Clutch vs non-clutch

[User Name]

We can't agree on a definition for a reason.

[Slasher9]

No, a fact is something definable and demonstrable.

 

You can't just show up and say "This is what it is because I said so". At least make a compelling point, like Kimmi.

 

i did. in my other posts. so now i am just stating the fact that Clutch exists.

[Slasher9]

The argument as to whether a player is clutch or not has been defined to be binary - either a player is clutch or he isn't. Unfortunately baseball doesn't work that way. Regardless of the situation the failure rate is much higher than the success rate. David Ortiz, "the greatest clutch hitter in Red Sox history" didn't come through every time - and yet he's being known as "clutch". The year Ted Williams hit .406 he allegedly passed on the chance to sit out the last game and assure himself of that .400+ BA. Do you think there wasn't any pressure there for him to get hits in that game? His response was clutch.

 

Using the "Pass/Fail" method to determine clutch is patently unfair in that it doesn't account for normalcy. If a player fails just once in a clutch situation statisticians will (rightfully) use that one AB as evidence that the player isn't clutch by the statisticians definition of clutch. "If he were truly clutch he would have gotten a hit that other time. Since he didn't get that hit he's not clutch". That's hogwash.

 

Even "Late and close" doesn't define clutch since many clutch situations happen when a game isn't "late and close". In fact, a clutch hit early in the game can often times be the reason a game isn't close late.

 

Some things aren't statistically quantifiable because they don't take into account unknown variables like a player's emotional/mental state at the time. Some things just ARE and should be accepted as such.

 

i'm currently giving you a virtual bro hug.

[User Name]

i did. in my other posts. so now i am just stating the fact that Clutch exists.

 

This is the stupidest thing I have ever read, and I've read some s*** on this site. Congrats!

[Kimmi]

Using the "Pass/Fail" method to determine clutch is patently unfair in that it doesn't account for normalcy. If a player fails just once in a clutch situation statisticians will (rightfully) use that one AB as evidence that the player isn't clutch by the statisticians definition of clutch. "If he were truly clutch he would have gotten a hit that other time. Since he didn't get that hit he's not clutch". That's hogwash.

 

That is not what statisticians do. This is a false premise.

[Kimmi]

Some things aren't statistically quantifiable because they don't take into account unknown variables like a player's emotional/mental state at the time. Some things just ARE and should be accepted as such.

 

Some things are very difficult to quantify, and statisticians acknowledge these shortcomings. At least they are attempting to see whether their is any validity to ideas and claims that have been accepted as true just because that's the way it's always been. You might have a problem with that, but I very much appreciate the research. To me, it's far better than accepting that things are because they just are. What if they're not?

[Kimmi]

Sort of, The problem is that someone else established the definition and then used the data to support the definition.

 

If we can't agree on the definition for it why should we believe it doesn't exist under someone else's definition?

 

Statisticians have acknowledged that one of the difficulties in proving whether 'clutch' exists or not is the lack of a precise definition of what clutch is.

[S5Dewey]

i'm currently giving you a virtual bro hug.

Aw, shucks. . Thanks.

 

)

[Bellhorn04]

My definition of a clutch player would be a player whose performance in high-leverage situations is equal to or better than their overall performance. I think it's the 'high-leverage situations' part that needs better clarification. High-leverage situations within games have been defined, but I think it would be more meaningful if there was also an additional weighting factor based on the leverage of the game itself.

[S5Dewey]

That is not what statisticians do. This is a false premise.

Then what percentage of the time is a player allowed to fail and still be considered clutch?

[S5Dewey]

My definition of a clutch player would be a player whose performance in high-leverage situations is equal to or better than their overall performance. I think it's the 'high-leverage situations' part that needs better clarification. High-leverage situations within games have been defined, but I think it would be more meaningful if there was also an additional weighting factor based on the leverage of the game itself.

 

Good post, Bell. If we're going to talk about clutch what we're talking about is how a person responds to pressure. Some people wilt under pressure, for some it has no effect, and others excel in pressure situations.

 

What we're talking about is how that person handles the pressure. Since pressure is self-inflicted - that is to say that the pressure we feel is pressure we put on ourselves - we can't know how much pressure a player is feeling unless we ask the player and even then we can't be sure of getting an honest answer. We're taught at an early age to be modest. When's the last time you heard a player say, "Yeah... I was feeling a lot of pressure and I stepped up delivered"? Instead they say, "Thankfully I got a pitch I could handle and fortunately I was able to square up on it and it fell in for a hit.". That's BS, but it sounds better.

 

How we perform in any given situation is both physical and psycological, a combination of our mindset and adrenalyn, which contributes to our success. Until we can measure the effect of those things on each individual person's performance as well as the pressure they're feeling we can't statiscally quantify "clutch". We simply have to accept it as something that IS (or isn't, as the case may be).

[mvp 78]

The Greatest Clutch Hitter in the History of the Boston Red Sox, David Ortiz, #34.

[Bellhorn04]

Players and coaches certainly don't like to talk about choking because it's a very delicate subject. But I did give the example earlier of Joe Torre talking about Clemens getting bombed in Game 3 of the 1999 ALCS and saying he thought 'the moment was too big' for Roger or words to that effect. An anecdotal suggestion of a player choking.

[mvp 78]

Players and coaches certainly don't like to talk about choking because it's a very delicate subject. But I did give the example earlier of Joe Torre talking about Clemens getting bombed in Game 3 of the 1999 ALCS and saying he thought 'the moment was too big' for Roger or words to that effect. An anecdotal suggestion of a player choking.

