Separate names with a comma.
Discussion in 'Miami Dolphins Forum' started by Pauly, May 30, 2016.
lol. You can't grasp it, ergo the concpet is stupid. Typical.
I was talking about "cold doesn't exist because it's just the absence of heat" not what you are talking about, which I am familiar with (the discussions re: no true vacuum etc.)
I can grasp it just fine. In fact, I grasped it so well I debunked it. Which I notice you have not rebutted.
Choking is the absence of clutch.
Look, I know all about cold doesn't exist as a silly thought experiment. Cold exists. It's not a primary quality but a secondary one. Just like your sense of smell. And your memories. And your emotions. Your undying love of Ryan Tannehill (which I know exists).
Not everything that exists must extend into physical space.
You didn't debunk it. Thinking you did proves you don't understand it. Again, typical. Keep spinning though.
I appreciate the explanation. But as I suspected, it sounds like the short answer is that there is no agreed or accepted scientific definition of clutch. I don't agree that yours is likely to become that for the reasons I have stated in the past. While it doesn't make us "right" necessarily, the fact that there are several of us here that strongly object to your definition suggests that it is far from having any universal acceptance and some future scientific consensus embracing your definition is unlikely.
I agree with the general concepts that if one is to study clutch you need to control for as many variables as possible. That's true of any scientific experiment of theory. IMO, that makes use of any QB stats as compelling evidence of clutch pretty hopeless. QBs do nothing on their own. Everything is affected by the OL, receivers, the completion, the play calling, the game situations, the coaching, the location (home v away), weather, etc.
I'm not a golf expert, but gold seems like a reasonable place in which to evaluate clutch. And it sounds like a bunch of studies have been done on golf, many of which agree with my position. I suspect there are others that don't. I haven't seen them so I can't speak to them. I think FT shooting in basketball and FG kicking in football are also reasonable places to look. There will be some variables, but I think they can probably be mostly controlled for. Serves in tennis (% of faults, % of aces, etc. in certain situations) might be another place to look. I don't know what kind of stats they keep for that. Penalty kicks in soccer are probably worth looking at too. Baseball has variables, but there is a wealth of clutch stats there and typically large sample sizes. Not perfect, but certainly better than QB stats.
Part of the problem with clutch and your notion of "fluid" definitions is that people seem to have ideas of about who is clutch based on reputation and isolated examples and then seem to look for stats that confirm that preconceived notion of clutchness. I don't think that is scientifically r4igorous or defensible.
The only thing I can tell you for sure is that If Ryan Tannehill what's top five on that list? You would have a couple of guys 100% believing all things "clutch"
Sent from my iPhone using Tapatalk
Actually, no. Cbrad has redefined "clutch" into doing better than the average in a situation. Thereby, a QB can perform worse than his norm, but still perform better than the average QB, and he considers that clutch. Choking is doing worse than normal in a situation. Therefore, his redefinition of clutch is actually your definition of choking (I'm assuming, as you haven't defined choking).
Anyway, it's beyond the pale to say that a QB performing worse than his norm in certain situations is actually "clutch" in those situations. You guys would all be gushing over the emperor's new clothes.
Also, cold is nothing like smell.
That's simply not true.
Further, as I said earlier, if someone is "clutch" all throughout a game, then it would no longer be "clutch." It would simply be his normal performance. Pauly, I think, tends to agree that good/great QBs are good all throughout a game, on average.
Which is exactly why good/ great QBs are also usually regarded as "clutch," because they play well in all situations...hence they are good/great.
Forget about what definition science may converge on in the future. I'm speaking from experience there but have no proof of what will occur.
There's no question the definition I proposed is a measure of differential response to pressure that works both for individual sports and team sports. There's also been no disagreement here that there's a spectrum where on one end it's "clutch" and the other end it's "choking" (that is, it's a spectrum, not just binary). So my proposed measure is quantifying the degree to which one athlete is more clutch than another or chokes more than another along that spectrum (quantifying how close to one or the other end of the spectrum one is). That's also not up for debate.
