I love sabermetrics. But there are some things that bother me. Just because the right tool to measure something doesn't exist doesn't actually mean the thing itself doesn't exist. Just because we don't have a bathroom scale that goes over 300 pounds doesn't mean that nobody weighs over 300 pounds.

Likewise, the fact that no sabermetric measure shows that there is such a thing as a clutch hitter proves that....sabermetrics is deficient.

Likewise, having watched baseball for 40 years I can tell you that Alex Gonzalez' fielding is way way way above average. If the numbers say otherwise, it proves that the numbers aren't worth looking at.

Am I just a zealot? Well I know what I know and what I see. There is no way there is no such thing as clutch in sports. Researchers have shown that exists with experiments in which people putt with varying pressure: some improve; some collapse.

Why not with athletes? Is there any question that Tom Brady and Joe Montana and Dennis Johnson and Dwane Wade and Michael Jordan have some ability to keep their wits about them that Peyton Manning and Rick Ankiel and Phil Mickelson don't?

But it doesn't happen with hitting? If that IS so it would be one of the great psychological mysteries .

Half the problem, pvm, is that it's hard to settle on a definition of clutch, as it means different things to different people.

Check out this thread, which looks at Close and Late.

interesting

Let me get this straight. You assume that a thing such as clutch exists. And because a statistical attempt to prove it exists finds none then the statistics just aren't looking at the right thing?

Is that the gist of it?

Or is it possible that it is a natural human reaction to want to glorify a great performance? Is it human nature to not attribute anything to luck?

I'm sure lots of craps players will tell you that they are clutch when it comes to throwing a 7, yet I think you will agree that it is highly improbable that such a skill exists (in a correctly run craps game).

But we all know that craps is purely random. Hitting a baseball --- partly random, partly skill. It certainly seems intuitive that "clutch" would exist. But Steve Brady is right; how does one even define "clutch," much less measure it?

Anyway, it is a misuse of statistics to say that simply because nobody's found a way to measure something, then it doesn't exist. You could say that the data does not support the existence of clutch, but that's a different thing from proving that clutch hitting does not exist. It's much harder to prove that something doesn't exist than to prove that it does exist.

But of course, it is also true that the mind finds coincidences and tries to assemble them into something that makes sense, even when what we really have is just coincidence or small sample sizes. We forget the cases where our theory isn't supported and remember the cases that support them.

And conventional wisdom can change. Randy Johnson was supposedly bad in the postseason until the 2001 WS. Barry Bonds was supposedly a postseason bust until 2002.

Of course, if clutch does exist, one might gain clutchness with maturity.

What we do know is, if there are clutch hitters, David Ortiz is one.