 g4 'ii'f 3+ 45 'it>h4 h6) 46 'it>h3 'ii'f1 +
44 . . . 'ii'c 2+ 45 'it>h3 'ii'b 2, and the a-pawn 47 <t>h4 'ii'e 2
queens.
Wh ite resig ned .
40 . . . �c7!
41 'ti'a4 Petrosian - Kholmov
If 4 1 i.xe6 �xd7 42 i.xd7 , then Black does Vil n i u s 1 95 1
not continue 42 . . . a2? 43 c6 a 1 i¥ 44 c7 with
a draw (in this variation the inclusion of the
moves 42 . . . e3 43 fxe3 does not change
anyth ing), but 42 . . . <t>f8! (or 42 . . . 'it>f6 ! ) 43 c6
�e7 . The c-pawn is halted , whereas the
bishop is u nable to stop the a3-pawn.
41 . . . "it'xc5
42 'ii'x e4
There is no point in playing on two pawns
down . If 42 i.xe4 , then 42 . . . 'ii'd 4 most
simply decides matters.
42 . . . �d4
43 'i*'f3 lt:Jg5
44 'it'b3 tt:Je4 Ratm i r Kholmov carried out a l ittle combi na­
tio n , enabling him to win a second pawn .
33 . . . 'ii'x b2 1 ?
3 4 'ii' x e4
34 i.xd3 exd3 35 lixd3? ii'b 1 + was com­
pletely bad .
34 . . . tt:Jxf2 ! !
35 .l:i.b7
With the fai nt hope of 35 .. ."�xb7? 36 "ilixd5
�b6?! 37 �d4 �xd4 38 lixd4, when the
black knight is trapped . 35 'it>xf2 .l:i.xd 1 would
have left Wh ite the exchange down , while
after 35 �xd5 tt:Jxd 1 he would not have time
to take the knight because of the threat of
45 f4 36 . . . �b6+ .
In the event of 45 i.xe4 'ii'x e4+ 46 'it>f1 (46 35 . . . .l:i.xd 1 +
'it>g 1 'ii'e 1 + 47 'lt>g2 'ii'a 1 ) there a re two 3 5 . . . tt:Jxe4 was also good .
1 94 w Grandmaster Technique

36 .txd 1 'ii'd 2 Pigusov - Atal i k

This leads to an ending that is easily won for Reykjavik 1 994
Black.
37 'i!Ve2 'i!Vxd 1 + 38 'itxf2 'i!Vxe2+ 39 'itxe2
l:ta8! 40 'it>d2 a4 41 'itc2 a3 42 Wb1 a2+ 43
Wa 1 g6 44 .Uc7 h5 45 .l:tb7 'itg7 46 �d7 .lla 3
47 .l:l.e7 h4 48 .l:l.e4 g5 49 l:!.b4 f5 50 l:tb6 f4
51 l:Ib5 �g6 52 l:Ib6+ 'it>h5 53 �f6 �e3 (of
cou rse, not 53 . . . .l:i.g3 54 �a6 llxg2? 55
.l:l.h6+! �xh6 - stalemate) Wh ite resigned .
But now I will explain why Black's 33rd move
was awarded not an exclamation mark, but
a '!?' sig n , expressing some doubt. The
point is that the opponent could have repl ied
34 l:!.e 7 ! . The only way to retai n the two extra
pawns is by 34 . . . f5 35 .txd3 l:Ixd3 36 .l:!xd3
'i!Vb 1 + 37 'it>h2 'i!Vxd3 , but then there follows A typical way of converting an adva ntage is
38 'it'c5 ! , intending 39 'i!Ve5 . to exchange the last pieces and transpose
i nto a pawn endgame. The point is that pawn
endings ca n usually be calculated right to
the end. If the resu lts of the calculation are
favou rable for you , it will no longer be
necessary to play the position move by
move (with a non-guaranteed outcome and
the danger of making a mistake at any
moment) - by forcing events, essentially we
simply demonstrate a n arrived-at solution.
In the given i n sta nce the exchange can be
carried out by switching the bishop to g2. Let
us check the variation s .
33 .tf1 ! 'it>e8
34 .tg2 .td5
I doubt whether, with the opponent's pieces 34 . . . .txg2 35 'it>xg2 is hopeless - Wh ite's
so active, Black will be able to exploit h i s outside passed pawn will a utomatically
material advantage. B u t t h e n t h e question decide the outcome. The attempt by the
arises : should he have carried out the opponent to avoid the excha nge also does
combinatio n , and wouldn't it have been not pose any problems: 34 . . . .td3 35 i..c6+ !
better to prefer the simple move 33 . . . .l:tfd 8 , 'it> d 8 3 6 b6 'it>c8 37 f3 followed b y 'it>f2-e3 .
retaining a great advantage? 35 .txd5!
The conclusion : when forcing events i n a A d ifficult move - usually it is not recom­
favou rable situation , it is important to check mended to improve the opponent's pawn
thoroughly the variations and eval uate their structure . Here this is justified by the
consequences - otherwise you can easily possibil ity of a pawn breakth rough on the
squander you r adva ntage ! kingside.
 Grandmaster Techniq ue tZJ 1 95

35 . . . exd5 e4 36 'it>f1 'it>e6 37 We2 'it>e5 38 'it>e3 i.c6 ,

36 f4 gxf4 and Wh ite is u n a ble to make any progress.
It is poi ntless to play 36 . . . 'it>d7 37 fxg5 hxg5 Wh ite's play can be improved with the
38 h4. suggestion by E rnesto l na rkiev: 34 Wf1 ! Wf6
37 g4! 35 'it>e2 'it>e5 36 'it>e3 Wd6 37 i.d3.

Wh ite succeeds i n promoti ng a pawn just in

I n a joint analysis with E rnesto we were
time. Some of the followi ng moves may be
unable to find a way for Black to save the
interposed .
game.
37 . . . d4
38 �1 d3 A) 37 . . . e5 38 i.e4 i.a6 39 b7 (39 'it>f3 h5 40
h4 is also strong) 39 . . .'it>c7 40 i.c6 (or 40
39 h4 e5
i.d5 e6 41 i.c6) 40 . . . i.f1 41 h4 i.h3 42
40 'it>e1 f3
�e4 (the same move can also be made
41 b6 �d7 after fi rst exchanging pawns on g5) 42 . . .
42 b7 'it>c7 i.g2+ 4 3 f3 g 4 44 'it>e3 '>t>b8 4 5 i.e4 i. h 3 46
43 g5 hxg5 f4 , and the wh ite king breaks i nto the
44 hxg5 e4 opponent's position .
45 g6 e3 B ) To 3 7 . . . i.g2 White does not reply 3 8 h4?
46 g7 f2+ gxh4 39 gxh4 e5 40 i.e4 i. h 3 ! , but 38 g4 ! .
47 Wf1 d2 The followi ng variations arise:
48 b8'i¥+ 'it>xb8 B 1 ) 38 . . . i.xh3 39 �f3 �c6 40 �g3 i.xg4 4 1
49 g8'ik+ �xg4 �b6 4 2 Wh5 �c5 43 �xh6 �d4 44
i.e2 , and Black ca n not approach the f2-
It is not so easy to find and accu rately
pawn ;
calcu late this variation. But to eval uate the
conseq uences of other ways of playing for a B2) 38 . . . e5 39 f3 Wc6 40 �f2 i.h 1 4 1 i.e4+
win is even more d ifficult. 'it>xb6 42 'it>g 1 i.xf3 43 i.xf3 'it>c5 44 �f2
lf 33 b6 ! ? Black replies 33 . . . i.b7 (33 . . . �e8? �d4 45 i.b7 e4 46 'it>e2 'it>e5 47 cot>e3 h5 48
34 i.xe6 is bad ), for example: 34 f4 gxf4 35 i.xe4 with a n easy w i n ;
gxf4 �f6 and 36 . . . e5, or 34 i.f1 e5! 35 i.g2 B3) 38 . . . i.d5 39 i.e4 i.c4 40 f4!
1 96 � Grandmaster Techniq ue

39 �c5 �e5 40 �b6 i.. d 5, or 37 'iti>c5 e6

(37 . . . e4?? 38 �b6) 38 �d6 (38 �b6 i.d5 39
i.. x d5 exd5 40 �c5 d4) 38 . . . e4, and now
both 39 b6 e3 40 �c7 i.. d 5 41 b7 i.xb7 42
�xb7 �e5 43 �c6 �d4 and 39 i..xe6 e3 40
i.. c4 �f5 ! ? lead to a d raw.

In the final example a strong impression is

made by the tech nique of Paul Keres, based
on a precise calculation of variations, and
typical of the Estonian grand master's play.

Keres - Geller
After 40 . . . ..if1 41 fxg5 hxg5 42 h4! or Budapest 1 952
40 . . . ..ia6 41 fxg5 hxg5 42 �f3 e5 43 h4 gxh4
44 g5 the passed g-pawn decides the
outcome. 40 . . . e5 also does not help: 4 1
fxe5+ (4 1 fxg5 i s n o less strong) 4 1 . . . �d7 42
..ig2 ..ia6 43 �d4 ..ib5 44 �c5 ..ia6 45
..ic6+ �c8 46 �d5 ..if1 4 7 �e6 ..ixh3 48
�xe7 ..ixg4 49 e6 .
The game went 33 f3? ! ..ixf3 34 �2 i.b7
35 �e3 .

Wh ite is a sound pawn to the good . The a6-

pawn is also attacked , and he ca n pin the
bishop by 26 l:!d 1 . A big advantage is
retained after practically any of the nu mer­
ous tempti ng conti n uations - the problem is
to assess how promising they are and
choose the opti mal one.
If 26 .l:!.d 1 the opponent will reply 26 . . . .l:!.e8!
27 i.. x a6 (27 l:td2 ! ? a5) 27 . . . ..txa2 . Pawn
After 35 ... �e8?? 36 ..ixe6 �d8 37 b6 the exchanges normally favou r the defending
position became hopeless. There followed side, which means that this way hardly
37 . . . i.g2 38 �d4 i..f1 39 �c5 i.. g 2 40 �b5 deserves preference .
i..f1 + 41 �c6 , and Black resigned . I n t h e event o f 26 i.. x a6 .l:. a 8 Wh ite can
But meanwh ile, by conti n u ing 35 . . . �f6 ! 36 tra nspose into a bishop ending with an extra
�d4 e5+ , Black would have saved the paw n: 27 i.. e 2 .l:.xa2 28 l:lxa2 ..txa2, but his
game. For example: 37 �e3 e4 38 �d4 �5 advantage may not be sufficient for a win -
 Grandmaster Technique t2J 1 97

here the assessment expressed reg a rding playing 29 . . . .l::!.e 4! 30 .l:!.xa5 .l:!.c4 . Only 29 f3 !
26 .:td 1 is again appl icable. And after 27 c4 l:tc2 30 .l:!.c5 is correct, when the loss of a
Black can choose between 27 . . . .i:!.xa6 28 second pawn is inevitable.
cxd5 .l:!.a5 29 a4 'it>f8 and 27 . . . i.. x g2 28 'it>xg2 27 l:!.d1
.l:f.xa6. As is wel l known , ' rook endings a re
27 a5?! is prematu re on acco unt of 27 . . . .:tc6.
never won ' - with a deficiency of just one
F i rst the rook m ust be tied to the defence of
pawn , the opponent has the rig ht to cou nt on
the bishop.
saving the game.
There is another idea which deserves to be
stud ied : to remove the a-pawn from the
attack by the bishop, i n order to retai n both
threats : ii.xa6 and .l:f.d 1 (as is wel l known ,
often 'the threat is stronger than its i m medi­
ate executio n ' ) .
If 26 a3?! B l a c k has both 26 . . . .l:!. d 6 27 l:td 1
'it>f7 ! ? , and 26 . . . a 5 ! ? 27 .l:!.d 1 .l:f.e8 28 .l:txd5
l:l.xe2 29 'it>f1 1Ic2 30 I1c5 a4 - a second
pawn is not lost and the chances of a d raw
are very rea l .
But i f t h e a-pawn is moved two squares
forward , thi ngs will be bad for Black i n the
rook ending, and also there will be the 28 l::i. x d5 l::i. x d5 29 ii.c4 is threatened - for
additional possibil ity of fixing the weakness this reason there is no time for 27 . . . a5. If
at a6 by a4-a5. 27 . . . 1i.b7 there follows 28 .l:!.xd6 cxd6 29 a5! ,
26 a4! .l:i.d6 and the bishop ending is undoubted ly won :
Accu racy would also have been demanded Black is not only a pawn dow n , but he has
of Wh ite in the variation 26 . . . a5 27 lld 1 .l:f.e8 also been saddled with a wea k pawn on a6,
28 .t!.xd5 .t!.xe2 . fixed on a square of the colo ur of his bishop.
I th i n k that Yefi m Geller should have tried
27 . . . I1e6 ! ? . The variations 28 �f1 �b3 29
.l::!. b 1 �d5 and 28 �g4 l:!.e8 (weaker is
28 . . . l:!.e7 29 a5) 29 f3 (29 a5 1i.b3) 29 . . . �b3
30 lld7 c6 (or 30 . . . 1i.xa4) do not seem
sufficiently convincing to me - at any event,
here Black can fight on. And i n the rook
ending after 28 .l:i.xd5 .l:!.xe2 29 Wf1 .l:f.a2 it is
not possible to win a second pawn . But
nevertheless here we have a more favour­
able version for Wh ite of the rook endgame,
compared with those exami ned earlier. By
conti n u i ng 30 .l:i.a5 .l:i.c2 31 .l:txa6 .l:txc3 32
We2 Wf7 33 a5 .l:f.a3 34 .l:f.a7 Wf6 35 a6 Wh ite
I n the event of 29 g3? (or 29 Wf1 ? ) Black is will most probably wi n .
able to avoid the loss of a second pawn by 27 . . .
1 98 � Grandmaster Technique

28 a5! 31 . . . lle5
The th reat of 29 �xa6 has been created . 28 32 .l:tc6 .Uxa5
l::tx d5? did not work: 28 . . .l:txd5 29 �c4 c.t>e6
30 cJi>f1 cJi>d6 31 Si.xd5 c.t>xd5 32 c.t>e2 'l!ic4 33
'it>d2 cJi>b3.
28 . . . .l:te6
If 28 . . . '1iie 6, then 29 f4 �c6 30 Si.g4+ is
decisive.

