Contemplating A.J. Burnett, Relief Pitcher

What’s there to say about A.J. Burnett?  He’s having an incredibly tough time right now, and this after looking like he’d made improvements from his career-worst 2010 campaign. These days, his ERA and FIP are both within a tenth of a run of his 2010 totals.  The Yankees’ “too many starters” dilemma has been a theme for a month now — since Phil Hughes returned from the DL and Ivan Nova came up from the minor leagues. On top of that, the Yankees get starts from ace CC Sabathia, Bartolo Colon and Freddy Garcia. So while some untimely weather and a finger injury to Garcia has put off Joe Girardi’s decision about whom to bump from the rotation, it looks like Burnett is still lined up to pitch against the Red Sox on Thursday. After his next start, it might not be a bad idea to try him out as a short reliever.

It feels like we’ve been saying it for a while, but one can’t help but wonder if Girardi has to pull the plug on Burnett if he has another clunker. After his disastrous nine-run outing against the Orioles on Friday, his August ERA is now at a tick under 12 and his FIP has been at 5.50 for the month. His normalized 4.16 xFIP is more reasonable, telling us that opponents have been smacking homers off of him at (probably) an unsustainable rate. So yes, he’s probably seeing some bad luck, but he’s still been pitching quite poorly.

With Hughes and Nova both pitching well, it doesn’t seem reasonable to kick one of them out of the rotation when the team needs to return to a five-man — though the Yankees have already seen Hughes excel out of the bullpen (in 2009, he gained 2 mph and saw his strikeout rates soar after being moved into a relief role). Given Burnett’s recent struggles, he’s looking like an obvious choice to be sent out, in spite of his veteran status and enormous contract. Something that’s become apparent is that Burnett has shown a tendency to lose some steam during games.  Taking all four-seam fastballs from his Yankees tenure (2009-2011) and averaging their velocities by pitch count, you can see that he’s not the same pitcher once he gets deep into his start:

A trend for plate appearance outcomes is less discernible, but lumping his plate appearances into two buckets (innings 1-4, innings 5+) indicates that his ability to get strikeouts and swings-and-misses deteriorates in the later innings.

+-------------+-------------+-------------+-------------+-------------+-------------+
|             | PA          |      K%     |     BB%     |    Whiff%   |  CU Whiff%  |
+-------------+-------------+-------------+-------------+-------------+-------------+
| Innings 1-4 |     1601    |     21%     |    12%      |    23%      |    44%      |
+-------------+-------------+-------------+-------------+-------------+-------------+
| Innings 5+  |     841     |     16%     |    11%      |    18%      |    33%      |
+-------------+-------------+-------------+-------------+-------------+-------------+

With this in mind, I’d like to see the Yankees try out Burnett as a one-inning guy for the rest of the year.  At this point, the Yankees have to think of any way possible to squeeze some value out of Burnett, and they have two basic options: keep pitching him every fifth day and hope he figures it out, or make a change. Given that that the Yankees already have five competent starting pitchers, Burnett’s stuff might play well in relief, and a move to the bullpen should be given serious thought.