Mar 28, 2007

Umps Effect on the Outcome of Games v. 0.9

AKA, Quantifying the umpire's effect (specifically, balls and strikes). In an attempt to quantify the effect of 'pitcher's strikes' and 'hitter's balls,' my plan is to analyze several games, pitch-by-pitch, and count the number of 'borderline' strikes and balls - calls that could've reasonably gone either way. Next, what is the result of the at-bat? What's the effect on the pitcher vs. the effect on the hitter? Do the calls even out over the course of a game? How often do they effect the outcome of the game?

My guesstimate is that on a given pitch, the outcome (strike, ball, hit into play) is controlled thus by responsibility: pitcher 70%, hitter 20%, umpire 10%. So the umpire will make a 'borderline' call every 10 pitches or so on average. That sounds about right to me, but I will test it against a few ST games (Yankees and Mets).

We know that hitters almost universally hit worse when falling behind in the count. And conversely, hit better when ahead in the count. What is the effect of calling the first pitch a ball as opposed to a strike? What about on a 1-1 pitch?

Ultimately, I would like to see umpires replaced by technological innovations that will never fall victim to human errors, e.g. giving an established pitcher (Maddux, Schilling) a strike solely due to renown, or vice versa for a hitter (Giambi, Ortiz).

As pitch counts for ST games are largely unavailable, I'll wait for the regular season. I expect to analyze a few early season games to determine how much effect different umps have on the outcome of games.

By the way, BP has an umpire report here. There are some large differences between umps, but I want to know more: do veteran pitchers/hitters get calls in their favor, and how many? Conversely, do rookies get close calls go against them? What about pitches where the catcher has to move his glove significantly? A lot of those calls go against the pitcher because the catcher's movement causes the ump to assume it's a ball.

No comments: