Hot-hand fallacy, agree or disagree?

[Fritz88]

The human mind's powerful ability to detect patterns often leads us to detect patterns where there are none.

Exhaustive research proved that prior shot performance did not change the likelihood of success on the upcoming shot.

The hot-hand fallacy or representativeness heuristic may be so well engraved into our decision processes that even scientific training may not fully eliminate it.

What do you think?

Do you think that a player who is having a hot night is your go to guy or is the other player who is not guarded but equally skilled may have a better chance of scoring?

http://en.wikipedia.org/wiki/Hot-hand_fallacy

[Shaid]

Hot hand, confidence does factor in.

[KC native]

Reversion to the mean is all I have to say.

['Hamas' Jenkins]

How exhaustive is the research and over how large of a sample-size? There are certain occasions when a player's shot is just locked in and other cases where a good shooter can't hit the ocean if falling out of a boat. Even players of equal skill are not equal consistently. If the player was able to perfectly mimic his mechanics every time I would be more likely to believe this, but the imperfection is key in this case, as there are days when our timing is simply better than others. That itself is an outlier, but it's an anomaly that leads to a predictable, beneficial outcome over the short term.

[ClevelandBronco]

A 40 percent lifetime shooter has a 40 percent chance of hitting his next shot.

[trndobrd]

If the study is of free throws or "a controlled shooting event", then the "hot hand" would be less relevant. In actual game play, when variables such as defenders, player energy level, delivery of the ball to the shooter*, and a hundred others, can give a player a decided "hot hand" on a given night.


*delivered to a post player where he is most comfortable or to a shooter in a position where he is square to the basket.

[Fritz88]

How exhaustive is the research and over how large of a sample-size? There are certain occasions when a player's shot is just locked in and other cases where a good shooter can't hit the ocean if falling out of a boat. Even players of equal skill are not equal consistently. If the player was able to perfectly mimic his mechanics every time I would be more likely to believe this, but the imperfection is key in this case, as there are days when our timing is simply better than others. That itself is an outlier, but it's an anomaly that leads to a predictable, beneficial outcome over the short term.

That's what I thought as well. I'd go with the hot hand in the short term but to quote the writer

"Exhaustive analyses show that it is gloriously, emphatically wrong!" In one analysis of the shooting of the 76ers and Boston Celtics.... researches found that immediately prior shot performance did not change the chance of success on the upcoming shot"

The writer himself says that this is one of the hardest findings that his audience would accept.

[Fritz88]

Reversion to the mean is all I have to say.

Even in a one game basis?

[cdcox]

According to an article cited in the wikipedia link in the OP, basketball streaks of hits and misses cluster about how you would expect them to if the outcome of each shot was completely random. That's pretty strong mathematical evidence that there is no such thing as a hor hand.

[Lex Luthor]

The Wikipedia article is a bunch of crap and is based upon a faulty assumption.

The "hot-hand fallacy" (also known as the "hot hand phenomenon" or "hot hand") is the fallacious belief that a person who has experienced success with a random event has a greater chance of further success in additional attempts.

There's nothing random about a basketball shot. It's a skill. Just because a player makes some shots and misses others doesn't mean it's a random event.

Whoever wrote the article is an idiot.

[cdcox]

The Wikipedia article is a bunch of crap and is based upon a faulty assumption.



There's nothing random about a basketball shot. It's a skill. Just because a player makes some shots and misses others doesn't mean it's a random event.

Whoever wrote the article is an idiot.

Someone who is more skilled will make a greater number of shots. To simplify things, let's just think about free throws. An 80% free throw shooter will make about 80% of his free throws. He is highly skilled. But the chances of making any given free throw is a random event, in that it has an 80% chance of going in and a 20% chance of missing. I am less skilled. I would do well to make 30% of my free throws. For any given free throw, it is also a random event with about a 30% chance of success.

