EJax in the stretch

by Brickhaus on Aug 29, 2008 1:22 PM EDT Comment 7 comments

I'm cross-posting this from Driveline Mechanics to see if people have an opinion on this.

I just want to get people's opinions on this. I've seen people bring up the theory that Jackson's mechanics are actually better out of the stretch than they are out of the windup, which is part of the reason why Jackson has outperformed his component ratios. His OPSA with the bases clear is .833, while his OPSA with men on is .667, leading to an unusually high strand rate. This is primarily because he's allowing fewer hits with men on base and because his walk rate is lower with men on (discussed more below).

The way I see it, there are four possible explanations here. The first is the sample size argument, and that it's just luck. The second is that the defense is in better position for Jackson's pitches when men are on base. The third is that he's actually better at pitching out of the stretch. The fourth is that he's just more mentally focused with men on base, and is better able to locate his pitches.

I don't think the second argument holds much water. Here's the splits for the Rays starters:

Kazmir - .763 with the bases empty, .608 with men on (but he has a higher K rate, a lower walk rate and much lower HR rate with men on, so it should be lower).
Shields - .668/.807. Most of the difference is because he allows a lot more doubles with men on.
Garza - .622 /.740. He seems to be discernably worse at pitching out of the stretch, as his walk rate increases significantly and his K rate drops to about 3 per 9 innings with men on.
Jackson - .833/.667, as discussed above
Sonnanstine - .758/.786

If it was because of the defense, you'd probably see a pattern here, but there really isn't one.

I'm not sure the fourth argument accounts for much of it either. If he was focusing more with men on, you'd see a pattern more like Kazmir's, where there's a dramatic change in the items the pitcher 'controls' because they're defense-independent (walks, strikeouts and home runs). Rather, his BAA drops dramatically from .251 to .206. His K rate increases slightly, from 12.3% to 15%. His walk rate decreases from 10.9% to 7.4%. His home run rate is almost identical. So while his component ratios are slightly better and probably accounts for a little of the reason why he's better with men on, they're not nearly better enough to account for the full difference.

That primarily leaves luck, or better mechanics (leading to better movement on pitches). Any thoughts?

0 recs | Comment 7 comments

Story-email Email Printer Print

Comments

BTW

If someone knows how to make one, I’d be interested to see a spray chart of where hitters are hitting the ball when men are on versus when the bases are empty.

Vogt early, Vogt often.

by Brickhaus on Aug 29, 2008 1:23 PM EDT reply actions 0 recs

I think you got your splits mixed up...

Kazmir – .763 (empty), .608 (men on)
Shields – .668 (?), .807(?).
Garza – .622 (?) .740 (?)
Jackson – .667 (men on), .833 (empty)
Sonnanstine – .758 (?), .786 (?)

by Vin on Aug 29, 2008 1:32 PM EDT reply actions 0 recs

Oops

Only Jackson was flipped around. The rest of them are (empty)/(men on)

Vogt early, Vogt often.

by Brickhaus on Aug 29, 2008 1:36 PM EDT up reply actions 0 recs

Fixed

Forgot you could edit the main post.

Vogt early, Vogt often.

by Brickhaus on Aug 29, 2008 1:37 PM EDT up reply actions 0 recs

SSS

This is a very interesting post/topic. But I’m inclined to attribute most (all?) of the difference between situations with runners on base and situations with none on to random variation.

We have to keep in mind that, for all pitchers, even one full season’s worth of stats is a small sample size. A guy could have the exact same ability for two seasons, and post very different ERAs. Justin Verlander comes to mind (look at his fundamental stats for this year and last, then notice the different ERAs). Jake Westbrook from a few years ago also comes to mind – he was extremely consistent for three years, but posted three very different ERAs.

Thus, situations within a season are even smaller sample sizes, subject to random variation that could easily be mistaken for skill (or lack thereof). It is noteworthy that Jackson’s underlying stats (BBs, Ks) are slightly improved when runners are aboard, but he’d have to do this consistently, over the course of several years, for this to truly be a skill. And if Jackson were to prove to be skillful in pitching with runners on base, he would be one of the very, very, very few pitchers who demonstrated this skill (I don’t know of any others that come to mind, actually. Maybe Tom Glavine?).

It’s possible that these splits are representative of a difference in skill in different situations. But it’s EXTREMELY unlikely.

by Peter Bendix on Aug 29, 2008 1:46 PM EDT reply actions 0 recs

Also

The only stuff I know about mechanics is what I read from people like Will Carroll, but I don’t know if better mechanics necessarily leads to more movement on pitches.

by Peter Bendix on Aug 29, 2008 1:53 PM EDT up reply actions 0 recs