Just how important is "clutch", really?

which QB is he?

 

We discussed this previously in this thread. The notion that Nate Kaeding is proof of clutch is nonsense.

Every player has stretches or examples of good play and bad play. That is true in the playoffs and the regular season. Kaeding had a bad stretch of 8 kicks that just happened to be in the playoffs. Michael Jordan had multiple bad shooting stretches of 5/19, 5/19 and 9/35 in the playoffs. Is he the definition of non-clutch? Of course not, because he also has stretches of greatness. Kaeding had many fewer playoff opportunities, so he didn't end up having those stretches of playoff greatness. But in a larger sample size of all games, he did show his greatness. Of course, this is why you have to look at the full record, not just cherry pick the great games or the bad ones.

 

Flacco

 

The magical fairy dusty is always there, with franchise QBs. They always will their teams to win, and pull touchdowns out of their magic sack whenever they need one.

 

He's both.

 

I do believe you can become clutch. It is being able to over come the "I hope I don't screw up here" mindset. If a person learns to control that they can be clutch.

 

I feel like you're talking about the opposite of what's been talked about.

I was told the adrenaline rush lets a person exceed their abilities and because of that they can be clutch, but now you're saying the adrenaline rush can hurt a player. I mean you're basically saying Thill gets rattled when up to this point the argument was more or less he doesn't get rattled enough.

 

But there's not really any strong evidence of that mindset. The vast, vast majority of players play the same in clutch situations as in non-clutch situations. The few examples offered to the contrary are small sample sizes, like Kaeding. But assuming it's possible, is there any evidence at all that Flacco had that mindset early in his career, or that he did something to get rid of it. I'm not talking about his performance itself, because we know it wasn't good early and then got good. But as I've said many times here that may just he how his performance distribution falls. Has he or anyone associated with him ever said that he feared failure early in his career and then did something to overcome it?

And if he did become clutch at some point, how/why does it seem he lost it? http://www.baltimoreravens.com/news...Troubles/5e5a9d45-c231-4c2b-8da0-f1f4d94034a2

 

Yes, like anything else, there is a good and bad to it.

 

How can you call 15 or so games, spread out over 4-5 years, in 2-4 game spurts, a "run?"

That's silly. That's not a run. That's a pattern. The pattern can reverse. But it isn't a run. It's completely spread out and separated by 60 games interspersed.

 

Yup, it's true the pattern could reverse in the future. But right now we can only use the data we have, and if you assume "playoffs" are a separate condition, then taking sample size into account there's only a 0.21% chance Flacco's playoff record was due to random chance.

What caused the relatively abrupt rise in ability? Who knows, but that doesn't need to be answered to argue against the claim it was due to random variation.

 

I don't think there is a quote from him saying he was scared early, but not he isn't. I know from personal experience you can over come those fears, thus I assume everyone can if they work on it.

 

That's the whole point. I don't assume or believe the playoffs are a separate condition. They are playing the exact same game under the exact same rules. Guys have good games and bad games whether its regular season or the playoffs. If you assume, as you seem to, that the playoffs are different then you have completely tilted the scales. You are then looking strictly at micro sample sizes and have essentially already concluded that good play in the playoffs equates with "clutch" and bad play in the playoffs equates with "choke."

 

Call it whatever you want -- run, pattern, streak, etc. It doesn't matter what you label it. If you like "pattern" I'm happy to use that. So what is the explanation for Flacco's pattern of vastly superior play (30+ rating points) against the NFL South as compared to any other division. That's spread out over his whole career. Statistical anomalies occur. We've talked about Robert Horry -- why was he able to shoot .485 from 3 against Golden State in 53 career games over a 17 year career, but only .341 overall for his career and only .226 against Detroit in his career? Those are spread out against different players, coaches, etc. But statistical anomalies occur and an apparent correlation doesn't mean or suggest any causal or logical connection. That's why sample size is important and that's why you don't disregard Flacco's first 5 playoff games. Even with those 5 games, it is still a pretty tiny sample size.

 

But there is no evidence of any fears whatsoever. Neither he nor anyone close to him has ever said that, to best of either of our knowledge.

