ATL - LMA retires with heart issue - pg 78

[Ron Swanson]

The argument would be that a player performing even up to their elite performance (Jordan, Lebron, Hakeem, etc.) during high-level competition and external factors (noise, defensive pressure, etc.) is the very definition of clutch. But there's plenty of instances of players performing above their normal level in late-game situations as well, otherwise I don't know how you're able to adequately explain why a player like Damian Lillard is so often able to make the most difficult shots on the final possession. Or how Tom Brady is consistently able to perform on a micro-level in late-game situations, despite the overall (macro) numbers sometimes not reflecting that, because you're stubbornly looking at the overall statistical model from a bird's eye view?

I find it hypocritical that we'd continue to just hand-wave those instances as "small sample size" and "variance" when they happen with enough frequency over the course of a player's career. That is, if we're going to ascribe data science towards something that's an entirely flawed and inconsistent variable such as the human mind and body. It's a sample size just like anything else.

[DingleJerry]

But someone like Dame makes those shots when it's not clutch time too. I'm kind of at the point that those clutch someone can be is to be unaffected by pressure and play his normal game. Think of like baseball. a hitter has played this game his whole life and is a .280 hitter, but all of a sudden if the game is on line he be 'clutch' and hit .400? I don't think that's really possible. But if he's clutch and doesn't get rattled by the moment, can he still be a roughly .280 hitter who gets a random small sample at the right time to look better, sure (or the opposite as well). If someone gets nervous and shaky though I can totally see how it turns his normal percents to go down though.

[coolhandluke121]

The argument would be that a player performing even up to their elite performance (Jordan, Lebron, Hakeem, etc.) during high-level competition and external factors (noise, defensive pressure, etc.) is the very definition of clutch.

That's not the definition of clutch that's being argued here, which is why I suggested that you jumped into the argument without knowing exactly what is being argued. I'm fine with that as a definition of clutch. My original post is literally exactly that, arguing that being "clutch" is just the absence of performing worse in the clutch.

But there's plenty of instances of players performing above their normal level in late-game situations as well, otherwise I don't know how you're able to adequately explain why a player like Damian Lillard is so often able to make the most difficult shots on the final possession. Or how Tom Brady is consistently able to perform on a micro-level in late-game situations, despite the overall (macro) numbers sometimes not reflecting that, because you're stubbornly looking at the overall statistical model from a bird's eye view?

I find it hypocritical that we'd continue to just hand-wave those instances as "small sample size" and "variance" when they happen with enough frequency over the course of a player's career. That is, if we're going to ascribe data science towards something that's an entirely flawed and inconsistent variable such as the human mind and body. It's a sample size just like anything else.

Anecdotes aren't good enough. There are literally thousands of players in the potential pool of "clutch" candidates, and variance guarantees that dozens of them will have significantly better stats in the clutch than they do in regular situations just because that's how variance works. But use those stats to try to predict who is going to over-perform in the clutch in the future, as opposed to using hindsight to build narratives, becomes very problematic. There's really no evidence that it works. Of course many of those guys will continue to be just as good in the clutch as they are in regular situations because their very presence on the list suggests that the pressure doesn't bother them, but if there is such a thing as being able to be even better in the clutch, someone should be able to provide actual objective evidence of it WITH predictive power. It's interesting that you use this as an opportunity to rail against the overreach of statistical sciences when this is precisely the kind of thing we really need stats for, because of the subjectivity of the anecdotal and experiential data of watching a game in a heighted emotional state and because of the tendency to build narratives from that. Brady and Jordan are great players who rarely fail in regular situations OR in clutch situations, but they fail in the clutch their fair share of times too because that's just sports.

[coolhandluke121]

But someone like Dame makes those shots when it's not clutch time too. I'm kind of at the point that those clutch someone can be is to be unaffected by pressure and play his normal game. Think of like baseball. a hitter has played this game his whole life and is a .280 hitter, but all of a sudden if the game is on line he be 'clutch' and hit .400? I don't think that's really possible. But if he's clutch and doesn't get rattled by the moment, can he still be a roughly .280 hitter who gets a random small sample at the right time to look better, sure. If someone gets nervous and shaky though I can totally see how it turns his normal percents to go down though.

