Fivethirtyeight.com - Nate Silver's new website and its sports page is a great addition to the world of data analytics in sports for the serious fan. Today they posted this article about umpire biases in three ball or two strike counts, as well as pitches right after a strike. They prove that on borderline pitches, umpires favor making the call to keep the at bat going, rather than calling a third strike or fourth ball.

 

The analysis is fascinating, however, I feel the story statistically confirms what most passionate baseball fans already know, which is that there is a built in bias in the game to see the ball put in play. Baseball players will largely recognize that on 3-0 counts, the umpire is going to widen the strike zone in favor of getting the ball put in play. The game counts on the ball getting hit and fielded, and the umpire biases his calls to favor this outcome. The same holds true for the converse. In 0-2 counts, the umpire is going to give the benefit to the hitter.

 

Removing this unspoken element of the game by using electronic methods would have a deep and profound impact, quite possibly negative.

 

What do you all think?

Removing this unspoken element of the game by using electronic methods would have a deep and profound impact, quite possibly negative.

 

I believe this to be the case. For good or bad, umpires are stewards of the game. They make bad calls on occasion but in many cases, they make the game more interesting to watch through their inaccuracy, whether it's conscious or not.

Data is from 2009-2011. Umps have only been using Pitchf/x to judge their strike calling since 2008. They do need to get better at calling strikes in 3-0 and 0-2, and I believe they will, but it will take time.

 

http://www.baseballanalytics.org/baseball-analytics-blog/2014/2/27/umpires-calling-balls-and-strikes-better-than-ever-before.html

 

If at some point the quality of umpiring plateaus where umps are repeatedly making wrong calls despite their best efforts to correct themselves, then I think you talk about having tech take over or expanding replay. Right now they are just figuring out those patterns.

I think it's psychologically pretty simple--we tend to aggregate our judgment towards the mean when we see our judgment swing hard one way (0-2, 3-0). Not the worse of predilections, I suppose.