If such fears are a real and meaningful thing, then they presumably don't just disappear. One would have to work to overcome them and that work would presumably be fairly difficult. But there's no evidence of any such work. In his last 3 playoff games before January 9, 2011, Flacco was 37 for 75, for 364 yards, 0 TDs and 6 INTs, for a passer rating of 30.1. If that was due to fear or nervousness, that was not just a small issue, but rather a major problem. His play in those games was horrible. Not just subpar or not up to his usual standards, but truly pathetic. But then, magically, his passer rating 115.4 in his very next playoff game. And was well over 100 over his next 8-9 playoff games. So if it had anything to do with overcoming fears and becoming clutch, it was a sudden and dramatic change. He went from being considerable worse than any QB in NFL history (passer rating of 30.1) to being the best and most clutch QB in NFL history. And it all happened seemingly overnight.

 

It's win or go home, of course it's a separate condition.

 

That doesn't make it a separate condition. It is the same game with the same rules. Teams and players are trying really, really hard to win every game, playoffs or not. Again, I don't think there is any real evidence of clutch.

 

Yeah I don't get having the position that the playoffs are "just another game"

Week 1, everyone is 0-0, is different than week 17 where you need to win to make the playoffs. It's even more different once you're in. Claiming "it's just another game" is convenient enough to support your position that there is no such thing as clutch.

 

Again, it's the same game with the same rules. Literally nothing changes. The stakes might be higher, but every player and every team wants to win every game. And they try really, really hard to accomplish that. I guess your position is that they only kinda-sorta try to win regular season games, but really, really try to win playoff games, but I don't think that's true and I don't think there's any evidence for it.

Again, for the umpteenth time, what is the evidence that there are really clutch players as compared to non-clutch ones? The anecdotal "evidence" you have provided is not compelling at all. We've talked about Flacco, but he only appears clutch if you exclude his first 5 playoff games and conclude that his "pattern" of great games thereafter must have been due to the fact that it was the playoffs. But that ignores (1) those first 5 playoff games, (2) the lack of evidence that his poor performance in those first 5 games had anything to do with nerves, (3) the lack of evidence of any steps he may have taken to so dramatically and instantaneously overcome those "nerves", (4) the fact that he had a similar stretch of great games in the regular season in 2010, (5) the fact that he has had a similar "pattern" of great play against the NFC South with no apparent explanation; and (6) that he appears to have now lost his clutch touch because he repeatedly failed to come through in the clutch last season.

There's the Nate Kaeding example, which really comes down to 5 kicks. A sample size that is really too small to be taken seriously. And the fact that he was good throughout his career on pressure kicks with a chance to win or tie. And he was fine in the playoffs before that 3/8 stretch. And the fact that virtually all of the reputedly "clutch" players in history, from Montana to Marino to Brady to Michael Jordan, etc., all put up some playoff stinkers that represented larger sample sizes and arguably worse performances than Kaeding's.

There's the supposed correlation between HOF QBs win% in close games, but virtually every one of those supposedly clutch guys actually had a higher win% in non-close games. And that one of the big reasons why they are HOF QBs in the first place is that their teams won a lot.

Is LeBron clutch? Many continue to say no, even though he's had some incredible playoff performances and his overall playoff numbers are outstanding. Of course, his non-playoff numbers are similarly outstanding. That's because he's a really good basketball player. Accounting for the fact that he averages roughly 8% more minutes per game in the playoffs and takes more shots there, his regular season and playoff stats are essentially identical. With the possible exception of rebounds, which on a Per 36 minute basis go up a little less than 1 rebound per game. That's a pretty small difference, but watching him very closely for many years, it is clear that he puts greater emphasis on his rebounding in the playoffs than he does in the regular season.

Michael Jordan's "clutchness" is taken as a given, but he too was the exact same player in the playoffs that he was in the regular season. If one wanted to be picky, one might say he was a teeny-tiny bit worse in the playoffs, but the difference is so negligible that I don't think that is fair to say.

 

Right.. being able to perform when it's really important to perform is what "clutch" refers to. You seem to think they have to perform better than they usually do. It's been repeatedly explained in this thread that what we're talking about is them performing better in some condition than what you'd expect based on the average drop-off in performance from one condition to another.

Flacco, as I mentioned earlier, is a rare exception in that he actually started to perform better in the playoffs than in the regular season. You can perform worse in the playoffs and still be "clutch" as long as your performance doesn't drop off as much as most others do.