Precisely. If players were really so damn good that they could actually be better in the clutch than they are in regular situations, they would apply that to regular situations too. The .400 example is a perfect one because that crap gets used in baseball all the time. Players are focused and determined to succeed at all times. Dame misses plenty of those clutch shots too but nobody talks about the misses for 3 days on SportsCenter.

Plenty of players are a lot better than other players in the clutch, but they're not better than themselves in the clutch. No evidence and no concrete explanation for why they would be = no theory. My understanding is that the Moneyball pretty much killed arguments about "clutch" players but I don't remember where I first read about this stuff. In any case, you don't hear much talk about it at all nowadays, and I think that's because it's such a ridiculous notion that people aren't even looking into it anymore, kind of like a scientist isn't going to waste time disproving astrology.

[Ron Swanson]

But someone like Dame makes those shots when it's not clutch time too. I'm kind of at the point that those clutch someone can be is to be unaffected by pressure and play his normal game. Think of like baseball. a hitter has played this game his whole life and is a .280 hitter, but all of a sudden if the game is on line he be 'clutch' and hit .400? I don't think that's really possible. But if he's clutch and doesn't get rattled by the moment, can he still be a roughly .280 hitter who gets a random small sample at the right time to look better, sure. If someone gets nervous and shaky though I can totally see how it turns his normal percents to go down though.

Baseball is a different beast. At most we're talking about, what? 30 at-bats a guy has per postseason? Even NFL playoffs is relatively unreliable until we get a bigger picture throughout a guy's career. But NBA? Nah. People call the NBA playoffs "small sample size" but it's really not. You have 7-game series and you have the best players with the ball in their hands the majority of each of those games. Hundreds of shots, hundreds of passes, and playing both ends of the court. Much easier to quantify who's able to exceed or fail to meet expectations. But's that's also why it's really only somewhat reliable for the best players in the league.

Robert Horry? I don't know if you can say he's inherently "clutch" based on a half dozen "big shots" he's made throughout a decade plus career given the volume of evidence. Certainly a guy I'd love to have on my team come postseason time, but I'll take Draymond Green with the ability to give me a clutch pass or defensive stop over Horry any day (because the underreported aspect of clutch is more than just making big shots).

[DingleJerry]

Just used baseball as an example due to being such a percent based thing so felt like it helped make the point clearly and to use the point about how I don't see a reason how someone would 'improve' at crunch time as opposed to just being unaffected. After the last few posts it does seem you CHL and I are generally on the same page though.

[emunney]

But someone like Dame makes those shots when it's not clutch time too. I'm kind of at the point that those clutch someone can be is to be unaffected by pressure and play his normal game. Think of like baseball. a hitter has played this game his whole life and is a .280 hitter, but all of a sudden if the game is on line he be 'clutch' and hit .400? I don't think that's really possible. But if he's clutch and doesn't get rattled by the moment, can he still be a roughly .280 hitter who gets a random small sample at the right time to look better, sure. If someone gets nervous and shaky though I can totally see how it turns his normal percents to go down though.

Precisely. If players were really so damn good that they could actually be better in the clutch than they are in regular situations, they would apply that to regular situations too. The .400 example is a perfect one because that crap gets used in baseball all the time. Players are focused and determined to succeed at all times. Dame misses plenty of those clutch shots too but nobody talks about the misses for 3 days on SportsCenter.

Plenty of players are a lot better than other players in the clutch, but they're not better than themselves in the clutch. No evidence and no concrete explanation for why they would be = no theory. My understanding is that the Moneyball pretty much killed arguments about "clutch" players but I don't remember where I first read about this stuff. In any case, you don't hear much talk about it at all nowadays, and I think that's because it's such a ridiculous notion that people aren't even looking into it anymore, kind of like a scientist isn't going to waste time disproving astrology.

There are so many confounding factors I have no idea how to even start to design an experiment to answer any of these questions. I personally don't believe it's realistic to expect a person's level of focus or effort to be 100% consistent, so the idea of "if they can do it in the clutch they'd just always do it" doesn't square with me. That said, how would we even know if this layer of "extra" effort/focus would necessarily help? Maybe it would hurt by virtue of being unusual. Maybe it doesn't play the same with everybody. Maybe it exacerbates other differences (if you're trying harder in the mental **** direction, doesn't that just make you **** up harder?) On it's face the idea of somebody being about to will a shot in seems like pure bull **** to me, but empirically I don't know what to do with it.

