The study, a collaboration between researchers at U-M and the universities of Illinois and Florida, looks deeper into the controversial argument over whether MLB umpires discriminate by calling more strikes for pitchers of the same race. It found little statistical evidence to support that claim, said Jason Winfree, associate professor of sport management at the U-M School of Kinesiology.
Winfree and co-authors Scott Tainsky of Illinois and Brian Mills of Florida,After analyzing millions of pitches between 1997 and 2008, the results suggest that findings of discrimination were questionable.
A draft of the earlier study that initially found evidence of discrimination was released in 2007 and published by the American Economic Review in 2011. National media reported on both the draft of the study and on its later publication.
In the U-M study, Winfree and his co-authors analyzed both their own data as well as that of the previously published study, but did not get consistent results when using different statistical methods and variables.
"Based on what we found, it's(discrimination) certainly not conclusive, and we could make an argument that there's actually reverse discrimination if you look only at averages," Winfree said. "Our point is (that) with something like this you want to look at the data a lot of different ways and see if you get a consistent result each time with each method. It's a pretty bold claim to say there is racial discrimination."
Winfree and colleagues found that the only specifications that suggested discrimination were when the analysis treated pitchers as completely separate players when pitching in stadiums where umpires were monitored. This seemed to drive much of the findings in the earlier study, they said.
The U-M study and others that look at discrimination in sports are significant not only for sports fans and franchises, but because it's very difficult to test for discrimination in most other occupations, Winfree said. However, because professional sports keep such detailed statistics, discrimination or lack of it is more quantifiable.
"This is a place where it's easy to test for discrimination, and if you find it here, it might be present in other work scenarios where you can't really test it," Winfree said.
The issue of discrimination among referees and umpires has raised debate in other sports as well. For instance, in 2010 the Quarterly Journal of Economics published a study that suggested racial bias among referees in the NBA.
The U-M study, "Further examination of potential discrimination among MLB umpires," appears online in the Journal of Sports Economics.
- Study abstract: http://jse.sagepub.com/content/early/2013/05/02/1527002513487740
- Jason Winfree: http://www.kines.umich.edu/profile/jason-winfree-phd
- U-M School of Kinesiology: http://www.kines.umich.edu