Debunking the hot hand is worth it just to show up blowhard color commentators.
Probably best to measure a feeling with a self-reported prospective yes/no question. Then you can use it in whatever analysis you want.
I think he gets that its a feeling, he says so.
I can believe hot hand, I've felt it on occasion- couldn't it be something as simple as whether you're well rested for a particular game?
The clutch factor is total bullshit though.
Why is the clutch factor bullshit? Do some players have a tendency to choke?
I'd guess both clutch and hot hand are hard to measure in the very best players, for obvious reasons.
5 - I thought the current sabermetric view was that clutch performance is a real thing (that is, there are a few players who perform particularly well over a long period of time in high leverage situations, suggesting that there really are players who are "good in the clutch") but (a) it's not a particularly important skill, totally swamped by other, more important skills and (b) it's often not possessed by players who have a reputation for being "clutch."
Dude, K-Dru, you can't measure the hot-hand statistically (at least not in any straightforward way) because it's a feeling.
I've thought about this, as somebody who plays basketball and thinks that math is generally a useful source of information. Here's what I would say:
1) Personally I'm inclined to believe that the "hot hand" is a feeling which doesn't correspond to any externally measurable effect -- even though that feels like a deeply counter-intuitive conclusion. But more on that in a minute.
2) If you accept that statisticians haven't been able to find evidence for a "hot hand" that implies one of two things (a) that there is no measurable "hot hand" or (b) that the statisticians haven't figured out the right way to measure a "hot hand." The tension between those two possibilities is why it's a question worth returning to.
3) The statistics don't say that people don't have hot or cold streaks, they just claim that hot or cold streaks don't occur any more frequently than they would if makes or misses are random independent events. That tells us nothing about the subjective feeling of being in a hot or cold streak.
4) So, I'm inclined to believe that the feeling of being in a hot or cold streak is mostly (but not completely) a retroactive interpretation of events (and subject to all of the problems with humans over-interpreting patterns, and with confirmation bias making us inclined to forget the times when our beliefs are contradicted and remember the times when they are supported). I say this knowing full well that I've had days when it felt like I couldn't miss, and days when it felt like I couldn't hit anything -- but I'm sure there were also days when it felt like I couldn't miss for 4 or 5 shots and then I did miss a few and just forgot about that feeling.
5) Also, as a participant you have access to more information than an observer would. The statistician just measures whether or not a shot goes in, you know how balanced you were going up for the shot, and how it felt leaving your hands -- I think this can add to a dynamic in which the shot feels good, you mentally think, "that's going in for sure" and then when it does go in it validates your prediction, and that also helps explain where there can be a subjective feeling of a "hot streak" even if those streaks don't occur any more frequently than they would by chance. It may be somewhat random (unpredictable) whether or not the shot is one that will feel good coming off the hand, but that tactile experience can make the outcome (after releasing the shot) predictable.
6) I don't know if anybody has done this study, but I think you might be more likely to find longer than expected hot or cold streaks among players with less experience -- when there is more of a feedback from a feeling (or lack) of confidence.
Does an unusually intense/stressful situation cause you to (1) perform worse than usual due to the additional stress ("choke"), (2) perform better than usual (which suggests you're not really playing at your peak all the time, which isn't surprising really), or (3) not impact your performance predictably (because your job is typically full of very intense/stressful situations, so moments of heightened intensity aren't a noticeable deviation)?
It would frankly surprise me if all three weren't real, observable and common reactions. Reaction (2), of course, means you're "clutch".
9 -- yes, there's a statistic in baseball to measure that pretty well. It turns out that few players are either unusually clutchy or unusually non-clutchy, and even for the ones who are it turns out not to matter much and it's hard to predict for a player for the future.
It turns out that few players are either unusually clutchy or unusually non-clutchy, and even for the ones who are it turns out not to matter much and it's hard to predict for a player for the future.
But, following on 8(6), I bet that's much more true for professionals than for players at lower levels, because among professionals reaction (3) so obviously dominates. Very few people will be professionals if their reaction to stress is to choke. And very few are good enough to be professioanls and yet not play at their best level (or very close to it) consistently. So almost all professionals fall in bucket (3).
I feel like I experience the hot-hand effect plenty of times when playing quizbowl, but probably it wasn't statistically significant. Also, I think that having this reaction makes me officially too dorky to take part in a sports thread.
I have the same skepticism sometimes about the concept of being a good test-taker versus being a bad test-taker. Are you sure the good test-takers aren't actually, you know, better at solving the problems on the test?
In practice I've barely ever run into a student who consistently performs awfully on a test, relative to how I believe their understanding to be.
Like everyone who plays a lot of sports, I've experienced the "hot hand" phenomenon many times and emotionally it seems crazy to deny it.