 

He pitched great in game 4 of the WS (when the Yanks were up 3 - 0). His postseason numbers are down from his regular season average, but I don't think he really choked or did terribly. He had a few memorable bad games, but was decent overall.

 

He just never had that Pedro 99 moment.

[sk7326]

He pitched great in game 4 of the WS (when the Yanks were up 3 - 0). His postseason numbers are down from his regular season average, but I don't think he really choked or did terribly. He had a few memorable bad games, but was decent overall.

 

He just never had that Pedro 99 moment.

 

No, although that 15 K one hitter at Seattle gets swept under the rug in the annals of great playoff performances

[sk7326]

My definition of a clutch player would be a player whose performance in high-leverage situations is equal to or better than their overall performance. I think it's the 'high-leverage situations' part that needs better clarification. High-leverage situations within games have been defined, but I think it would be more meaningful if there was also an additional weighting factor based on the leverage of the game itself.

 

Exactly. And also - this is the tricky part of clutch - these players are all awesome. Yes, Ortiz is better than most hitters in high leverage spots conceptually ... but he also better than most hitters, period. His clutch lore - deserved - is also built by being on (generally) a really good team which created those spots for him a lot.

 

I mean even notorious choker ARod was a monster in the 2009 postseason.

[Bellhorn04]

Exactly. And also - this is the tricky part of clutch - these players are all awesome. Yes, Ortiz is better than most hitters in high leverage spots conceptually ... but he also better than most hitters, period. His clutch lore - deserved - is also built by being on (generally) a really good team which created those spots for him a lot.

 

I mean even notorious choker ARod was a monster in the 2009 postseason.

 

Yes, he was...I guess Kate Hudson was really good for him.

 

I think the concept of 'preponderance of data' applies with clutch as with any other measurement...

[Slasher9]

This is the stupidest thing I have ever read, and I've read some s*** on this site. Congrats!

 

good post.

[harmony]

No, although that 15 K one hitter at Seattle gets swept under the rug in the annals of great playoff performances

I was at Roger Clemens' one-hitter on October 14, 2000:

 

http://www.baseball-reference.com/boxes/SEA/SEA200010140.shtml

[User Name]

good post.

 

It respects the laws of grammar at least. Fact!

[Kimmi]

Then what percentage of the time is a player allowed to fail and still be considered clutch?

 

It's not a certain percentage that a player is allowed to fail. It's having a year to year correlation that says that clutch is a repeatable skill. Or having data that statistically shows that a batter is doing something special beyond what would be expected in a randomly generated sample.

[Bellhorn04]

The statistical clutch researchers haven't quite figured out how to define clutch or how to find it, that's all. They will. They have been making strides in other areas. They don't quit, those guys. I have confidence in them.

[iortiz]

Well, to me RBI players and SV players are clutch players. Plain and Simple LOL!

[Kimmi]

The statistical clutch researchers haven't quite figured out how to define clutch or how to find it, that's all. They will. They have been making strides in other areas. They don't quit, those guys. I have confidence in them.

 

That's the thing. They won't quit.

 

I don't mean to offend anyone with the following (though I probably will). Non stat baseball fans have been allowed to claim that things are true and are fact for decades without any proof that they are. It's what they know, it's what they see, it's the way that it's always been, and so it must be true.

 

The reality is, much of what has been accepted as true really isn't. At least the stat geeks are making an attempt to understand what's really going on in the game, not just accepting what has been assumed for all those years.

[notin]

It's not a certain percentage that a player is allowed to fail. It's having a year to year correlation that says that clutch is a repeatable skill. Or having data that statistically shows that a batter is doing something special beyond what would be expected in a randomly generated sample.

 

When it gets boiled down to a percentage, good s nd bad hitters all succeed within a very tight range as it is. .

[S5Dewey]

Thus far we've gone from "clutch doesn't exist" to "we don't know if clutch exists or not because we can't even define it". And it's taken 26 pages to get there.

 

Wouldn't it have been easier to have just said that in the beginning rather than start with the premise that it doesn't exist and then try to build a case from that point?

[Hugh2]

What if you're a "clutch hitter" but every pitcher you ever got a hit against in those "clutch" situations had a whip, bb/k, h/9, and ERA that was 4 times as worse in those "clutch" situations???

[Bellhorn04]

What if you're a "clutch hitter" but every pitcher you ever got a hit against in those "clutch" situations had a whip, bb/k, h/9, and ERA that was 4 times as worse in those "clutch" situations???

 

You're right in the sense that there are a pile of variables in the data, and the 'strength of opposition' has to be factored in. Analyzing this much data is a massive undertaking, that's for sure.

[notin]

Thus far we've gone from "clutch doesn't exist" to "we don't know if clutch exists or not because we can't even define it". And it's taken 26 pages to get there.

 

Wouldn't it have been easier to have just said that in the beginning rather than start with the premise that it doesn't exist and then try to build a case from that point?

 

Depending on whose posts you read, that arguably already happened.

[a700hitter]

Thus far we've gone from "clutch doesn't exist" to "we don't know if clutch exists or not because we can't even define it". And it's taken 26 pages to get there.

 

Wouldn't it have been easier to have just said that in the beginning rather than start with the premise that it doesn't exist and then try to build a case from that point?

And there have been several posts discussing the strong statistical evidence disproving something they don't think can even be defined.