The only issue you and some others here can point to is whether the origin on the axis should be at "average drop-off". You have to put the origin somewhere and that seems like the most natural place, but you don't like that.. so what? We're measuring relative degrees of clutch or choking no matter where on the axis you are. In the end, your (and others') critiques are really so minor it can be dismissed as irrelevant to the main argument that some players are statistically speaking more clutch than others (that is.. not due to random variation).
There's plenty of disagreement on the spectrum. I don't believe clutch is a thing.
You have the normal, or average, play of the QB. You have evidence that QB rating drops, on average, in certain situations. Sometimes you see a QB make a terrible play, i.e., choke, in a pressure situation. That doesn't mean, however, that if he doesn't choke, that he is now "clutch."
1) Let's note that Fin D flat out disagrees with you in post #756 where he agreed that clutch and choking are two different ends of a spectrum. He was of course wrong about my analysis not using that spectrum, but he does think that the spectrum exists.
2) We're not talking about whether anything satisfies the definitions of "clutch" or "choking" here, just what those words tend to mean in sports. And whoever you talk to (except you maybe) people say "clutch" and "choking" are opposites of each other in sports and that there's a degree to which you are one or the other. So your disagreement I'm dismissing because it's not how we generally use those words, regardless of definition.
Clutch can only happen in pressure situations, so that's a flawed analogy.
Peyton Manning is a great QB, but I wouldn't call him clutch, he has a tendency to succumb to pressure.
Joe Flacco is no where near as good a QB as Peyton, but is clutch.
I'd take Flacco in the playoffs over Peyton in his heyday.
No, there is plenty of question about whether your definition works for any sports, individual or team. I absolutely disagree that everyone who performs better than average in a pressure situation is clutch and I have said so numerous times in this thread. I also strongly disagree with almost all of the attempts I have seen to take team stats inherently influenced by numerous players, coaches and conditions and try to make them a measure of clutch for a single player. I have said that numerous times as well.
Theoretically, there is that spectrum between clutch and choking, but as I have said numerous times, I think it essentially disappears in the realm of professional sports. That is because the pressure situations one sees in professional sports are essentially all within the peak performance portion of the pressure/performance curve. If you held a gun to the head of an NBA player and credibly threaten that you would shoot him and his entire family if he misses a free throw, that would likely be beyond that peak performance portion of the curve and I expect most players would probably shoot worse than normal. But that kind of pressure doesn't exist in professional sports.
No, average dropoff doesn't seem like a sensible measure. Take a 87% free throw shooter (Player A) who shoots 10 free throws in clutch situations and makes 8 of them. Then you have another guy (Player B) who normally shoots free throws at 52%, but makes 6 of 10 of his clutch free throws. What standard are you using for the "dropoff"? The player's own normal FT% rate or the league average? If its the player's own normal FT%, then your definition would say Player B is more clutch. But is he? He's still missing more clutch free throws than Player A, so I don't really see how he's more clutch. On the other hand, if the standard for measuring dropoff is league average, then Player A is probably still a better FT shooter in clutch situations than the league average and Player B is still worse. Does that make Player A more clutch? I don't think so at all. He's just a better free throw shooter, whether its a clutch situation or a non-clutch situation. So your definition doesn't work for me either way. And in this example, and many, many other sports examples, we are dealing with a very small sample size of just 10 FTs. In reality, both guys are making about the same number as you would expect in those clutch situations. An 87% guy would be expected to make 8 or 9 out of 10. He can't make 8.7 out of 10. When he makes 8 it is well within the expected range. And even with that expected range, if the guy shoots 10 sets of 10, he'll probably have a set or two when he misses 4-5 and probably have several sets when he makes all 10. Same concept for the 52% guy.
When you take those inherent problems and apply them to playing QB, an activity inherently influenced my a myriad of variables and the results are even less reliable as a measure of clutch. And that's for actual passing stats. When you move into attributing wins and loses in close games to the QB and doing so based on small samples, you start getting into ridiculous territory.
Fin D is entitled to disagree with Resnor who is entitled to disagree with me and vice versa. We all have our own thoughts and opinions. That we don't agree on anything does not strengthen your position at all.
