Symposium on Statistics and Operations Research in Baseball

California State Univ., East Bay
Valley Business and Technology Center Presentation Room
Hayward, CA
Wednesday, July 11, 2007
Need more information?
Contact Mitch Watnik
ASA/SIS and CSUEB Dept. of Statistics and Biostatistics

Lineup of events

Note that times and order are subject to change.
10:00 - 10:05 AM Introductory Remarks Mitch Watnik, Symposium Organizer
10:05 - 11:45 AM Panel Discussion Moderator: Mike Siano, Senior Producer for the Baseball Channel and Host of Fantasy 411, MLB.com
Panelists: Chris Long, Jeffrey Ma, Sig Mejdal, Cory Schwartz, Nate Silver, Joel Sokol, Hal Stern, Mitch Watnik
11:45 AM - 1:15 PM Lunch Break
1:15 - 1:35 PM From the Box Score to PECOTA: A Brief History of Statistics in Baseball Henry Reichman, Professor, Department of History, CSU East Bay
From Henry Chadwick, the so-called "father of baseball" in the mid-1800s to Alan Roth, Branch Rickey's "numbers man," to Bill James, SABR, and Project Scoresheet, followers of baseball have sought to define players' contributions in quantitative terms. Focusing on a few key individuals, this talk will try to demonstrate how, while the methodologies may have changed and grown more sophisticated, the kinds of questions baseball statisticians ask have not really changed all that much over the years.
1:35 - 1:55 PM The Baseball Practice of Statistics Cory Schwartz, Director of Statistics, MLB Advanced Media
My responsibility is towards the data capture, validation and distribution of the stats for all MLB, minor league and winter league baseball games, plus other MLB special events such as the World Baseball Classic. The goal is to provide the most detailed, timely and accurate data possible to the fans, clubs and media who follow our game and its players. As a kid I always wanted to write the backs of baseball cards-- "Jim enjoys hapkido karate"-- so now I help produce the stats!
1:55 - 2:15 PM Cardinal Numbers Sig Mejdal, Senior Quantitative Analyst, St. Louis Cardinals
2:15 - 2:25 PM Break
2:25 - 2:45 PM Applying Operations Research Across the Baseball Decision Hierarchy J. Eric Bickel, Assistant Professor, Department of Industrial and Systems Engineering, Texas A&M University
Operations research can be applied at all levels of baseball decision making-from the field to the front office. In this talk I will present two examples spanning this range of application. At the field level, I will consider the decision of whether or not to take a particular pitch. For example, should a batter take on 3-0? Why? What about 2-0? From the perspective of the front office and player valuation, I will discuss a new method to determine how much a particular starting pitcher increases his team’s chance of winning. For example, by how much does Roger Clemens increase the Yankees’ chance of winning in games he pitches? These examples illustrate the breadth and power of operations research and highlight its ability to improve tactical and strategic decision making in baseball and beyond.
2:45 - 3:05 PM PECOTA Under the Hood Nate Silver, Columnist, BaseballProspectus.com
In its five year existence, PECOTA has gone from being a side project that saved me from boredom at my day job to perhaps the most widely recognized "brand" in baseball forecasting models. We'll take a whirlwind tour of PECOTA's philosophy, methodology, origins, strengths and weaknesses, and address the question of whether it really matters in the first place.
3:05 - 3:20 PM Statistics, Baseball, and Me Mitchell Watnik, Assistant Professor, Dept. of Statistics, CSU East Bay
Baseball is one of the reasons I become a statistician. A model for players' salaries inspired my dissertation. I will discuss how I have used baseball in my research and in my classes to show what the sport has taught me and my students about Statistics. I will also discuss what those in the sport-- both in the business and in fantasy-- can learn from Statistics.
3:20 - 3:40 PM Chris Long, Senior Quantitative Analyst, San Diego Padres
3:40 - 3:50 PM Break
3:50 - 4:20 PM Measuring the pros and the PROTRADE market Jeffrey Ma, Vice President of Research and
Mark Kamal, Statistical Analyst, Protrade.com
Creating the first real stock market for sports was PROTRADE's original mission statement and with that we created the most state of the art statistical tools in terms of quantifying an athlete's ultimate worth. Initial efforts centered on assessing an athlete's contribution on a play by play level to his team's ability to win. As we have evolved the business, our core market has been built less on real world value but more on statistical measures easier for the common consumer to understand. However, we have continued to develop exciting statistical measures which have relevant applications to major media presences and have been featured on ESPN, in the NY Times, Wall Street Journal, MLB.com and Sports Illustrated.
4:20 - 4:40 PM Supply Chains, Investment Portfolios,
and Roster Management
Joel Sokol, Associate Professor, School of Industrial and Systems Engineering, Georgia Tech
Operations research (OR) techniques are commonly used to successfully determine investment portfolios and coordinate supply chains. Many of the same basic supply chain and investment concepts also are important in the long-term management of baseball rosters, implying that OR techniques can be helpful tools for roster management. At the same time, baseball rosters differ from supply chains and financial investments in some ways, requiring the development of new OR methods to account for those differences. In this talk, we discuss some of the major similarities, describe how our current research is attempting to account for the differences, and suggest that an opportunity exists for a beneficial collaboration between the OR and baseball communities.
4:40 - 5:00 PM Baseball Statistics Meets Mathematical Statistics Hal Stern, Professor and Chairman, Dept. of Statistics, UC Irvine
Numbers and statistics have long been central to baseball players, professionals and fans. The scientific field of statistics is concerned with using quantitative information to learn about a subject. It is no surprise then that there is a long history of statistical research related to baseball. This work concerns a variety of subjects including the evaluation of players and the evaluation of baseball stategy. A number of examples are presented here focusing on the potential application of modern data analysis strategies.

Click here to return to the Symposium home page.