Projecting X 2.0: How to Forecast Baseball Player Performance

Projecting X 2.0: How to Forecast Baseball

Player Performance

By Mike Podhorzer

Copyright © 2016 by Mike Podhorzer

All rights reserved. No part of this publication may be reproduced, distributed, or transmitted in any form or by any means without the express written permission of the publisher except in the case of brief quotations embodied in critical articles and reviews.

Printed in the United States of America

Second Edition, Smashwords Edition: January 2016

www.ProjectingX.com

Table of Contents

Introduction 

Projection Systems

Projection Methodology 

Spring TrainingNew Technology and Statcast 

Chapter 1: Projecting Hitter Performance 

Plate Appearances (PA)Doubles and Triples (2B and 3B) 

Strikeout Percentage (K%)

Ground Ball, Line Drive, Fly Ball Rates (GB%/LD%/FB%) 

Batting Average on Balls in Play (BABIP) 

Home Run per Fly Ball Rate (HR/FB) 

Runs Batted In and Runs Scored (RBI and R) 

Stolen Bases (SB) 

Chapter 2: Projecting Pitcher Performance 

Innings Pitched (IP) 

Strikeout Percentage (K%)

Walk Percentage (BB%) 

Ground Ball, Line Drive, Fly Ball Rates (GB%/LD%/FB%) 

Batting Average on Balls in Play (BABIP) 

Home Run per Fly Ball Rate (HR/FB) 

Saves (SV) 

Chapter 3: Creating an Excel Spreadsheet to Project Hitter Performance 

Chapter 4: Developing a Hitter Projection 

Chapter 5: Creating an Excel Spreadsheet to Project Pitcher Performance 

Chapter 6: Developing a Pitcher Projection 

Chapter 7: Putting a Bow on It

Chapter 8: Frequently Asked Questions 

About the AuthorSupport Me!

Introduction

Welcome to Projecting X 2.0, the follow-up to the original Projecting X. Think of this edition as the pizza with even more of your favorite toppings, the ice cream sundae with an extra drizzle of delectable syrup, or the baseball game that treats you to a rare fireworks display after the final out is made (because, ya know, we all love fireworks, right?).

Though this version is not completely revised, it is most certainly updated with statistics through the 2015 season. Furthermore, and this is where your toppings, syrup, and fireworks make their highly anticipated appearances, 2.0 is chock full of new research, new metrics, new ideas, and new methods for projecting baseball player performance. And of course, the new metrics and methods are all explained in the included step-by-step guide to creating your own projections in Microsoft Excel.

Throughout these digital pages, you will find many links to websites, articles, and research studies. I encourage you to visit the content I have linked to in order to expand your knowledge base even further. The following sites play a vital role in my projection process and research:

FanGraphs – my primary source for statistics

Baseball-Reference – for several advanced pitching metrics

Pro Sports Transactions – injury and disabled list data

Baseball Heat Maps – hitter batted ball data

Brooks Baseball – PITCHf/x player cards

Roster Resource – depth charts and projected lineups/batting orders

Projection Systems

ZiPS. Steamer. Marcel. PECOTA. These are not the names of my pet hamsters or the latest fad diet trends. They are the names of some of the most well-known baseball player projection systems around. All of them derive their forecasts by mixing historical data in a blender and cooking up a projected performance line. These systems are pretty darn good, but for the majority of them, precious little is shared about their inner workings, making them essentially black boxes.

Accordingly, at some point, you might get the urge to project players yourself. And because you are reading this book, that urge may have already been felt. Why might you decide to project players on your own rather than rely on one of the aforementioned systems? Well for one, you may not agree with every projection a system generates. If you were to project a player yourself, you would hold full control over every individual statistical measure in the forecast (strikeout and walk percentages, BABIP, etc.) and be granted the opportunity to instantly discover how a change in one metric affects another.

Furthermore, the forecasting process requires a combination of both art and science. Since the projection systems are built using sophisticated computer models, they have the science part down quite well. However, that also means that these models involve no manual human intervention and are completely objective. But, the art, or subjective, component falls a bit short.

Computers have come a long way since their invention. But sadly, they are still unable to watch the games or read the news and promptly parse the gobs of data out there to incorporate any new actionable pieces of information into a player’s projection. As a result, these systems are prone to miss various dynamics that could affect a player’s projection.

There are several categories of examples in which a human has the advantage:

1) Lineup Spot Changes – A hitter changing spots in the lineup could greatly affect his counting stats, but a computer is unaware of where a hitter is slotted to bat. If it’s the same spot as the previous season, fantastic, but if not, its various counting stat forecasts will require tweaking.

2) Changes to Team’s Lineup – We all know that a hitter is going to score more runs and drive in more runners in a lineup surrounded by above average players rather than replacement level players. So if a team signs the top offensive free agent on the