Yesterday I sat looking at the Nats 2012 roster trying to find a positive regression candidate. When most people thinking of the term regression they think of a decline. Of stats moving in a negative direction, but that isn’t what the term means. The entire term is regression to the mean and it is a theory that over a given period of time a player will return closer to their career averages. That after any hot or cold stretch the most likely thing to happen isn’t for the hot or cold stretch to continue but for the player to return to being themselves. A great example of this is 2011 Jayson Werth. Werth for his career has hit .267/.362/.462. In the first half of 2011 Werth wasn’t good as he hit .215/.319/.362 but in the second half of the season his stats were much closer to his career averages at .255/.345/.426. Werth regressed in a positive direction, but because of such an awful first half his season numbers were never going to look good even though he was much closer to the Werth that should have been expected in the season’s second half.
While I was scanning the Nats 2012 roster I was obviously missing the biggest positive regression candidate. The Nats have so many players returning that I never even bothered to think of players who weren’t on the Nats in 2012. The biggest is Dan Haren who last season had one of the worst years of his career with a 4.33 ERA and 4.24 FIP in 176 2/3 innings. When looking at Haren from a three true outcomes perspective it becomes quite obvious as to why he had a down season. For his career Haren has a 7.60 K/9 and in 2012 a 7.23, he is a career 1.89 BB/9 pitcher and in 2012 1.94, neither of those are far enough off to make a real difference, but Haren has a career 1.05 HR/9 and in 2012 it was 1.43. When batters were making contact against Haren it was harder contact. This could have something to do with the fact that Haren has dropped velocity on his fastball, but he has dropped velocity at the same rate over the last several seasons.
The best explanation is that while Haren suffered a back injury he lost a bit of movement or that it is simply the random variations that can be expected season to season. If Haren’s homerun rate returns to his career norms then there is little to no reason to doubt that he can be the pitcher that he was before the 2012 season. Look at it this way. For his career batters have hit .252/.294/.407 and in 2012 they hit .275/.312/.463. The increase in SLG is by far the most troubling change, but consider that Haren has a career .291 BABIP against and in 2012 it was .305. It isn’t much of a difference in hits allowed on balls in play, but a homerun isn’t a ball in play, and BABIP doesn’t count the quality of the hit.
According to most batted ball stats Haren was the exact same pitcher in 2012 as he had been for most of his career. His line drive rate rose minimally from 19.9% for his career to 20.7% but his ground ball rate decreased from 43.2% to 39.6% and his fly ball rate increased from 36.9% to 39.6%. Haren wasn’t giving up that many more line drives, but his ground balls were becoming fly balls and more of those fly balls were leaving the yard. In total his HR/FB increased from 10.5% to 12.8%. None of this seems all that alarming. It is one small area in Haren’s game that has to improve and it can be deduced that all it would take for that improvement to happen is for Haren to continue pitching.
If you’re still not convinced think of it this was. In 2012 the average AL team hit 179 homeruns and the average NL team 152. Haren no longer has to face the DH and that is going to make a difference. In 2012 the average starting pitcher in the AL had an ERA of 4.37 and in the NL 4.04. If Haren doesn’t do anything differently at all his HR/FB rate and ERA should both decrease simply because he no longer has to face a DH. The move from the AL to the NL is rarely a bad one for a pitcher. Haren is widely regarded as not just a talented pitcher but a smart one as well. He has the right mix of stuff and control to be successful wherever he pitches. In 2012 he had a bad season due mostly to giving up more homeruns than he typically has in a season. That should regress naturally to the mean and perhaps should improve due to the move from the AL to the NL. Haren might not simply return to his career norms, but by virtue of pitching in the NL could better them in 2013.