Where Have A.J. Burnett’s Strikeouts Gone?

Following Wednesday night’s disastrous start (3.1 innings pitched, two homers, and more walks than strikeouts) A.J. Burnett has now faced 382 batters on the season. Two hundred batters faced is the amount Pizza Cutter established as the threshold for strikeouts/plate appearances to become a reliable metric during a single season. That’s worth noting because Burnett’s strikeout percentage sits at 16.5%. Over the last three seasons, he struck out 23.7% of batters faced; over the last five that number is 23%; for his career it’s 21.7%. As it stands, the only season in which Burnett struck out batters at a lower rate than he is currently was way back in 2000. That year, he struck out 15.7% as a 23-year-old

When Burnett signed with the Yankees, the expectation was that his numbers could actually improve. He was battle tested in the rough American League East. More important than battled tested, though, he was successful. His xFIP had ranged from 3.55 to 3.64 over his three-year stint with the Jays, and a move to the Yankees meant he would no longer have to face a lineup consisting of multiple Hall of Famers on numerous occasions each season. That hasn’t been the case, though.

It was probably unrealistic to expect Burnett to improve on his whiff rate, which was over 10% in 2008, but that figure declined to 8.2% in 2009, and sits at 7.1% in 2010. A once raging fastball, sitting in the upper 90s on some occasions, has retreated in velocity. Pitchfx has its average velocity sitting just about 93 miles per hour, with empty swings occurring under 6% of the time. The only pitch Burnett is actually having batters miss often is his knuckle curve (12.5%). Compare that to last season and the only noticeable changes have nothing at all to do with his velocity being down. He’s actually getting more whiffs on the fastball and fewer on the curve.

That would suggest the decline is not entirely due to his velocity, which leaves predictability and location as the other potential culprits. The curve becomes Burnett’s go-to pitch when he reaches two strikes. With the exception of 3-2 counts, Burnett tosses his bender more than 50% of the time on each count. This doesn’t differ too much from 2009, when he threw it 60+% of the time on 0-2 and 1-2 and 48% on 2-2. This usage pattern actually resembles his 2008 pattern more than anything.

This would lead to a question of whether it’s Burnett’s location of the curveball that’s problematic. In 2009, the pitch was swung at under 40% of the time and resulted in a strike nearly 60% of the time. In 2010, however, the swing rate has remained static, while the strike rate has dipped to a little over 51%. That’s not a radical shift, mind you, but a big enough one that it alters Burnett’s ability to succeed. If batters are able to hold off on the inevitable avalanche of curveballs that dive below the zone on two strike counts, that means Burnett will have to beat them with his fastball or sinker, a combination of which hasn’t resulted in a positive run value since 2007.

The most troubling statistic associated with Burnett is that 40% of his strikeouts have come against the Orioles and Indians. Those starts account for a little more than 20% of his starts this season.




Print This Post





20 Responses to “Where Have A.J. Burnett’s Strikeouts Gone?”

You can follow any responses to this entry through the RSS 2.0 feed.
  1. Dirty Water says:

    “The most troubling statistic associated with Burnett is that 40% of his strikeouts have come against the Orioles and Indians”

    Isn’t it also true that CC has only beaten the Orioles? So, essentially, the Yanks are relying on Pettitte, Hughes and Vasquez to win important series?

    Oh, oh. Someone better go inform Kevin. Looks very ugly.

    Vote -1 Vote +1

    • Zack says:

      He’s picked 4 of his 7of Ws against the O’s, but we all know by now that W/L records are flawed. He’s had good games against the Phillies, Red Sox, A’s, Rangers, and Rays. And then 3 games where he’s given up 5ER, 1 game where he gave up 6ER, and 1 game he was pulled at 4.2 IP because of rain delay.

      Vote -1 Vote +1

      • Dirty Water says:

        True, but don’t forget this is FG where you’re posting. CC’s K rate has gone down 3 straight years, and that means he’s godawful, the end is near.

        Vote -1 Vote +1

    • frank says:

      This is purely anecdotal from watching him pitch last year… but he seemed to throw more curveballs for strikes early in counts last year (I have no idea if the stats back that up – I’m not talking just # of curveballs but curveballs in the zones).

      If he’s missing with the curveball early in counts, it’s easier for hitters to lay off it when behind or even in the count (1-2, 2-2). If he shows a team he can command it for strikes early in the count, then when he throws it with 2 strikes batters have to give it a bit more respect.

      He also seems to be throwing more 2 strike curves way out of the strike zone (bouncing them to the point where hitters don’t see the pitch as potentially a fastball strike out of his hand). He’s at his best when the catcher is catching that low curve just barely above the ground (especially to lefties) or trying to do a backdoor curve over the outside of the curve.

      Vote -1 Vote +1

  2. delv says:

    He has fewer Ks because he has thrown the 2-seamer more, inducing more contact, and his velocity is down from last year.

    Vote -1 Vote +1

    • The Tom says:

      pitching to contact is a myth!!!

      Vote -1 Vote +1

      • frank says:

        The previous poster said he was inducing more contact because he’s throwing more 2 seamers… you of course twist the logic to say that is a pitch to contact myth… but he didn’t say he was intentionally pitching to contact.

        Here’s a crazy thought..- maybe he just has better command of his 2 seamer…which leads to him throwing it more…which has the byproduct of more contact.

