Tuesday, January 31, 2012

Is There Hope For A.J. In 2012?

(It's the dreaded return of A.J. Two-Face)

After the additions of Michael Pineda and Hiroki Kuroda, it seemed like a forgone conclusion that A.J. Burnett's days as a starting pitcher for the New York Yankees were over.  It had already been reported that Cash shopped him around at the Winter Meetings while offering to eat $8 million of his remaining contract, and that was before the Yankees were flush with the rotation depth that they now have.  Once Pineda and Kuroda were in the fold, it seemed obvious that Ninja Cash already had the next move lined up to rid himself of one of his biggest mistakes.  And yet, we found ourselves still talking about A.J. Burnett as a current member of the Yankees, which leaves the frightening door of opportunity open for him to somehow be the 5th starter in 2012.  It is a door that many of us do not want to see A.J. walk through, but we may have no choice if Cash can't find a deal to move A.J. that works for him and the team.  That being the case, we might as well prepare ourselves for what A.J. has to offer, which, given his performance up to this date, won't be much.  While most of A.J.'s 2011 stats could be used to create optimism for a better 2012, there are some trends within the numbers that make me think 2012 and beyond will be more of the same.

The biggest problem with A.J. is that his value is derived almost entirely from the quantity of innings he has pitched, not the quality of the innings (save for Game 2 of the '09 World Series).  Innings eaters are a valuable asset to have in the rotation, but for $82 million most people would expect better than the back end production A.J. has put up in those innings, and justifiably so.  His 5.26/4.83 ERA/FIP split in 2010 didn't pass the eye or the smell test, and neither did 2011's 5.14/4.77 follow up.  A.J. has been consistently erratic and inconsistently effective, marred by bouts of problems with his mechanics, inability to repeat his delivery, and little or no command of either of his two pitches.  He has tried to combat this with the attempt to develop a changeup, but mid-30s is hardly the right time for a pitcher with command problems to attempt to master a new pitch.

In order to be a successful two-pitch starter, a pitcher better have a damn good fastball, and for the early part of A.J.'s career he has.  But over the last handful of years A.J. has seen his fastball velocity steadily decrease, from 95.9 MPH in 2007 with Toronto (the earliest year PITCHf/x tracked it on FanGraphs) to 94.4 in '09, 93.1 in 2010, and 92.7 this past season.  As the velocity has decreased, the value of A.J.'s fastball has plummeted with it, both by standard pitch value measures (-14.1 in '09, -16.2 in '10, -34.0 in '11) and as measured by PITCHf/x (-12.4, -10.1, -28.1).  It's also no secret that the majority of A.J.'s fastballs tend to be located in the middle to upper part of the strike zone (see heat map below for an example), which is not a good place to live when you're losing your heat on the pitch.  Even when paired with a curve that was surprisingly above average in 2011 (8.9 rating standard/10.7 PITCHf/x), a low-90s fastball with little movement that consistently hits the heart of the plate is going to do more harm than good for a two-pitch pitcher, and it certainly hasn't done A.J. any favors.

 (Courtesy of FanGraphs)

When a pitcher's throwing a flat fastball with decreasing velocity down the meaty portion of the plate, the expected result would be a lot of contact.  Surprisingly enough, this actually wasn't the case for A.J. in 2011, as his contact rate dropped from 81.6% in 2010 to 76.5% in 2011.  His in-zone and out-of zone contact rates, 54.3% and 90.4% respectively, also decreased from their rates of 2010.  An explanation could be that hitters were laying off more of A.J.'s offerings that they didn't like, especially his curveball when he couldn't locate it, but his 2011 swing rates don't support that theory.  In direct contrast to his contact rates, A.J. generated more swings (43.8%) and more out-of-zone swings (30.3%) than he did in 2010, and his Swinging Strike rate of 10.0% was a high for him as a Yankee.

These trends last season certainly help to explain A.J.'s elevated K/9 total (8.18), but they should also lead to a better ERA and FIP result.  A pitcher generating more swings, more swings and misses, less contact, and striking out more batters while walking fewer than he did the year before should result in much more than the marginal improvements in ERA and FIP that A.J. experienced in 2011.  The answer for why that didn't happen lies in the contact A.J. is generating, specifically the quality of contact, and the insight that leads into what might be happening when batters face A.J.

