Vicente Padilla has been outstanding so far in 2008. This may be my attempt to jinx him, but he has been effective at the “bend, don’t break” strategy the Rangers have been working on (personally I think it should be “don’t bend or break”, because that leads to fewer collapses like the ones we’ve had the last few days in Boston). Today, I want to look at what he was doing last year and compare it to this year, using Pitchf/x data.
Through mid-June 2007 he was 3-8, 6.69, and went on the DL. He came back for eight starts in August and September and was 3-2, 3.86. When I was checking these numbers, it surprised me, I did not think he had pitched that well in the second half. Combining that with his early numbers in 2008, and it strongly lends itself to the idea that he was injured for the first half of 2007, which caused his bad numbers.
After four starts in 2007 he was 0-3 with a 6.00 ERA, and stinking up the place. His last four starts in 2007 were 2-1, 2.12. This year, with four starts down, he is currently sitting at 2-1 with a 3.12 ERA. I decided to use these three group of four starts to look at what he has changed in that time, to get clues as to his improvement. His pitch counts for the groups ranged from 284 to 335, so they are all in a similar quantity which should help us note differences visually.
Note that I wrote about the Rangers rotation back in June 07, just before he went on the DL, and those charts include the first half of last year’s data for him. The data I am working with now is the same, but utilizes several techniques that Pitchf/x researchers have developed more recently. In particular, the spin charts shown were first used (I think) by Alan Nathan and then Mike Fast, and it was a spreadsheet posted by Mike (created by Tangotiger, I believe) that contained the spin chart that I have adjusted for my use here. Much credit to all of them for their work on this.
First, this polar plot maps the spin speed vs the spin angle. The angle is shown simply by it’s rotation around the circle. The speed is shown as distance from the center. In this case (and the others in this blog post) the center of the circle is 50mph, and it increases by 10mph at each step away from the center, until you get to the outer ring which is 100mph.
In my review last year, linked above, I said that Padilla was showing a fastball, curve, and slider, although the results were so mixed that it was hard to tell what was what. Reportedly he has a changeup as well, but I could not pick it out from this data. In fact, here you can see I have split the fastballs into 2 and 4 seam versions, that is something I was not able to do until working with the polar plot and some other tools.
Not only can you see much more distinction between the groups of pitches, you see the numbers have clearly changed. The groupings are much tighter than before, and the sliders (red) are clearly differentiated from the fastballs. The one thing I do have trouble with here is the curve, it appears that there are two distinct groups, suggesting one is a different pitch. None of the other charts I have show this difference, so I have left it all as a curve, but at two different speeds (the group closest to the center is around 60mph, the other is between 70 and 80 mph, which is kind of fast for a curve so is more likely to be a different pitch).
Here you see a chart much more similar to the late 2007 than early 2007 one. If anything, the groupings are even more compressed this time around, and the distinction in the curve has pretty much disappeared. Of course, the curve has almost disappeared too.
So, let’s look at the differences:
Much tighter groupings of pitches as time goes on. The suggestion I would have for this would be mastery of his pitches – as he has learned to throw each one, he’s gotten better at it and is more consistent. This would be more believable if he hadn’t been pitching in the majors for ten years – if he was 22 or 23, I might find this a convincing argument.
Ratio of pitch types has changed considerably. The slider has stayed almost constant at 12-14& of pitches, the fastball (both kinds) has grown from 73% to 81%, and the curve has disappeared, from 13 and 14% last year to 5% in 2008. The ratio of 4-seam to 2-seam fastballs has stayed roughly 2-1 in favor of the 4-seam, although late last year he threw more 2-seamers than in the other time periods.
Fastball speed is down. In late 07 and early 08, his top speed was 96.5mph, second best was 95.8. He beat 95.8mph 31 times in early 07, peaking at 98.6. This drop of peak speed is about 2mph, which curiously is not reflected in the overall average for the fastballs, which stayed just about the same for all groups, in the 91-92 range. He had a much bigger spread for the fastballs, but the average remained the same. Again, this leans toward someone learning how to throw more consistently.
The slider was more focused around the 80mph mark, instead of being spread around in the mid 80s. Like the fastball, he got better when he threw more consistently.
If I give you this link, you can go back and check out his release points for the first half of 2007. It contains a very ugly image, but the gist is that he was releasing over a very wide area, well over a foot square, and on a slope from bottom-left to top-right.
You see here a much cleaner release area. About 3/4 of a foot wide by a little over a half a foot tall. It is also very circular in pattern, compared to the angled release last year.
My conclusion is simple: Padilla was hurt early last year, and after a couple of months on the DL in the middle of the year came back and pitched much better at the end, and much better to begin 2008. He is releasing the pitch in a smaller area, suggesting he is not trying to compensate for where it hurts to move his arm. He is producing much clearer patterns of pitches, suggesting he has more confidence in what he is throwing, and better ability to throw it how and where he wants.
All-in-all, the much-maligned (especially by me) Vicente Padilla is probably a poster child for pitching healthy. I remember a lot of talk about how he wanted to pitch even when he was hurt, because he felt that was what he needed to do to be part of the team. This study shows that when pitching hurt, he (and presumably other pitchers) are not as effective, to the extent that they are hurting their team as much as themselves. Repairing the damage is a better option than pitching through it. As I recall, there were a number of complaints from his teammates about how slow he was at pitching, which disrupted their own rhythms of being in the game. Could it simply have been a case of him having to step off the mound after each pitch for long enough that his arm would stop hurting enough so he could throw again?
I’m still not convinced about Padilla’s ability, because every time he pitches I’m still expecting the roof to fall in at any moment. It will take a while for me to lose that fear, if ever. But at least when things are going wrong for him these days, I have some hope that it’s explainable, that he can work through it, and not just because he’s injured. This use of Pitchf/x was very helpful to me to understand why something happened, and not just accepting my own opinion that Padilla sucks.
Coming soon, I’m going to give the same treatment to Millwood, who has shown a surprisingly similar pattern to Padilla from last year to this. It will be interesting to see if we can discern something that explains his struggles and improvement as well as we could for Padilla.