33 .Ub6!
Accu racy to the end! After 33 .l:!.xa6? llc5 the
c3-pawn would have been lost.
33 . . . .i.c2
34 llxa6 .l:tc5
29 .i.f1 ! 35 lla3 .U.d5

Not 29 Si.xa6? .i.b3. It is also not possible to 36 f3 Itd1

win a second pawn in the variation 29 37 'liif2 llc1
.Uxd5?! .Uxe2 30 'iii>f 1 l:!.a2 31 .l:!.c5 c6 . 38 h4 i.g6
29 . . . Si.b3 39 Si.c4 'l!ie7
30 lld7+ cJi>f8 40 g4 h6
31 .l::tx c7 41 i.d5
White has increased his material advan­ With his last few moves Wh ite has strength·
tage. Now 31 . . . .Ue1 32 f3 lla 1 33 .l:!.c5 is ened his position to the utmost, and fu rther
hopeless for Black. resistance is pointless. Black resigned .
 CtJ 1 99

Artur Yusupov, Mark Dvoretsky

Analys i s of a Game

Dchampionship Vasya
voretsky. I n the 1 990 world u nder- 1 4 8 cxd4 f6
Emelin fi nished 9 exf6 lt:Jxf6
second , behind only the famous J u d it 1 0 ttJf3 i.. d 6
Polgar. He annotated in deta i l one of his
11 0-0 'Wic7
games, played agai nst the Roman i a n player
Gabriel Schwa rzma n . Today we are going to 1 2 �g5 0-0
analyse this game together. 1 3 il.h4
Why this game in particular? Wel l , fi rstly, it is D. Note the open ing va riati o n . Wh ite has
very i nteresti ng to deal with a genui nely chosen one of the most dangerous plans
conscientious analysis . You know, when you against the system chosen by his opponent.
read a phrase l i ke 'such-and-such a move In particular: without the inclusion of the
ca me i nto consideration ' , it says noth i n g . moves 1 2 lt:Jc3 a6. I fi rst saw it in the game
Yes, i t no doubt came i nto consideration . It is Zapata-Chern i n from the I nterzonal Tou rna­
a qu ite d ifferent matter, if the com mentator ment (Su botica 1 987).
tries to investigate what was correct, and Wh ite is th reatening the advantageous
what was wrong. One can agree with exchange of the dark-sq uare bishops by 1 4
Emel i n 's assertions or d ispute them , but il.g3 , for example: 1 3 . . . lt:Jg4 1 4 .tg3 (but not
here at least there is someth ing to th i n k 1 4 h3? .l:i.xf3 ! ) . The sta ndard reaction
about. 1 3 . . . ttJh5 is d u bious i n view of 14 'Wic2 , when
Secondly, it fits i n well with the basic theme 14 . . . g6? 1 5 .txg6! hxg6 1 6 'ikxg6+ lt:Jg7 1 7
of our session . From the open i n g , play went lt:Jg5 is bad for Black. All that remains is the
directly into a favourable ending for Wh ite . central freeing advance . . . e6-e5, but then
The problem of the tech n ical conversion of Black is saddled with a n isolated d5-pawn ,
his adva ntage faced Emelin throughout the wh ich in the approach ing endgame will
enti re game. become a weakness.
13 . . . eS
Emelin - Schwarzman 1 4 dxeS lt:JxeS
Fond d u Lac 1 990 1 5 lt:Jxe5 .txeS
French Defence 1 6 �g3 �xg3
1 e4 e6 Emel i n . In one of the previous rounds of the
2 d4 dS world championship I reached the same
position . My opponent D.Zifroni played
3 ttJd2 tt:if6
1 6 . . . �g4 . He excha nged on e2 and I was
4 e5 ttJfd7 soon able to press on the d 5-pawn .
5 c3 cS 1 7 'Wib3 ( D : 1 7 .l:i.c1 'Wid6 1 8 f3 i.. d 7 1 9 'Wid2
6 �d3 lt:Jc6 with the th reat of 20 f4 was strong)
7 lt:Je2 cxd4 17 ... .txe2 18 il.xe2 �xg3 19 hxg3 l:!.ac8
200 w Analysis of a Game

(D: 1 9 . . JlVb6 ! ? ) 20 ..if3

1 977), Wh ite captu red on g3 with his kn ight,

Here my opponent sacrificed a pawn , but d id which is somewhat weaker. It was 1 7 hxg3!
not gain sufficient compensation. that at one time I analysed with Yusupov,
20 . . .'it'c4 2 1 1\Vxb7 �f7 22 'ii' b 3 "it'xb3 23 and I remember that we did not fi nd a clear­
axb3 l:!.c2 24 l:!.ab1 l:!.b7 25 l:!.fe1 ! I ntending cut way to eq ual ise.
26 l:!.e3 ; Black can not take on b3 because of Yu . Probably B lack should avoid further
.lii.d 1 . exchanges and keep play i n the middlegame.
25 . . . l:!.d2 26 .l:i.ed 1 .l:i.xd 1 + 27 ..ixd 1 ltJe4 Say, 1 7 . . . 1Ld7 and then . . . �d6 or even
Here I went wrong: 28 i.g4? I should have 1 7 . . . 1\Ve5! ? , hoping if possible to develop an
played 28 l:.c1 ltJd2 29 l:.c3 l'!e7 30 .ig4 attack by . . . ltJg4 . Here too Wh ite's position
.l:Ie 1 + 31 '1t>h2 lL'lf1 + 32 'it>h3 'it>f7 33 i.f5 g6 is preferable. But it would appear that in the
34 l:.c7+ 'it>f6 35 i.d3 with an obvious endgame h i s advantage is g reater, and in
advantage . add ition the opponent has no chances of
Yusupov. Why did you p l a y d ifferently i n creati ng cou nterplay.
t h e game, a n d what was t h e reason for the 17 . . . �b6
mistake? 1 8 �b3 '*'xb3
E. I thought that th is was better, but in my 1 9 axb3 .lii. d 7
calculations I simply overlooked some el­
E . 1 9 . . . a5 20 b4 1eads to the loss of a pawn,
ementary deta i l .
for example: 20 . . . b6 21 bxa5 bxa5 (21 . . l:i.xa5
.

22 l:!.xa5 bxa5 23 .tf.a 1 ltJg4 24 f4) 22 b4 a4

1 7 hxg3! 23 ltJc3 .lii. d 7 24 b5 l:.fb8? ( D : after 24 a3 . . .

things are not so simple) 25 l:.xa4 l:.xa4 26

(see diagram) lbxa4 , and 26 . . . 1i.xb5? fails to 27 .l:i.b1 .
20 b4!
D. Strangely enoug h , this natu ral recaptu re
would appear to be a novelty! Both in the E. I prevent . . a7-a 5.
afore-mentioned Zapata-Chern i n encoun­ Yu . This is an importa nt point. If Black were
ter, and i n two earl ier games which I able to place h i s pawn on a5 u n p u n ished , he
managed to d iscover, played by Rad ulov would solve h i s problems.
and Smyslov against Vaganian (Len ingrad 20 . . . a6
 Analysis of a Game ttJ 201

E. The opponent is afraid that, by playing tem pting fi rst to strengthen the position on
b4-b5 , I will fix his pawn on a7. G iven the the kingside: 29 g4. H owever, the choice
opportun ity, he i ntends to exchange the here is a matter of taste.
bishops on b5. It was possible to defend Let's just go back a l ittle, to the position after
differently: 20 . . . l::Ifc8 , then . . . �c7 and . . . b7- 23 l:ta6 . Black also has another idea -
b6 , but a l l the same this would not h ave 23 . . . l:tf8 ! ? (instead of 2 3 . . . .1i.c8) . It is desir­
changed the evalu ation of the positi o n. able to defend the a7-pawn with the rooks
Yu . Don' t rus h . I n the endgame it is always from the side. Of cou rse, here too problems
important how the pawns a re a rranged . remain : 24 l:tfa 1 lt:Jg4+ 2 5 Wg 1 lt:Je5 26
Show us the variations that you analysed at �xa7 �xa7 27 l:txa7 lt:Jxd3 28 l::Ix d7 lt:Jxb4.
home. Wh ite stil l stands better - h i s rook is more
E. 20 .. J�fc8 21 f3 .l:Xc7 22 �f2 b6 . active, and the opponent has more pawn
weaknesses . But perhaps this is how B lack
should have defended?
21 lt:Jd4
D. Wh ite blockades the isolated pawn . But,
as Bent Larsen remarked i n his time, one
should always also consider the more d i rect
plan - the attempt to captu re it. I n the given
i n sta nce: 21 l:ta5!? followed by .l:!.d 1 and
lt:Jf4 .
21 . . . llac8
22 f3 l:r.fe8
23 �f2 .l:i.eS

23 .l:i.a6 ..tc8 24 .l:i.a3 a5 25 .l:i.fa 1 .l:i.ca7 (25 . . .

.l:i.b8 2 6 bxa5 bxa5 2 7 .l:i. 1 a2 .l:i.a7 2 8 .l:i.xa5
.Mxa5 29 l:.xa5 .Uxb2 30 l:.a8 and wins) 26
'it>e3 Si.d7 27 'it>d4 'it>f7 28 lZ'lc3 'it>e6 29 i.b5!
- after the exchange of bishops it is not clear
how to defend the pawn weaknesses .
D. I n the variation found by Emelin there a re
several instructive points as rega rds the
tech n ique of converting a n advantage. For
example, the timely centralisation of the
king , and the transformation of a n advan­
tage at the end (the exchange of the
opponent's passive bishop for the sake of
'processing' his pawns). 2 3 .l:i.a6 ! is a typica l 24 .Mfe1 ?
move . The rook ca n be placed immediately E. I wanted to play my king to the centre .
on a3, but it is u sefu l fi rst to l u re the black But this move is not the best; 24 l:tfc 1 ! was
bishop to a n i nferior square . stronger. If 24 . . . l:tee8 , then simply 25 l:tc5
Yu . It is not essential to p l a y 29 i.b5. It is with the advantage . The attempt to gain
202 � Analysis of a Game

counterplay with 24 . . . l:i.f8?! does not suc­ a rrange the pawns . The fate of the game
ceed . Wh ite repl ies 25 J::!. c7 . The check on may depend on the pawn structu re chosen
g4 clearly does not g ive anyth i ng , and by Black.
25 . . . .ib5 26 ltJxb5 axb5 27 l:txb7 ltJe4+ 28 If one proceeds, so to speak, from strictly
�g 1 ltJxg3 29 l:i.aa7 l:tg5 30 �xb5 is also structu ral considerations, then 26 . . . h5!? de­
bad . If 25 . . . �c8 there follows 26 l:i.ac1 , then serves serious considerati o n . For Wh ite it
the king retreats to g 1 , and it is not clear why would be usefu l to stretch the opponent's
Black has allowed the rook onto the 7th defences and create a target on the kingside.
rank. In this respect the g3-g4 advance is very
Yu . Here a simple principle operates: i n the u n pleasant. The move . . . h 7-h5 prevents it.
endgame the open fi le which is fu rther from After . . . h 7-h6 Black has more problems in
the king is more important (in the middlegame the knight endgame - after the exchange of
it is just the opposite). Therefore Wh ite bishops it will be hard to d rive the kn ight
should have fought for the c-fi le, and not from f5 .
exchanged the rook on e5, which is rather P robably Wh ite should h ave advanced his
stupidly placed . pawn to g4 on the previous move , instead of
24 . . . .i:lxe1 26 Wd2 .
25 'it>xe1 'it>f7 E. I n the event of 26 . . . h 5 I simply strengthen
26 �d2 my position by 27 .l::!. e 1 with the th reat of 28
lle5 .
Yu . Yes, y o u a re suggesti ng t h e m ost
natural pla n . Let's have a look. Black can
probably offer the exchange of rooks by
27 . . . l:r.e8 .
E. Then I play 28 l:i.c1 , a n d if 28 . . . �c8 29
.l:l.xc8 .ixc8 30 .if5 .
D. This is not dangerous i n view of
30 . . . �xf5 3 1 ltJxf5 ltJe8 an d then 32 . . . '1t>e6.
P robably Wh ite shou l d n 't exchange on c8 -
29 �c5 ! is stronger. If 29 . . . We7 , then 30 .if5
now gains in strength . Black should con­
sider 29 . . . g 5 ! ? , removi ng his g7-pawn from
a possible attack.
26 . .
. g6?! Yu . Vasya analysed . . . h 7-h6. Let's check
D. Amusingly, Black has placed all his his analysis.
pawns on squares of the colour of his own E. If 26 . . . h 6, then after 27 g4 rt;e? I
bishop. Do you remember that at the exchange bishops on f5 . The knight reaches
previous session of the school we analysed f5, from where it can not be d riven away.
the game Polugayevsky-Mecking (Mar del Sooner or later Wh ite will get to the weak g?
Plata 1 97 1 )? There Mecking defended i n the or b7-pawns. Therefore my opponent de­
same way, and this did not turn out wel l . cided to cover the f5-sq ua re immediately.
Yu . The position i s worthy of more thorough D. I s it real ly all so clear? I n you r notes to
consideration. Here we encou nter a very the game you give the variation 28 �f5 .ixf5
important endgame problem - how to 29 ltJxf5+ Wf8 30 Wd3 . Let's conti nue it:
 Analysis of a Game ttJ 203

