If you missed yesterday's Volume 1, I loaded all of the PitchFX data, via Brooks Baseball, for Ross Ohlendorf from the Yankees in 2008 and the Pirates in 2009 into an Excel spreadsheet, made some charts, did some rudimentary statistics, and decided that his fastball has definitely lost about 3 mph this year, while gaining a small amount of vertical break (about an inch) and losing a larger amount of horizontal break (about three inches) without an appreciable change in strike percentage.
This only starts to tell the story on Ohlendorf, though. He's defintely been a better pitcher for the Pirates this year than I expected and he's outpitched his projections so far. How, exactly, is he doing this with a fastball that's not as good as the one he's thrown in the past? That's what I set out to do today, so follow along after the jump for more charts and technical jargon as I attempt to unravel the mystery of Ohlendorf.
The first off-speed pitch of Ohlendorf's that I really wondered about was his slider. It seems, anecdotally, to me that on nights when Ohlendorf is on, he mixes the slider in very well and catches a lot of hitters off guard, while on the nights that he's off, he leans much more heavily on the fastball.
Like the fastballs yesterday, I've charted the sliders with the Yankees in blue and the Pirates in black, vertical break on the y-axis and horizontal break on the x. I'll again post the SOSH guide chart for reference.
It's pretty easy to see that his slider has been much more like a slider this year, breaking harder away from righties. The numbers support this: the average H-break on his slider in 2008 was -1.4 while this year it's 0.2. That indicates about an inch and a quarter more break away from right-handed hitters. Like his fastball, the velocity on his slider has dropped off about 3 mph. His average slider with the Yankees clocked at 83.1, while with the Pirates it's clocked at 80.4. This speed is intriguing to me; normally a slider is tough to hit because it's close in speed to a fastball and looks similar out of the pitcher's hand, only to break away (from a righty throwing to a righty) at the last second. Ohlendorf's slider clocks in almost ten miles per hour slower than his fastball, which seems, to me at least, to make it more of a curveball that breaks like a slider. This will be important to remember later on in the post as I round towards my conclusion.
The next logical step for me was to check and see how often he's throwing his slider. In 2008 out of the Yankees bullpen, Ohlendorf threw 397 fastballs out of 606 pitches (this total isn't exactly accurate because of the vagaries of pitch classification; for ease on my part I'm only couting fastballs that PitchFX counts as four-seamers, which probably make up 95% of Ohlendorf's fastballs), or 65.5%. With the Pirates this year, 750 of his 1546 pitches have been fastballs, which is only 48%. He's clearly mixing things up a bit this year.
To my surprise, it's not because he's throwing the slider more. In fact, he's throwing it considerably less. I counted 164 sliders for Ohlendorf with the Yankees in 2008, which is a little more than 27% of the 606 total pitches. This year, he's only got 280 sliders in 1546 pitches, which is just a hair over 18%. He is throwing more of them for strikes, though. Only 56.7% of his Yankee sliders went for strikes, while this year they're crossing the plate at a 60.1% rate.
So I guess we can maybe assume that Ohlendorf is throwing a better slider, even if he's not throwing it more often. I'm hesitant to make a stronger statement because the contact rate on his slider is up this year (about 14% from about 8%), but it's much harder to quantify outs and hits, from the raw data, which means this could just be happening because he's throwing more strikes and inducing more outs with it. Whatever the case, it's clear that a big change for Ohlendorf this year is the addition of a changeup to his repertoire. His slider and fastball made up more 90% of his pitches in 2008 with the Yanks. This year, they only comprise about 70%.
In 2008, Ohlendorf threw 14 changeups in 606 pitches. This year, he's thrown 292 changeups in 1546 pitches. He's essentially gone from a two-pitch pitcher in 2008 with the Yankees to a three-pitch pitcher in 2009 with the Pirates. There's no use comparing his year-to-year changeups, but we can compare it to his other pitches to try and see how effective it's been.
You can see that for the most part, despite some probably misclassified pitches, his changeup breaks exactly the same way his fastball does. The average speed of these changeups is 82.6 mph and he's thrown it for a strike almost exactly 60% of the time.
Now, keeping in mind that we're leaving about 10% of Ohlendorf's pitches from this year on the table (PitchFX's pitch classification isn't perfect, as the chart above illustrates, and some fastballs get called two-seamers or sinkers, some sliders get called curveballs, etc.; here I've done the best I can to look at as many pitches as possible), I've got some guesses as to what's causing his velocity to drop, besides his permanent move to the rotation. Out of the hand, the fastball, slider, and changeup should theoretically look very, very similar because the strength in both of the offspeed pitches is that they look like a fastball, but either don't move like it, or move much slower.
Since that's the case, let's compare his release points on all five pitch types (fastball and slider with the Yankees, fastball, slider, and change with the Bucs). This graph is a bit crowded, so I'd definitely recommend right clicking and viewing at full size on another page.
You can see that last year there was a clear delineation between where his fastball and slider were released. There is this year, too, but the changeup obscures that. It also looks like Kerrigan (or someone) moved him over to the other side of the rubber and got him raise his release point just a bit.
When Ohlendorf is on, I've heard more than one opposing announcing crew mention how he throws three pitches from basically the exact same deliver. That makes it hard know whether the pitch will be a 90 mph fastball, an 82 mph change that looks just like the fastball, or the 80 mph slider that will suddenly dive away from a right-handed hitter. I'm guessing that what Ohlendorf worked on the most this year was disguising those pitches and tweaked his mechanics a bit (hence the release point changes from last year to this year) to accomplish that. In the process, he's lost a bit of gas on his fastball. This isn't in any way a bad thing; the changeup makes him a far more complete pitcher than he was before he got to Pittsburgh.
I don't have the energy right now to go and compare pitch selection with start outcome, but as I went through his starts to pull in the data I noticed that in a lot of his outings in May and June he was relying much more heavily on his fastball than in other starts. This was about the time of year he was struggling the most. From watching him this year, it certainly seems to me that the nights that he's had the most success (save Sunday's bad outing against the Marlins), he's mixed all three pitches effectively and used them to keep hitters off balance. If he can do that more consistently, he doesn't need a 96 mph fastball.