April 21, 2019

Clustering

David Laurila’s weekly notes column contains a discussion with Red’s catcher Tucker Barnhart about how the data he now receives helps him call pitches. This part stood out for me:


“You can put in a database, ‘Luis Castillo versus hitter A’ and come up with his mix, and what’s been successful against that specific guy,” Barnhart told me. “You can find which pitchers in the league are as close as possible to him, and look at their history against certain guys. There are a lot of things out there that I wasn’t privy to until fairly recently.”

FanGraphs.com

About 15 years ago I helped a team with a something like this. Without revealing anything important, they had ratings for pitchers, and wanted to use those to expand the sample size of the batter/pitcher match-up. So if player X was 3 for 10 against that day’s starter, but 4 for 30 against all pitchers like that starter, maybe you sat him that day. It’s an example of how long it takes some organizations to delve into the data.

Leave a Reply

Your email address will not be published. Required fields are marked *