Clutch

[CBL-Commish]

Originally Posted by majesty95

Some of you are so arrogant and elitist that it’s not even worth interacting with. Just because you can’t quantitate something means it doesn’t exist. Yet Eddie Murray suggests it does exist. But if we argue that I a thing then you find reasons to poo poo it like sample size or variance.

If there was ZERO clutch ability in the universe you would end up with people whose clutch performance matched Eddie's just because of random variation. Even over a long career randomness doesn't even out.

Take... Nick Markakis. He has almost 10000 plate appearances now. With nobody out and the bases loaded he has a .737 OPS. With one out and the bases loaded he has a .952. Is that because he's particularly clutch with one out but not with none? No, that's silly. It's because of nothing. Stuff just happens. Nick has a .876 against the Rangers, and a .683 against the Giants. Is there some deep, underlying cause there? Probably not. He's lucky against the Rangers, and against the Giants he was mainly older and not as good after he came to the NL.

There have been 20,000 major leaguers. Some of them will be 70 points of OPS better in high leverage situations just because. There is essentially no difference between the distribution of clutch performances in real life, compared to models that do not account for clutch at all.

[Anyone]

Agreeing with Syd but putting it differently,

Given that some sabermetricians like Tom Tango have found support for some level of clutch ability likely existing, and many like Bill James who once insisted there was no such thing essentially now say they aren't sure...

I'd say the most likely possibility is that clutch ability differs among MLB players but is small enough to be very hard to find in statistics. This may be "not choking as much as most players do" or it could be that certain characteristics of clutch situations (like with at least 2 men on, the pitcher has more pressure to throw strikes) are helpful to some hitters, but to some degree it probably exists, but its effect is quite small, small enough that it's really hard to study with enough detail to prove it exists. I think Tango's conclusion was that it exists but is almost definitely smaller than the platoon advantage.

So, if it's a factor in OOTP but quite a small one, then OOTP has it right.

[Number5]

The point isn't, that a factual difference between players in certain situations doesn't exist. It certainly does, plenty of players perform in situation x better than in y. I believe therefore some players might tend to perform better in some situations that would lead to them being described as "clutch".


The problem is, how to recognize those players, what situations are considered "clutch" and what bonus to give them to them.


Let's go back to Rivera. He performed great in the postseason, this is seen as "clutch". However, he performed below average against his biggest rival, the Red Sox. This is "unclutch". So what is he now - clutch, unclutch, averagely clutch? Lets say another closer converted a lot of one run saves (clutch) but bottled many 4 out saves (unclutch). What now?


Postseason perfomance is associated with clutchness, but postseason baseball has a lot of factors (End of season, cold weather, extra rest days) that make it objectively distinct from regular season baseball as a whole. Being good there doesn't make you better in other situations, and if lets say some guy would be consistently better in april, most people would react "huh, that's interesting" but if somebody has the same quirk in october, its "HE'S A CLUTCH GOD!!!!" - and saying this postseason performance proves he's better in hitting with RISP on a may night in ****ing Tampa Bay.
And this has also the big caveat that the "postseason clutchness" isn't just a statistical fluke with small sample sizes.


If there is data that a bunch of players consistently performs better in certain situations with a reasonable sample size and controlled for other factors (LIKE SACRIFICE FLIES THAT MAKE EVERYONE HAVE BETTER AVERAGES WITH RISP), statminded people would be open to seeing that.

We also want to quantify why Ortiz bats .455 and Bumgarner has a .25 ERA and Rivera a .99 ERA in the WS.

That said, those performances doesn't mean that those players deserve a clutch rating that helps them perform in arbitrary situations some people describe and some not as clutch or unclutch. It may mean that some people MIGHT deserve a "postseason performer" bonus, but even then, you can only quantify this after a long and successful career for a good team and then only if you're a big name guy cause nobody notices utility infielder Jones going 5-12 in the WS when usually he bats .210



And there's a fair chance that there is a parallel universe where people hail clutch postseason hero Kershaw and spit on the names Ortiz, Rivera and Bumgarner as postseason bottlers, just because the coin flipped on heads and not tails in the other universe.


So we say: This is 99% percent speculative bull****, and we'd rather have no clutch ratings than asspulled clutch ratings and if you can quantify what clutch is and when does it provably apply, we will be willing to listen to you. If you say something exists, then it's on YOU to prove us this exists and cherrypicked stats by cherrypicked players prove **** all in the millons of stats the baseball world produces each year.

[majesty95]

Y’all really have no idea what you’re really arguing. I’ll say the last post was the most reasonable until the asspulling and coin flip logic.

Where is the data that says “captains” lead to more wins and better chemistry? How do you quantify what a captain is? Are you in the locker room to see every interaction? Where is the data that says movement leads to more ground balls or “stuff” leads to Ks? Where’s the data that egotistical players ruin clubhouses and lead to less wins?

