Using Pitchfx to analyze this stuff will probably lead to something like a Schrodinger’s equation, but the research does point to individual pitcher variability in BABIP, so that throughout their careers, some pitchers, like Tim Hudson, will have consistently lower BABIP than the .300 norm, yet still have sizeable variability in other performance metrics.

“That means if you get just one extra flare a week – just one – a gorp… you get a groundball, you get a groundball with eyes… you get a dying quail, just one more dying quail a week… and you’re in Yankee Stadium.”

When JH is victimized in almost this exact manner by bad luck this year, he will certainly feel he was unlucky…and he will be right.

Here’s what I wonder: Would JH be willing to agree that hitters who faced him last summer were unlucky. After all, they did not benefit from the usual number of dying quails, gorps, etc. I’m guessing he might, although he clearly feels the defense behind him is impenetrable.

For pitchers, over the course of hundreds of innings, this difference should even out, which is what Voros McCracken found. Is Jeremy Hellickson’s career .230 BABIP over 225 innings a Small Sample Size Theater production or a real, repeatable skill? Stay tuned.

I should have recognized his name in your post.

If some pitchers do have some control over results, then we need a better model.

We do, but unfortunately this is the best we can do with the information at hand. It was far easier for Voros, 10+ years ago, to look at the basic information available, and be peer reviewed, and have people unable to disprove his theory.

Now with the proliferation of pitch f/x, and those who have access to hit f/x, we are able to come to more refined conclusions than before, and end up with articles like Mike’s that I posted above [unfortunately Mike was an excel wiz and pitch f/x guru who was hired away from BP to the Astros, so he won't be writing any more articles like that =\].

However, too many people look at BABIP as some magical predictive stat and make claims like: Pitcher X’s BABIP was far below normal, he’s going to regress the following year and be worse. Or Pitcher Y’s BABIP was far above normal, he’s going to regress the following year and be far better.

Either way, people are looking at the BABIP solely and trying to draw conclusions based on it, rather than look at the underlying causes. Does the individual have a usually lower/higher BABIP than normal? Does the pitcher generate a ton of groundballs/popups? Did the pitcher change his repertoire? All of these could have direct effects on BABIP not just regression.

This really is a stupid stat

Why is it a stupid stat?

Always appreciate your responses.

but he’s probably right

Except he’s sort of not. 10 years ago Voros McCracken showed, via DIPS Theory, that most pitchers really don’t have much control. One case in McCracken’s point is the back to back absurd seasons Pedro Martinez put up in ’99 and ’00. Look at his peripherals:

’99 – 13.2 K/9, 1.56 B/9, 1.74 ERA, .323 BABIP
’00 – 11.8 K/9, 1.33 BB/9, 2.39 ERA, .236 BABIP

Both of those are the extremes (full season stats) of his career. After Voros, Tom Tango refined DIPS into FIP which fangraphs uses.

Now, the latest research shows that certain pitches have more control over the outcome than others (Rivera for instance). However, it’s not that some pitchers have 0 control and those, like Rivera, have 100% control. The pitchers who tend to produce weak contact tend to have lower than expected BABIPs. For a more indepth look, here’s Mike Fast’s Baseball Prospectus article from 11/22/11:

http://www.baseballprospectus.com/article.php?articleid=15562

His conclusion:
The reasons that balls fall for hits are complex. We saw in the previous article that pitchers have significant control over how hard batted balls are hit. We saw here that the speed of the batted ball interacts with the vertical launch angle to determine whether the ball is likely to fall for hit….

Further research is also needed to better understand and quantify what pitchers and batters are doing to create weakly-hit or solidly-hit batted balls. There are many potential avenues of investigation. Certainly, factors such as pitch location, type, and speed could be important. Batter stances, swing planes, and swing speeds could also be important factors, though they are more difficult to measure. Ultimately, a model of the ball-bat collision utilizing data about the incoming pitch trajectory and outgoing batted-ball trajectory could be the most useful explanatory tool.

This doesn’t mean that Hellickson will be worse this year than last year however, because it doesn’t take into account a normal growth curve for a pitcher his age.

I think his bigger problem is that he walked way too many hitters given that he doesn’t strike enough out. Jose Molina will help him get more strike threes, though. As good as any other catcher in framing the ball.

Now if my FIP was a run lower then my ERA, I would believe in it big time.

I don’t claim to be much into the statistics of it; not sure I would be that interested in anything but the results. But BABIP is one I have more problems buying into without some data and explanations.

http://www.bradenton.com/2012/03/05/3917408/tampa-rays-pitcher-jeremy-hellickson.html