Again, I think you are being very lose with your "scientific" definitions of clutch and what you say they tend to mean in sports. There are many who disagree with your position, including several in this thread. Every player with any reasonable number of attempts will have good moments under pressure and bad moments under pressure. Not every good moment shows clutchness or makes that player clutch. And not every bad moment is a choke and makes that player a choker. Yes, people may use those labels, but I think they are essentially meaningless. For many of us who don't believe in "clutch" as some inherent characteristic, part of the reason is that you just don't see players who consistently and systematically perform better in pressure situations than they do in non-pressure situations in any decent sized sample. And you don't really see players who systematically perform notably worse in those situation with any decent sample size. Every example I've ever seen offered can be and has been picked apart as any kind of proof of clutch or choke. Usually, as time goes on and the sample size increases those numbers revert to the player's overall averages. Hot streaks get exposed as just that and the same kind of hot streaks that the same player had many times in non-pressure situations. And chokes get exposed as just a cold streak at an inopportune time.
You're not understanding my argument.
First, as you say there is theoretically a spectrum between clutch and choking. Now.. forget completely whether there are any chokers or clutch performers in sports. That continuum exists independent of whether anyone satisfies the definition.
Second, for any two points on that continuum X and Y, if X<Y, then Y is more clutch than X, or equivalently Y is less of a choker than X. Replace X and Y with numbers and you can see the same thing: Y is more positive than X regardless of whether both X and Y are positive numbers, negative numbers or a mix.
Thus, ANY measure using that continuum will separate players into "more clutch" and "less clutch", or "chokes more" and "chokes less". What I'm saying is.. don't say X = choker and Y = clutch. Say Y is more clutch than X, or X is more of a choker than Y.
Finally.. my proposed measure operates on that continuum, and of course works for team or non-team sports. Thus, my measure is measuring the degree to which any athlete is more or less clutch than another. All that remains is to make sure the result was unlikely to be due to chance, but we covered that already.
So yes, this approach works once you stop thinking "he is clutch" or "he is not clutch", and just replace it with "he is statistically more clutch" or "he statistically speaking chokes more".
You missed the point completely. Fin D often says one thing but then acts like he didn't say it. That was bait.
Not sure how accurate these stats are, but assuming its true (and allowing for the fact they may be completely wrong)
People said the defense carried Tom Brady those first few years vs. Peyton Manning. But I find this interesting. Through their respective first 24 playoff games:
So why during their early years was Manning's team success not so great, and Brady's success so great? This is where I think it is folly to look at stats to argue clutch. Both their playoff passer ratings were similar, but Manning had wild swings in passer rating per game, while Brady was more consistent.
This is what is hidden by overall stats. Manning has a much wider standard deviation when you plot his individual game ratings.
Also keeping in mind, Brady plays a lot more Championship games and Super Bowls, while Manning played less of those, and more wild card games. Not all playoff games being equal. Manning played more games against weaker teams while Brady played tougher teams and in turn, usually (not always) tougher defenses.
Manning played in 7 wild card games vs Brady's 3 according to these stats. So his 102 in the WC round props up his playoff passer rating pretty highly.
So in the end, you say, "oh they have similar playoff passer ratings" but that's not the whole story.
Stat nerds, nerd out in this reddit thread (and tell us if it's accurate)
You ignored the rest of my post explaining why I don't think it is really sound to say Y is more clutch than Y or vice versa. And what stat or measure are you using in your continuum? As I explained, whether it is compared to the player's own norms or the league norms it is flawed in many ways. And when applied to team sports it can quickly venture into the absurd. One of the big problems (for both team and individual sports) is that it gives very, very different answers depending on the stat you choose. Different "clutch" stats suggest that different players are clutch. Some players may appear clutch by one measure but not several others. And that completely undermines the notion that clutch is some kind of inherent trait that is consistently exhibited by players. If clutch was a real thing and consistently exhibited by clutch players, we would see the same players show up as clutch according to virtually every measure. But we don't see that at all.
Post #759, where he pretty much explicitly says clutch didn't exist.
Now, of course "clutch" would be the opposite of choking, just like "cold" is the opposite of "hot." However, he's also arguing, correctly, that "cold" is simply the absence of heat...it exists when we remove heat. Just the same, he's arguing that "choking" can be real, and would be the opposite of "clutch," but that doesn't mean that "clutch" is something that exists. Someone appears to be "clutch" simply because they don't choke, even if what they did is what they always do.