Those are all things accepted to be true on some level in this game but because it’s not a “statistic” nobody argues it and it’s included for realism. Yet “clutch” will NEVER be proven because sample size will always be argued. Yet these “stat” guys want to vehemently shoot it down while at the same time blindly going along with controlling managers and spark plug personalities with ZERO statistical reference. The real question should be why? These guys are like religious zealots and can’t possibly accept their may be flaws in their religion because their fragile psyche built around a superiority complex can’t take it.

If you want to say you don’t believe it exists because there’s no conclusive data to support it, fine. But miss us all with this were intellectually superior to you because we don’t believe in this nonsense and you all are so ignorant that y’all believe in this mystical clutch fairy BS.

[CBL-Commish]

Originally Posted by majesty95

Y’all really have no idea what you’re really arguing. I’ll say the last post was the most reasonable until the asspulling and coin flip logic.
...

If you want to say you don’t believe it exists because there’s no conclusive data to support it, fine. But miss us all with this were intellectually superior to you because we don’t believe in this nonsense and you all are so ignorant that y’all believe in this mystical clutch fairy BS.

You're going to struggle to gain traction with your case when it hinges on your word against objective evidence. Evidence almost always wins. And data almost always trumps "obviously I'm right, everyone knows that (except all the people who have rigorously studied the subject)."

[CBL-Commish]

Originally Posted by Anyone

Agreeing with Syd but putting it differently,

Given that some sabermetricians like Tom Tango have found support for some level of clutch ability likely existing, and many like Bill James who once insisted there was no such thing essentially now say they aren't sure...

I'd say the most likely possibility is that clutch ability differs among MLB players but is small enough to be very hard to find in statistics. This may be "not choking as much as most players do" or it could be that certain characteristics of clutch situations (like with at least 2 men on, the pitcher has more pressure to throw strikes) are helpful to some hitters, but to some degree it probably exists, but its effect is quite small, small enough that it's really hard to study with enough detail to prove it exists. I think Tango's conclusion was that it exists but is almost definitely smaller than the platoon advantage.

So, if it's a factor in OOTP but quite a small one, then OOTP has it right.

And like I mentioned before, the magnitude is small enough that even if you knew with certainty that someone was clutch you'd almost always take a good hitter who wasn't over a mediocre hitter who was when it really mattered.

If Mookie Betts choked by 50 points of OPS in the clutch and Brock Holt upped his game 50 points in the same situations, you're still using Betts all day every day.

[Syd Thrift]

Originally Posted by CBL-Commish

And like I mentioned before, the magnitude is small enough that even if you knew with certainty that someone was clutch you'd almost always take a good hitter who wasn't over a mediocre hitter who was when it really mattered.

If Mookie Betts choked by 50 points of OPS in the clutch and Brock Holt upped his game 50 points in the same situations, you're still using Betts all day every day.

...and Tom Tango said as much. But that’s a different argument than “clutch doesn’t exist” and it’s moving the goalposts.

[zappa1]

StratOmatic uses "clutch" and they just got a great review. No? They think it exists. Not that I care either way.

[CBL-Commish]

Originally Posted by Syd Thrift

...and Tom Tango said as much. But that’s a different argument than “clutch doesn’t exist” and it’s moving the goalposts.

It's like a lot of things. Like pitcher's ability on BABIP, or DIPS. The initial data said "doesn't exist". Further study indicates it does but it's vastly smaller in magnitude than commonly believed. Which isn't "doesn't exist" but is small enough that you can't use it for decision making the vast majority of the time so it might as well not exist.

[CBL-Commish]

Originally Posted by zappa1

StratOmatic uses "clutch" and they just got a great review. No? They think it exists. Not that I care either way.

Do you think the people reviewing StratOmatic know about the research into clutch performance? And more importantly, do you think they based their reviews mostly on that small aspect of the game?

[Sweed]

I think I've asked this before in these clutch threads but never got an answer.

Has there been a study of players to determine how many maintain there "normal" performance vs players that have their performance "go down" in clutch situations?

Is there evidence of a "choke" factor? If so should\could players be rated for choke?

I, like a couple others here, have been of the mind clutch is the ability to not shrink in a clutch situation. Some P's may have a tendency to overthrow etc. while a batter may expand the zone while a "clutch player" doesn't become better, he stays in his normal frame of mind.

[Syd Thrift]

Originally Posted by CBL-Commish

It's like a lot of things. Like pitcher's ability on BABIP, or DIPS. The initial data said "doesn't exist". Further study indicates it does but it's vastly smaller in magnitude than commonly believed. Which isn't "doesn't exist" but is small enough that you can't use it for decision making the vast majority of the time so it might as well not exist.

...and like BABIP, the "initial data" is from the late 1990s, and the other stuff has been around for more than a decade, and if you came in saying "pitchers have no effect on BABIP" and then people said "Charlie Hough and Mariano Rivera" and you said "like I said, most pitchers only have a small effect", then you would also be accused of moving goalposts in that case.

[Hoiles]

Originally Posted by CBL-Commish

And like I mentioned before, the magnitude is small enough that even if you knew with certainty that someone was clutch you'd almost always take a good hitter who wasn't over a mediocre hitter who was when it really mattered.

If Mookie Betts choked by 50 points of OPS in the clutch and Brock Holt upped his game 50 points in the same situations, you're still using Betts all day every day.

