Analysts: Tough to determine if there is such a thing as clutch

[snopes]

Isn't Derek Jeter one of baseball's top clutch performers, the guy who comes through for the New York Yankees when, as so many Yankees fans believe, Alex Rodriguez doesn't? Mr. November to Jackson's Mr. October?

Maybe, if clutch players exist, an issue debated by traditional baseball minds and statistical analysts.

http://www.usatoday.com/sports/baseb...9-clutch_N.htm

[pob14]

Looking into players' eyes might be as good a way as any to find clutch players.

Now THERE's a way to speed up the games. Don't bother to play them; just look in the players' eyes and see who the clutchiest ones are, and give them the win.

[RealityChuck]

This has always been an issue by Sabermatricians. Bill James has recently changed his mind from "it doesn't exist" to "it's possible."

The problem is defining what "clutch hitting" is. The statistics analyzed assumed a definition of clutch that is not what most people mean when they say it -- Late innnings (even if the game is way out of reach), for instance.

Also, the sabermetricians all agree that there is such a thing as clutch pitching (though their definition also doesn't match what people say when they refer to it). For it to exist for pitchers and not batters seems unlikely.

[pob14]

Originally Posted by RealityChuck

This has always been an issue by Sabermatricians. Bill James has recently changed his mind from "it doesn't exist" to "it's possible."

The problem is defining what "clutch hitting" is. The statistics analyzed assumed a definition of clutch that is not what most people mean when they say it -- Late innnings (even if the game is way out of reach), for instance.

Also, the sabermetricians all agree that there is such a thing as clutch pitching (though their definition also doesn't match what people say when they refer to it). For it to exist for pitchers and not batters seems unlikely.

I don't recall seeing anything like that; what sabermetricians have studied "clutch pitching"?

And James has said on his website (I won't quote, because it's a subscription service) that the evidence says that, whatever clutch performance may be out there, it can't possibly be large enough to make a (say) 80 RBI man better than an 85 RBI man. He does say that the issue deserve study rather than outright rejection, but he pretty much says that about everything. He's less irascible than when he was younger.

[Jay Temple]

Originally Posted by RealityChuck

The problem is defining what "clutch hitting" is. The statistics analyzed assumed a definition of clutch that is not what most people mean when they say it -- Late innnings (even if the game is way out of reach), for instance.

I'd refine it a little to a phrase I heard earlier this year, or maybe last: "late and close". I know at ESPN.com you can filter batting average to only those at-bats in innings 7-9. I'd define it as 7th inning, 2 out through the end of the game, including extra innings. For close, I'd take anything from the batter representing the tying run through his team already leading by no more than two.

Also, the sabermetricians all agree that there is such a thing as clutch pitching (though their definition also doesn't match what people say when they refer to it). For it to exist for pitchers and not batters seems unlikely.

Unlikely? It would have to exist. A clutch batter would be one who fares better than average against pitchers who are in clutch situations.

[Johnny Slick]

Baseball's weird like this. If there is such a thing as clutch, it shows up like 20 times a year per player at the most. If clutch was measured by LIPS situations, stats for those periods wouldn't look so random (good hitters tend to hit better in LIPS than bad ones, of course). I am on the fence about clutch. It's mainly something you can look back and point to rather than having any predictive value. Kirk Gibson and Joe Carter were clutch in 1988 and 1993 for sure.

Interestingly, I think basketball has a much, much better case for clutch existing. In the last minutes of close games, you need guys who can either create their own shot in the face of double-teams or get off a quick jumper off of a screen. Those are specific skill sets one can point to that add up to clutch in a way I don't see any specific skill sets for baseball (fastball hitting?) . But any basketball fan worth his salt can tell you that Horry and Bibby are clutchmonsters and the stats that are out there prove it.

[Mateus]

What they need to do is pick a reliable statistician, sit him/her down, and give him/her a two week deadline to figure it out. Base answers accordingly.

[videoguy]

Originally Posted by Johnny Slick

Baseball's weird like this. If there is such a thing as clutch, it shows up like 20 times a year per player at the most. If clutch was measured by LIPS situations, stats for those periods wouldn't look so random (good hitters tend to hit better in LIPS than bad ones, of course). I am on the fence about clutch. It's mainly something you can look back and point to rather than having any predictive value. Kirk Gibson and Joe Carter were clutch in 1988 and 1993 for sure.

I think you have a very different definition of "clutch" than most people. Gibson and Carter hit big homeruns in the World Series, but one at-bat doesn't tell you anything about how clutch a player is. It just means they got *a* hit in *a* clutch situation. Kirk Gibson, by the way, hit just .154 in the 1988 NLCS, and was 1 for 1 in the World Series.

[Johnny Slick]

Right. My point is that study after study can't find that "clutch" that people think exists. My "definition" would be #2 if it was in the OED: a person who produces clutch hits. It's circular and completely useless in determining actions or probabilities in advance, but it's a method one can use without stirring up too much controversy.

As to what skill sets might induce clutch ability, I was thinking...

- Bad Ball Hitters: Vlad Guerrero has a career OPS of 967 in LIPS. Is that a coincidence? He's really good in non-LIPS, to be sure, but if I were thinking of a skill set that equals clutch, I would think of the ability to make good contact with pitches that are not designed for you to make good contact with.

On the other hand, I would guess that guys who wait for their pitch and don't go after anything until they get it would wilt a little bit. Barry Bonds is the best example I can think of this, but my own 2001 Mariners were a team built around selectivity and who "choked", as were the Moneyball A's. There's probably an easy way to study this instead of just naming teams, though.

- Fastball Hitters: The thing is about this one is that it's counter-intuitive in a way. Getting around on a grade-A fastball is, more than anything else, a function of bat speed, which is one of those abilities that young players have in spades and vets lose as they get older. Yet, you rarely hear a youngster has clutch. It's nearly always a grizzled old vet.

- Speed: I'm not sure whether this adds to clutch or detracts from it. On the one hand, a fast player who puts the ball in play requires an infield to be perfect or nearly so to get him out. Ichiro is just plain awesome at slapping the ball and capitalizing on small mistakes to get on base. On the other hand, in real clutch situations teams are going to be putting their best men in the field, so this would be mitigated, I think. On top of that, BABIP and bad-ball hitting are very closely linked, so counting speed may be counting another skill twice.

- Intangibles: The whole point of intangibles is that you can't measure them. Still, stuff like sign-stealing (see: Bobby Thomson's shot off of Ralph Branca in 1951) or getting into a hitter's/pitcher's head can make a huge difference if trotted out specifically for big-game situations.