What to Do with Andy Sonnanstine?

I have finally joined the 2008 and got on Twitter. I cannot promise anything great from my feed, but I like how easy it is to follow baseball news and read other analysts’ passing thoughts. It has already paid off in the form of the idea for this post. This morning Sky Kalkman tweeted:

Andy Sonnanstine: trade bait, rotation candidate, bullpen filler, or AAA veteran?

I had sort of forgotten about Andy Sonnanstine, but the tweet reminded me what an interesting pitcher he is. Sonnanstine was worth over 3.5 wins for the Rays in 2008, but then things fell apart in 2009. Part of that was bad luck on his BABIP and HR/FB, but part of it was also based on his performance. Sonnanstine doesn’t strike out many batters or get that many ground balls, so he needs to have a great walk rate to succeed. That is what he did in 2008 — walking just 1.7 — but in 2009 it increased to 3.0, erasing much of his value.

The increase of walks was not from missing the strike zone — his pitches were in the zone just as much — but, rather, from batters swinging less often at his out-of-zone pitches. Here are those numbers by pitch type:

             O-swing        O-contact
           2008   2009     2008   2009
Fastball   0.23   0.23     0.83   0.80
Cutter     0.22   0.19     0.73   0.72
Slider     0.40   0.35     0.55   0.69
Curve      0.29   0.26     0.57   0.86

The rates on his fastball were essentially the same, but for his cutter, slider and curve the O-Swing rates were way down. This turned many more plate appearances into walks. Another big problem was that the O-Contact rate on his slider and curve were way up, although this was not responsible for the increase in walks it does show these pitches were easier to hit.

Interestingly Sonnanstine also threw his cutter much more often in 2009. According to my pitch classifications, it went from 28% of the time in 2008 to 44% in 2009. Mostly this change came at the expense of his fastball which went from 35% to 24%. It could be that hitters do better on the cutter after seeing it more often or because they are expecting it. But I do not see evidence for this on an at-bat level. That is there was no trend for batters to do any better on the second or third cutter they see in an at-bat than the first cutter they see in at-bats against Sonnanstine.

Getting back to Sky’s question, I think I would take a little from column a, a little from column b and a little from column c. That is start him off in the pen as a long reliever — since the Rays have five better starting options — but with the eye to trading him if anything of value comes along or moving him to the rotation if needed. He has a relatively small platoon split, doing fairly well against LHBs, so deploying him as a long reliever would be a nice way to leverage that talent. Also his very deep repertoire of pitches plays well as a long reliever where he might have to face batters multiple times. These skills also mean he might be better suited as a starter if he can get things back together, which might mean throwing his fastball a little more often.