And I don't have to agree with everything someone says to agree with them on "clutch" being mostly a fabrication of the media, combined with people wanting to believe in a superstar athlete somehow possessing this seemingly superhuman trait. People want to believe in greatness.
Also interesting is number of GWD/4QC. Ideally you'd have none because you'd win all games by 30. But it really tests your mettle to lead teams on GWDs and 4QCs. With all of the playoff games Manning has played, and all those losses, he's had 2, one of them being this past year.
Brady has 7.
It is clear, Peyton just isn't as superb as he is normally in the playoffs, and in pressure situations. And he's one of the GOATs. Also keeping in mind, Manning gets to play home games in a heated dome, while Brady is outside in the cold.
But then again, Brady is just plain better even if you consider all of Manning's volume stats. Playing 50% of his games in a dome does wonders.
Other than Randy Moss and Gronk, Brady has played with no name Running backs (Corey Dillon was pretty good) and no name mostly midget wide receivers. Consider that when you compare playoff stats.
This article was written in 2014 (before 2015) but consider this:
Again, Brady and Manning define clutch, and choker.
You seem to have ignored the initial premise of your post -- "People said the defense carried Tom Brady those first few years vs. Peyton Manning" -- by failing to address the playoff performance of either the Colts defense or the Pats defense during their respective careers (or at least their first 24 playoff games). I'm not sure what those numbers will say, but in light of their similar passer ratings, if you are using win% as a measure, it seems like taking the defense into account would be a good thing to do.
As for those 4th Q and OT when trailing in the playoff stats, Brady does not stay in line with his normal production. His rating drops to 84.3, which is 12.1 pts lower than his regular season career rating. And no, it is not the case that all or most QBs have similar drop-offs in those situations.
I'm showing 88 for Brady.
Again, you are using volume stats. Click my second link to do a season by season analysis of the defenses they played with. Peyton Manning reserves his worst game of the year, for the playoffs usually. 2005, #1 defense in the league. 3 points through 3 quarters, one and done that year.
But go ahead and click that second link and go down to the bottom where they go year by year and game by game in the playoffs.
As for the "dropoff" have you accounted and weighted these numbers to other QBs who played so many games against the top teams in Championship and Super Bowl teams?
I showed you how Manning has a 102 rating in the WC phase because he gets to beat up on crappier teams. Brady is playing Seattle in the SB (#1 defense) and Broncos last year (also #1 defense) etc. etc.
But again, I stress the faults of looking at mass stats over a career to judge clutch.
You need to go through each game and see how he performed.Did an exceptionally poor performance by the QB lead to the team's exit from the playoffs? Did he have a chance late to win or tie and blow it? That's what you need to look at.
Yeah I was trying to make sure the argument about the validity of saying "more clutch" or "less clutch" regardless of where you are on the continuum was first settled because that issue really shouldn't crop up again (that is, where the origin of the axis is doesn't matter). And you can use any stat where you have data in "less pressure" vs. "more pressure" conditions using my definition.
Now.. getting to your example in the 3rd paragraph of post #857, I'll agree that the measure I'm proposing gives a result you'd ideally not want. That is, it would say player A is more clutch than player B even if A's absolute level of performance in pressure situations is less than that of B, provided that A has a smaller drop-off than league average in pressure situations than B.
So.. what to do? First of all, one has to note that what seems like the easy solution of weighting two measures: my definition of clutch + measure of absolute performance, probably doesn't work because I'm pretty sure you'll get MORE undesirable results that way (too many cases where you're not sufficiently weighting differential response to pressure).
So all I can say is that if you can think of a definition that works better, I'm all ears. Keep in mind it has to operate on the continuum described, so it can only measure differential response to pressure (because that's the common underlying concept here), and it has to work regardless of whether anyone satisfies the definition in reality or not (we can just create hypothetical examples like you did to test how well it corresponds to intuition if necessary).
My point is it doesn't need to be real to determine whether a definition is a good one or not.
No I don't. WTF is wrong with you guys? I mean seriously......I could not be more clear and specific.
AND you are not using that spectrum. I've laid that out clearly as well.