30 . . . .Uc4 31 b5 tt:'ld7! - Black gains cou nter­ rather more accu rately. Say, 28 b3 �d6 29
play. Besides, the exchange on f5 is not i.f5 . Black has to move his knight from f6 (it
essential - there is also the i mmed iate is not doing anyth i n g there ) , but where to?
28 . . . .Uc4 ! . Let us sum up. After both 26 . . . h6 and
Yu . I n itself the exchange o f bishops does 26 . . . h5 Wh ite reta i n s the better chances,
not yet win the game, althoug h it creates but Black can defe n d . It seems to me, that
dangerous threats . It also has d rawbacks - out of all the possible pawn moves on the
vul nerable points appear in Wh ite's position ; kingside, he chose the least successfu l .
for example, the c4-square i s wea kened . Note that i n m a n y variations t h e defensive
E. Wh ite should probably play more plan i nvolves playing the king to the centre ,
accu rately. I suggest 28 .l:!.e 1 + �d6 29 b3. to d6, or exploiting the open c-file. All these
The th reat of 30 �f5 is renewed . resou rces appeared as a result of the
D. Black has to reply 29 . . . .l:!.e8 30 .ti.c1 (30 exchange of the 'wrong' rook on the 24th
.ti.xe8 tt:'lxe8) 30 . . . .l::!. c8 . move!
E. But then 3 1 .l::!. xc8 i.xc8 32 i.f5 . N ow let's retu rn to the game.
D. Never m i n d , for the moment there is a 27 g4 tt:'le8
defence: 32 . . . i.d7 33 i.xd7 'it>xd7 34 tt:'lf5 Yu . On f6 the knight has no futu re - it must
tt:'le8 . be moved from there. The only question is
E. After 35 'it>d3 Wh ite has an obvious whether to do this immed iately, or fi rst
advantage. prevent g4-g5.
D. Black's position is i ndeed u npleasant, E. Black would have d o n e better t o choose
but he still has some cou nterplay. 35 . . . <;i.>c6 27 . . . h6. O n h 7 the pawn is weaker than on
suggests itself, with the i ntention of attack­ h6. Then I was intending to play 28 .ti.h 1 'it>g7
ing the wh ite pawns on the b-file. 29 tt:'lb3.
Yu . It transpires that the move b2-b3 had
not only virtues!
D . The exchange of bishops on f5 is a
double-edged decision , since Black's 'bad'
bishop is exchanged. Of cou rse, i n return
Wh ite gains some important sq ua res and
attacks the enemy pawns. But if Black can
parry the d i rect th reats, h i s position may be
improved .
E. There is one more try. I will not g ive a
check on e 1 , but play 28 tt:'lb3.
Yu . We seem to have tal ked you out of the
bishop exchange. Very wel l , let's check this.
For the moment Black's actions a re obvious:
28 .. .<it>d6 29 tt:'lc5 i.c6 , and if 30 .t:!.e 1 , then If 29 . . . .Ue8 there follows 30 tt:'lc5 �c8
30 . . . �e8 . If 30 i.g6 I play 30 . . . l:tc7 , i n order (30 . . . �c6 is bad in view of 3 1 .l:!.c1 .t!.c8 32
to have the move . . . .ti.e 7 . tt:'lxa6 bxa6 33 i.xa6 �c7 34 b5) 31 �c3
Even s o , the exchange o f bishops on f5 is a tt:'ld7 32 �d4 . In the event of 32 . . . ttJxc5 33
good idea ; only, it must be put i nto p ractice bxc5 i.e6 34 .l::!.e 1 the position is completely
204 � Analysis of a Game

won (b2-b4 , l:!.e5 , i.. d 3-c2-b3). And if E. If 30 . . . d4, then 3 1 l:!.e 1 tt:Jd5 32 lle4 is
32 . . . tt::lf6 , then 33 l:!.c1 followed by tt:Ja4, and possible.
the rook invades on the c-file. Yu . Black is forced to reply 32 . . . tt:Jxb4 33
Yu . It is very obvious that Wh ite's advan­ .l:i.xd4 tt:Jxd3 . I would exchange pawns - this
tage has sharply increased . But active is an achievement for Black. 33 .l:te7+! Wf6
defence by Black should also be consid­ 34 .l::i. x d7 .l:!.xc5 is more dangerous for him.
ered : 32 . . . tt:Je5 with the th reat of 33 . . . tt:Jc6 + . The rook ending after 35 .ll x b7 tt:Jxd3 36
E. T h e check can b e prevented b y 33 b5 ! . 'it>xd3 is, of cou rse, i nferior, but by no means
Yu . Yes, after 3 3 . . . tt:Jxd3 34 tt:Jxd3 axb5 35 defi n itely lost.
�e 1 Wh ite has a decisive advantage. What D . And yet a nother plan of defence should
else ca n be devised? Let's try 33 . . . b6 ! ? 34 be examined: 29 . . . .i.b5 ! ? . For example, 30
tt:Jxa6 tt:Jxd3 35 'it>xd3 �d7. Now 36 'it>d2 tt:Jc5 i.. x d3 31 'it>xd3 a5.
�xb5 37 tt:Jc7 l:!.e2+ is pointless. In the E. Then 32 tt:Je6+ 'it>f7 33 tt::ld 4 .
event of 36 'it>d4 the pawn ca n not be take n ,
D. Of cou rse, after 33 . . . axb4 34 .llx h6 the
b u t 3 6 . . . �e2 is possible. U nexpected ly
position favou rs Wh ite, but 34 . . . .l:!.c1 retains
Black has gai ned cou nterplay. All the time
some cou nterplay. True, after 3 1 . . . a5?!
Wh ite has to reckon with active possibil ities
there is a far more u npleasant reply: 32
such as th is.
l:Ic1 ! . Therefore Black should try 31 . . . b6!?
D. Wh ite stil l retains a g reat advantage by 32 tt:Jxa6 h 5 33 gxh5 tt::lx h5 or 33 g5 tt::l h 7 34
36 tt::l c7 .l::i. c8 37 l:!.c1 i.. x b5+ 38 'it>d4 or 38 f4 tt::lf8 with some cou nterplay. I n difficult
'it>d2 . A check on e6 is th reatened , and the situations it is someti mes worth defending in
d5-pawn is under attack. But if he doesn't
this way - sharply change the pattern of the
want to go in for complications, he ca n
play, and go i n for material or positional
simply play 32 tt:Jxd7 (instead of 32 'it>d4)
concessions for the sake of activating you r
32 . . . i..x d7 33 'it>d4 .
forces.
E. Another set-up is no better for Black -
Wh ite ca n double the opponent's pawns : 30
29 . . . .l:tc7 30 tt:Jc5 i.. c8 3 1 'it>e3 b6 32 tt::l a 4 . . .
i.. x b5 axb5 31 tt::lc 5. I had i n mind 3 1 . . . b 6 32
D . Stop, stop, you ' re overlooking 3 1 . . . a5! . tt::le 6+ 'it>f7 33 tt::l d 4 .l:tc4 34 'it>d3 h5.
Yu . I n add ition Wh ite has to reckon with Yu . Unfo rtu nately, after 35 b3! Wh ite has a
30 . . . d4 !? (instead of 30 . . . i.. c8 ) . 31 . . . tt:Jd5 is g reat advantage .
threatened .
D . Yes, that's true. B u t th is means that the
D. Remember: earlier, in the analysis of
best p l an of defe nce is nevertheless
26 . . . h6, Vasya underestimated the rook
29 . . . .l:tc7 ! .
move to c4 , wh ich gave Black excellent
counter-chances. When you sta nd better in I have some dou bts about 29 tt::l b 3. Should
an endgame, you must all the time keep an the knight be moved from the excellent
eye out for sudden activity by you r oppo­ square d4? In my view, consideration should
nent. I th in k it is typical of Vasya to be g iven to 29 .l:te 1 'it>f7 30 .l:te5 .l:te8 (3 1 g5
underestimate his opponent's possibilities. was th reatened ) 3 1 .l:txe8 and 32 'it>e3 .
This is dangerous, and is l iable to cost many Yu . Whatever d ifficulties Black would have
poi nts, especially when trying to convert an subseq uently faced , it is clear that 27 . . . h6
advantage. You overlook something - im­ should have been played . The prospect of a
med iately cou nterplay flares u p , and of your kingside bind by g4-g5 is just too u npleas­
advantage noth ing remains. ant.
 Analysis of a Game l2J 205

D . It is good that such a n a nalysis has been

made! It would be simplest to stop here and
say 'Wh ite has the adva ntage ' . Yes, the
advantage , but is it sufficient for a win? I n
over-the-board situations there i s not usu­
ally any particular point i n seeking an
answer to this q uestion - it is sufficient to
u nderstand whether o ur position has im­
proved or deteriorated, and whether we
have extracted the maxi m u m possible. But
when there a rises a position which can be
eval uated exactly, you should endeavour in
analysis to establish the truth .

28 .l:ta5? ! 34 rJ;; e 3 .i.f5 35 �e2 'it>e5 36 f4+ 'it>d6 37

.i.f3 b6
E. I should first have played 28 g5! , fixing
E. The pawns should be moved off the light
the h7-paw n . And then thought where to put
squares.
the rook: manoeuvre it to c5 or place it on
h1 . 38 'it>d4 .i.e6 39 g3 a5 40 bxa5 bxa5

28 . . . �e6 Now I must seize control of the h 3-c8

diagonal with my bishop and beg i n adva nc­
E. Another possibil ity is 28 . . . tLlc7 . Then 29
ing my kingside pawns.
g5 rJ;,;e7 , and here Wh ite does best to retreat
41 .i.g2 (zugzwa n g) 41 . . . .i.f7 42 .i.h3 .i.e8
his rook to a1 with the th reat of 31 .Uh 1 . If he
43 .i.c8
plays the i naccu rate 30 .l:lc5 ? ! , then after
30 . . . rJ;;d 6 31 .Uc1 tLle6 ( 3 1 . . . 'it>e5 32 rJ;; e 3 Yu . It would be desirable for Black to
tLle6 33 f4+ ! , and 33 . . . tLlxf4? 34 tLlf3+ is not d islodge the bishop from c8 by 43 . . . 'it>c7 , but
possible) 32 .Uxc8 .i.xc8 33 tLlxe6 .i.xe6 this then there fol lows 44 �e6. If only this same
leads to a won bishop ending. position could be obtained with the bishop
on f7 . . . But I don't see how this can be
achieved .
D. If 42 . . . .i.g8 (instead of 42 . . . .i.e8 ), then 43
f5! .
43 . . . .i.f7
E. The opponent has to reckon with f4-f5 .
For example, if 43 . . . .i.a4, then 44 f5 .i.c2 45
f6 .i.b3 46 f7 rJ;,;e7 47 �e6 and 48 �xd5 is
decisive .
44 g4 .i.e8 45 .i.b7 .i.f7
After 45 . . . .i.d7 46 .i.xd5 .i.xg4 4 7 �g8 Black
loses a pawn (if 47 . . . rJ;,;e7, then 48 'it>c5 is
decisive ) .
46 f5 � g 8 47 � a6 .i. f7 48 .i.d3 .i.e8 49
Yu . Here you have a very deep and .i.b1 .i.f7
interesting analysis. Please show us it. Bad is 49 . . . a4 50 .i.c2 with zugzwa n g .
206 � Analysis of a Game

50 i.a2 i.g8 51 .tb3 i..f7 52 .ta4 Black's objective is to g ive u p h i s bishop for
Again zugzwang. the g-paw n . Wh ite has two moves: 61 i.xd5
and 6 1 i.f5 .
52 . . . i.g8 53 fxg6 (53 i.. e 8 '3;e7) 53 . . . hxg6
54 i.e8 i.. e6 (54 . . . i.. h 7 55 i.f7) 55 i.xg6 A) 6 1 i.f5 i.xf5 (6 1 . . . �xb3 62 g6) 62 �xf5
i..xg4 56 Ji..f7 Ji..f3 d4 - the pawns promote simu lta neously.

