In Praise of Speed
Back in 2003, Scott Hatteberg was the best hitter in baseball–OK, maybe not the best but apparently most mispriced–and the arbitrageur Billy Beane was eating free lunches almost daily. OPS was what mattered. Of course, then OPS became overpriced, and Billy started buying based on other statistics, ones that I don’t know of because he smartened up and didn’t let Michael Lewis put his new ones in a book. Intuitively, I still like OPS, but I think I have improved it. Here’s AOPS:
OBP + (1B + 2 x 2B + 3 x 3B + 4 x HR + SB – CS) / AB
Okay, simply, I just adjusted OPS (OBP + SLG) by adding the bases a runner steals and canceling out his hit when he gets caught stealing.* I ran the numbers for all 329 hitters with more than 200 AB in 2008. It’s not very interesting to look at the top ten hitters before and after the adjustment. They stayed the same, except for Youk falling from 9th to 11th, which is because the hitters with the biggest OPS and AOPS numbers will be the homerun hitters, and won’t benefit much from including steals (also, there’s a long tail factor at play, which means that ranks won’t change much). But my guess is that the Mets and Sox fans reading this will be interested to know that four of the biggest beneficiaries of this tweak to OPS are Jacoby Ellsbury, Jose Reyes, Luis Castillo, and (surprise) Coco Crisp.
Jacoby’s jump is huge. Before the adjustment, he’s comparable with crappy players like Jose Bautista, Troy Tulowitzki, most of the KC lineup, and old friends Mark Loretta, Kevin Millar, and Lastings Milledge. After: Mike Lowell, Russell Martin, Cliff Floyd, and Ichiro. His jump is from the 38th to the 61st percentile of players. Reyes’s jump may be even more impressive, though it is much smaller percentile-wise. After adjustment, his comparables are quite nice:
Sorry to include so many players, but I wanted to get Ryan Howard in there for the Mets fans. Anyway, the biggest risers** are included below.
*I realize that a single followed by a steal is not as valuable as a double, which often scores the non-David-Ortiz runner on first, but I’d say that this effect is balanced by the fact that a single followed by a caught stealing is not as bad as a strikeout, because that single and caught stealing probably scored the runner on second and almost certainly the one on third.
**The biggest fallers don’t tell us that much because they’re the players who never try to run and are getting passed by those who do. As such, I’ve not included them.