"science", other people say: Clutch is the absence of choking
YOU say: Clutch is choking slightly less than a league average.
THOSE. ARE. TWO. DIFFERENT. GODDAMN. THINGS.
Why should that issue not crop up again? I don't think it's a continuum at all. It's more like a flow chart. Start with a box that says "pressure situation" if "Yes" continue to performs "within normal average" or "below normal average." If "within normal average," continue to box that says "normal." If "below normal average," continue to box that says "choke."
If "No," return back to the box that says "pressure situation."
You simply can't say that a guy playing within normal averages is "clutch."
Really, you've redefined "clutch" to mean something else, so that you can now prove "clutch" exists.
Tim Tebow is regarded as clutch...
lol.. glad you responded this way. I think everyone can see you agreed clutch and choking are at the opposite ends of the same spectrum in post #756. Can't get clearer than that. Means I can dismiss your future responses on this topic if you're not even going to stay consistent at such a basic level.
That IS a continuum as long as there is variability within each category. That is.. do all players within the category "within normal average" have precisely the same level of ability? If there's any difference whatsoever, what you've done is place thresholds on a continuum. The thresholds defining the category boundaries are fine with me, but the continuum exists.
You have to compare each player to his normal average range. Comparing to the rest of the league is pointless. I mean, I understand why you want to do it, as you're looking at QB As numbers vs a defense to see how his 82 rating compares to other QBS. However, if he's usually carrying a 105 rating, I really don't care how that 82 compares, he doesn't qualify as clutch.
Unless you redefine clutch.
For instance, let's say QB A carries a 105 rating on the year. For three quarters, he's at an 84. Let's say the average QB in the league carries an 80 against this defense. Then, in the fourth, he plays to a 101 rating. You'd define that as clutch. Actually, you have to define the whole game as clutch. But I'd argue that he played to his normal average range in the fourth.
QB B plays to a 95 on average. He plays the same defense the entire game at a 90, and his team loses. In the fourth, he plays to an 85. Is he " clutch" even though he lost?
That's another example of media pumping this idea of "clutch."
Even though the article is stating that kickers kick worse in those situations, they call them "clutch."
If you don't compare to other QB's, what's the criteria for saying the QB is "clutch"? The problem is you'll have an arbitrary threshold, and you won't know what that threshold should be for different stats.
The utility of comparing to league-wide drop-off is that you get that "threshold" (you can have multiple ones too btw if you want multiple categories) objectively.
WHAT ARE YOU TALKING ABOUT?
You stopped making sense a few posts ago.
I said "science" (which isn't me) and "other people" (which is by default, also not me). So again....
WHAT ARE YOU TALKING ABOUT?
Me saying there's no such thing as clutch only an absence of choke is what started you and others losing your collective minds about cold and heat. I'm not backtracking and I'm not going against anything I've said.
All I'm pointing out with you, is that you CBRAD, are not saying the same thing about clutch that you claim "science" says. I haven't wavered or faltered from that.
Also, you don't want to get into a fight about who is contradicting themselves...considering you say that your definition is basically the same as "science's" in some posts, then claim its different and that's ok because definitions are like hypotheses in other post.
You say you can change the definition of clutch here:
Then here you make the argument you're not changing the definition:
Then in the same damn post, you make the argument for changing the definition again.
Then on this page, you say again how your definition and s"science's" definition are basically the same.
If he were so "clutch" he'd still be playing in the league. Again, I believe Tebow to be a product of media sensationalizing him in his wins.
Did he suddenly throw a better ball late in games? No. Tebow, bottom line, was not a good QB. Skip Bayless with "all he does is win" garbage is exactly what I'm talking about.
People want to believe in something bigger than themselves. They latch on to "clutch," but it really isn't anything.
Quite simply, it is pointless. I'd say a guy is clutch if over his career, he played better late in games than in the other quarters. That would be evidence, perhaps, of upping his play.
I made no comment about the article. I was simply providing the source.
The important fact is that a University study shows a 6% reduction in kicking accuracy in high pressure situations. Which goes back to the point repeatedly made by cbrad in this thread.
And, they seem to be using cbrad's definition of clutch too.