Yeah, but if you had two players of around the same OPS but one was deemed clutch and one wasn't who do you choose? Does clutch still not exist?

[CBeisbol]

Originally Posted by Hoiles

Yeah, but if you had two players of around the same OPS but one was deemed clutch and one wasn't who do you choose? Does clutch still not exist?

What handedness is the pitcher?
And the two hitters?

[Syd Thrift]

Originally Posted by Sweed

I think I've asked this before in these clutch threads but never got an answer.

Has there been a study of players to determine how many maintain there "normal" performance vs players that have their performance "go down" in clutch situations?

Is there evidence of a "choke" factor? If so should\could players be rated for choke?

I, like a couple others here, have been of the mind clutch is the ability to not shrink in a clutch situation. Some P's may have a tendency to overthrow etc. while a batter may expand the zone while a "clutch player" doesn't become better, he stays in his normal frame of mind.

I honestly think the mental aspect is a no-go in terms of finding the mechanism. Sure, an individual player might just plain lose it in a big-time setting (John Rocker comes to mind) but how can you predict that? People are complicated and a guy who's normally very level-headed might suddenly, for one series, and for no particular baseball-related reason, suddenly lose it or find another level of concentration that ups their game. I'd be all for including that kind of thing in a game but it's not exactly "clutch", and I think that sort of thing probably happens more often than at least the mental aspects of "clutch".

I'm convinced that a larger issue is that some players are better at getting around on very good fastballs, hitting very good curves, and so on, and that's what (most of) "clutch" is in baseball. Yes, over the course of a season it all evens out, and by and large these are the guys who are better hitters overall, but in crunch time you're much more likely to face top-tier pitchers and so this effect is heightened. Ironically, this *probably* makes younger, less seasoned players better in those situations - I'd go so far as to say that if you didn't account for that effect and found no difference between vets and rookies in clutch situations, that is probably a sign of a (small) effect for the mental aspect of clutch hitting.

As noted before, in sports where there's clearly a "clutch" effect, it's based on physical attributes, not "mental toughness". In basketball you want to go to guys who can create their own shots or release very quickly off of a screen because easy shots go away when a game is on the line. In football, quarterbacks with strong arms and who can extend plays lead comebacks way more often than QBs who don't (and who are otherwise fine players - Kirk Cousins, for instance, is not a bad QB but he's simply not in the same league as Russell Wilson when it's late and you need to get down the field). I think a lot of the time sportswriters tend to post facto describe good clutch players as being mentally tough, too, but there are things that remain - you'll never see a Cedric Ceballos style garbage collector described as "clutch" (the closest I can think is how people thought of Rodman, but that was all based on defense and rebounding), and there are very clearly "game manager" QBs who otherwise put up fine numbers who don't generally lead their team from behind.

[Syd Thrift]

Originally Posted by Hoiles

Yeah, but if you had two players of around the same OPS but one was deemed clutch and one wasn't who do you choose? Does clutch still not exist?

Even Tom Tango said that the effect is so small that it's likely that some other factor will swamp it, even if "everything else is equal", which it never is.

[Hoiles]

Originally Posted by CBeisbol

What handedness is the pitcher?
And the two hitters?

Both hitters have the same handedness. Considering there's only two choices, it's a pretty common situation.

Originally Posted by Syd Thrift

Even Tom Tango said that the effect is so small that it's likely that some other factor will swamp it, even if "everything else is equal", which it never is.

Yes, but it's fairly common to not have enough info to distinguish between two players, especially in the dugout when you have a split second to make a decision. Given two hitters of same handedness with a <0.010 OPS difference on the season, but one had a big series in the NLCS 3 years ago, and the other was a rookie, who do you choose? Keep in mind if you choose the rookie based on that 0.010 OPS difference, you're not accounting for random variance.

[PocketsAintFull]

Let’s meet halfway and agree on ‘clutch’ being performing optimally in higher pressure situations? Statistically unquantifiable but psychologically likely? Call it a draw?

Interestingly re: OOTP, it probably shouldn’t be there if numbers can’t cover it. The only way I could think was if a player *happened* to come up big in ‘big situations’, they could develop some sort of confidence trait? I suppose it would be more of an ‘experience’ mechanic really

[CBeisbol]

Originally Posted by Hoiles

Both hitters have the same handedness. Considering there's only two choices, it's a pretty common situation.

Point was,
I'd like to see the Venn Diagram for people who argue strongly for clutch and people who decry platooning

One of them has a strong observable effect, the other is clutch

[Syd Thrift]

Originally Posted by Hoiles

Both hitters have the same handedness. Considering there's only two choices, it's a pretty common situation.



Yes, but it's fairly common to not have enough info to distinguish between two players, especially in the dugout when you have a split second to make a decision. Given two hitters of same handedness with a <0.010 OPS difference on the season, but one had a big series in the NLCS 3 years ago, and the other was a rookie, who do you choose? Keep in mind if you choose the rookie based on that 0.010 OPS difference, you're not accounting for random variance.

I take the guy who hits that pitcher better.