[coolhandluke121]

But someone like Dame makes those shots when it's not clutch time too. I'm kind of at the point that those clutch someone can be is to be unaffected by pressure and play his normal game. Think of like baseball. a hitter has played this game his whole life and is a .280 hitter, but all of a sudden if the game is on line he be 'clutch' and hit .400? I don't think that's really possible. But if he's clutch and doesn't get rattled by the moment, can he still be a roughly .280 hitter who gets a random small sample at the right time to look better, sure. If someone gets nervous and shaky though I can totally see how it turns his normal percents to go down though.

Precisely. If players were really so damn good that they could actually be better in the clutch than they are in regular situations, they would apply that to regular situations too. The .400 example is a perfect one because that crap gets used in baseball all the time. Players are focused and determined to succeed at all times. Dame misses plenty of those clutch shots too but nobody talks about the misses for 3 days on SportsCenter.

Plenty of players are a lot better than other players in the clutch, but they're not better than themselves in the clutch. No evidence and no concrete explanation for why they would be = no theory. My understanding is that the Moneyball pretty much killed arguments about "clutch" players but I don't remember where I first read about this stuff. In any case, you don't hear much talk about it at all nowadays, and I think that's because it's such a ridiculous notion that people aren't even looking into it anymore, kind of like a scientist isn't going to waste time disproving astrology.

There are so many confounding factors I have no idea how to even start to design an experiment to answer any of these questions. I personally don't believe it's realistic to expect a person's level of focus or effort to be 100% consistent, so the idea of "if they can do it in the clutch they'd just always do it" doesn't square with me. That said, how would we even know if this layer of "extra" effort/focus would necessarily help? Maybe it would hurt by virtue of being unusual. Maybe it doesn't play the same with everybody. Maybe it exacerbates other differences (if you're trying harder in the mental **** direction, doesn't that just make you **** up harder?) On it's face the idea of somebody being about to will a shot in seems like pure bull **** to me, but empirically I don't know what to do with it.

I'm not really even suggesting that effort and focus are 100% consistent, just that the inconsistencies are there in clutch situations just like they are in all others. If it's like being in a zone or whatever, that can happen at any time. And I would definitely agree that trying to focus more and play better in the clutch is at least as likely to hurt as it is to help, so again there's just no mechanism that would explain "clutch" players and hold predictive explanatory power. Great execution in the clutch is probably more a product of preparation than anything, and that preparation helps at all times during a competition.

But again, if there is such a thing as "clutch" players who actually play better in the clutch, there should be some evidence that reliably identifies those players and predicts future performance instead of having to rely on anecdotal evidence. I'm just not aware of any at all, but I've seen several analyses that completely dispel the notion by showing that how clutch reputations are rarely backed up by evidence and aren't reliably correlated from year-to-year, with routine regression to the mean almost always ruining the party (but not always, because random variance will ensure that some players will stay with elevated performance a little longer, but nobody can reliably predict which ones will). And again, this is in reference to guys actually playing better in the clutch, not just guys who have proven they can handle the mental aspect of it and continue to execute as they normally would, give or take normal allowances for variance.

[emunney]

Precisely. If players were really so damn good that they could actually be better in the clutch than they are in regular situations, they would apply that to regular situations too. The .400 example is a perfect one because that crap gets used in baseball all the time. Players are focused and determined to succeed at all times. Dame misses plenty of those clutch shots too but nobody talks about the misses for 3 days on SportsCenter.

Plenty of players are a lot better than other players in the clutch, but they're not better than themselves in the clutch. No evidence and no concrete explanation for why they would be = no theory. My understanding is that the Moneyball pretty much killed arguments about "clutch" players but I don't remember where I first read about this stuff. In any case, you don't hear much talk about it at all nowadays, and I think that's because it's such a ridiculous notion that people aren't even looking into it anymore, kind of like a scientist isn't going to waste time disproving astrology.