Why would he put greater emphasis on something in the playoffs if (according to you) playoffs are no different than regular season games? Maybe because playoffs are "win or go home?"

You only need to see how baseball managers manage games in the playoffs vs. regular season to see they don't manage players the same way. Playoffs = different condition.

 

I haven't read most of this, but there is absolutely something to being able to play in tense/challenging/pressure situations.

It's evident in sports and in every day work environments. Some people have the ability to think, problem solve, and work their way through situations that others can't, whether they have shown previous competency to do so or not. The moment at times can be too big, or just another moment to people depending on that ability.

 

But there isn't a general dropoff in performance in the playoffs. Overall, the level of play in the playoffs is about the same as in the regular season. And as you would expect, some guys' numbers are a bit better in the playoffs and some guys' are a bit worse, but virtually nobody is notably better or worse over any kind of decent sample size. Flacco's 3.9 pt uptick in passer rating in the playoffs is negligible. As I have said, it is roughly one extra 12 yard completion in a 36 pass attempt game. Hardly even noticeable. And no, it is not that rare for a QB to have a higher passer rating in the playoffs. Some are higher and some are lower. As their playoff sample size increases, they tend to revert to their regular season means.

One would probably have to ask LeBron why he seems to focus on rebounding more in the playoffs. Again, we are talking about a pretty small difference -- less than 1 rebound per 36 minutes. LeBron is a guy who can literally do everything on a basketball court. He's also a guy who is well known for reading the game and seeing what his team needs from him and doing some more of that and perhaps less of some other things. So it may be circumstance or just his read of the games. When he was on the Heat, the Heat was a weak rebounding team and was playing more small-ball in the playoffs. That may have made him realize the team needed more rebounding from him. In 2014-15, Kevin Love was out and he may have thought he needed to pick up some of the rebounding slack. This year, as in 2014-15, they played the Warriors in the Finals and their small-ball lineups may have created more rebound opportunities.

The baseball manager example tells me nothing, as it doesn't say anything about how players perform in pressure situations. Baseball players with large enough sample sizes over representative portions of their careers tend to perform the same in the playoffs as in the regular season. Derek Jeter is supposedly Mr. Clutch, but his regular season BA is virtually identical to his playoff BA (actually 2 pts higher). Some guys have slightly better postseason numbers and some have slightly worse post season numbers, as would be expected by any random distribution.

 

We're not talking about regular people in stressful situations. We are talking about the elite of the elite of the elite in their sports who play in front of tens of thousands of live fans and millions of tv watchers every game. You say it is evident in sports, but other than just riding the result and claiming someone who played well in the big moment must therefore be clutch, what is the evidence? The athletes most commonly associated with clutch in almost every sport actually perform almost identically in clutch moments as they do in the regular season -- Michael Jordan, Derek Jeter, Tom Brady, Joe Montana, Larry Bird, Magic Johnson, etc. They're great players so they play great in the playoffs and in big games. Just like they play great in the regular season and in small games.

 

Well, it's obviously true you can explain the same data in many different ways. You prefer one explanation I prefer another. Thankfully world peace doesn't depend on pinning this down!

But when it comes down to it, the reason for my preference is simple: there's massive amounts of evidence in many different contexts that people perform better under pressure up to a point, after which they perform worse. That's a property of humans and should apply everywhere. It's just really hard for me to believe (can't prove it obviously) that somehow professional athletes are all immune to this.

And if they're not immune to pressure, then some will crack under it and others will not. Those that don't = "clutch". The data can be explained this way for sure. But yes you're right there is no way to truly prove differential response to pressure is what led to the stats.

 

I am not sure what planet you are living on, but through out the history of sports people and players have talked about pressure and how it picks up in the playoffs.

Now IF a person can play that game like it is just another game, then they won't feel that pressure and thus they are more likely to be clutch.

I totally agree they should play the game like it is just another game, but that just isn't reality when everyone in attendance is jacked up with extra energy.

 

I find that it's rarely the case with professional athletes outside of golfers. Most of what athletes do is muscle memory. The game at the professional level is just too fast for people to be in their own head. They have to just react. That reaction comes down to the muscle memory that's been ingrained through repetition. And studies I've seen come to the same conclusion. Given a large enough sample size, just about all professional athletes (excluding golfers) tend to perform to their averages in stressful and non-stressful situations.

