A study in shows that umpires unconsciously incorporate racial/ethnic bias into calling the strike zone. What does that mean for baseball? For society?
Major League Baseball umpires exhibited racial/ethnic bias in calling strikes between 2004-08, which translated to wage discrimination against Hispanic pitchers, according to a study published in the American Economic Review last month.
The study found that home plate umpires were less favorable in judging the strike zone with pitchers of a different race/ethnicity than their own, while pitchers matching the race/ethnicity of home plate umpire benefited slightly.
The 27-page findings of the study, Strike Three: Discrimination, Incentive and Evaluation, established positive and negative biases by home plate umpires by analyzing over 3 million pitches over four seasons worth of games between 2004-08, and found evidence that pitchers were cognizant of them.
The result: tailored pitching strategies to compensate for the effects of widened or narrowed strike zones. Pitchers benefiting from the bias were found to exploit it by targeting the fringe areas of a given strike zone, where umpires use the most discretion and where pitchers would gain the most competitive advantage over batters.
Those pitchers given less favorable calls in what the report described as "fuzzy" regions adjusted to the bias, and threw more easily judged pitches to avoid it to the detriment of their performance, a striking discovery according to one researcher.
Does evidence of umpires incorporating a racial/ethnic bias into calling balls and strikes bother you?
"You’ll throw pitches that are obviously strikes or balls, pitches that are not difficult to call. We found that a minority pitcher facing a white umpire would throw fewer curveballs, sliders and change-ups, and less likely toward the edge of the plate.
“Pitchers are adjusting their behavior to the umpires’ discrimination, but do it in a way that makes it harder for the pitcher. If you throw it toward the middle of the plate, a batter is more likely to hit it out of the park.”
Hamermesh believes that managers could apply the findings toward strategy, by considering a pitcher's race/ethnicity as much as any split situational statistic before inserting him into a given game, though neither he nor the study provided evidence suggesting that it already happens.
“It should even affect the behavior of managers," he said. "If I'm a manager, I should try to match pitchers’ and home plate umpires’ race/ethnicities where possible."
The study also cited a measured effect on baseball's labor market and analysis of it, particularly in studies of racial/ethnic wage disparity.
Due to the effects of the bias, Hispanic pitchers' earnings potential was undercut by as much as $500,000 per player per season, according to Hamermesh.
“You better believe it affects front-office decisions. If I’m a minority pitcher, my stats aren’t going to look as good. And front offices look at stats like how much to pay, who to trade, and so on, this means that in all kinds of labor market circumstances, the minority pitcher would be discriminated against and not do as well.”
The Elias Sports Bureau does not track player race/ethnicity or contract information.
Hamermesh, who contributes to Freakonomics.com, the same-named website of the best-selling series of books, did not believe any of the findings resulted from conscious prejudices.
“No, I don’t think so at all," he said. “I’m sure umpires don’t say, ‘Oh, I’m going to discriminate against Hispanics.'"
"It exists in the outcomes from the time," he said. “That’s enough.
“Let’s say I talked to pitchers and they said, ‘I’ve never thought about (the bias). But that wouldn’t mean it didn’t happen, however unconsciously. You’ve got to look at behavior. Watch what we do, not what we say.”
Hamermesh also cautioned against assuming that the bias targeted Hispanic pitchers specifically, given the absence of other representative minority subpopulations for comparison renders further social analysis inconclusive.
The percentage of Hispanic pitchers in Major League Baseball International was 29 percent in 2010, compared to the five percent representation of African Americans, and two percent of Asian Americans, according to the Institute for Diversity and Ethics in Sport.
That gap has widened since the 2004-08 period, when only 22.88 percent of pitchers were Hispanic, 3.04 percent were African American and 3.21 percent were Asian American, according to the study. Baseball employed 102 home plate umpires during that time, of which 89 percent were white, 4.90 percent were Hispanic and 5.88 were African American.
No Asian Americans served as home plate umpires between 2004-08.
MLB did not respond to an email requesting the race/ethnicity of its umpires, and comment.
Researchers gathered variables from PITCHf/x, a computerized technology that measures the trajectory of balls thrown, tracking movement from the pitcher's hand to beyond the plate. The program records the pitch type; the location of where the pitch crosses the plate, relative to center-front of home plate; and the top and bottom of the strike zone, determined by the program operator.
Analyzing the data found little discrepancy between strike calls for balls crossing within the strike zone's "inside" region, an ellipse with a 1.6 foot diameter. Only a 0.3 percentage point swing was observed for unmatched pitcher-umpire pairings.
Pitches thrown to the outside region, the outermost area defined by the program, "showed no difference at all," according to the study.
Eighty percent of all pitches recorded crossed at one of these two regions, split evenly at 40 percent each.
Hamermesh called that discrepancy significant enough to change statistics and outcomes of games, and consequently, Hispanic pitcher salaries.
“It means that if you’re a minority facing a white umpire you’ve got a problem.”
The study notes exceptions to the observed bias, which is negated by the presence of formal and informal monitoring.
That can be institutionalized, like in the one-third of major league stadiums between 2004-08 with QuesTec PITCHf/x technology, used to cross-reference umpires’ discretion with a standardized, digital strike zone.
It can also be casual, like the urgency of regular or postseason games that draw high fan attendance, and consequent scrutiny. The study defined "well-attended" games as those just under 70 percent of stadium capacity.
Decisive situations, like potential third strikes and fourth balls, also eliminated any notable bias, according to the study.
Hamermesh noted the recent popularity of automated strike zone technologies in MLB stadiums beyond that of the analyzed seasons, and expects the bias found in the study to be less prevalent because of the increased monitoring.
MLB did not respond to another email regarding the use of strike zone technologies in stadiums, and toward evaluating umpires.
“With the new systems in place everywhere, I would bet there’s much less of this going on,” Hamermesh said. “That would be a great study for someone else to look into.”