        Maybe he’s under the “myth” of trying to get ahead in counts by throwing strikes (and he’s throwing the pitch that he’s more likely to get a strike with) or avoiding walks by throwing a pitch he can command when behind in counts. Maybe is not a conscious decision to pitch to contact but an attempt to throw more strikes which just so happens to result in more contact.

        Vote -1 Vote +1

    • vivaelpujols says:

      His two seamer is his best pitch! Same velocity as is his fourseamer and with amazing tailing movement. Why would increased usage of that result in less strikeouts?

      Vote -1 Vote +1

      • aweb says:

        Because strikeouts are not simply a matter of velocity+movement. Location matters too. The four-seamer is probably a better pitch to combine with his curveball; perhaps they look more similar on the approach to the plate. I would think a good two-seamer, often down in the zone, would not be an ideal pitch to combine with a curveball. If a pitch starts at the bottom of the strikezone, and the pitcher is throwing a lot of curves and two-seamers, it’s easier to let it go, while a four-seamer might stay in the strikezone.

        Hard two-seamers/sinkers do not lead, necessarily, to a lot of strikeouts. I would imagine this is due to the physics of batting – it’s very difficult to swing higher up than you started as opposed to lower down.

        Also, aren’t batters trying to hit the ball slightly below centre? This would give a larger margin of error (if contact is the only goal) on pitches moving down, as opposed to staying up.

        Vote -1 Vote +1

      • Daern says:

        Strikeouts are quantifiable, therefore it is “fewer strikeouts”, not less.

        On a less grammar-nazi note: The pitch may be better, but looking at his numbers that isn’t reflected. While his fastball hasn’t been a positive-value pitch since ’07, it is pretty damn bad this year, already worth -5.8 after being worth -14.4 all of last year. Not only that, but he’s throwing it a lot more, 71.6% of the time compared to 65.9% last year, an increase of almost 6%. Unless his two-seamer accounts for the extra 0.3% of unclassified pitches he thrown this year (0.8% compared to 0.5% last year) then all I’m seeing is he’s relying on his fastball more and it is doing worse.

        Vote -1 Vote +1

    • Matt says:

      Of course with the way pitches are identified being constantly tweaked, I wouldn’t be so certain that he’s actually throwing his 2 seamer more often.

      Vote -1 Vote +1

  3. Roger says:

    I dont know about his two seamers or unclassified pitches or positive value pitches I just know he is not very good right now and all the excuses or evaluations in the world cant fix that.

    Vote -1 Vote +1

  4. niiiiiiigggggaaaa says:

    he has nipple rings niggas

    Vote -1 Vote +1

  5. Kyle says:

    He hasn’t had an ERA under 4 going on 3 years now. His best season(other than 2002 where he struck out a guy per inning and had his lowest WHIP and ERA), it was over 4(where he won 18 games.) Strikeouts do matter when the guy is known for walking a lot of guys, and also giving up homeruns.

    It’s been 7 years after he had Tommy John surgery, and usually a pitcher goes into decline around that time. A guy who is known for being unhittable at times, but will walk 6 guys in a game and throw 2 wild pitches is going to rely on k’s far more than pitchers like Maddux or Halladay.

    The Yankees looked at his strikeouts and wins in 2008, and paid Burnett like an ace. The guy has had one shutout in the past 5 seasons(06), and has never made an allstar team.

    Vote -1 Vote +1

  6. frank says:

    “The most troubling statistic associated with Burnett is that 40% of his strikeouts have come against the Orioles and Indians. Those starts account for a little more than 20% of his starts this season.”

    PROBLEM 1 (basic math skills)
    – AJ has 4 starts against the Orioles and the Indians (out of 14 total)
    – That would be 28.6% … this is a “liitle” more than 20%? Since you like to round let’s say a little less than 30% ( I assume you counted 3 starts instead of 4 and rounded 3/14 down to 20%?)

    Would 40% of his strikeouts in 30% of his starts seem troubling? (most would call this variation when you look at a t-test or try to put any confidence levels on this size of a data set)

    PROBLEM 2: (basic reasoning skills)
    Would his % of strikeouts be correlated to % starts? At a very crude level wouldn’t innings be far better? (and on a better level batters faced?)

    – Innings against BAL/CLE = 30, total innings = 87.1
    – % of innings against 34.4%

    So is it still “troublesome” that AJ has accounted for 40% of his strikeouts in ‘just’ 34.4% of his innings? Would this still be significant in your view?

    Look the math error is one thing – it happens… but the statement 40% of his strikeouts in 20% of his starts is not a “statistic”, it is an OBSERVATION and an observation that is not grounded with any basic statistics (like a confidence level to determine if the observation is meaningful). Can you correct the error at a minimum, or better yet not try to loosely correlate % starts to % strikeouts, or better yet understand that even if the #’s were 21.4% and 40% is still might not be meaningful unless you put a confidence level on it to account for sample size?

    This is an interesting article but like some other recent ones… the sensationalist observations (2-2 fastball rate with Blanton) are being used as statistically significant without bothering to check if that really is the case.

    Vote -1 Vote +1

  7. Milan Borge says:

    Awesome topic, I concur with u blogs should have comments on as blogs are presenting a view of the person and after getting talks from people, he can get a better version of the point of view it has presented.

    Vote -1 Vote +1

  8. Lisa says:

    It doesn’t matter A.J. is the best! He does his best every single game, I have faith in him.

    Vote -1 Vote +1