But there's so much demand for data showing a hot hand effect, and so little actual such data, that I'm fairly sure there is no hot hand effect. The same way I disbelieve all studies "proving" the existence of gender differences on principle because there's such high demand for studies proving the existence of gender differences.
The "good test-taker" effect I definitely believe in, though. Knowing how to guess and knowing how to BS for maximum partial credit are real things.
Possibly math tests are less subject to that kind of thing, and possibly I just get snowed by good test-takers into believing they've got inflated ability.
--Surely someone, somewhere, has linked to the Samuel L. Jackson reaction to Sam Rubin's question? Surely!
Former colleague (African American) posts three old pictures of himself on Facebook. Former colleague II (white guy) comments: In photo No. 2, you look just like LeVar Burton!
No. He did not look like LeVar Burton.
Need to do fMRI to get at the emotional part but unfortunately the only athletic thing you can do in an MRI is have sex.
Actually, if y'all RTFA I am pretty sure I linked to a statistical study back during the Maverick's Championship run that showed the Mav's had three of the league top ten clutch shooters. 4th quarter, close games, shooting percentage.
NBA Database 2010-2011 to play with
Here are the 2010-11 league leaders in clutch plus-minus:
Kidd, DAL +119
Terry, DAL +118
Nowitzki, DAL + 116
Chandler, DAL +92
Westbrook, OKC +90
Here are the 2010-11 league leaders in clutch plus-minus per 48 minutes:
Marion, DAL +38
Nowitzki, DAL +38
Kidd, DAL +37
Terry, DAL +37
Williams, CLE/LAC +36
Chandler, DAL + 34
Here are the 2009-10 league leaders in clutch plus-minus:
James, CLE +116
Nowitzki, DAL + 102
Williams, CLE +95
Kidd, DAL +87
Terry, DAL +71
Halford will be along to say it was the coaching.
20: Lee has gotten so many comparisons to random black celebrities, but our favorite is the woman who said, "Please don't take this the wrong way, but I think Lee looks exactly like Angelina's daughter Zahara because in every picture of Zahara she's scowling exactly like Lee does!" That was a resemblance we could see once she pointed it out and that I've enjoyed ever since. Will Smith, not so much.
23: My AA former colleague is a very good sport about ignorant things said by well-meaning white people. Don't ask me how I know this.
Halford will be along to say it was the coaching.
One simple explanation for how it could be coaching: I remember seeing that one reason why the Mavs did better in the playoffs than people expected was that they had played a variety of inefficient line-ups during the regular season, and that if you looked at the regular season point differential for the line-ups that they used in the playoffs they looked like a much better team.
If Carlisle was playing his best line-ups in crunch time then everyone in those line-ups would have better on-court stats in crunch time than they did in an average minute, even if they didn't improve their play, just because they were playing with a better unit.
Also, what exactly is "+/-" being used for there? Comparing crunch-time line-ups with Dirk to crunch-time line-ups without Dirk? When would they not have Dirk on the floor during crunch time?
When they don't have the ball, of course.
25.last: link #2
I am not pasting any more from it.
Or link #1, which has a +/- column, and I am sure some explanation.
Since ogged fails me, I have to go to Drum to get to the original article.
||
So LinkedIn is constantly suggesting that I "connect" with people I already tried to connect with.
Does this mean:
- those people already "ignored" me, but LinkedIn is unable to keep track of that fact or wants me to be annoying?
- those people never saw my invitation, because they don't check their accounts?
- those people never saw my invitation, because I didn't do the invitation properly?
- the invitations never went out, because even though the people popped up as connections, LinkedIn doesn't think they're close enough connections for me to be able to contact them?
|>
Speaking of 5-man lineups, is there a website where I can get a list a nice list of each team's most used 5-man lineup with their points per position and opponents points per possession? At 82games it requires way too much clicking through, and at NBA stats (which can be filtered by minutes played) they only give per game stats not per possession (I'd settle for per minute!).
Or link #1, which has a +/- column, and I am sure some explanation.
Reading a little further I think that's "on-court" performance. Which makes complete sense, I'm used to seeing used to generate an on/off court differential, but that isn't necessary there.
I do notice that somebody in comments makes the point that seeing all those Mavericks on the top of the list just means that they were, collectively, the best crunch time unit in the league, not that each of them was individually the best crunch time player in the league. Which goes along with the point I was making about line-ups.
Also is there a good place that predicts playoff series win percentages between different teams based on some simple extrapolation from their best 5-man units (which play more in the playoffs)?
Speaking of 5-man lineups, is there a website where I can get a list a nice list of each team's most used 5-man lineup . . .
B-R.com has "lineups" data for each team. For example, here is OKC this season.