[Fritz88]

My main issue is how is this applicable to events where the person can influence the outcome such as sporting events. I can see how this theory works in black jack and gambling. But Hamas makes a good case for going with the hot hand in the short term.
Posted via Mobile Device

[Ebolapox]

Someone who is more skilled will make a greater number of shots. To simplify things, let's just think about free throws. An 80% free throw shooter will make about 80% of his free throws. He is highly skilled. But the chances of making any given free throw is a random event, in that it has an 80% chance of going in and a 20% chance of missing. I am less skilled. I would do well to make 30% of my free throws. For any given free throw, it is also a random event with about a 30% chance of success.

statistically speaking, I agree with you... I just wonder if there's some unquantifiable factor that confidence plays... in other words, if you're on a cold streak, your mechanics may be jacked and you'll be more inclined to miss.

[cdcox]

statistically speaking, I agree with you... I just wonder if there's some unquantifiable factor that confidence plays... in other words, if you're on a cold streak, your mechanics may be jacked and you'll be more inclined to miss.

But the streaks of hits and misses of basketball players statistically resemble streaks heads and tails of a weighted coin. If there was an extra human contribution to the streaks (confidence or mechanics of the moment), one would expect the human to be more streaky than a weighted coin.

[Ebolapox]

But the streaks of hits and misses of basketball players statistically resemble streaks heads and tails of a weighted coin. If there was an extra human contribution to the streaks (confidence or mechanics of the moment), one would expect the human to be more streaky than a weighted coin.

quite possible. I've seen the point debated by those much more statistically inclined than I am, with absolutely no convincing evidence being presented that there is any role played by a human factor-I'm inclined to cede the point and agree fully. for some reason, however, it just 'feels' like there should be a role played, which is (I believe) why a lot of people are so reticent to fully accept the statistical explanation.

[Fritz88]

But the streaks of hits and misses of basketball players statistically resemble streaks heads and tails of a weighted coin. If there was an extra human contribution to the streaks (confidence or mechanics of the moment), one would expect the human to be more streaky than a weighted coin.

Well said.
Posted via Mobile Device

[AdumbGuy]

everybody who has actually played basketball disagrees.

If I shoot 10 free throws over the course of a game at random intervals, I'll make 6 or 7 of them. If I shoot all 10 in succession, I'll make 9 or 10. Context matters.

The definition of a hot hand in basketball is also very different from the definition they set forth. If somebody makes a shot, they're only going to claim they have a hot hand if they're a douchebag. If you make three or four in a row - now you've got a hot hand.

And it makes sense then, that if you've made three or four in a row, you're more likely to make then next one. You're calibrated, essentially.

Anybody that's played long enough has experienced it - that zone, where motherfucking everything just works - it's a beautiful thing.

[suzzer99]

I'd like to see this study on high leverage situations only. I can believe that averaged out over all NBA games at all times during the game, the "hot hand" doesn't necessarily predict any better or worse shooting percentage then that player's average.

However, you can't tell me when Reggie Miller was dialed in late in the game, that he was just shooting his average three point percentage, and running really well. Maybe that's the case. But I want to see a study with a huge sample size that only looks at high pressure situations late in games.

[Pasta Little Brioni]

Anybody that has played enough competitive sports knows that somedays you are just "on". It's hard to explain, but when you are in the groove, your chance of success increases tenfold.

[Canofbier]

A 40 percent lifetime shooter has a 40 percent chance of hitting his next shot.

I disagree. It means that if he takes a whole bunch of shots, you expect that he's going to make 40% of them. Means aren't predictive of individual outcomes.

[2bikemike]

How exhaustive is the research and over how large of a sample-size? There are certain occasions when a player's shot is just locked in and other cases where a good shooter can't hit the ocean if falling out of a boat. Even players of equal skill are not equal consistently. If the player was able to perfectly mimic his mechanics every time I would be more likely to believe this, but the imperfection is key in this case, as there are days when our timing is simply better than others. That itself is an outlier, but it's an anomaly that leads to a predictable, beneficial outcome over the short term.

This here makes sense. I believe the "Hot Hand" is just another way of saying someone is "In The Zone" anybody who plays sports or does athletic things whether it be golf or skiing or whatever. There are just sometimes when everything comes together for you.