Now 57 g6 is prematu re : 57 . . . i.h5 58 i.e8 B ) 6 1 i.xd5 i.c2 (or 6 1 . . . i.g6 62 �f6 i.c2),
i.g4 59 g7 i.e6 60 i.g6 i.g8 61 i.f5 '3;e7 and how can Wh ite improve his position?
62 �e5 d4! with a d raw. D . Perh a ps instead of 59 We5 Wh ite should
change pla n : 59 i.f7 and 60 g6. After all,
57 i.g8! i.e4 58 b3
now the black bishop does not manage to go
via f3 to h 5 . I n the event of 59 . . .'it>b5 I had in
mind 60 i.xd5 i.g6 61 i.e4 i.f7 62 �c3
'it>c5 63 g6 i.e6 64 i.d3 and 65 i.c4 .
E. 60 g6 i.xg6 6 1 i.xg6 �b4 (6 1 . . . a4 62
i.e8+ ) 62 i.c2 is even simpler. 60 . . . �b4 61
g7 i.h7 62 i.xd5 is completely hopeless ­
the wh ite king goes to h 6 .
Yu . Yes, that's true. But I ca n also change
the plan of defence. If 59 i.f7 I play
59 . . . 'it>d6 ! 60 g6 �e7 61 �e5 ..tc2 with a
d raw.
It appears that here mutual zugzwang
positions beg i n to a rise. Wh ite ca n try 59
i.e6 (seeing as after 59 . . . 'it>b5 60 i.xd5 we
58 . . . i.c2 (58 . . . i.f3 59 g6; 58 . . . �c6 59 �e5 have fou n d a w i n ) 59 . . . 'it>d6 60 i.f7 .
with the th reat of Ji.. g8-e6-f5) 59 i.xd5 a4
D. Black replies 60 . . . i.f3 . We have reached
60 bxa4 Ji.. xa4 61 g6 i.c2 62 g7 i.h7 63
a position wh ich we a l ready had after the
i.a2 '3;e7 64 'it>e5 '3;e8 65 '3;f6, and Wh ite 56th move, only the wh ite pawn has moved
wins. to b3.
D. A remarkable analysis. The wh ite bishop Yu . So, as yet we have n 't fou n d a win. It has
marches around the entire board . The to be investigated whether Wh ite's plan can
length of the main variation is more than 30 be improved . Vasya , please do this at home,
moves! to complete you r analysis.
Is it really all correct? In one of his a rticles D . Later Erne/in found a simple improve­
Bent Larsen asserted that long variations ment. Let us return to the position after
always have some mistake in the m ; when Black 's 56th move.
he sees them, it awakens i n him the i n stinct
of a killer, a striving to immed iately bury the (see diagram)
entire analysis. In the variation examined by him 57 g6 1i.h5
Yu . Let's return to the position after 58 b3. 58 .tea ..tg4 59 g7 ii.e6 60 ii.g6 i.gB 61
The white b-pawn is vulnerable - this is ii.f5! 'it>e7 62 �e5 d4! White does not
suspicious. Since everything else loses, capture the pawn, but makes a waiting move
let's try allowing the wh ite king to go to e5. with his bishop (63 i.c2 or 63 i.e4), and the
58 . . . 'it>c6! 59 'it>e5 'it>c5 60 i.e6 'it>b4 ! . opponent finds himself in zugzwang.
 Analysis of a Game ltJ 207

D . I was watch ing this game when it was

played . From the side, of cou rse, you don't
delve i nto details, but some general impres­
sions nevertheless rem a i n . It seemed to me
that the wh ite rook had strayed off cou rse
and was cramped among the black pieces
and pawns, as a result of which the
opponent later acq u i red excellent saving
chances. If Black did not h ave a knight, the
rook would be excellently placed on b6,
tying the enemy rook to the defence of the
b7-pawn . But the knight on d6 seriously
restricts the activity of the wh ite rook.
3 1 .l::t a 5! was m uch stronger. The rook goes
Now we will contin u e the analysis of the
either to a?, or via a1 to h 1 . First, of cou rse,
game.
it will be necessary to cover the c4-square
29 b5?
by playing b2-b3. 3 1 . . .lk4 32 tt:'Jc2 is
E . Here I was over-hasty. Aga i n I should pointless, while i n the event of 31 . . . tt:'Jc4+ 32
have fixed the enemy pawn s by 29 g 5 ! . li.xc4 l:txc4 33 �d3 ( i ntending 34 :b5)
Yu . You began action on t h e queenside, Wh ite has a n undisputed advantage.
without completing you r work on the kingside. 31 . . . cJ;; e 7
E . I miscalculated : I considered 29 . . . axb5 32 g5 ..td7
30 .l:i.xb5 tt:'Jd6 3 1 l:tb6 �e7 and though that I
33 'it>e3 .)lc6
would win a pawn by 32 tt:'Jb5. I overlooked
the reply 32 . . . l:rc6 . D. Schwarzman conducts the entire game
very passively. I would have preferred
Yu . Yes , and after 32 . . . tt:'Jxb5 a pawn is not
33 . . . l:rc1 !? . The rook should pester the
lost (33 lixb7+? tt:'Jc? ) .
opponent, not allowing him to calmly
29 . . . axb5 strengthen h i s positi o n . Rook activity is one
30 .l:!.xb5 tt:'Jd6 of the important principles of playing end­
ings.
E . Wh ite would have replied 34 tt:'Je2 ,
prepa ring �d4 and tt:'Jf4 . 34 . . . l:rd 1 is not
possible because of 35 tt:'Jc3.
D. Fi rstly, th is is not so in view of 35 . . . d4+ ! ,
when 3 6 cJ;;x d4 tt:'Jf5+ leads t o a d raw. I n
additio n , 3 4 . . .l�h 1 ! 3 5 �d4 ..tc6 3 6 tt:'Jf4 l:!.d 1
or 36 tt:'Jc3 .l:.h4+ (36 . . . .l:i.h2) is possible. You r
pieces a re nicely placed , b u t it is not easy to
make prog ress - the black rook h inders this.
E . The check on h4 ca n be prevented by
36 f4.
D . Then , say, 36 . . . .l:!.d 1 , and Wh ite does not
31 .l:!.b6? have 37 tt:'Jc3? tt:'Jf5+.
208 � Analysis of a Game

Yu . Here it is not a matter of specific moves. D. You see, with your roo k on the a-file the
It is clear that Wh ite has strayed from the bishop wou l d n 't h ave bothered you at all.
correct cou rse . His rook on b6 is i nactive , Whereas here the bishop restricts the rook.
merely attacking the b7-pawn , which is But even so, you shouldn't have excha nged
secu rely defended by the m i nor pieces. If it.
the rook had been on a 1 , none of this 34 . . . l:!.xc6?
counterplay would have arise n ; on the
E. Good d rawing chances were offered by
contrary, it is the wh ite rook which would
34 . . . bxc6 ! . After 35 'it>d4 the most accu rate
have created th reats from h1 or e 1 .
reply is 35 . . Jlc7 ! .
An advantage is usually accu mulated l ittle­
35 . . . 'it>d7 is weaker: 36 'it>c5 ( 3 6 'it>e5 l:!.e8+
by-little, but it can also be lost l ittle-by-l ittle .
37 'it>f6?? lle7 and the king is in a mating
Thus here Wh ite h a s lost t h e g reater p a rt of
net) 36 . . . l:!.c7 37 l:!.b8 .
his advantage as a result of such a 'trifle' as
the poor position of his rook. I n this ending D. Even so, the king move looks natural - it
the rooks are the strongest pieces, and their frees the rook from the defence of the c6-
activity is of enormous sign ifica nce . Both pawn . I suggest checking 35 . . . 'it>d7 36 'it>c5
players u nderestimated the i mportance of tt::lf7 ! . The enemy king on c5 is too strong -
this factor. Black must try to evict it. If 37 .l:tb7+ , then
37 .. Jk7, while if 37 f4 1 reply 37 . . . tt::l d 8 38 f5
E. It is not essential to play 35 �d4 - the
'it>c7 (or 38 . . . .l:ta8). Wh ite retai n s the better
rook can be brought out via b4 .
chances, but the play is not just in one
D. But then you are forced to lose ti me. d i rection , and the situation becomes rather
Besides, 'it>d4 forces the black bishop to tense.
take up a passive position on c6 . Whereas i n
Yu . Perhaps Wh ite should nevertheless
the event o f 3 5 l:!.b4 you constantly have to
exchange rooks: 37 .l::I b 7+ llc7 38 l:!.xc7+
reckon with an exchange of minor pieces on
�xc7 39 f4 .
f5.
D. I must play my knight to b7. But not
necessarily via d8 - I will try 39 . . . tt::ld 6,
restra i n i n g f4-f5 .
Yu . Then 40 g4 tt::l b 7 + 4 1 'it>d4 'it>d6 42 f5 .
D. The g-pawn has to be given u p , but in
return Black ca n become active in the
centre : 42 . . . c5+ 43 'it>e3 c4 44 fxg6 hxg6 45
..txg6 'it>e5 followed by . . . d5-d4+. If he can
manage to exchange on the queenside, the
d raw will be not far off.
E. I n the minor piece endgame I also
considered a nother defensive idea: at the
point when f4-f5 is played , to answer . . tt::lf7 ,
.

captu re on g5, and block the remaining

34 tt::lx c6+? wh ite pawn with the knight.
E. I should probably have withd rawn my Yu . A good pla n . Apparently Black does
rook. I was simply tired of playing against i ndeed have good d rawing chances.
this bishop and I decided to exchange it. E. I also do not see how Wh ite can win after
 Analysis of a Game CZJ 209

35 . . . l:i.c7 . For example: 36 l:tb8 .l:tc8 37 l:txc8

tt:Jxc8 38 �c5 �d7 39 f4 li:Jd6 40 b4 <t;c? 4 1
ii. b 1 'it'd? (weaker i s 4 1 . . . tt:Jf7 4 2 b 5 cxb5 43
ii.a2) 42 g4 We? 43 Wd4 (43 f5 tt:Jf7 )
4 3 . . . � d 7 4 4 �e5 �e7 .
Yu . Of cou rse, you should n 't have taken o n
c 6 , al lowing Black t o con nect his pawn s -
after this it would appear that the game
should have ended i n a d raw.
E. Schwarzman a l ready had l ittle time left,
and with h i m about to get i nto ti me-trouble I
wanted to change the position somehow.
Yu . Such an approach is by no means
always justified . As a ru le, when the oppo­ 41 . . . .l:tg1
nent is in time-trouble you should still a i m to E. The roo k ending is hopeless: 41 . . . tt:Je4+
make the strongest moves . This is a more 42 �xe4 dxe4 43 l:tb6+ Wf7 44 .l:txb 7 + <t;e6
effective strategy. (44 . . . �g8 45 l:i.e7) 45 .l:tb4 I:te 1 46 Wd4 e3
Show us what happened i n the game. 47 �b3 e2 48 .l::t e 3+ <t;d6 49 �e4 .I:tg 1 50
35 l:tb4 l:i.c7 l:txe2 l1xg4 51 .l:!. h 2 .
36 �d4 �e6 D . T h e black king s h o u l d have moved
towards its kingside pawns: 48 . . . �f7 (in­
37 l:i.b6 .l:tc1
stead of 48 . . . Wd6?) . For example: 49 �e4
E . My opponent apparently believed me,
�g7 50 �f3 (50 b4 l:i. b1 51 I:txe2 .l:!.xb4+)
that after 37 . . . l:i.c6 the position was lost. I n
50 . . .l:tf1 + 5 1 �xe2 l:i.xf4 with a d raw. After
fact that i s what h e should have played . But
49 �e5 Wg7 50 b4 ! (50 We6? .l:!.f1 ; 50 f5
he was i n serious time-trouble.
�f7 ! or 50 . . . .U.g 1 ) 50 . . . .l:tb1 5 1 .:Xe2 .l:txb4 52
Yu . Were you both short of time? l:i.d2 l:tb7 (52 . . . .l:tb5+ 53 .l:!.d5) 53 <t;e6 Wh ite
E . No, I stil l had some time left. p robably wins, but this entire variation is
Yu . And on what move was the time contro l , rather complicated , and in addition Black
t h e fortieth? could have ta ken control of the e6-sq uare
E. The fiftieth . beforehand by 50 . . . �f7 ! ? , and only then
played 51 . . . .l:tb 1 .
Yu . Wel l , then he's is a bad way. To hold
such a position i n time-trouble is al most Wh ite probably wins more simply with 47
impossible. �e4 (instead of 47 l:!.b3 ), and if 47 . . . e2, then
48 �f3 .
38 f4 l:td 1
42 ii.c2 l:!.c1
39 g4 �e7
Yu . 42 . . . l:!.xg4 loses to 43 l:!.b6 . But isn't it
40 l:tb3
possible, by playing 42 . . . l:tg2 , to try and trick
E . Th reatening 41 <t;xd5. Wh ite? If 43 l:!.e3+ 'it'd? 44 ii.a4 + , then
D. Wh ite also had another i nteresti ng idea: 44 . . . b5. In the event of 43 ii.d3 the rook will
4 1 b4! ? (zugzwa ng ! ) 41 . . . �d7 42 �c5 , return to g 1 . For the moment the eval uation
transposing i nto a favourable rook ending. of the position seems unclear to me; a
40 . . . �e6 defence is sti l l possible.
41 �c5 E . All the same Wh ite must be better.
210 � Analysis of a Game

Yu . Yes, but the question is now on another 46 . . . b6+

plane: is there a forced win or can Black E. 46 . . . lt:ie4+ 47 �xe4 dxe4 could also
successfu lly defend? Hi s rook has become have been tried , but I have time to capture
active and is attacking you r pawns . You can the g6-pawn and stop the passed e-pawn.
no longer say that Wh ite has a n obvious For example: 48 .l:!.h6 e3 49 :!.xg6+ 'it>f7 50
adva ntage - you must specifically check .l:!.f6+ 'it>e 7 51 f5 .l:!.g2 52 .l:te6+ 'it>f7 53 g6+
whether it is possible to crack the oppo­ �g7 54 .l:!.xe3 .l:!.xg4 55 .l:!.e7+ �f6 56 .l:!.f7+
nent's defence. 'it>e5 57 g7 .
43 .:f.c3 .l:!.e1 D. Wh ite adva nced his pawns with gain of
44 .l:':.h3 .l:!.e2? tem p i , exploiti ng the position of the enemy
Yu . What for? The rook should have returned king . 49 . . . 'it>d7 50 f5 .l:!.g2 should be checked.
to c1 . What would you have done then? And i n stead of 52 . . . 'it>f7 Black had the more
tenacious 52 . . . 'it>f8 . Perhaps here too he is
E. Defended the bishop with 45 .l::!. h 2 .
lost, but perhaps not.
D . T h e opponent can reply 45 . . . lt:ie4+ 4 6
Yu . This was certainly Black's last chance.
'it>d4 (otherwise 4 6 . . . .l:!.f1 ) 46 . . . 'it>d6 ! ? , and if
47 ii.xe4 , then 47 . . . .l:!.c4 + . 47 Wc6 tt::i c 8