28: The other day I logged in to LinkedIn and there were like 15 link requests from random people going back over a year that I had never seen before. It was odd.
With math tests, to me it felt less like "having a hot hand" and more just that adrenaline made me much faster. Team Mathcounts was 4 people working on 10 problems, and we had one person who handled the assignment logistics but basically when you solved one problem you got one of the new problems other people hadn't solved. In practice it'd be maybe 4/3/2/1 or something, and then in competitions I'd do like 6 or 7.
Off-topic, but then, I suppose this is a 'various' thread: Valentine's Day trolling in the WSJ. Oy.
32: Like 82games though that's all the lineups for one team. I'd like to look at the most important lineups for all the teams without clicking through to 28 separate webpages.
couldn't it be something as simple as whether you're well rested for a particular game?
Way back when, I saw myself as a clutch performer in HS sports until one of my friends said "you're not very good but you're a little better when you concentrate."
I'd like to look at the most important lineups for all the teams without clicking through to 28 separate webpages.
Perhaps the Lineup finder can help. I haven't used it much but would this work?
Way back when, I saw myself as a clutch performer in HS sports until one of my friends said "you're not very good but you're a little better when you concentrate."
That's what seems logical to me. Some players concentrate all the time, and some don't concentrate when it's not important. So the latter, who may seem to be lazy, are the real clutch performers, though they really reach that status by being non-clutch anti-performers.
Well, the post was about "hot hand" rather than clutch, anyway. My mistake.
Umm, hot hand. Well, even after watching that maniac Terry for years I tend to doubt it. Except.
If a lack of confidence in your swing or shot can reduce your percentages, why wouldn't confidence help? Maybe it is a long run thing. Are they really saying that having hit in 50 straight games, Joe Dimaggio had no better than usual chance of hitting in the 51st? Was Ted Williams a great hitter because he thought he was? Bobby Fischer was on a fucking roll, dude.
Of course, conditioning, effort, training skill and practice are more important. I think.
When I play pool, I have a hot hand as a quadratic function of how many drinks I've had.
41: I'm skeptical. The function falls off a lot more rapidly than it rises, in my experience.
When I play pool, I have a hot hand as a quadratic function of how many drinks I've had.
This is bowling, for me.
38: Perfect, thanks! I bow to your Basketball-Reference-foo (when I looked there I couldn't even find where the lineup data was kept).
41: For me the function peaks at 2 beers but my perception of my own awesomeness at the game continues to rise for some time.
20/23/24: Remember how George Stephanopoulos misidentified Bill Russell as Morgan Freeman? This has become a running joke in our house. Whenever there's a middle-aged AA man on TV, one of us asks the other if it's Morgan Freeman (or Bill Russell). White people are terribly stupid sometimes.
As far as football goes, just want to be on the record that Michael Sam was originally projected to go in the 8th round. That was bumped up to 3-5 by his fans in order to show a quick drop.
Sam is too short and can't switch to linebacker. He tried. He is exactly the kind of player that can bust from college to NFL. Let this truth control the forthcoming narrative.
Cosby roofied women? I heard "scandal" but didn't go look. Never can understand such stuff, but then for me sex sucks unless the partner is attacking and devouring you like a tiger on meth. Lucky for me, that happens once a decade or so. Okay, not recently.
Desiring sucks, but being desired is great. In that way it is the opposite of love. Loving is good, but being loved is a PITA.
Listen to the old man, kids, From the perspective of surviving to 60+ years I have attained such wisdom and experience that I can pass on to you the lesson I have learned:
"Why did I even bother?"
I was once at a wedding where the night before we went out. We stopped to play pool. I was warned beforehand that that the brother of the bride was "really good". I happened to have the optimal amount of drinks just as we started playing, and I crushed him. He had one turn, and then I ran the table. Then we did some other stuff, so we only played that one game.
I like to think that somewhere out there there's a guy who now has a completely inflated notion of how good of a pool player I am.
I thought there was some stat for Jason Kidd showing he shot free throws better at the end of close games. He wasn't a great free-throw shooter overall.
One time, on my birthday, I crushed all my friends at dominoes.
35: That woman wrote that same op-ed in the Princeton University paper while her son was attending.
35, 51: An appropriate response. (via the other place)
The link in 52 is great. This, in particular:
Look, did you ever watch "Ally McBeal"? Remember that dancing baby? Well, if you do, bad news: YOU ARE TOO OLD TO GET A MAN. GO BACK TO GRADUATE SCHOOL AND SET UP A BIG BEAR TRAP AND PRAY.
What a nicely done bit of joke-misdirection!
I've heard of Petri before but now I can't remember for what. She's very funny.
Petri has a regular gig at the WaPo. She's sort of hit-or-miss but when she hits she really hits.