Along with his contact and swing rates that trended in the right direction last year, A.J. also experienced the type of GB and FB rate trends in 2011 that a pitcher in Yankee Stadium would want to see.  His GB Rate increased from 44.6% in 2010 to 49.2% in 2011 while his FB Rate decreased from 37.5% in 2010 to 32.3% in 2011.  One rate that didn't trend the positive way last season, however, was A.J.'s LD rate, which at 18.5% represented a high for him as a Yankee.  And the rate that lingers above all this discussion like a bad odor in the room is A.J.'s 17.0% HR rate from last year.  A rate that high is typically associated with bad luck, and A.J.'s 3.86 xFIP last season would agree with that.  But given everything else that's been discussed in this post and how it all adds up to A.J.'s subpar performance, I think A.J.'s HR rate is not only the smoking gun in the case of his crappy 2011, but also the sign of the overall trend that has plagued him over his 3 years in pinstripes and will continue to do so moving forward.

As A.J.'s GB/FB rates have been trending in the right direction over the last 3 years, his HR rates have continually increased, from 10.8% in 2009 to 11.6% in 2010 to the aforementioned 17.0% last season.  This trend, combined with A.J.'s declining fastball value and his uptick in line drives in 2011, says to me that hitters, while making less contact off of A.J. last year, generally got more out of the contact they did make.  And when that contact is being made on either a fastball that's elevated in the zone or a poorly-located curve that grooves over the heart of the plate, hitters are more than likely going to get more bang for their buck.  Since A.J. has been on this general trend of missing up and losing velocity for a while, hitters could now be familiar with it and could be scheduling their approach against A.J. accordingly.  It's not so much the two nasty curveballs that they swing and miss at in the at-bat that matter any more; it's the fastball A.J. grooves with 2 strikes that they're squaring up on and driving for power.  That's what these numbers suggest that hitters are doing on A.J. now, squaring up on him and driving his misses all over (and out of) the park  And that's how a pitcher who the majority of the peripherals say isn't pitching that bad can end up having such bad results.

Of course this is just my interpretation of the numbers, but this explanation is really the only one that makes sense to me in explaining A.J.'s past two seasons and predicting his future.  Sure, he might be generating more swings in general and less contact in general, with the majority of that contact being the type that typically helps a pitcher succeed.  But when he's still making the same mistakes up in the zone with his rapidly-declining fastball, and he's still missing his location enough with his curveball that he can get in trouble when he does, the results of those specific pitches and those specific at-bats are what carry the weight in determining A.J.'s outcomes as opposed to the entire general landscape of an outing.  It's almost as if hitters aren't up there looking for a particular pitch to hit so much as they're up there waiting for the one or two mistakes by A.J. that they know they can destroy.  If that's the case, then peripherals be damned, it's going to be another long year in 2012 for A.J. whether he's a Yankee or not.

At some point in his career, every pitcher is going to lose the juice on his heater.  It's what he does to combat that loss in velocity that determines how successful he can be moving forward.  In A.J.'s case, the decline of his fastball over the past three seasons has basically crippled his ability to be an effective starter anymore.  Even in a year when almost all of his peripherals suggested he should have had a better season, A.J.'s 2011 ended up almost a carbon copy of his 2010.  That decrease in his fastball velocity combined with the continued increase in HR rates points to A.J. no longer being able to fool hitters with his stuff, and as a pitcher with little to no consistent command, that doesn't give reason for there to be a whole lot of hope for him moving forward.


Larry Koestler said...

"It's not so much the two nasty curveballs that they swing and miss at in the at-bat that matter any more; it's the fastball A.J. grooves with 2 strikes that they're squaring up on and driving for power."

Having done a fair amount of research on Burnett myself -- not to mention a post dedicated to his splits -- I was curious to see whether this statement was actually true. It turns out that A.J.'s problem isn't necessarily a two-strike issue -- his tOPS+ with two strikes last year was 36, while his tOPS+ while ahead in the count was 16 -- but, as you noted, it does appear to be some sort of sequencing issue, as his tOPS+ in counts favorable to the hitter was a ridiculous 208. So it seems like his problem isn't so much screwing up when ahead, but digging himself an even deeper hole when falling behind.

Unknown said...

Great find, Larry.

Truth be told, I never even thought to dig that deep and see if there was a trend to be found in those numbers. I wrote the "with 2 strikes" thing off the top of my head as a random example, but it's good to see that there are some trends to support the idea that hitters are murdering A.J.'s mistakes.

Pitchers always make mistakes. But when those mistakes become more common and you stop being able to get away with those mistakes, that's when you get in real trouble as a pitcher and that's what seems to be the case with A.J. now.

Larry Koestler said...

I'm actually doing a follow-up on this for my Friday post at RAB; thanks for the inspiration!