Yu . Black also h a s t h e reply 45 . . . lt:ic4 . It is 48 f5+ 'it>e5

true that after 46 .l::!.e 2+ it is bad to play 49 fxg6 �b3
46 . . . 'it>d7? because of 47 i.a4 + . But he can 50 i.f5 .l:tc3+
retreat 46 . . . 'it>f7 , when 47 'it>xd5? l:xc2 ! is 51 'i!id7 b5
not possible. The result of the game 52 g7 tt::i b 6+
becomes problematic. With 44 . . . .l:':.e2? the
53 'it>e8 .l:!.a3
opponent really played i nto you r hands,
al lowi ng you r to activate you r bishop. Had it 54 g8�
not been for his ti me-trouble, I th i n k you Black resig ned .
would have had to pay for the positional
errors made earlier: 28 .l::!. a 5?, 3 1 .l:!.b6? and D. Wel l , what are you r impressions?
34 tt::ixc6?. Yu . The endgame turned out to be rather
45 i.d3 .l::!. x b2 instructive, and in it there were several
46 .l:!.xh7 interesting poi nts.
The first problem which both players en­
cou ntered was how to a rrange their pawns .
First there w a s a clash on t h e queenside.
With b3-b4 Wh ite threatened to cramp his
opponent, and the latter did not find anything
better than to reply . . . a7-a6. In principle,
with a lig ht-squa re bishop it would have
been better to keep the pawns on dark
squares.
Then a similar problem arose with the
kingside pawns. Black should certa inly have
prevented the activation of the wh ite pawns
with g 3-g4 , by playing . . . h 7-h5 . But he
 Analysis of a Game 4J 21 1

chose what was probably the worst set-up formative , but I gai ned the impression that
and allowed h imself to be cramped . towards the end Vasya became a l ittle tired ,
There was an interesting point i nvolving the and stopped d rawing attention to resou rces
exchange of rooks. It is importa nt to remem­ for the opponent.
ber than in the endgame you r rook should I n cidentally, i n positions of this type, when
aim to occupy a fi le which is remote from the you have a slight adva ntage and the
enemy king , so that it does not prevent a n opponent has no cou nterplay, it is very
invasion on t h i s fi le. important to watch for h i s possible activity,
The game could have turned i nto a g raphic and not al low h i m to i n itiative double-edged
demonstration of the principle of two weak­ clashes. Excellent examples of this can be
nesses, but Vasya did not fix in time the found in the games of Anatoly Karpov. There
second weakness on the kingside by g4-g5 is no way that he would have allowed the
(the fi rst weakness is the isolated pawn in black rook to go to c1 .
the centre ) . If he had done this before D. I n the g iven instance the underestima­
playing b4-b5 , he would certainly have tion of the opponent's possibil ities was seen
stretched the opponent's defences more mainly i n the variations and less i n the game
convincingly. At the same time he violated itself, possibly because the opponent played
the principle 'do not h u rry ' , which demands passively. But in other games from the same
that the position should be improved as tou rnament this deficiency g reatly h i ndered
much as possible before turn i ng to active Emel i n . Remember, for example, his game
measu res changing the character of the agai nst Zifro n i , the fi rst pa rt of which we
play. have see n . A clearly better ending with an
The main theme of the fu rther cou rse of the extra pawn was even lost! I n the last rou n d ,
game was the activity of the rooks, which is after excellently outplaying his opponent,
extremely important not only i n rook end­ Vasya blu ndered , missed a win , and as a
ings. Wh ite took his rook to b6, where it was result he fi n ished half a point behind J udit
hardly doing a nyth ing at a l l . In tu rn , Black Polgar.
delayed the activation of his rook.
When I was watch ing the game, I gai ned the
As usually occu rs i n the playing of endings, impression that Wh ite was trying to convert
on several occasions there was the need to his adva ntage in a non-method ical way. But
assess the advisabil ity of various piece the impression could have been fa ulty, and
exchanges. And by no means always did the to check it I found it very i nteresting to look
two players act correctly. at his analysis . Now we have seen that
A strong impression was made by the White did indeed make a number of positional
analysis of the bishop endgame. And it errors. I n my view, the conversion of an
doesn't matter that in it we discovered a adva ntage is one of the weakest aspects of
vul nerable place. I n the solving of compli­ Vasya's play, as usually happens with
cated problems, such errors are practically players who are incli ned to underestimate
i nevitable. their opponent's resou rces . He should do
In general the commenta ry was very in- some serious work on this problem.
212 �

PART IV

Artur Yusupov

From Games by P u pi ls of the School

Aprevious books, one of the most impor­

s we have already mentioned i n our Bai kov - Zviag intsev ( 1 4)
Moscow 1 990
tant resou rces for the improvement of a
player is a serious analysis of his own
games. The examples g iven below, a n no­
tated by our pupils, became a topic for
discussion at joint or individual lessons. The
young players' analyses were critically evalu­
ated , and the results of these d iscussions
form the basis of the present chapter. I hope
that the readers will fi nd much that is usefu l
and i nteresting in these endings played by
young players - they conta i n both success­
ful decisions, and typical m istakes.

Opposite-colour bishops
Two examples from the games of Vad i m
Zviagi ntsev provide a good add ition t o the material advantage may prove i nsufficient
chapter on the theory of endi ngs with for a wi n .
opposite-colour bishops.
Let's try to point out the special features of
(see diagram) the g iven position . Two details help Black to
defend:
49 .i.xf7+ 'iit b4! 1 ) The passed a-pawn may d ivert the white
Black has to defend passively, si nce bishop or restrict its mobil ity - thereby it will
49 . . . 'iit b 2? loses to 50 f4 a4 5 1 e4 a3 52 e5 i nd i rectly defend the kingside pawns which
a2 53 .i.xa2 'iit x a2 54 'iit c4 .i.a3 55 g4 'iit b 2 are situated on l ight squares.
56 f5 gxf5 57 gxf5 �c2 58 f6 .i.b2 59 f7 .i.a3 2 ) The corner sq uare h8 is i naccessible to
60 e6, and Black has no defence agai nst the the opponent's bishop. This factor enables
advance of the king to d7 followed by e6-e7. Black to d raw with a lone king against king,
The resulting situation is rather i nteresting . bishop and h-pawn - a n i m porta nt resou rce
It is difficult to g ive it a clear eval uation in many endings.
immediately. Wh ite hopes soon to obta i n I n principle, it is a l ready possible to guess
two con nected passed pawns. On t h e other the fu rther development of events . Wh ite will
hand , we know about the strong drawing adva nce his f- and e-pawns, if necessary
tendencies of such endings: a m i n i mal supporti ng them with the g-pawn . Black
 From Games by Pupils of the School ttJ 21 3

must try to halt this advance and ideally set breaks through to the g8-sq uare.
up a blockade on the dark squares. How­ Black must play 56 . . . '0tc6 ! , approaching the
ever, his king is badly placed and for the passed pawns with his king. It transpires
moment it is not taking part i n the defence . that even the two con nected passed pawns
50 f4 a4 are insufficient for a w i n .
51 e4 57 f5 'it d 7 58 'Otf4 'it d 8 59 f6 h6! 60 'Ote4
White is intending 52 e5 fol lowed by 'Ote4 , cotes 61 i.. b 3 'itd8 62 'itd5 (or 62 'itd3 i.. g 5
g2-g4 and f4-f5 . 5 1 i.. g 8 would not have 63 'Otc2 a2 ! ! 64 i.. x a2 i..f4 65 e6 i.. e 5)
given anyth ing in view of 51 . . . a 3 . 62 . . . i.. g 5 63 'Ote6 cotes 64 i.. a 2 i.. h 4 65
i.. b 1 ! ? cotta 66 'itd7 i.. g s 67 'Otc6 'itt? 68
51 . . . i.. g 1 1
i.. a 2+ cotes 69 i.. b 3 'itd8 70 cotes i.. h 4
A defensive idea which is a l ready familiar to (70 . . . '0tc7? 71 f7 .i.e?+ 72 'itd5 and 73 'Ote6 )
the reader (from the chapter on opposite­ 7 1 'Otb4 a2 ! ! 72 i.. x a2 i.. g 3 73 e6 i.. e 5 74 f7
colour bishops): 'pawns under attack' . (74 e7+ 'Ote8 ) 74 . . . '0te7 with a n obvious
52 h3 d raw.
52 e5 looks more logica l , but after 52 . . . a3, It is usefu l to note the d iverting pawn
according to a nalysis by Zviagi ntsev, Black sacrifice, which enables Black to create an
can still hold the position . Let us look at h i s impreg nable fortress. This typical idea ,
variations. wh ich occu rred i n o ur exam i nation of
Timman's study, is a good illustration of the
principle nuances in the position are
more important than material.
52 . . . i.. h 2?!
In the game this move fully justified itself,
althoug h Black should have reckoned with
the rep ly 53 f5! . If 53 . . . g5 or 53 . . . a3, then 54
'itd4 is u n pleasant. After 53 . . . gxf5 54 exf5
cotes (54 . . . i.. e 5 55 'Ote4 i..f6 56 'itd5 a3 57
'itd6) 55 'Ote4 'Otd6 there follows 56 f6 . Safer
was 52 . . . a3 53 e5 cotes , transposing i nto
variations considered earlier.
53 'Ote3?! cotes
54 i.. a 2 a3
53 h4 h5!? 54 'Otc2 (54 'Ote4 i..f2 55 f5 gxf5+ 55 g4
56 ..t>xts i..x h4) 54 . . . cotes 55 i.. x g6 i.. h 2 56
Better practical chances were promised by
i.. x h5 i.. xf4 57 e6 'itd6 58 i..f7 i.. g 3 59 h5
55 'Otf3 and 56 g3 .
i..f4 60 'Otb 3 i.. c 1 , or 56 e6 'itd 6 57 f 5 i.. g 3
55 . . . cote&!
58 i.. x h5 i.. x h4 59 'Otb3 cote s and 60 ... i.. e 7
with a d raw; 56 e5
53 h3 cotes 54 'Ote4 i.. f2 ! (if 54 . . . h5, then 55 Neither 56 h4 h6! nor 56 g5!? was sufficient
f5 ! is strong) 55 i.. a 2 i.. h 4 56 g4 (56 cotf3? for a win. The move i n the game allows
'itd4 ! ) . Now 56 . . . i.. e 1 ? is bad : 57 f5 i.. h 4 58 Black to simplify the position i m med iately.
f6 i.. g 5 59 i.. b 3 i.. h 4 60 'Otf4 h6 61 f7 .i.e? 62 56 . . . g51
h4 followed by 63 h5, and the white king 57 fxg5 i.. xe5
214 � From Games by Pupils of the School

58 �e4 �d6 Fi rstly, Wh ite has to h u rry with this attack,

59 �f5 �g 7 since with the pawn on a4 it is pointless to
60 h4 �e7
threaten the h 7-pawn i n view of the reply
. . . a4-a3 . Here , however, there is no contra­
61 h5 � c3
diction with the afore-mentioned principle.
62 g6 ' Do not h u rry' by no means sign ifies marking
62 h6 �d2 ! . ti me. The essence of the principle is that
62 . . . h6 before making decisive changes to the
63 g5 hxg5 position you should try to squeeze the
maxi m um out of the existi ng structure.
64 h6 �f8
The second comment relates to endings
And the players agreed a d raw.
with opposite-colour bishops. We know that
Note should be made of the largely compe­ for the defending side it is normally
tent actions by Zviagi ntsev, who q u ickly advantageous to arrange his pawns on
coord inated his forces and frustrated the squares of the colour of his own bishop.
opponent's plan by a timely attack on his The g iven example is i nteresting for the fact
pawns. But did Wh ite make use of all his that it shows: one should not blindly and
resou rces? Let us return to the position after literally follow rules without taking into
Black's 49th move . account the features of the specific
position. Wh ite's plan is to advance his
kingside pawns . The fact that the g6-pawn
will be undefended will force Black to
advance or exchange it, conced ing the key
f5-sq uare to the opponent.
This was probably that small deta i l wh ich
was lacking in Wh ite's wi n n i ng mechanism.
50 ... h6 51 f4
The hasty 5 1 �f7 g5 52 �e4? was weaker
in view of 52 . . . �c3 ! . But now, if Black
defends as in the game, he will no longer be
able to set up a fortress: 51 . . . a4 52 e4! .ig1
53 e5 a3 54 h 3 �c5 55 g4 �c6 56 f5 gxf5 57
gxf5 �d7 58 'Oti>e4 �c5 59 f6 �e8 60 'it>f5
I n the chapter 'Converti ng an advantage' 'Oti>f8 6 1 �c4 �b4 62 e6 h5 63 ..ti>g5 etc.
Mark Dvoretsky d rew attention to the impor­ H i s position is also not eased by 5 1 . . . .id6
tant pri nciple ' do not hurry!' One of the 52 g3 (with the th reat of e3--e4--e5) 52 . . . g5,
aspects of this principle is attention to after which , if there is noth ing better, there
'trifles' . Don't neglect even the slightest can follow 53 f5 �c5 54 �e4 �c6 55 �b3 !
opportunity to strengthen your position �c5 56 �e5! �xe3 57 �e6 a n d 58 f6 .
or worsen the opponent's. Even so, the defensive resou rces a re not
I n the diagram position Wh ite could have yet exhausted (again remember about the
weakened the opponent's pawn chain by 50 'drawing tendencies' of endings with oppo­
�g8 ! . Here I should l i ke to make two site-colour bishops). Let us try 51 . . . 'iti>b5 ! .
comments: I n t h e event o f 5 2 �f7 g5 53 f5 �c6 the
 From Games by Pupils of the School 215

black king j ust in time joins i n the battle

against the passed pawns. For example: 54
f6 a4 55 'lt>e4 'it>d7 55 g4 .

The primitive 55 g4? 'lt>c6 56 f5 gxf5+ 57

gxf5 Wd7 58 f6 'lt>eB 59 i.c4 a3 60 'lt>f5 (60
'it>d5 'lt>f7 ! ) 60 . . .'1t>f8 61 e6 i.d6 62 'lt>g6 i.b4
63 'it>xh6 �c5 leads to a d raw. The wh ite
The simplest way to d raw is by 55 . . . i.b6! 56 bishop can not take control of the impo rtant
i.c4 'it>e8 ! , when Wh ite gets nowhere with e8-square , and therefore the black king
either 57 'it>d3 i.dB or 57 Wf5 i.xe3 58 ..t>g6 easily prevents any attempt by the oppo­
i.d4 nent's king to help its pawns.

