The Red Sox and Stolen Bases

Teams that rely heavily on sabermetrics are usually cast aside as teams that dislike the stolen bases. The perception is that the two are lip locked in an inverse relationship. Something like this:

Of course, most of this perception comes from Moneyball, in which Ray Durham is treated like a prisoner upon his arrival. Bemoaned not to steal, otherwise something bad would happen. Obviously this isn't true, instead sabermetric teams simply pick and choose their steals in situations more prone to success. Something I found interesting is the Boston Red Sox and their steal attempts since Theo Epstein became General Manager.

Epstein took over after the 2002 season, Mike Port was the GM during that year, and I'm including him only to show where the Sox were when Epstein took over. The Red Sox stole 80 bases in 108 tries that season, good for 74%. In 2003 the steal attempts increased by 15, but the success rate dropped to 72%. In every year until 2007 the attempts dropped, and with the exception of 2005, so did the success rate. That doesn't make a ton of sense. A smart team steals less, therefore they should be successful more often, right? Well, apparently not.

For whatever reason, things changed in 2007, and the Sox not only ran more, but were successful more often -- the most in Epstein's tenure actually. In 2008 the steals increased again, and the success rate stayed high. So, what in the world was going on here? I had two common sense theories:

A. The Sox had lower OBPs in the years they stole less -- which seemed unlikely, but who knows.

B. The Sox were built as slugging/homerun hitting teams during the years they stole less.

Neither of those had much of a correlation with attempts, which lead me to the other theory: change in philosophy. Epstein could only make so many changes to the 2003 team, and they ended up stealing more. 2004 was more of his team, and then he left in 2005 before returning after a few moves were made. 2006 and since were, again, his teams.

I suppose this post isn't as much about Epstein, stolen bases, or the Red Sox, and more about the perception that statistical teams never adapt to new philosophies. I guess some don't, and those usually aren't successful for much longer. The best ran teams are the ones who are willing to change, and those who are finding the change.