Table of Contents

Tuesday, September 25, 2007

Could Jared Burton be another Mariano Rivera? And other PITCHf/x stories...

John Walsh was incredibly gracious and this morning sent over PITCHf/x plots like those I profiled here on six additional Cincinnati Reds. I requested plots on guys that I felt are "key" members of next season's pitching staff, in that if they do well the Reds might be contenders, and if they don't ... well... we'll likely see yet another losing season. As the title of this post suggests, he found some interesting stuff on one Jared Burton, but I'll get to him further down the page.

Let's start with perhaps the most important unanswered question of next season:

Homer Bailey
To re-orient you to these plots, they are drawn from the perspective of the catcher and show the break of pitches relative to a theoretical ball with zero spin thrown in a vacuum. No information on pitch location can be gleaned from this plot. Units are in inches.
We see from this figure three basic pitches. First, in the upper left, is his fastball, which has excellent "rise" and is consistently in the "pink" 90-95 mph range (with a few blue 95+ mph pitches for good measure). We also see his change-up, which is typically coming in about 10 mph slower than the fastball and accounts for the red dots in that upper-left cluster. The change doesn't quite have the vertical "rise" than the fastball does, and might be something that Bailey could improve upon.

Lastly, we see his big curveball, which has excellent negative vertical break, more so than any other Cincinnati Red I've seen to date. It also comes in a good 15 mph slower than his fastball. It would seem to me that a hard rising fastball and a big breaking curve ball that differ this much in velocity would make for a pretty challenging combination. One thing to note, however, is that none of Bailey's pitches seem to break much in the horizontal plane.

If he can be a solid #3 or better, the Reds would seem to me to be contenders without only incremental improvements from some of the other young players. Last week's start in San Francisco was a step in that direction. But if Bailey can't harness his control, the Reds' rotation will likely be in a world of hurt.

David Weathers
Weathers was extremely solid through July, but has struggled since then. I've been predicting a decline from this guy for a long time, and it might be that his peripherals are finally catching up to him. Nevertheless, we might be able to see in this plot part of what has allowed him to be effective.
Weathers is reportedly a sinker/slider type of pitcher. Sinkers have typically been very difficult to pick up in these sorts of plots, but essentially they are fastballs that don't "rise" as much as a traditional 4-seam fastball. With Weathers, we see that his fastball (presumably all the dark green pitches) is extremely variable in how it breaks. Some have terrific vertical rise (nearly 15 inches), while others only rise five inches compared to a default pitch. It'd be neat to run a cluster analysis on his pitches to see if he's throwing distinctly different pitches in that blob 'o fastballs.

We also see his slider, depicted in the red blob. Like a lot of sliders, it actually follows a path fairly similar to what you'd expect from a spinless ball thrown in a vacuum, and comes in ~5 mph slower than his fastball. There might also be a fourth off-speed pitch in there--maybe a variation on a slider?--but it's a pitch he doesn't show very often.

Jared Burton
Ok, enough teasing--let's look at Jared Burton. Here's what John Walsh wrote when he sent over these figures:
Burton has a very unusual profile. Look how symmetric his plot is -- I can't even tell if he's lefty or righty. Ok, so he's righty -- that means that pink blob with horiz>0 seems to be a cutter -- compare to Rivera's plot in my article. Is he known to throw a cutter, as far as you know? It certainly seems to be the pitch he throws most often (from this plot).
Let's look at that plot, along side a plot of Mariano Rivera:
As Tom Tango pointed out in response to Walsh's article, Rivera's profile actually looks more like a lefthander than a righthanded pitcher, explaining at least in part why he defies the typical platoon split and is so tough on left-handed hitters (0.606 OPS vs. right, 0.526 OPS vs. left).

Interestingly, Burton's plot has a lot of similarities with Rivera's, especially in terms of the pink blob on the right side of the graph. Looking back on the original Cinci Post article on Burton's acquisition, the Reds' brass did cite his cutter as being a good pitch. He throws more fastballs (the pink blob on the left) than Rivera does, and for some reason has a bit more variation in velocity (hard to tell if a splitter and/or a slider might also be present), but sifting through Walsh's PDF's from the original article, there aren't a lot of other pitchers that are comparable to these guys. As always, I'd love to hear feedback from those of you who can actually evaluate pitches with your eyes...

Burton hasn't faced enough hitters to make evaluation of his left/right splits particularly meaningful, but currently he does show a rather small left/right split (0.527 vs. right, 0.544 vs. left), though it's not reversed like Rivera's. Burton is already 26, so it's hard to project a long bullpen career from the guy...but on the other hand, Rivera didn't break in with the Yankees until he was 25 years old either...

Let me be clear: the realistic chances of Burton turning into the second coming of Mariano Rivera are remote. But these data do indicate that Burton has a pitch that is at least similar in profile to Rivera's devastating cutter, which can't help but bode well for the young reliever...assuming he can control it. If Weathers falters next season, Burton is the obvious choice right now to step in and try to fill that role.

