Clutchiness

In response to Dejan Kovacevic's clutch study, Charlie argues that the Pirates are wrong and Dave Pinto follows Jason Bay and argues that it's not clutchiness that exists but chokiness.

There's no doubt that the situation affects a player's ability to hit. With all respect to Charlie, I am not persuaded by something like career batting averages and postseason batting averages. The hitting environment is just not stable, and the data is far too incomplete (making sample sizes much too small) to say with any authority that the numbers prove anything in the debate about "clutch." It could be coincidence that Joe Rudi's postseason numbers delivered in the clutch match his overall regular-season numbers. Joe Rudi does not dent the argument either way.

Think about the variance in the hitting environment from at-bat to at-bat. Some players are affected by defensive alignments. Sometimes you'll see a shift on, for example, and both the shortstop and the second baseman will be on the right side of the infield. That affects a player's ability to get on base in the at-bat. Or maybe some lucky genius had the right-fielder cheat just far enough toward the line that a usual double is caught for an out. These situations are not easily sorted from the rest of the player's at-bats. And there is no "true" or consistent "regular" hitting environment. There is enough variance from at-bat to at-bat to make each one unique and each one its own too-small sample size--when, that is, we get into debates about one of the endless variety of at-bat subsets.

So it's hard for us to say what a player's "true" clutch hitting ability is--it depends, in ways we cannot articulate with the current stats made available, say, from ESPN or USAToday, on the kinds of situations in which they bat. I am not saying that season-long, month-long, or career-long averages are meaningless. I am saying they are too messy and complicated to neatly subdivide into statistically authoritative arguments about subsets of at-bats.

And not all "close and late" situations are the same. There are games where both teams look tired and appear to be going through the motions, and there are games where the playoffs are on the line and the stadium is packed and the crowd noise is intimidating.

The data is not complete enough to say much of anything decisively at this point.

Proof for me that hitting is heavily influenced by the immediate hitting environment are the abysmal numbers most players post in pinch-hitting situations. Very few players are good pinch-hitters, as I discovered, to my surprise, a few years ago. If hitting ability was truly some kind of ideal constant, we would not see a difference between PH numbers and everyday play numbers. This, for me, is proof that something like "clutch" surely exists--but what that is would depend on how it's defined and even then, the numbers are not likely to be very helpful in distinguishing one player's clutchiness from another's. Fine distinctions may be impossible to quantify.

Further proof that clutch exists is anecdotal. Even Charlie concedes it in Elway and Jordan. But for me, this anecdotal evidence carries as much or more weight than the kind of statistical evidence we have at hand today. I have no disdain for statistics, mind you. Rather it's a serious respect for their present limitations. Because I have thought so long and hard about them, I am of the opinion that they do not tell as much as we wish. So I expect the study of baseball to be one that is further characterized by innovations in approach and further scholarly explanations of how everything you previously thought was all wrong. (For example.) Though I do think that some sabermetricians adopt a tiresome "Mr. Enlightenment" tone toward the observation of anecdotal evidence, I am no bleeding-heart Romantic arguing for the conviction of reason-transcending intuition. My belief in clutch is not an article of faith, either.

My advice to the Pirates would run along the lines Adam LaRoche advances in the piece. You certainly can't depend on some perception of clutch ability. The season is too long. Not every game will have a playoff atmosphere. It's just not possible.

If the Pirates want to score more runs, they have to get on base more often. Any way they can get on base is a good way to get on base. As I've said before, I'm not so frustrated when they load the bases and fail to score--as they have too often done this season. The best solution to such trouble "in the red zone" is practice in the red zone. If they keep loading the bases, they will score more runs.

So I believe in clutch and I believe that people choke, but I would not trust in clutch. I would trust more in consistency. I'd rather have a #2 hitter who can draw a walk in his sleep than a #2 hitter who appears to have an ability to deliver a hit in a clutch situation. Why? Because the first player will have far more opportunities to exercise his skill than the second player. As a result, the team would score more runs. And I'd also be skeptical that we can be certain that this or that player truly possesses more clutchiness than the others.

And one more thing on the lineup. I'd stick with what works. If the Pirates win big against mainly right-handed pitching on Monday and face a right-hander on Tuesday, I'd run that same lineup out there again. If they can win often with Jack Wilson in the #2 spot, then don't fix what's not broken. The standings, however, suggest that Jim Tracy would do well to keep tinkering with the lineup until he finds a consistent winner.