Follow Datanami:
April 8, 2014

Come On, Blue! Data Reveals Umpires’ Biases

George Leopold

It turns out that Major League Baseball umpires really don’t need glasses when calling balls and strikes. Instead, empirical research using pitch-location data revealed a variety of subjective judgments and biases that result in MLB umpires getting it wrong 14 percent of the time when the batter didn’t swing at a pitch.

Researchers Brayden King of Northwestern University and Jerry Kim of Columbia University crunched the numbers – 700,000 pitches thrown during the 2008 and 2009 seasons – and discovered that factors like the pitch count, the pitcher’s status, even race, can increase or shrink the strike zone.

Their results, to be published in the journal Management Science, first appeared in the New York Times. The pitch-location data was culled from high-speed cameras used to grade the performance of MLB umpires working behind home plate.

The researchers said they were surprised to find “such random errors” since big league umpires are “keenly aware that their ball-strike calls are being scrutinized….”

Wikimedia Commons

One predictable finding is that umpires tend to widen the strike zone for the home team. After all, umpires and other professional referees prefer to leave a stadium or arena without being accosted by angry fans. This home team bias resulted in umpires widening the strike zone for the hosts 13.3 percent of the time, King and Kim found.

The leisurely pace of big league baseball also plays into umpires’ judgments. For example, Thursdays on the big league schedule tend to be travel, or “get-away” days for players and umpires. Hence, the games are often played in the afternoon. The desire to end the game and make a flight to the next city on the schedule tends to widen the strike zone for both teams. The home plate umpire wants hitters to swing rather than “work the count” to draw a walk.

Hence, balls become strikes.

A pitcher’s reputation also influences the dimensions of the strike zone. All-Star pitchers tend to get strike calls when they “paint the black,” that is, throw pitches just off the edge of home plate. Wild pitchers and rookies tend not to get these strikes calls until they establish their reputations with umpires.

The researchers found that umpires are an estimated 16 percent more like to erroneously call a ball a strike on behalf of star pitchers. “The strike zone did actually seem to get bigger for All-Star pitchers and it tended to shrink for non-All-Stars,” the researchers reported in the Times.

However, umpires tend to favor hitters with a reputation for “knowing the strike zone,” particularly the strike zone of individual umpires. The researchers didn’t address the issue directly, but the fact that an All-Star hitter didn’t swing at a pitch on the outside corner frequently results in umpires calling the pitch a ball.

Good hitters also know how to “work the count,” and the researchers found that the count influences umpires’ perception of a pitch. For example, they found that umpires mistakenly call a ball a strike on a 3-0 count 18.6 percent of the time. Conversely, the strike zone tends to expand when the count is 0-2. They concluded that umpires simply do not want to end an at-bat. This tendency is amplified at the end of a close game.

Perhaps the researchers’ most troubling finding was that race plays a role in determining the dimensions of the strike zone. Umpires were 10 percent less likely to expand the strike zone for African American pitchers. The researchers did not report findings for the growing number of pitchers from Latin American countries.

Race, of course, has always been a subtext in Major League Baseball. That’s why it’s especially gratifying this week to be observing the 40th anniversary of Henry Aaron’s record-breaking 715th home run.

Umpires were acutely aware that “Hammerin’ Hank” knew the strike zone.

Related Items:

Playing Ball with MLB’s New Analytic Data Feed

Inside Major League Baseball’s “Hypothesis Machine” (HPCwire)

When Data Analytics Goes Horribly Wrong: A Sporting Example

 

 

Datanami