Tuesday, April 20, 2010
“I was gonna lay down the bunt, but then I got high”
Posted by Nick SteinerI was watching a Cardinals game a couple of days ago (I believe it was the Sunday night one against the Mets) when one of the announcers mentioned that it was harder to bunt pitches that were elevated in the strike zone. The idea is that, since a successful bunt is almost always going to be on the ground and there is a very strong relationship between vertical location and ground ball percentage across all pitch types, it's harder to bunt pitches up in the strike zone. That makes sense, and I've heard it a million times in games; however there is the whole question of whether or not it is actually correct.
My method of investigation was pretty simple. I looked at all attempted bunts by pitchers in 2009 and measured whether or not they resulted in a sacrifice or a hit. Those were considered "successful" bunts. There were 2244 attempts last year overall and 63% of them were successful. I then grouped that by the vertical height of the pitch, in intervals of .33 feet. Here is a graph of what I found with the height of the pitch (in feet) on the x-axis, the percentage of bunt attempts what were successful on the y-axis and the standard error range in gray:
As you can see there is actually a slight upward trend in the graph. In other words, the higher the pitch was the in the strike zone the more often it was bunted successfully. The 2008 data shows a similar trend as well.
Now this brief analysis leaves out one key aspect of bunting - foul balls. Most people would consider a foul ball on a bunt attempt a failure; however, foul bunts with less than 2 strikes are not recorded by the GameDay operators. Only at bats that ended with a bunt are. So while based off of this analysis we can say that when a bunt is put in play there is little to no relationship between it's vertical location in the strike zone and its odds of success, that doesn't not necessarily mean that applies to all bunt attempts. Still, it appears that there isn't much truth to that bit of conventional wisdom, at least in regards to pitchers.
Nick is a Cardinals fan living in LA. He also writes for Viva El Birdos. He welcomes questions and comments via email
Nick - Retrosheet and MLBAM record all foul bunts in their pitch sequences as a pitch type “L”. There were 645 by pitchers last year of which 82 resulted in a third strike.