The success and failure of A.J. Burnett’s repertoire on a month-by-month basis

A lot is riding on that right arm of yours this season, Mr. Burnett (Photo c/o Baseball Canadiana)

After reading Moshe’s post analyzing Ken Rosenthal’s piece comparing A.J. Burnett to Josh Beckett the other day, I was inspired to take a deeper look at some of A.J.’s PitchFX data to see if there were any observations that might portend a better 2011 for Burnett. While every projection system across the board has A.J. outperforming his wretched 2010, no system is going to project a matching 5.26 ERA the very next year for a pitcher with Burnett’s track record, and so it seemed like it’d be useful to see what specifically A.J. might look to do to right himself in 2011.

Before looking at the data I’ve collected, I’d also like to take a moment to point out the comprehensive and detailed Burnett pitch analysis Lucas Apostoleris at Beyond the Boxscore published last month. Ideally this piece would function as something of a sister post to Lucas’. I would also be remiss if I didn’t point out our pal Jason’s excellent Burnett piece from last month, in all its heat-map goodness.

Anyway, here’s a chart featuring key data from the last three seasons on A.J.’s fastball and curveball, his two primary pitches, broken down by month. All PitchFX data is courtesy of Joe Lefkowitz’s wonderful PitchFX site:

I realize there’s a lot of data here, so I’ll try to make this as easy as possible.

In 2008, Burnett posted the best season of his career in terms of fWAR (5.5). His average fastball (94.5mph) broke about six inches inside on righthanders and rose nearly eight inches. Of the three seasons shown in the above table, A.J. generated both his highest whiff% against righthanders in 2008, at 20%, and against lefthanders, at 9%.

In 2009, his fastball (avg. vel. 94.2mph) broke even further inside on righthanders, with -7 inches of horizontal break, but stayed roughly the same vertically. His whiff% v. RHB fell to 13%, and against LHB to 7%. However, as we know A.J. was still very effective that season, even if he did lose 2.1 fWAR.

Now here’s where things get somewhat interesting. In 2010, clearly the worst season of his career, his average fastball lost about one mph, and was also considerably closer to the plate than it had been, averaging -5 inches of horizontal break — or two inches less than in 2009 — and also rising nearly an inch higher, averaging 8.6 inches of vertical break. However, his whiff% against RHB actually improved, to 16%, while the rate stayed the exact same against LHB.

Of course, none of this tells the full story. Per Fangraphs, A.J.’s wFB the last three seasons are as follows: -5.9, -14.4 (7th-worst among qualified MLB starters) and -14.3 (9th-worst). So while Burnett was busy getting torched in 2010, his fastball was essentially exactly the same as it had been in 2009. Meticulous readers will ascertain that part of this can probably be explained by Burnett’s .295 BABIP in 2009 compared to his career-high .319 in 2010; although without having access to BABIP breakdowns by pitch, I can’t tell you just how many more balls fell for hits on the fastball in 2010 than in 2009, but based on what we know I think we can assume a decent amount.

The other side of the Burnett conundrum, of course, is his vaunted curveball. In 2008, it was the 4th-most valuable in baseball, at 13 runs above average. In 2009, it was once again the 4th-best in baseball, and was even more effective, at 16 runs above average. Of course, in 2010, he posted the worst wCB value of his career, at -3.9 runs above average.

What happened to this one-time dominating pitch? In 2008, A.J.’s curve (81.2mph) averaged 5.96 inches of H-break and -6.72 inches of V-break. In 2009, both of these numbers fell by nearly an inch each, to a 4.96 average H-break and -5.97 average V-break. And last season, the average H-break shrank almost another inch to -4.13, though the V-break remained relatively flat, falling 0.37 inches.

So with A.J.’s average curveball catching more of the plate than it ever had before, batters were missing it less, as A.J.’s Whiff% vs. RHB plummeted from 47% in 2009 to 41% in 2010, and vs. LHB it fell from 50% to 39%. Less whiffs of course not only resulted in more balls in play, but more effectively hit balls in play due to an easier-than-ever curveball to square up on, hence the career-high BABIP.

If you look at A.J.’s monthly breakdown, it’s clear that his curve ideally needs to break closer to 5 inches off the plate and also drop at least -5.5 inches to return to ruthless effectiveness. In June and August — his two worst months in 2010 — he had decent H-break in the former (4.91) but it broke too low (-6.18); while in the latter month he had absolutely nothing, with an average H-break of 3.45 and V-break of -4.17, unsurprisingly resulting in much lower Whiff rates (32% vs. both RHB and LHB) that were way out of line with his recent seasonal averages. A.J. may as well have been putting his curve on a tee in June (.351 BABIP) and August (.344).

Ultimately, the fact that Burnett actually hasn’t generated a positive wFB since 2007 is troubling in and of itself, but that his curveball control completely abandoned him more or less explains why he was absolutely creamolished in 2010.

However, I tend (or at least try) to be an optimist when it comes to improved performance from players I like, and while I have no idea why I’ve always been an A.J. Burnett fan, and so I not only agree with the projection systems, but I think he can be even better than the ~4.50 ERA most systems have him pegged for. The early returns out of camp have been positive, and while the usual spring training/small sample size caveats must apply to any discussion involving Burnett in March, it’s nonetheless encouraging to hear that he’s both feeling and performing better. Hopefully this portends a string bounceback year for A.J. Burnett.

Follow Me On Twitter