The routine 55 . . . a3?! is less accu rate in view P u rely study-like su btleties a rise i n the
of 56 i.a2! WeB (56 . . . i.b6 57 Wf5 ! i.xe3 58 variation 55 e6? ! 'it>c6 56 'it>e5 i.g3 ! (other­
..t>g6 i.d4 59 f7 ..t>e7 60 ..t>xh6 or 58 . . . '1t>e8 wise 57 e7 'it'd? 58 'it>f6) 57 �f7 �h4 58
59 ..t>xh6 Wf8 60 'it>h5 i.d2 61 h4 gxh4 62 �xg6 (58 g3 a 3 ! ) 58 . . .'it'c7! (but not 58 . . . a3?
59 �b1 'lt>c7 60 f5) . N ow after 59 i.c2 the
Wxh4) 57 'it>d3 ! ! 'it'd? 58 e4 'it>d6 59 'lt>c4,
obvious 59 . . . a3 60 �b3 '1t>d8 loses to 61
and there is no defence against 60 e5+ .
'it>f5! �g3 ( 6 1 . . . 'it>e7 62 'lt>g6) 62 'it>f6. Black
I ncidentally, after 56 i.b3? ( i nstead of 56
should not cl ing on to the pawn - the
i.a2 ! ) a defence would have been fou n d :
blockade is more important: 59 . . . ..t>d8! 60
5 9 . . . a2! 60 i.xa2 i. a 3 .
i.xa4 'it>e7 61 ..t>f5 i.e 1 62 i.b3 i.d2 with a
Nevertheless, as Alexander Motylev indi­ d raw. The most da ngerous is 59 g 3 ! a3! 60
cated , Black also does not lose here, if i n gxh4! (60 i.b1 i.xg3) 60 . . . a2 61 e7 a 1 �+
reply to 56 i.a2 ! he chooses a cou nter­ 62 ..t>f5 �b 1 + 63 ..t>f6 , but I somehow don't
attacking plan: 56 . . . '1t>d6! 57 'it>f5 i.xe3 58 see how Wh ite wins after the approximate
'it>g6 ..t>e5 ! 59 f7 �c5 60 Wh5 i.e? 6 1 �b3 63 . . . �b4 ! 64 f5 "iWxh4+ 65 ..t>f7 "iWc4+ 66
..t>e5 with a d rawn position . 'lt>g7 'it'd4+ 67 f6 �d7 68 Wf8 (68 �f5 �eB
5 2 e4! i.g1 ! 5 3 e 5 ! (53 h 3 ? i.h2) 5 3 . . . i.xh2 69 i.e6 '1t>d6 ; 68 �f7 �g4+ ) 68 . . . 'ifd6 69
54 '1t>e4 a4! 'lt>g8 (th reatening 70 eBltJ+ ! ) 69 . . . 'ii'e 6+ (or
Otherwise 55 �f7 g5 56 f5 wins easily. 69 . . . '1t>b6 ) .
55 i.f7 ! a3 5 6 e6! a2 5 7 e7 a1 'i!V 5 8 e8'ii'+
(see diagram)
(see diagram)
216 � From Games by Pupils of the School

been demonstrated - this dema nded an

a nalysis 35 ( ! ) moves i n length (done
together with Dvoretsky). Such lengthy
variations a re rarely without mistakes, and
therefore it is q uite possible that the readers
will fi nd either a defence for Black, or a
shorter way for White to win .
[As was already mentioned earlier, it is
advantageous for Black to exchange all the
kingside pawns: the position with white
pawns on e6 and f6 against a black pawn on
a3 is drawn. This factor suggests the correct
plan of defence.
This position would appear to a rise by force. 50 it.gB h6 51 f4 a4 52 e4 j_g1 53 e5, and
It is not possible to mate the black king , and now Black should play 53. . . 'it>c5 54 h3 h5!,
therefore Wh ite's aim is to exchange the achieving the exchange of pawns after g2-
queens, then capture the g6-pawn and win g4- Dvoretsky] .
the bishop for the f-pawn . This plan is q u ite
realisable, although not without some d iffi­ Zviagi ntsev ( 1 7) - On ischu k
culty. Berl i n 1 993
58 . . . 'it>c5 59 'ii'c 8+ 'it>b6 60 'i!Vb8+ 'it>c5 61
�c7+ 'it>b5 62 'i!Vb7+!
Noth ing is g iven by 62 it.e8+ 'it>a6! 63 �c6+
'it>a7 64 'it'c5+ 'it>b8 ! .
6 2. . . 'it>c5 6 3 'it'd5+! 'it>b6 6 4 'iWd6+ 'it>b7
(64 . . . 'it>b5 65 it.e8+ is bad for Black) 65
it.d5+ 'it>c8 66 it.e6+ 'it>b7 67 'i!Vd7+ 'it>b6 68
'it'd8+ 'it>c5 69 1\Vc7+ 'it>b5 70 it.d7+ 'it>b4
Now moving to a6 loses (as in the similar
position with the bishop on e8): 70 . . . 'it>a6 71
�c6+ 'it>a7 72 'ii'c 5+ 'it>b7 73 it.c6+ 'it>c7 74
it.b5+ 'it>b7 75 'ifc6+ 'it>b8 76 'ii'd 6+ 'it>b7 77
it.c6+ 'it>b6 78 it.d5+ 'it>b5 79 ifc6+ etc.
71 �b6+ 'it>a3 72 'ir'a5+ 'it>b2 73 'ir'xa 1 +
How should Black defend: 35 . . ..l:Ixa2 or 35 . . .
'it>xa1 74 j_e8 'it>b2
.U.e3 ? I n other words, is h i s position lost after
Or 74 . . . g5 75 f5 it.d6 76 j_h5 'it>b2 77 f6 'it>c3 35 . . . .U.e3 ? In the game Black replied to this
78 'it>d5 j_a3 79 f7 'it>d3 80 g4 followed by q uestion in the affirmative, by choosing 35 . . .
'it>e6-d7-e8. .U.xa2 3 6 .l::t d 3! .l:t a 1 + 3 7 'lt>g2 l::.a 2+ 3 8 'it>h3
75 it.xg6 'it>c3 76 f5 �c4 77 f6 'it>c5 78 f7 i.c4 39 .l:td8+ 'it>g7 40 j_xf4 j_f1 + 41 'lt>g3
it.d6 79 j_h5 it.f8 80 'it>f5 'it>d6 81 'it>g6 'it>e7 .l:tg2+ 42 'it>h4 .l:tf2 43 g 5 ! , and Wh ite retained
82 'it>h7 'it>f6 83 'it>g8 'it>e7 84 g4, and Black both his extra pawns, since if 43 . . . nxf3 he
is i n zugzwang. wins by 44 j_e5+ 'lt>g6 45 'it>g4 . The game
Thus a win for Wh ite would appear to have ended i n a win for Wh ite on the 1 OOth move.
 From Games by Pupils of the School lLJ 217

Let us examine the consequences o f the O f course , i t is adva ntageous for Wh ite to
rook exchange. After 35 .. J:te3 36 l::l.xe3 fxe3 retai n hi s passed a-paw n. A very important
it would be a blunder to play 37 .ltxe3?? principle i n the conversion of a n advantage
.ltxa2 38 �f2 - despite Wh ite's two extra - the principle of two weaknesses - also
pawns , the position is d rawn . applies in endings with opposite-colour
If he defends passively: 38 . . . i.. b 3 39 �g3 bishops (more details about this principle
i.. d 1 ('pawns under attack' ) Black has to are g iven i n the chapter 'Converti ng an
be aware of certa in dangers , as the follow­ advantage ' ) . The passed a-pawn and the
ing variations demonstrate : passed pawn on the kingside stretch the
A) 40 h4 �g7 4 1 h5 �h7 42 .ltd4 .lte2 43 g5 opponent's defences. The fact that Wh ite
i.. d 1 44 i..f6 .lte2 45 'it>f4 i.. d 1 46 �e4 .ltc2+ g ives up some of his extra material does not
4 7 'it'e5 .ltd 1 48 �f4 .lte2 49 h6 �g6 50 'it>e4 play any sign ificant role: nuances in the
i.. d 1 51 f4 .ltc2+ 52 �e5 i.. b 1 . Now noth ing position are more important than mate­
is g iven by 53 f5+ �h7 54 i.e? .ltc2 55 �f6 rial. After 37 a4 i.. xf3 Wh ite has a choice
i.. b 1 56 i.. b 4 .ltc2 57 i.. d 2 i.. b 1 58 �e5 .ltc2 between 38 a5!? and 38 h 3 .
59 g6+ fxg6 60 f6 i.. b 3 61 �d6 �g8 . 53
�d6, threatening to play the king to g 8 , is
slig htly more cu n n i n g . However, Black par­
ries the threat by 53 . . . �h7! 54 <tJe7 <tJg8 55
.ltc3 .ltc2 56 �f6 i.. b 1 57 f5 <tJh7 ! .
B ) 4 0 f4 ! ? � h 7 4 1 f5 .lte2 4 2 �f4 �h6 4 3
�e5+ �g7 (with the g iven structure the king
must not be allowed to go to f6 , since then
Wh ite advances g4-g5-g6) 44 g5 i.. d 1 45
h4 .lte2 46 i.. d 4 i.. d 1 47 .ltc5 .ltg4 (47 . . . .ltc2
is perfectly possible, since if 48 h5 there is
48 . . . i.. d 1 ! 49 h6+ �h7 50 'it>f6 .ltc2 ) 48 .ltb4
i.. d 1 49 �d6 .ltc2 ! 50 i.. c 3+ �g8 5 1 �e5
�g7 52 h 5 i.. d 1 ! 53 h6+ <tJh7 54 �f6 .ltc2
with the same d raw as in the previous
After 38 h 3 e 2 39 i.. d 2 i.. b 7! 40 'it>f2 .lta6
variation .
Black places his bishop ideally and Wh ite
The simplest cou rse is the construction of encou nters serious d ifficu lties.
a fortress - 38 . . .f5 ! . There can fol low:
Thus 4 1 g5? �g7 42 h4 'it>g6 43 �e3 is bad
39 �g3 fxg4 40 fxg4 �h7 41 h4 i.. e 6 42 h5 i n view of 43 .. .f6 ! 44 gxf6 �xf6 45 'lt,Jd4 �e6
.ltd? 43 �h4 .lte6 44 g5 i.. f7 ; 46 �c5 'lt,Jd7 47 �b6 i.. d 3 48 a5 (48 �b7
39 h3 �f7 40 �g3 fxg4 4 1 hxg4 �g6 42 f4 .lte4+ 49 'it>b8 �c6) 48 . . . '1t,lc8 49 <tJa7 �c7
.lte6; 50 h5 .ltc4 51 h6 i.. d 3 52 i.. e 1 'lt,Jc8 53 �g3
39 g5 �g7 40 h4 'it>g6 41 �g3 i.. b 3 42 �f4 .lte4 54 a6 i.. d 3 55 'it>b6 e1 'iii' 56 .ltxe 1 �b8.
i.. d 1 ; 41 �e3 f6! ? 42 �d4 'it>f7 43 �c5 i.. b 7 44 a5
39 gxf5 i.. b 1 40 f6 �f7 4 1 .ltd4 i.. h 7 . is correct, and if 44 . . . .ltg2 45 h4 �f3 46 a6!
Subsequently, depending on the situation , .ltxg4 47 a7 i..f3 , then not 48 �b6? �g6 49
Black manoeuvres with this king between �c7 �h5 50 .lte 1 f5 51 �b8 f4 52 a8'i!V
the squares f7-g8 or with his bishop along .ltxa8 53 �xa8 f3 54 <tJb 7 �g4 55 �c6 �h3
the b 1 -h7 diagonal or the squares h7-g 8 . 56 h5 �g2 57 h6 f2 with a d raw, but 48 �d4!
218 � From Games by Pupils of the School

�g6 (48 . . . 'lt>e6 49 �e3 �d5 50 h5 ) 49 'it>e3 analyse recent games, and devise new
�b7 50 �xe2 'it>h5 51 �e1 with an easy win ideas in order to su rprise their opponent.
- the king again heads towards the a7- In modern chess, ope n i ng d isputes some­
pawn . times conclude only after two or even th ree
44 . . . 'it>e6! is a more tenacious defence: 45 dozen moves deep i n the endgame. That
�e 1 ! (45 'it>b6? i.g2 is similar to a variation was also the case in the game g iven below.
given above) 45 . . . i.g2 46 h4 i.f3 47 a6
�xg4 48 'it>d4 �f3 49 'it>e3 i.c6 50 a7 f5 5 1 Kirjakov (1 5) - Svidler ( 1 4)
'it>xe2 f4 5 2 �d3, and Wh ite stil l has to Daugavpils 1 990
overcome some tech n ical d ifficu lties .
1 d4 lbf6
38 a5! is stronger: 38 . . . i.xg4 39 a6 i.f3
2 c4 g6
After 39 . . . e2 40 �f2 i.f3 4 1 i.d2 �c6 42 h4 3 tt:Jc3 i.g7
Black has no defence agai nst the oppo­
4 lDf3 d5
nent's pla n : advance one pawn to h6, the
5 cxd5 tt:Jxd5
other to a7, captu re the e2-pawn and take
the king over to the queenside. It is 6 e4 tt:Jxc3
importa nt that Wh ite's bishop defends its 7 bxc3 0-0
own pawn and prevents the advance of the 8 l:!.b1 c5
opponent's passed pawn along the same 9 �e2 lbc6
diagonal. 1 0 d5 lbe5
40 a7 (40 i.xe3 is also good ) 40 . . . 'it>f8 1 1 tt:Jxe5 i.xe5
Black's only hope is to take his king across 1 2 �d2 b6
to the a7-pawn ; i n th is case it will be 1 3 f4 �g7
sufficient for h im to g ive up his bishop for the
1 4 c4 e5
h-pawn .
1 5 i.b2
41 h4 e2 (4 1 . . . 'it>e8 42 h5 �d7 43 h6) 42
The alternative is 1 5 0-0 .
'it>f2 , and Black is unable to defend against
15 . . . exf4
the afore-mentioned plan by Wh ite .
Another possibil ity, 1 5 . . . �d6, occu rred in the
Thus we have come to the concl usion that
game Komarov-Smejka l , Bad Mergentheim
the exchange of rooks would have lost,
1 989, published i n lnformator Volume 48.
althoug h it would have demanded a certai n
accu racy on t h e opponent's part . 1 6 �xf4
1 6 i.xg7? is bad , since Black i nterposes
1 6 . . . �h4+, and if 1 7 g3 fxg3 1 8 'ii h 6, then
From th e open ing i nto t h e endgame
1 8 . . . g2+ ! .
The following training game was played at
16 . . . 'ille 7
the second session of the school, devoted to
opening preparation . We suggested that the 1 7 0-0 .i.d7
young players should ' reveal their ca rds' - 1 8 .i.d3 .l:!.ae8
inform their opponents beforehand what The latest word in th is variation is the
opening variation they were i ntending to immediate exchange of bishops. In the
choose. Then they had to familiarise them­ game Sakaev-Ftacn ik (Dortmund 1 992)
selves with the theoretical recommenda­ after 1 8 . . . .i.xb2 1 9 �xb2 f6 20 .i.c2!?
tions on the plan ned opening variation, �ae8 ! ? (with the idea of . . . f6-f5 ) Black
 From Games by Pupils of the School tLJ 219

achieved equal ity. I n stead of 2 0 . . . ltae8 , Igor A) 26 . . . f5 2 7 d6 'ii? x h7 28 d 7 ;

