PDA

View Full Version : Mr. Clutch



RedsManRick
12-09-2010, 04:52 PM
Fangraphs just rolled out a splits feature on their leaderboards.

In "High Leverage" situations, Joey Votto had the 2nd best wOBA (.542) in baseball -- 152 qualified players. Nelson Cruz was #1.

#152? Rajai Davis. #151? That would be Derek Jeter. ARod was 8th.

If "clutch" exists as a repeatable skill and not just a descriptor of circumstance, it sure as heck manifests itself in weird ways...

camisadelgolf
12-09-2010, 05:01 PM
That is in no way sustainable. Then again, I was saying it for the entire second half of this season, and time and time again, Votto produced. I'd normally tell Reds fans to get ready for a drop-off, but I first need to be convinced that Votto is human.

TheNext44
12-09-2010, 05:32 PM
I wonder if we looked at clutch stats the same way we look at defensive stats, looking at three years of them instead of year to year, if they would be more predictive. I have no idea, but it would be interesting.

RedsManRick
12-09-2010, 05:52 PM
I wonder if we looked at clutch stats the same way we look at defensive stats, looking at three years of them instead of year to year, if they would be more predictive. I have no idea, but it would be interesting.

There have been numerous studies that look at big samples (full careers) and the bottom-line take away is this:

The distribution of "clutch" performance* looks exactly the same as we would expect the distribution to look if "clutch" was completely random. That doesn't mean we don't observe "clutch" players over large samples, but rather that there's so few of them who had statistically significant differences that we can't say with any confidence that there is a real skill.

*defined as performance measured by OPS in clutch situations relative to performance in non-clutch situations

It's like if you had 1,000 people flipping coins 1,000 times and 3 people ended more than 550 heads and 3 people with less. For any given person, you'd expect 500 heads. But if you look at enough people, you're going to find a few who racked up a lot more purely by luck. Just because a few guys showed a big difference from what we would expect if there was no such skill doesn't mean that there actually is a skill. Random variation produces a few outliers. And the clutch studies turned up as many outliers as we would expect if only random variation were at play.

Now, we can't draw a conclusion in the other way either. Those studies don't disprove clutch. They just say that if it exists, we can't know when we're seeing it and not just random variation.

Brutus
12-09-2010, 07:05 PM
There have been numerous studies that look at big samples (full careers) and the bottom-line take away is this:

The distribution of "clutch" performance* looks exactly the same as we would expect the distribution to look if "clutch" was completely random. That doesn't mean we don't observe "clutch" players over large samples, but rather that there's so few of them who had statistically significant differences that we can't say with any confidence that there is a real skill.

*defined as performance measured by OPS in clutch situations relative to performance in non-clutch situations

It's like if you had 1,000 people flipping coins 1,000 times and 3 people ended more than 550 heads and 3 people with less. For any given person, you'd expect 500 heads. But if you look at enough people, you're going to find a few who racked up a lot more purely by luck. Just because a few guys showed a big difference from what we would expect if there was no such skill doesn't mean that there actually is a skill. Random variation produces a few outliers. And the clutch studies turned up as many outliers as we would expect if only random variation were at play.

Now, we can't draw a conclusion in the other way either. Those studies don't disprove clutch. They just say that if it exists, we can't know when we're seeing it and not just random variation.

The issue, really, is how to define it. That's the crux of the problem.

But I take a Justice Potter Stewart approach to clutch. In the famous 1964 landmark Supreme Court case regarding pornography where he said, "I shall not today attempt further to define the kinds of material I understand to be embraced...and perhaps I could never succeed in intelligibly doing so. But I know it when I see it."

Redhook
12-09-2010, 07:33 PM
In certain sports, clutch can be the difference between a very good player and a great one. The best rise to the occasion more often than not when the game/match/tournament is on the line.

blumj
12-09-2010, 08:01 PM
In certain sports, clutch can be the difference between a very good player and a great one. The best rise to the occasion more often than not when the game/match/tournament is on the line.

I'm not much of an expert on any sports, but you'd think it would be completely different for sports with time limits, the game changes strategically, one team is trying to do what they usually do faster than they usually have to do it, while the other team is trying to waste time.

RedsManRick
12-09-2010, 08:27 PM
In certain sports, clutch can be the difference between a very good player and a great one. The best rise to the occasion more often than not when the game/match/tournament is on the line.

But do they rise to the occasion because they have some extra talent called clutch beyond their general ability or because they're the most talented guys on the field/court?

Name me some famous clutch players -- guys who did it on a regular basis (not just 1 or 2 memorable ones). How many guys we label clutch are otherwise just mediocre or good?

It seems to me that the guys who are reliably clutch and tend to "rise to the occasion" are the guys who rise to every occasion, the guys who are simply the best, period.