Bray, Coffey, and Majewski
I'll do the last three with one figure:
Bray and Coffey look pretty similar to one another (though Bray's profile is reversed because he is left-handed). Hard throwers with a fastball/slider combination. Bray has a bit more "rise" on his fastball, and has thrown far more breaking pitches than Coffey. There was talk that Coffey was to stop throwing his splitter, and that's supported by this figure: splitters usually show up as a slightly slower pitch directly "below" the fastball in these plots, and there's nothing fitting that description here.

Finally, Majewski was best known--at least when he was effective--for his ability to induce ground balls via his sinker. Majewski's fastball plot looks a lot like Weathers' and Arroyo's, so that's probably the sinker we're seeing there. He also seems to be mixing in a few variations on a slider, which shows up in the 70-80, 80-85, and 85-90 mph ranges. It's possible that the slowest pitch is a curve ball (it has negative vertical break), though I haven't heard that he throws one.

I continue to be really interested in what this new set of statistics can tell us. For the time being, this stuff is basically just descriptive. There's not much here that can really inform us about why a pitcher is effective and/or ineffective. But I think that with additional work, perhaps by comparing the same pitcher when they are effective and ineffective, these data can help us to quantitatively describe changes in the actual pitches thrown and connect those changes to differences in performance on the field. Exciting stuff.

Once again, a huge thank you to John Walsh for spending the time and energy to create the above figures!

Ok, now to get back to my profile on the Cleveland Indians.


  1. That's cool stuff. Thanks to John for letting Justin at your data. Letting the wider community at your information really helps to expand upon the system of knowledge.

    And I'm going to start telling everyone I know that Burton is the next Mariano Rivera. Don't worry, I'll give you credit...unless it turns out to be true. :)

  2. Letting the wider community at your information really helps to expand upon the system of knowledge.

    That's one of the best things about John's original post--that he made plots of 150 pitchers for us to look at! :) The fact that he was willing to do a few more was just awesome.

    And I'm going to start telling everyone I know that Burton is the next Mariano Rivera.

    It's nice to dream, eh? I keep getting this wistful little smirk on my face whenever I let myself think about having a decade with a 2.00 ERA closer...

  3. As far as the Homer Bailey stuff.... I am guessing that doesn't include last nights start right?

    I would be a lot more interested in seeing what his stuff was doing before June 26th and since he returned, in his healthy starts.

    Just to the eye, he has looked incredible the past two starts.

  4. Doug, I don't think this includes either of his last two starts, though I didn't confirm that. But there's only one visible 95+ pitch on that plot, and he was regularly breaking 95mph in the start I listened to last week (didn't listen to last night's game yet!!).

    Maybe if I get some time to sit down and figure out how to extract this stuff in the offseason I'll be able to look at the before and after starts to see if anything but velocity is different. I'd predict, for example, that he's improved the consistency of his release point via more consistent mechanics now that his groin has healed....

  5. This might be a dumb question, but how does negative horizontal break occur and how does one read that when looking at the pf/x data?
    Understanding that Homer is a RHP, and the data are from a right hand batter.

    I cannot find a synopsis on how this is entered into the scatter plot data.

  6. Hi Taylor,

    I'd recommend that you read John Walsh's tutorial on these plots, which I mentioned in this post.

    Basically, negative horizontal break would be a ball that, from a right-handed pitcher, tails back in on the hands of a right-handed pitcher. Most fastballs do this, for whatever reason. In these plots, which are done from the perspective of the catcher, you'd see balls plotted to the right side of the zero line (again, we're not plotting location, we're plotting direction of movement relative to a theoretical pitch with no spin thrown in a vacuum).

    Homer doesn't have much horizontal movement on any of his pitches, at least as depicted in these plots. Tons of vertical movement though.

  7. Taylor, negative horizontal break means that the RHP put sidespin on the ball when he threw the pitch. This can happen if he drops down from a perfectly overhand 12 o'clock delivery or if he applies pressure or spin to the ball toward the third base side either with his wrist or fingers as he releases the pitch.

    A four-seam fastball gets this kind of sidespin that produces negative horizontal break because the pitcher drops down toward a 1 o'clock delivery.

    A two-seam "sinking" fastball gets more of this kind of sidespin because the pitcher also applies pressure to the ball with his middle finger as he releases the pitch. There are probably dozens of variations on how to throw the two-seamer, but that gives you the basic picture.

    A standard changeup has negative horizontal break for similar reasons as the four-seam fastball, but some pitchers also purposely apply some sidespin to the ball.

    Those are the three pitches that you will usually see with negative horizontal break.

  8. Thanks for your help guys. I think I am figuring this out now.

  9. Big thanks to Mike for his post. That really helped me ground some of this stuff in reality. -j