Stohl's recommendation i n lnformator Vol­ B ) 26 . . . .l:!.d8 27 'liff2 .l:!.e5 28 a 4 a6 29 l:!. b 1
ume 48 - 20 . . . 'ii'e 5 21 �xe5 fxe5 is less (Sto h l ) ;
accu rate, si nce after Kirjakov's suggestion
C ) 26 . . J�e5! ? (with t h e i d e a o f . . . b6-b5 or
22 �fb 1 ! Wh ite retains the better prospects
. . . f7-f5) 27 g4 b5 28 Si.f5 bxc4 29 'it>f2 ! ? .
in view of the threat of a2-a4-a5 .
With t h e move i n t h e g a m e Black does not
1 9 ..if6 ! ?
solve the problems facing h i m .
T h i s is more accu rate t h a n 1 9 Si.xg7 'li?xg7
22 . . . lle5
20 a4 f5 with equal ity, as in the game
Vaisser-Stohl (Biel 1 989). 23 l:16f3
19 . . . Si.xf6 The . . . il.f5 th reat acts on Kirjakov's nerves
and he incorrectly withd raws his rook from
20 'ii'xf6 'it'xf6
its active position . 23 l'::t 1 f4! was more
21 l:!.xf6 'it> g7
accu rate , when 23 . . . Si.f5 would be simply
22 l:!.bf1 ! ? answered by 24 l:!.c6 .
Less good is 22 .Ud6 Si.a4 23 l:!.f1 f6 - Sto h l . 23 . . . il. g4
Wh ite wants to beg i n play on the queenside
by 24 Si.c2 , and then 25 a4 or 25 .U.a3. The
main problem of Black's position is that his
nominally good bishop is not taking an
active part i n the game. The only target for
cou nterplay is the e4-paw n . But 23 . . . f5?
does not work in view of 24 g4 . Therefore he
should have thought about switching his
bishop to g6 with 23 .. .f6 followed by . . . g6-
g5 and . . . il.e8-g6.
24 ll3f2 ..id7?!
Here too it was not too late for 24 . . . g5! with
the idea of . . . Si.h5-g6. Black's waiting
tactics might not have proved justified .
I n t h e resulting e n d i n g Wh ite h o l d s the 25 a4
i nitiative. Of cou rse, i n itself the pressu re on
25 il.c2 ! ? .
the f-file is not too u n pleasant for Black. The
real problem is that the opponent has a 25 . . . l:tee8
simple plan of improving his position on the 26 il.c2
queenside. By playing Si.c2 and a2-a4-a5 If 26 a5 there is 26 . . . bxa5 27 lla 1 (27 l:tb1
Wh ite wants to create a second weakness l:tb8) 27 . . . a4 28 il.c2 lle5 . But 26 h4 came
in Black's camp . Therefore serious consid­ i nto consideratio n . Lul led by his opponent's
eration should have been given to changing und isti ngu ished actions, Kirjakov wants to
the character of the play, as suggested by play 'with every comfort ' , and he u nderesti­
Sto h l : 22 . . . ..if5 ! ? 23 exf5 'it>xf6 24 fxg6+ 'it>g7 mates a freeing breakthroug h .
(24 . . .'it>e5?? 25 g7) 25 gxh7 .Ue3. If now 26
..if5 , then 26 . . . lld8 27 l:tf4 .l:!.ee8. After 26 (see diagram)
.Ud 1 Black has a choice :
220 � From Games by Pupils of the School

26 . . . f5! 39 . . . b5 40 g4 a6! ! (or 39 . . . a6 40 g4 b5! , but

A subtle solution to Black's defensive prob­ not 40 . . . <t>a3?? in view of 4 1 c5! ! and wins)
lems, the evaluation of which depends on a 41 h4 (4 1 cxb5 <t>xb5) 41 . . . <t>a3 , and in the
pawn ending that arises by force. queen ending arising after 42 c5 b4+ 43
27 exf5 .i.xf5 <t>d2 b3 44 c6 b2 45 c7 b 1 'it' 46 c8'it' Black
should be able to avoid defeat.
28 i.xf5 l:txf5
35 . . . <t>xd6
29 .l::txf5 gxf5
36 <t>d3 'it>e5!
30 l:.xf5 .l:Ie4!
Now 36 .. .<it>c5? loses to 37 g4.
31 d6! l:td4
37 g3 h6!
32 .tld5 <t>f6!
38 h3 h5!
32 .. J�xd5? 33 cxd5 <t>f7 would have lost i n
view o f 3 4 g 4 <t>e8 35 g5 <t>d7 3 6 h4 <t>xd6 39 g4 hxg4
37 h5 <t>xd5 38 g6 hxg6 39 h6. 40 hxg4 <t>f4
33 nxd4 cxd4 41 <t>xd4 'it>xg4
34 <t>f2 <t>e6 42 <t>d5
35 <t>e2 42 a 5 <t>f4! 43 a6 (43 axb6 axb6 44 'it>d5
Wh ite does not exploit all his chances. The <t>e3 ) 43 . . . <t>f5 44 <t>d5 <t>f6 45 '>t>c6 '>t>e6 46
opponent's task would have been more <t>b7 <t>d7 would also have led to a d raw.
difficult in the event of 35 <t>f3 <t>xd6 36 <t>e4 42 . . . <t>f4!
<t>c5 37 <t>d3 . After 37 . . . <t>b4 38 <t>xd4 <t>xa4 But not 42 . . . <t>f5?? in view of 43 a5! bxa5 44
39 <t>c3 ! it appears that things a re bad for c5 and wins.
Black. However, he is saved by an unusual 43 <t>c6 <t>e5
defence.
44 <t>b7 <t>d6
45 <t>xa7 <t>c7
(see diagram) And the players ag reed a d raw.
 From Games by P u pils of the School ttJ 221

Exchanging 28 . . . 'it> d 7 29 .l:lxc8 'it>xc8 30 'i!tg2 � d 7 3 1

Should I exchange q u eens a nd go i nto a n 'it>h3 'it> e 7 32 'it> h 4 'it>f7 33 'it>h5 'it> g 7 3 4 f3 h 6 !
e n d i n g ? H ow should t h e conseq uences of is d raw n : 35 g4 'it>h7 36 g 5 hxg5 37 fxg5
the rook exchange be eval uated? Similar �g7 . 34 . . . 'i.t>h8? ( i n stead of 34 . . . h6 ! ) 35 'it>h6
questions often have to be answered d u ring 'it>g8 36 g4 'it>h8 is i n correct i n view of 37 h 3!
a game. It is not surprising that i n the ( b ut no t 37 h4? Wg8 38 g 5 e5! ) 37 . . . 'it> g8 3 8
endgame, when there are a l ready few h4 (zugzwan g ) 38 . . . 'i.t> h 8 39 g5 fxg5 40 hxg5
pieces left, it is especially i mporta nt to solve and wins.
correctly the problem of what to exchange. 27 fxe5

M ugerman - Makariev ( 1 4)
Moscow 1 989

Despite the material equal ity, Black's posi­

tion is critica l . H ow can he defend against
the march of the wh ite king towards the
weakened pawns on the kingside? In the
Black should take accou nt of the fact that his
game Black was unable to solve this
kingside pawn structu re is spoiled , a nd i n
problem and after 27 ... .Uc1 +? 28 'it>g2 'it>c7
t h e endgame t h e pawns can b e attacked by
29 f4 1la 1 ?! 30 a3 .l:tc1 31 Wh3 l::.c 5 32 'it>h4
the enemy king . Therefore he should not
l1d5 33 l1c2+ 'it>d8 34 Wg5 he fi n ished up i n
have excha nged queens. After the correct
a hopeless ending.
25 . . . 'it>a8 26 l:txd2 'i!Vc6 (intending 27 . . . 'it'c1 + ,
27 . . . 'iWf3 or 27 . . . a6) Black h a s cou nterplay. The evaluation of the position largely de­
pends on whether Black ca n take his king to
25 . . . 'iic 7? the kingside. To do th is he is forced to allow
26 llxd2 'i¥xe5?! the exchange of rooks .
Aga in a poor decision . The exchange on e5 27 . . . �c7
improves the opponent's pawn structu re - it 28 llc2+
u ndoubles his pawns, and moreover the
28 'it>g2? llg8 with an acceptable position for
wh ite e5-pawn will hold back two black
Black.
pawns. 26 .. .f6 ! was stronger, when Wh ite
28 . . . 'it>d7
should reply 27 'iie 3, reta i n ing somewhat
the better chances. The pawn ending after 29 11xc8 Wxc8
27 'ii'x c7+ 'it>xc7 28 llc2+ (28 'it>g2 ! ? ) An i nteresting pawn ending has been
222 � From Games by Pupils of the School

reached . Wh ite takes his king towards the

opponent's pawn weaknesses , and the
black king h u rries to their defence.
30 �g2 �d7
31 �h3 �e7
Active cou nterplay is too late : 31 . . . 'it>c6 32
'it>h4 �d5 33 f4 '.te4 34 �g5 �f3 35 'it>f6
'it>g2 36 �xf7 'it>xh2 37 �xe6 �xg3 38 'it>xf5
h5 39 e6.
32 'it>h4
Now Black has a choice between 32 . . . �f8
and 32 . . . f6 .
32 . . . 'it>f8 33 'it>h5 �g? 34 'it>g5 (but not 34 Now noth ing is g iven by either 35 Wg5 h6+
f3? in view of 34 . . . f6! ) 34 . . . h6+ 35 'it>h5 '.th? 36 'it>h5 (36 'it>f4 Wf6) 36 . . . e5, or 35 f3 Wf6!
36 f3 (as will become clear from later 36 g4 fxg4 37 fxg4 We5.
variations, it is more methodical to include 35 h3
the moves 36 b4 b5) 36 .. .f6 (in the event of
The most logical cou rse - Wh ite strength­
36 . . .'it>g7 37 g4 fxg4 38 fxg4 'it>h? 39 g5 hxg5
ens his position on the kingside.
40 'it'xg5 the presence of the outside
passed pawn decides the outcome) 37 35 . . . h6
exf6 e5 38 g4 e4 (38 . . . f4 39 g5 e4 does not As will become clear from the variations
save Black in view of 40 g6+ �g8 41 'it>xh6 g iven below, any advance of the black
exf3 42 g? f2 43 'it>g6 and 44 f7 - mate ! ) 39 pawns on the q ueenside merely ma kes
fxe4 fxe4 40 �h4 'it>g6 41 Wg3 'it>xf6 42 �f4 things easier for the opponent. For example:
e3 43 '.txe3 'it>g5 44 �f3 �h4 45 'it>f4 'it>h3 35 . . . a5 36 a4 h6 37 g4 fxg4 38 hxg4 '.th? 39
46 g5 hxg5+ 47 �xg5 'it>xh2 48 �4 'it>g2 49 f4 '.tg? 40 g5 hxg5 41 fxg5 e5 42 �h4 rttf7
'it>e5 'it>f3 50 �d6 'it>e4 5 1 We? b5 52 'it>b? ! 43 '.tg3 'it>g6 44 'it>g4 e4 45 'it>f4 e3 46 <t>xe3
(not 52 c.1tc6 b4 53 �b5 b3 54 axb3 Wd3 55 Wxg5 4 7 '1td4 'it>f5 48 'it>c5 '>i?e4 49 Wb6 'iitd4
'it>a6 Wc2 56 b4 'it>b3 or 54 a4 Wd3 55 'it>b4 50 '>i?xb? <t>c5 5 1 'it>a6 Wb4 52 b3.
Wc2 56 Wa3 a5 - stalemate ! ) 52 . . . �d3 36 g4
(52 . . . a5 53 Wb6) 53 'it>xa? b4 (53 . . . 'it>c2 54
But not 36 f3?? e5 37 g4 f4 .
b4 'it>c3 55 a3) 54 'it>b6 'it>c2 55 b3, and
White wins. 36 . . . fxg4
37 hxg4 'it>h7
32 . . . f6! ?
Here things are more d ifficult for White .
3 3 exf6+ 'it>xf6
34 'it>h5 'it>g7 (see diagram)
34 . . . 'it>e5 35 'it>h6 �e4 36 �xh? Wf3 37 'it>g6
'it>xf2 38 'it>f6 '>i?g2 39 'it>xe6 Wxh2 40 Wxf5
Wxg3 41 'it>e5 transposes into a variation
just examined , but 36 f4 ! 'it>f3 37 'it>xh? Wg2
38 h4 wins more qu ickly.
 From Games by P u pils of the School ltJ 223