There are so many confounding factors I have no idea how to even start to design an experiment to answer any of these questions. I personally don't believe it's realistic to expect a person's level of focus or effort to be 100% consistent, so the idea of "if they can do it in the clutch they'd just always do it" doesn't square with me. That said, how would we even know if this layer of "extra" effort/focus would necessarily help? Maybe it would hurt by virtue of being unusual. Maybe it doesn't play the same with everybody. Maybe it exacerbates other differences (if you're trying harder in the mental **** direction, doesn't that just make you **** up harder?) On it's face the idea of somebody being about to will a shot in seems like pure bull **** to me, but empirically I don't know what to do with it.

I'm not really even suggesting that effort and focus are 100% consistent, just that the inconsistencies are there in clutch situations just like they are in all others. If it's like being in a zone or whatever, that can happen at any time. And I would definitely agree that trying to focus more and play better in the clutch is at least as likely to hurt as it is to help, so again there's just no mechanism that would explain "clutch" players and hold predictive explanatory power. Great execution in the clutch is probably more a product of preparation than anything, and that preparation helps at all times during a competition.

But again, if there is such a thing as "clutch" players who actually play better in the clutch, there should be some evidence that reliably identifies those players and predicts future performance instead of having to rely on anecdotal evidence. I'm just not aware of any at all, but I've seen several analyses that completely dispel the notion by showing that how clutch reputations are rarely backed up by evidence and aren't reliably correlated from year-to-year, with routine regression to the mean almost always ruining the party (but not always, because random variance will ensure that some players will stay with elevated performance a little longer, but nobody can reliably predict which ones will). And again, this is in reference to guys actually playing better in the clutch, not just guys who have proven they can handle the mental aspect of it and continue to execute as they normally would, give or take normal allowances for variance.

Generally speaking I've been less than impressed by the rigor of most statistical sports studies I've seen; you probably feel the same way. You have to control for so many things and end up with an estimation with very little power anyway, just not enough total observations in most cases. I'm sure they do better than "Michael Jordan actually shot 40% in the clutch!" but how much better? Are they able to control for opponent, shot type/quality, officiating (we generally understand that officiating is looser in tight games and big moments, but how exactly? Bear of a measurement problem that could really affect estimates).

Like you said, ultimately whether it exists or not, if you can't identify and predict based on it, what's the point? Just give me the guys who play smart and together and roll the ball out.

[DrWood]

But someone like Dame makes those shots when it's not clutch time too. I'm kind of at the point that those clutch someone can be is to be unaffected by pressure and play his normal game. Think of like baseball. a hitter has played this game his whole life and is a .280 hitter, but all of a sudden if the game is on line he be 'clutch' and hit .400? I don't think that's really possible. But if he's clutch and doesn't get rattled by the moment, can he still be a roughly .280 hitter who gets a random small sample at the right time to look better, sure (or the opposite as well). If someone gets nervous and shaky though I can totally see how it turns his normal percents to go down though.

It's possible that some elite player is bored unless the game is on the line and performs slightly better during those situations. the available evidence shows that's rare to nonexistent, though.

[coolhandluke121]

Generally speaking I've been less than impressed by the rigor of most statistical sports studies I've seen; you probably feel the same way. You have to control for so many things and end up with an estimation with very little power anyway, just not enough total observations in most cases. I'm sure they do better than "Michael Jordan actually shot 40% in the clutch!" but how much better? Are they able to control for opponent, shot type/quality, officiating (we generally understand that officiating is looser in tight games and big moments, but how exactly? Bear of a measurement problem that could really affect estimates).

Like you said, ultimately whether it exists or not, if you can't identify and predict based on it, what's the point? Just give me the guys who play smart and together and roll the ball out.

I agree completely. I suppose the lack of evidence of "clutch" performance isn't proof that it's impossible to find in a well-designed study, but I would think they would have something by now if there was such a thing - especially in baseball, for example.

On a separate point, MJ or Kobe hitting 40% of their shots in the clutch wouldn't necessarily be bad per se if you controlled for situation because I would assume a lot of those plays are ones where you want someone to shoot as the clock expires, which means they have to hold the ball and work against the other team's best defender, who knows when they're going to shoot, all of which makes it normally a terrible shot, but it's still a good option because you can basically guarantee the opposing team won't have a possession.

[blazza18]

Not wanting Chris Paul looks to be such a bad decision from Giannis/Bucks.