Anyways, I didn't mean to go down this road. It's certainly well traveled elsewhere on RedsZone. Rather I wanted to highlight just how influential labels and narrative can be. Ask most fans who's more clutch and they still talk about Jeter. And really, that's how fans experience the game. Fans don't really care if Jeter is truly clutch by some definition -- all they care about is the ability to label him as such, experience his exploits through that frame and have it justified by results on occasion.

<rant>
What gets people like me frustrated is when casual fans (and most announcers) treat their narrative as fact. They want to take an experiential description and use it as proof make analytical statements such as "so and so is clutch", clearly implying that the guy is more likely to come through in clutch situations in the future -- and using it as a wedge against some other player who may have failed in a clutch situation in the past. It just feels good to have knowledge, to have understanding. It's fun to have an experience like watching a guy come through in the clutch and build an emotional bond on that. That sense of knowledge is a powerful emotion. And when the guy comes up again, it makes us feel good to call on that emotional bond ("he's my guy") and to project it back on to the player ("dude is clutch... watch this"). And when he does come through, that bond is reinforced many times over. And when he doesn't, we sort of it ignore it and discount it as a data point in our mental calculus.

But statements of fact can be verified. People like me who crave verifiable "truth" want to see more than selective, circumstantial evidence. Most people don't. Most people see that as work and maybe even counter to the emotional experience they're having watching the game.

I love the narrative. The narrative is what makes the game enjoyable. But it really gets under my skin when people confuse narrative with real knowledge -- and especially when they turn around and use their supposed knowledge to belittle those who would call them out on the weakness of their "proof". If a person wants to believe Jeter has a talent called clutch, I can't and won't stop them from doing so. But if that person is attached to a belief based solely on subjective personal experience and intuition and is unwilling to used an objective standard, we'll just going to have to agree to disagree on our respective conclusions.</rant>

kaldaniels
12-09-2010, 08:56 PM
Good points Rick...do you think there is such a thing as being anti-clutch...that is, do some players consistently wilt under pressure?

kpresidente
12-09-2010, 09:51 PM
Good question, kal. I'd like to know that. It seems a lot more reasonable to think some guys just cave in pressure situations than the alternative. Reasoning goes that players are always performing pretty much at the top of their game, so there's not much room for them to "turn it up a notch" in clutch situations, yet there's a lot of room for them to crumble. So the mental aspect might have more effect going in that direction.

kpresidente
12-09-2010, 10:01 PM
I'm not much of an expert on any sports, but you'd think it would be completely different for sports with time limits, the game changes strategically, one team is trying to do what they usually do faster than they usually have to do it, while the other team is trying to waste time.

Some games are a lot more about effort, and a lot less about skill. Being a good defensive tackle is not the same thing as hitting a 95 mph fastball. Heart plays a big role in the former, the latter is mostly muscle memory.

RFS62
12-09-2010, 10:05 PM
BCubb defined it beautifully, a few years ago in a discussion....

"Clutch is the absence of choke"

Roy Tucker
12-10-2010, 12:13 AM
I think good hitters are good in the clutch because they are good hitters. Just like lousy hitters are lousy in the clutch because they suck as hitters.

Just call me genius.

TheNext44
12-10-2010, 01:16 AM
The issue, really, is how to define it. That's the crux of the problem.



Exactly. Bill James, a few years ago, devised a formula for ranking the importance of each AB by effect it had on the outcome of the game and the importance of the game itself in the standings. Late in the season games between Cards and Reds last season were more "clutch" than early season games between the Pirates and the Nationals.

He found that there were players, using that definition of clutch, that were able to repeat higher or lower levels of performance year to year, and that the number of those players was much larger than what it would be if only luck were a factor.

Now, to be fair, James never released the details of that formula, and he was working for the Red Sox at the time, and his work showed that David Ortiz was the best clutch hitter in the game. So I'm not sure how valid it was, but it was an interesting take on the subject.

RANDY IN INDY
12-10-2010, 02:11 AM
BCubb defined it beautifully, a few years ago in a discussion....

"Clutch is the absence of choke"

:)

Ron Madden
12-10-2010, 03:42 AM
I think good hitters are good in the clutch because they are good hitters. Just like lousy hitters are lousy in the clutch because they suck as hitters.

Just call me genius.

You are a Genius.

IslandRed
12-10-2010, 12:30 PM
In regards to the "everyone can handle pressure if they reach the big leagues" argument... I don't recall which RedsZoner said it -- I'm too lazy to search the archives at the moment -- but, paraphrasing, he wrote "saying a guy doesn't have a problem with pressure if he reaches the majors is like saying he doesn't have a problem with talent." There's a bar to clear, but that still doesn't make everyone the same.

As for me, I don't think clutch exists in the sense of guys playing better; as mentioned before, baseball is a skill sport and an extra shot of adrenaline usually doesn't do much good. But choke definitely exists. We've seen plenty of it in every other sport under the sun, so I don't know why baseball would be exempt.