From the point of view of the first plan (the

creation of an outside passed pawn ) the
situation has not changed . But for the
second plan the strengthening of the posi­
tion proves sign ificant.
44 g5! hxg5
45 'lt>xg5 Wf7
46 'it>f4
47 �e4
47 . . . e5 also does not help: 48 Wd5 'lt>f5 49
Wd6 �f4 50 �c7 'it>f3 51 'it>xb7 Wxf2 52 'lt>c6
e4 53 b7.
48 �e5 �e7
The d i rect 38 f4? does not succeed . After
49 f3 !
38 . . . 'it>g7 39 g5 hxg5 40 fxg5 e5 4 1 �h4 �f7
It is useful to note that Wh ite wins only
42 �g3 'it>g6 43 �g4 e4 44 'it>f4 e3 45 �xe3
tha n ks to the existen ce of two reserve
�xg5 46 'it>e4 'it>f6 47 �d5 �e7 the wh ite
tempi.
king ca nnot break through to the queenside
pawns. Now it is clear why the a7- and b7- 49 . . . �d7
pawns should remain i n place. 50 'it>f6 'it>d6
A draw also results from 38 g5? hxg5 39 Or 50 . . . 'lt>c6 51 Wxe6 'lt>xb6 52 f4 We? 53 f5
'lt>xg5 �g7 40 Wf4 'lt>f6 4 1 We4 e5 42 �d5 'it>d8 54 �f7 b5 55 f6 b4 56 Wg8 b3 57 f7 ,
Wf5 43 b4 (if 43 f3 Black has either 43 . . . Wf6 , and the white pawn queens with check.
or 43 . . . 'it>f4 44 We6 �xf3 45 'it>xe5 �e3 46 51 f4 '>t>d7
'lt>d6 Wd3 ) 43 . . . b5 44 �c5 a6 45 Wd5 e4 46 52 �f7 �d6
a3 �f4 4 7 'it>e6 �g4 ! 48 �e5 �f3 . 53 �e8
Let us remember about the principle 'do not After gaining the 'horizontal opposition' ,
hurry! and try fu rther improving the position the wh ite king performs an ' outflanking'
by the advance of the queen side pawns. manoeuvre .
38 b4! 53 . . . Wc6
This move could also have been i ncluded 54 We7 'it>xb6
earlier. Or 54 . . . 'lt>d5 55 'it>d7.
38 . . . �g7 55 �xe6
Black is forced to stick to waiting tactics - 56 f5
otherwise Wh ite wins by creati ng an outside 57 �f7
passed pawn on the kingside (f2-f4 and g4- And Wh ite wins.
g5).
[ Ten years later grandmaster Viorel Bologan
39 b5 �h7
suggested a different plan in the pawn
40 a4 �g7 endgame, which wins more quickly:
41 a5 �h7 38 a3!? �g7 39 a4 �h7 40 Wh4!? �g6 41
42 b6 axb6 �g3 (intending 42 �f4) 41 . . . e5 (41 . . . h5 42
43 axb6 �g7 gxh5+ Wxh5 43 �f4 �g6 44 �e5 'it>f7 45
224 � From Games by Pupils of the School

�d6; 41 . . . �g5 42 f4+ �f6 43 <t.tf3, and if B ) 28 �g2 fxg3 29 hxg3 (29 �xg3 l:tg8+ and
43 . . . e5, then 44 f5) 42 �h4! a5 (42 . . . e4 43 30 . . . �c7) 29 . . . .Uc5.
b4 or 43 �g3 �g5 44 b4 with zugzwang -
this is why White needed the pawn on a4) 43
f3 (zugzwang) 43 . . . �f6 44 �h5 �g7 45 g5.
Here the plan involving the 'expansion of
the bridgehead ' (an exchange of pawns
with the aim of breaking through with the
king to the opposite wing) is especially
effective, since the e5-pawn is immediately
lost- Dvoretsky.]

Thus 27 . . . �c7 would have led to an objec­

tively lost pawn end ing. At the board it was
hardly possible to calculate all the variations
exactly, and the probabil ity of a mistake by
White was q uite h ig h . In any case 27 . . . �c7 30 f4 <t.tc7 3 1 <t.t h 3 (or 3 1 �3 h 5 ! ? 32 .l:!.h2
would have g iven more chances of saving �d7 ) 31 . . . h6!? 32 �g4 .l:tc1 , and the wh ite
the game than 27 . . . .l:.c1 +?, the move cho­ king cannot break through to the black
sen . pawns;
However, Black has one more possible 30 .Ud7 l:txe5 3 1 .ll xf7 h 5, i ntending 32 . . . l:te2
defence. By playing 27 f4! ? he would h ave
. . .
or 32 . . . a5;
changed the character of the play, a n d , as
30 l:td8+ �c7 3 1 �f8 .Uxe5 ( 3 1 . . .l:tc2 is
shown by the analysis g iven below, he could
weaker: 32 .l:txf7+ �c6 33 .l:!.e7! .l:!.xb2 34
have successfully defended :
.l:!.xe6+ <t.td5 35 .l:!.e7) 32 .l:!.xf7+ <t.tc6 33 .l:txh7
.l:te2 or 33 �f3 h 5 .
[ White can also t ry 28 .l:!.d l!?. Then the
obvious 28 . . . f3? leads to a hopeless position
after 29 h4 l:tc1 + 30 �h2 .Uc2 31 g4 '1J.xf2+
32 �g3 !lxb2 33 l:t.xfl - the threatened
advance of the kingside pawns is just too
strong. The pawn ending after 28 . . . fxg3 29
hxg3 .l:r.c7? 30 1id8+ l:.c8 31 .U.xc8+ �xc8 32
�g2 �d7 33 �f3 is also lost: 33. . . �c6 34
�g4 �d5 35 f4 <t.te4 36 �g5 <t.tf3 37 �f6
�xg3 38 <t.txfl <t.txf4 39 �xe6 h5 40 �f6 h4
41 e6 h3 42 e7 h2 43 eE!fl h1'ili 44 'iVe5+
�g4 45 'ilig5+ followed by the exchange of
queens, or 33. . . �e7 34 �g4 �f8 35 �h5
A) 28 gxf4 l::t c4 29 l:td8+ �c7 30 l:th8 l:txf4 �g7 36 g4 h6 37 f3! �h7 38 f4 <i;g7 39 f5,
3 1 Z:.xh7 �c6 32 �g2 �d5 33 �g3 �xe5 34 and Black is in zugzwang. His only hope is
.l:!.h5+ f5! 35 l:th4 l:txh4 36 �xh4 �f4 37 �h5 for success in the complications arising after
�f3 38 �g5 f4 39 h4 e5 with a d raw; 28 . . . fxg3 29 hxg3 ltc2- Dvoretsky.]
 tb 225

In d ex of P l aye rs a n d A n a lysts

Alekhine 1 5 , 1 6 , 74, 1 1 5, 1 20, 1 30 , 1 35 , Dolmatov 67, 1 33

1 59, 1 63 Donchenko 88
Anand 1 80, 1 85 Dorfman 60, 1 83
Andersson 23, 24, 1 20 Dreev 1 8
Andrianov 1 36 Dvoretsky 30, 3 1 , 47, 1 00 , 1 2 1 , 1 22, 1 26,
Atalik 1 94 1 28 , 1 52 , 1 73, 1 9 1
Averbakh 35, 70
Ehlvest 1 36
Bag irov 63 Elkies 1 1 9
Bahr 82 Emel i n 1 99
Baikov 1 28, 2 1 2
Bakulin 47 Flohr 1 34 , 1 37
Bannik 1 43 Franco 23
Barbero 50 Frid man 39
Barcza 1 49 Fridstein 1 4
Barlov 60 Ftacn i k 1 80 , 1 83 , 1 85, 2 1 8
Bastrikov 1 48 Fuchs 73
Belavenets 25, 29, 1 23
Berger 69 Gall iamova 25
Biryukov 1 65 Gdanski 1 65
Bogoljubow 1 6, 74 Geller 1 96
Boleslavsky 1 38 Gheorg h i u 1 02
Bologan 6 1 , 1 5 1 , 223 Gl igoric 9
Bondarevsky 75, 1 34 Gragger 1 49
Botvinnik 85, 1 26, 1 5 1 , 1 69 Grigoriev 1 08
Bron 35 Gulko 1 9
Bronstein 1 84 Gurevich , D . 45
Browne 1 1 8 , 1 90 G urgen idze 1 09
Burgess 9 1
Butnoris 1 2 1 Henneberger 66
Hort 1 68
Capablanca 2 1 , 28, 30, 1 20 , 1 59 Hortov 36
Chekhover 79 Horwitz 1 4
Chern i n 1 99
Cooper 1 26 l na rkiev 1 95
lvanchu k 1 65
Dautov 1 75 Kaidanov 1 42
Delchev 1 65 Kamsky 1 80
226 w Index of Players and Analysts

Karolyi 1 88 Maizelis 1 0
Karpov 42, 62, 1 1 2 , 1 29, 1 37 , 2 1 1 Makariev 22 1
Kasparian 93 Makarychev 62, 1 1 6
Kasparov 1 60, 1 77 Makogonov 1 34
Kavalek 62 Mandler 1 1 0
Kempinski 1 65 Mar6czy 1 5
Kengis 1 2 M a rtinovic 60
Keres 1 37, 1 96 Mata novic 85
Kharlov 75 Mecking 202
Khau nin 39 Menchik 2 1
Khenkin 75 Mestel 24, 55, 1 1 3
Kholmov 60, 73, 1 93 Mi les 45, 1 48
Kidambi 1 87 Moiseev 63
Kikiani 30 Montaigne 1 02
Ki�akov 2 1 8 Moravec 35
Moskalenko 1 8
Kiselyov 1 48
Motylev 2 1 5
Kling 1 4
Mugerman 221
Koberl 1 48
Mu l l e r 48, 74 , 99, 1 67
Kolesnikov 35
Kolterman 69
Najer 1 65
Komarov 2 1 8
N i kolac 1 48
Korchnoi 1 2 , 1 29, 1 34
N imzowitsch 80, 1 36
Kotov 1 3 1 , 1 34
Norl i n 79
Kovacevic 1 37
Notkin 25
Krantz 1 92
Nunn 91
Krogius 73
Kunitz 1 9 1
On ischu k 2 1 6
Kuzmin 55 Ostvath 1 89

Lamprecht 74 Pachm an , L. 1 3 1
Lapin 1 6 Pachm an , V. 36
Larsen 62, 1 20, 1 7 1 , 201 Peckover 36
Lasker, Ed. 74 Peterson 84
Lasker, Em. 62 Petrosian 6 1 , 62 , 1 20 , 1 28, 1 39, 1 43 , 1 93
Leonhardt 1 23 Pigusov 1 94
Li Ruofan 1 66 Polgar, J . 1 85
Li Zunian 1 03 Polugayevsky 202
Ljubojevic 49 Popov, L . 1 1 3
Loginov 24 Portisch 6 1
Lutikov 1 4 Potkin 1 7 1
Lutz 1 75 Privorotsky 3 1 , 84
 Index of Players and Analysts ttJ 227

Rachels 45 Thomas 1 35
Radulov 200 Timman 53, 67, 1 77, 1 90
Ragozin 28 Timoshchenko 76
Reti 1 3, 1 1 0. 1 24 Torre 1 20 , 1 48
Rinck 60 Trabatton i 60
Romanovsky 40, 60, 1 24 Tseitl i n , Mark 1 1 6
Rowson 1 65 , 1 66 Tseshkovsky 52
Rubinstei n 1 20 Tu kmakov 44
Tylor 1 1 5
Sadler 1 48
Sakaev 2 1 8 Udovcic 1 37
Salov 1 53
Utyatsky 1 6
Samisch 1 30
Schlosser 1 43
Vaganian 200
Schwarzman 1 99
Vaisser 60, 2 1 9
Seirawan 1 85
Vakh idov 76
Shipov 74
Vancura 40
Shirazi 1 9 1
Vasyukov 62, 1 9 1
Sideif-Zade 24
Veresov 1 5 1
Skembris 1 48
Vidmar 1 35
Smejkal 2 1 8
Vogt 1 49
Smirin 1 49
Vol kov 1 73
Smyslov 9, 97, 1 5 1 , 200
Vulfson 88
Sokov 36
Sozin 1 4
Spassky 1 34 , 1 39 Wolff 1 1 8
Speelman 1 32 , 1 33
Spielmann 1 23 Xie J u n 1 7 1
Stean 1 68
Stein 1 4 1 Yandemi rov 1 65
Stein itz 1 39 Yates 30
Stohl 2 1 9 Yudovich 1 92
Suetin 1 59 Yusupov 42 , 49, 50, 52, 53, 55, 1 02 , 1 03 ,
Sveshn i kov 1 73 1 1 2 , 1 53 , 1 87
Svidler 1 73, 2 1 8
Szabo 44 , 1 48 Zaitsev 6 1 , 1 20, 1 49, 1 5 1 , 1 8 1 , 1 90
Szily 1 89 Zakharov 1 22
Zapata 1 99
Taimanov 1 9 , 1 4 1 Zifron i 1 99
Tal 97, 1 29 Zilberstei n 1 52
Tarrasch 1 5, 40, 79, 80, 1 59 Zlotnik 88
Tartakower 1 5 Zviagintsev 67, 1 69 , 1 73 , 1 76, 2 1 2 , 2 1 6