 

The reason I didn't contest Fineas' statement is because of the debate Fineas and I had earlier in this thread. He likes to look at average play, combining stats from all QB's as if they were a single QB, while in this debate I prefer what you're doing Pauly in that we look at differences per QB. So really both you and Fineas are correct, but talking about different things (or using as baseline a different stat).

First of all, here are the average passer ratings in the playoffs vs. regular season from 2002-2015, that is, combining all "completions", "attempts", "yards", "TD's", and "INT's" as if they are from a single QB in each season, meaning you're looking at the average performance of a playoff vs. regular season QB here:
https://i.imgsafe.org/95f7257a2a.png

As you see, average passer rating in the playoffs tends to vary around average regular season passer rating, with more variance of course because sample size is smaller. Over that period of time, average passer rating in the playoffs is 1.4% greater than average regular season passer rating.

So that supports what Fineas is saying. However, it's not answering the question many of us are asking, which is what the average drop-off (if any) you get per QB from regular season to post-season is. That's what Pauly's HoF QB stat is showing and I think it's more relevant to this discussion because we're talking about what happens to each individual QB, not average overall performance where the QB's that are competing change (many regular season QB's don't make the playoffs or play just a few playoff games in their career).

Also note that both Pauly's stat and the stat I just posted supporting Fineas' statement are consistent with each other because you tend to have better QB's in the playoffs, meaning if the average QB performance is the same in both, then the better playoff QB's must on average be performing worse in the playoffs, which is what many of us are saying.

Final point: IF someone does decide to compile a QB-by-QB comparison of drop-off from regular to postseason, it's imperative that sample size is taken into account. That is, you don't really want to just state what the average passer ratings are in each condition because stating that alone obscures sample size (though I think with HoF QB's you might be in safe enough range because they tend to have a larger number of playoff games). You want to also compare the set of passer ratings of that QB in the regular season to the set of passer ratings in the post-season and do something like a 2-sample t test for unequal sample sizes to test whether both sets of numbers came from the same distribution, statistically speaking. For Flacco they don't, but that's the only one I tested.

 

Wait. We're not including non-playoff qb passer ratings in the regular season average are we?

Why would we do that?

 

The only way the claim that average passer rating doesn't change from regular season to playoffs holds is if IF you look across all QB's.

As I pointed out though, the data is still useful because the only way you get similar averages is if the better QB's (the playoff QB's) are on average performing worse in the playoffs than in the regular season.

 

Some of those numbers are wrong. Joe Montana, for example, has a 95.6 rating in the playoffs and 92.3 in the regular season. And then there are some notable omissions, like Terry Bradshaw, whose playoff rating is 12.1 pts higher than his regular season rating. And Kurt Warner, whose playoff rating is 8.9 pts higher than his regular season rating. And Ken Stabler, whose playoff rating is 8.9 pts higher than his regular season rating. Etc.

Also, another factor in considering QB playoff ratings is weather. There is obviously a much higher percentage of bad/cold weather games in the playoffs (Dec/Jan) than in the regular season. Those games are generally not ideal passing conditions. With playoff sample sizes so small to begin with, a career playoff passer rating can easily get skewed by one or two bad games in terrible weather conditions. There's also the home/away issue. QBs on non-great teams tend to be on the road in the playoffs. Most QBs (and teams) are better at home. So a guy who has 5-6 career playoff games, all on the road, might tend to have less impressive playoff stats, but they might be as good or better than his regular season stats on the road. There are a lot of factors to consider and the NFL, buy its nature of 1 game per week, tends toward small sample sizes.

 

As I mentioned, I suspect weather is a factor that should be taken into account when comparing regular season and playoff passing stats. The playoffs are in Dec/Jan and there are more bad weather games then. Those games generally aren't conducive to great passing numbers.

 

Hmm.. pro-football-reference doesn't have easy to parse stats to see how passer rating changes by week across the league. However, this site does list average ratings for the last 3 games of the season, as well as average for the season, for each NFL team:
https://www.teamrankings.com/nfl/stat/average-team-passer-rating?date=2016-02-08

The stuff pastes into Excel easily (hate using Excel and never do use it for data analysis, but I must admit the one thing it's good for is pasting stuff into readable format!!). From 2003-2015, the average drop-off per team in passer rating over the entire season vs. last three games is 2.1245 passer rating points, with a small standard deviation of 2.1881.