[Matches Malone]

How can we get Beal in the offseason? He's rotting in DC

Read on Twitter

[emunney]

Richie Tenenbaum having a tough one

Read on Twitter

[emunney]

I sympathize with Love for his mental health issues and appreciate his openness about it but like retire, stop cheating your teammates.

[Bmaasse]

Then why isn't there consistency in who performs better in the clutch than they do the rest of the time from year-to-year? You're suggesting explanations of why something is true when there's not even evidence of it being true. I'm explaining a phenomenon that actually happens (choking in the clutch) whereas you're offering what you consider a plausible explanation of a phenomenon that doesn't actually happen. That's why I say again that you need data AND an explanatory mechanism to have an actual theory. You can barely suggest a plausible mechanism and even that is a bit of a reach.

Instead of comparing how a player performs in regular season games vs. the playoffs, did you scompare their shooting splits for the first 3 quarters vs. the 4th, or even the last 5 minutes?
If you only look at the box score from last night, you would think that Middleton had a much better game than Lou Williams, but who's performance actually had a greater impact on the overall outcome of the game. Lou made shots when it mattered most, Lou was clutch last night.

And sometimes Khris will have bad games but make a bunch of shots in the 4th quarter. Stop building stupid narratives out of random variance.

ETA: I've been giving you way too much credit. Should have known this was going to boil down to the same narrative-building garbage it always does, all based on the emotions of the 4th quarter with no evidence or reason to back anything up. I'm out.

No one is clutch but players do choke.

Player A hits tough shots occasionally during the regular season so when they do it to win a championship it wasn't clutch, it was just them being them.

Player B bricks free throws occasionally during the regular season, so when they do it in the playoffs and it causes their team to lose, they absolutely choked. Or maybe they maybe choked? Or maybe they just missed their throws?

What player B did can be proven and predicted, while what player A did cannot.

Stress only serves as an "explanatory mechanism" when it causes players to play bad.
No one ever plays better under duress, even though your fight or flight response at its essence is a performance enhancer.

I'm trying to rid myself of stupid narratives and be more like you. What else am I missing?

[tonyreyes123]

I sympathize with Love for his mental health issues and appreciate his openness about it but like retire, stop cheating your teammates.

Kevin Love is acting like a whiny bitch, the mental health angle is getting trite nowadays. Larry Sanders played hard for a month to get a new deal and as soon as the ink is dry I see him in the clubs but he’s too “anxious” to play anymore

[coolhandluke121]

Instead of comparing how a player performs in regular season games vs. the playoffs, did you scompare their shooting splits for the first 3 quarters vs. the 4th, or even the last 5 minutes?
If you only look at the box score from last night, you would think that Middleton had a much better game than Lou Williams, but who's performance actually had a greater impact on the overall outcome of the game. Lou made shots when it mattered most, Lou was clutch last night.

And sometimes Khris will have bad games but make a bunch of shots in the 4th quarter. Stop building stupid narratives out of random variance.

ETA: I've been giving you way too much credit. Should have known this was going to boil down to the same narrative-building garbage it always does, all based on the emotions of the 4th quarter with no evidence or reason to back anything up. I'm out.

No one is clutch but players do choke.

Player A hits tough shots occasionally during the regular season so when they do it to win a championship it wasn't clutch, it was just them being them.

Player B bricks free throws occasionally during the regular season, so when they do it in the playoffs and it causes their team to lose, they absolutely choked. Or maybe they maybe choked? Or maybe they just missed their throws?

What player B did can be proven and predicted, while what player A did cannot.

Stress only serves as an "explanatory mechanism" when it causes players to play bad.
No one ever plays better under duress, even though your fight or flight response at its essence is a performance enhancer.

I'm trying to rid myself of stupid narratives and be more like you. What else am I missing?

EVIDENCE.

[paulpressey25]

How can we get Beal in the offseason? He's rotting in DC

They're 10-2 in their last 12.

I wasn't aware of that either until I heard SImmons and House discussing on their podcast. Russ apparently has gotten his act together.

And Daniel Gafford, the through in center they got in the deal with the Bulls is playing like a poor man's Ben Wallace.

And don't get me started on RoLo. He's doing great also.

[M-C-G]

Man, the fall of John Wall. Life comes at you pretty quick.