Baseball Think Factory links to a developing debate between Joe Posnanski and Sean Forman about the defensive adjustments to pitcher WAR at Baseball Reference. As someone who dabbled in measuring defense, figuring out how much some defense helped or hurt a particular pitcher is a tough problem.

The Posnanski piece wonders why rWAR rates Justin Verlander so much higher than Rick Porcello. The difference comes down to a defensive adjustment. The Red Sox rate as an excellent defense, the Tigers as a poor defensive team. Posnanski’s argument is that the Tigers played great defense behind Verlander, as his BABIP was .256 versus the team BABIP of .302. Forman’s argument is that the team’s defense is the team’s defense, and there’s no reason to think they got good because Verlander was on the mound. There is a lot of random variation in baseball, because nearly every set of individual stats is a small sample size.

When I worked on the Probabilistic Model of Range, one thing I liked to do was look at the expected outs a pitcher produced versus the actual outs recorded. Here’s a post from the end of the 2011 season, when Verlander won the Cy Young Award:

Notice that Justin Verlander and Josh Beckett both produced high expected DERs, but the defense magnified that by turning so many more batted balls into outs. Verlander pitched very well, but his defense gave him that extra push into the Cy Young.

Maybe that happened again. Maybe Verlander is just so good at inducing poor contact that even a bad defensive team looks good. I would love to know the expectations for Verlander’s and Porcello’s balls in play.