So yeah, there apparently is an effect of playing in December/January but it's too small to account for most of the drop-off seen with some of the other stats posted.

If I were trying to argue clutch doesn't exist, the big question for me for the playoff data wouldn't be how cold weather affects play but instead how strength of competition affects it. That's probably a pretty big effect, but I can't do the analysis without putting in some real work (which I won't do right now) because none of these sites have the data in the right format to make it easy to analyze.

Either way, let's keep one thing in mind: whatever identifiable conditions (e.g. opponent record, weather, etc..) help explain playoff record vs. regular season record, that alone isn't enough if you're trying to argue "clutch" doesn't exist because there are many other stats (e.g. 4th quarter rating, or rating while trailing in last few minutes, etc..) where many of these arguments won't work. Like I said earlier, I doubt one can pin down anything, but in any case weather alone won't do the trick for playoff vs. regular season differences.

 

Cool. You can name the greatest players of all time.

What about Cody Ross? What about Andre Iguodola? What about Scott Brosius? What about Danny Green? What about Joe Flacco? What about Kenny Rogers? What about Bryan Bickell? All guys that had unbelievable moments, won MVPs in the final rounds of their respective sport tournaments, etc.

On the flip side - what about Cristiano Ronaldo? Alex Rodriguez? Clayton Kershaw? Dan Marino? Peyton Manning? Kevin Love? Guys that are great, HoFers but don't have the greatest stat lines in their respective sports postseasons.

You're in bananaland if you don't think players look at specific situations as different.

You've probably played sports like I have. You can't tell me that you didn't have a different feeling/testosterone/jolt maybe even fear as certain times.

Personally, my ABs in the College World Series were completely different than an opening series in February. To act like it's not is just dumb.

I understand the argument due to stats but that time of the game just can't be analyzed through stats. That's where the human element comes in.

You're a stat guy in that regard. I'm not. We will never see eye to eye on this.

 

Stabler has been voted into the HOF. He's as HOF as Favre, who was on your list. Are you suggesting there's some relevance as to whether they have actually been through the induction ceremony? C'mon.

 

Stabler has been voted into the HOF. He's as HOF as Favre, who was on your list. Are you suggesting there's some relevance as to whether they have actually been through the induction ceremony? C'mon.

 

Those lesser players you mention have all had great moments and stretches in the regular season too. And they happened to have had some in the playoffs. Iguodala's career playoff Per 36 numbers are slightly down as compared to his regular season numbers, albeit not by much. Danny Green's are effectively the same -- a bit higher in some categories and a bit lower in others. But if there's some innate clutchness to Danny Green, why did he such so furiously in the 2015 playoffs (.344 from the field and .300 from 3)? We've talked about Flacco do death and his career playoff numbers are not appreciably better than his career regular season numbers. But if he has some innate clutchness, why did he such so bad in clutch situations last year? Scott Brosius is a career .257 hitter who hit .245 in the playoffs. His regular season slugging% is .422 and its .418 in the playoffs. Those are minuscule differences, but if anything suggest non-clutchness according to your theory. Kenny Rogers had a career regular season win% of .584, but it dropped to .500 in the post-season. His ERA was pretty much the same (4.27/4.15).

Your great players with lesser playoff production are pretty much guys who had some great games and some bad ones in the post-season. But if there was truth to the clutch/choke thing, there should be more consistency. Clutch guys should consistently be better in the playoffs and chokers should consistently choke. But that's really not what we see. I'm not a soccer fan, so I won't address Ronaldo. I just don't know enough about it, nor do I know where to find all the stats. Marino had a mixture of great playoff games and duds. He had 5 playoff games with a passer rating over 107. How could he have done that if he was a choker who routinely succumbed to the pressure of the big games? Same for Manning. Kevin Love's playoff career is pretty much just 2016 (and a few games in 2015) on a team on which he is a supporting piece with 2 very high usage isolation players. And he's just not the same guy he was in Minnesota, for whatever reason.