[-King-]

I call Bullshit. How the hell do they explain the numberous streaky shooters in the league?
Posted via Mobile Device

[BlackHelicopters]

No one, repeat no one, has a hotter hand than Alex Smith.

[-King-]

I disagree. It means that if he takes a whole bunch of shots, you expect that he's going to make 40% of them. Means aren't predictive of individual outcomes.

This.
Posted via Mobile Device

[beach tribe]

If you have ever been in the zone then you know damn well it is not a fallacy.

[J Diddy]

I disagree. It means that if he takes a whole bunch of shots, you expect that he's going to make 40% of them. Means aren't predictive of individual outcomes.

This.

[beach tribe]

This here makes sense. I believe the "Hot Hand" is just another way of saying someone is "In The Zone" anybody who plays sports or does athletic things whether it be golf or skiing or whatever. There are just sometimes when everything comes together for you.

Beat me to it.

[milkman]

I'll say this.

The one guy that I would like this study applied to is "The Microwave" Johnson.

['Hamas' Jenkins]

Isn't a randomness fallacy an assumption that outcomes are always distributed and dispersed randomly?

[Rausch]

Failed attempt to apply statistics to human nature is still failed...

[tk13]

I get the idea, and in general it's probably true. The statistical probability of a 40% shooter making his next shot is 40%. It's possible the term is overused. But these guys also aren't robots either. What happens when Nate Robinson goes off for 29 points in the 4th quarter and OT against the Nets the other week? What is that called? It's certainly something.

[cdcox]

I get the idea, and in general it's probably true. The statistical probability of a 40% shooter making his next shot is 40%. It's possible the term is overused. But these guys also aren't robots either. What happens when Nate Robinson goes off for 29 points in the 4th quarter and OT against the Nets the other week? What is that called? It's certainly something.

Nice example. Let's look at it. His shooting sequence in the fourth quarter and OT were:

1 0 1 1 1 1 0 1 1 1 1 0 1 0 1 0 0 0

That's 11/18. This is considerably better than his typical 43% shooting from the field. I quickly ran 100 random simulations and had 3 cases where a 43% heads coin would come up heads at least 11 times in 18 tries. There were probably another 10 times it came up with 10 heads. So this kind of streak is rare, but not unexpected.

We can interpret this as about a top 3% performance from a player of his caliber. But given the thousands of player-games that take place every NBA season, we expect these kind of hot streaks to occur. It would be more strange if they didn't occur.

Another point: after his first 11 shots in this sequence we would say he has the "hot hand". He then went on to hit only 2 of his next 7 shots.

[BlackHelicopters]

Francoeur always has a hot hand.

[Rausch]

The human mind's powerful ability to detect patterns often leads us to detect patterns where there are none.

Exhaustive research proved that prior shot performance did not change the likelihood of success on the upcoming shot.

The hot-hand fallacy or representativeness heuristic may be so well engraved into our decision processes that even scientific training may not fully eliminate it.

What do you think?

Do you think that a player who is having a hot night is your go to guy or is the other player who is not guarded but equally skilled may have a better chance of scoring?

http://en.wikipedia.org/wiki/Hot-hand_fallacy

...

http://24.media.tumblr.com/e46d17a471a42a0a39e2df30228989b7/tumblr_mhtkiwbhmi1r4u11so1_500.jpg

[alnorth]

I'd like to see this study on high leverage situations only. I can believe that averaged out over all NBA games at all times during the game, the "hot hand" doesn't necessarily predict any better or worse shooting percentage then that player's average.

I think this is a pretty good idea, though it would take an enormous amount of work. Say, playoff basketball games, 4th quarter or later, where neither team is leading by more than 5 points in the beginning of the quarter. Take every one of those games, hundreds of quarters, thousands of players, and see if they still behave like weighted coins.

My bias is probably to presume it is a fallacy, but the argument that someone who is hot might be more comfortable, loose, and confident, is good enough for me to want to see real numbers.