Mark Pankin Baseball Page

Retrosheet - collects, computerizes, and distributes accounts of major league games played before 1984.

Markov Models/Batting Order Optimization - mathematical models based on the probably of moving among the base runners and numbers of outs situations. These models can be applied to strategy analysis, but I have concentrated on using them to a) estimate how many runs a given batting order will average, and b) determine the best batting order for a given nine players.

June 1, 2006 article in Philadelphia Daily News about the best batting orders for the 2006 Phillies.

Who's Right: La Russa or Gant? - Presentation given at SABR 29 in Scottsdale AZ in June 1999 discussing whether or not La Russa was correct to put the pitcher 8th in his batting order after the All-Star Game in 1998. He did so to try to get more runners on for Mark McGwire hoping the McGwire would be walked less frequently as he pursued the home run record. Card's outfielder Ron Gant, after being traded to Philadelphia, was critical of this tactic and said it may have been good for McGwire, but it hurt the team. Situational analysis of McGwire's performance and the Markov model are employed to answer this question.

Does Speed on the Bases Win Games? - Of course it does. The real questions are how many additional runs result from faster runners and what is the tradeoff between speed and better hitters. This presentation given at SABR 30 in West Palm Beach, FL in June 2000 studies the topic using the Markov model. In a nutshell, it doesn't take much of an improvement in hitting ability for a team to overcome the effects of slower runners.

To see or print the following presentations made at SABR conferences, you need Acrobat Reader. It is probably already on your computer, but if not, it is a free download from Adobe web site.

Do Base Stealers Help the Next Batters? - Slides and some notes from presentation given at SABR 32 in Boston in June 2002. The link is to a fairly large (about 1.3 Mb) Adobe Acrobat file.

When Should Bonds be Walked Intentionally? - Barry Bonds has had the two greatest offensive seasons in baseball history. Only Babe Ruth in 1920-21 was even close to Bonds in 2001-02. That has caused opposing teams to sometimes walk him intentionally in non-traditional situations. When is this a good idea? The link has the slides and some notes from my presentation given at SABR 33 in Denver in July 2003. The link is to a fairly large (about 1.0 Mb) Adobe Acrobat file.

Relative Value of On-Base Percentage and Slugging Average - Presents research inspired by a comment in the book Moneyball that an extra point of OBP is worth three extra points of slugging average. Is that really the case? How did Oakland's GM Billy Beane do in applying this idea when trying to replace Jason Giambi and Johnny Damon who left as free agents after the 2001 season. The link has the slides and some notes from my presentation given at SABR 34 in Cincinnati in July 2004. The link is to a 600Kb Adobe Acrobat file.

Additional On-base Worth 3x Additional Slugging? - This is a follow-up study to the one above that examines the relative marginal values of on-base percentage and slugging average for major league play over the period 1901-2005. The focus is on seeing how the ratio of the marginal values varies as scoring levels have changed over time. Also studied are a few extremely high scoring and pitifully low scoring teams and the individual players in the 1927 Yankees lineup. The link has the slides and some notes from my presentation given at SABR 36 in Seattle in July 2006.

A Markov Model of Baseball: Applications to Two Sluggers - Talk given at national meeting of INFORMS (Institute for Operations Research and Management Science) in Pittsburgh in November 2006. Talk mainly combines "Who's Right: La Russa or Gant?" and "When Should Bonds be Walked Intentionally?" talks given at SABR. Discussion and notes are a bit different with some added material.

Should Pitchers Bat 9th? - Pitchers usually are not hitting once a National League game gets into the late innings. For most teams their replacement hitters (pinch hitters or those who enter as the result of a double switch) are better than the teams' number eight hitters. Would it make sense to bat the pitcher higher in order to get their replacement hitters up sooner in the later innings of the game? The link has the slides and some notes from my presentation given at SABR 37 in St. Louis in July 2007.

Bragan's 1956 Pirates Unconventional Lineups - Bobby Bragan was in his first year as a major league manager, and his Pirates after getting off to a fast start in 1956 had faded and were well out of contention in August. He had some interesting ideas about better batting orders that were contrary to conventional wisdom, and he used them from August 18 through Labor Day. How valid were his ideas, how well did they work, and why did he stop implementing them? The link has the slides and some notes from my presentation given at the regional SABR meeting in Columbia, Maryland on January 26, 2008.

What do Rock/Paper/Scissors and Baseball Have in Common? - Game Theory, a mathhematical analysis method, applies to both. R/P/S is very simple in that regard, and the solution is not very interesting. Baseball, on the other hand, is far more complicated and has many facets to which game theory can be applied. The presentation, which has virtually no math in it, looks at the possibility of bunting and defending against bunts with a runner on first and none out. Most analysis has shown that bunting, except by very weak hitters such as pitchers, is not an effective way to try to score. Game theory says that being too predictable is not a good idea, and the analysis shows that bunting some fraction of the time, particularly when it is not expected, may be helpful. The link has the slides and some notes from my presentation given at SABR 39 in Washingtion in July 2009.

Did No Lights at Wrigley Field Hurt the Cubs? - The Cubs were the last team to play home games at night. The first was in August 1988 while all other major league teams had lights by 1949. Many former players and some analysts have said that playing all day games at home hurt the team, and it is "famous" for some dramatic late season collapses, particularly in 1969. This paper uses Retrosheet data for 1950-2010 to examine the question and see if things have changed since Wrigley Field has had lights for 22 full seasons. I lived five blocks from the park my last two years in graduate school and enjoyed going there. The link has the slides and some notes from my presentation given at SABR 39 in Long Beach, California in July 2011.

How Large is the Advantage of Batting Last? - Baseball is unique among the major sports in that one team knows it will get the last chance to go on offense if needed. This is clearly an advantage, but how large is it. Examiming most of the major league games played since 1901, this presentation develops an estimate and relates to the typical home field advantage that sees the home teams win between 53% and 55% each season. The 1894 season when home teams often batted first is presented as a special case.


SABR Home Page - SABR is the Society of American Baseball Research. The home page has information about the organization, data of interest to researchers, and links to other baseball web sites.

Sean Lahman's Baseball Page - loaded with great data and stats, some that are hard to find anywhere on the Web, and some essays.

Skilton's Baseball Links - hundreds of (maybe over 1000) links to baseball sites