Ball Four: Or is It?

I can't recall how many times I've watched a pitcher get a little wild, walk a couple of guys, and then can't catch a break from the umpire. Almost every pitch that borders between a strike and a ball is called a ball. This inevitably forces the pitcher to start aiming his pitches, which usually translates into him either getting even more wild and walking even more batters or grooving one right down the middle of the plate, which the batter then launches into one of the gaps and drives home numerous runs. It's as if once a pitcher becomes a little wild, the umpire assumes that the next pitch will be a ball. And unless the next pitch is clearly a strike, the umpire calls it a ball. Of course, I have had no data by which to test this hypothesis. Data on how Little League, high school, and college umpires call pitches do not exist.

But, they do exist for professional baseball. Pitch location data have been collected by Major League Baseball for several years now, and a recent study Brayden King, associate professor at the Kellogg School of Management at Northwestern University, and Jerry Kim, assistant professor at Columbia Business School, indirectly appears to confirm my suspicions ("What Umpires Get Wrong"). They not only found that umpires call about 14 percent of non-swinging pitches incorrectly (i.e., calling a strike a ball and a ball a strike), but they are heavily influenced by unconscious preconceptions of what the next pitch will be. For instance, umpires were 16 percent more likely to call a ball a strike if the pitcher had appeared in an All-Star game than if he hadn't, and they were 9 percent less likely to mistakenly call a strike a ball if the pitcher had appeared in an All-Star game than if he hadn't. Similarly, a pitcher who has a reputation for not walking a lot of batters is much more likely to have a ball called a strike than a pitcher who is known for being wild.

Thus, it isn't too much of a leap to suppose that after a pitcher has walked a couple of batters, umpires are much more likely to call the next pitch a ball than if he hasn't. What's interesting is that I can actually test this hypothesis. I think I'll wait until I get tenure before I do, however.