ArsTechnica: The three point "hot hand" doesn't exist

ArsTechnica (a tech blog) looks into a recently released statistical analysis regarding NBA players, and having a player shoot a lot of three's because they have "the hot hand".

----

The authors' test of a hot hand was very simple: for every three-point shot attempted, the authors tracked the results, and the results of the next consecutive shot. When analyzed this way, hands appeared to get cold. If a player took another three-pointer after making one, their success rate was 6 percent lower than if they took it after having missed one. In other words, you were more likely to make a three pointer if you'd missed on the previous one. In contrast, there was little difference in the success rates of ensuing two-point shots.

6 months ago Timmay! 8 comments 0 recs |

Story-email Email Printer Print

Comments

Counter argument:

Wes Matthews

66-0 baby

by thomasikehara on Dec 8, 2011 12:46 PM PST reply actions

This was discussed back in 2009 at the MIT Sloan Sports Analytics Conference.

http://sports.espn.go.com/nba/columns/story?columnist=hollinger_john&page=PERDiem-090309

Huizinga studied high-volume shooters over the past five seasons and took note of players who had made a jump shot on the previous trip. On the subsequent trip, they shot about 16 percent more often and converted 3.5 percent less of those shots. The finding was statistically significant and pointed to a tendency by players to act as if they were hot after one made J … and a counterproductive tendency to feed the allegedly hot hand by both the player and his teammates.

Proving that something doesn’t exist is always vastly more difficult than proving it does, but Huizinga’s analysis hammered the hot-hand theory from so many angles that its proponents are reeling. And as luck would have it, the scenario played out at the end of the Boston-Orlando game: The Celtics ran a play to feed their hot hand, Ray Allen, and he missed a difficult 3-pointer after a strong contest by Dwight Howard that would have tied the game.

Interestingly, Huizinga’s study showed this phenomenon was almost entirely confined to perimeter players — most likely because it’s far more difficult for post players to ignore the offense and call their own number. (Though I suspect Zach Randolph and Dirk Nowitzki may be exceptions).

"You know, when you are in the game, you hear 20,000 people behind you, you don't feel anything."
- Nicolas Batum on playing through his shoulder injury during the 2010 playoffs.

by halo_on on Dec 8, 2011 12:59 PM PST reply actions

Yea, I've heard this one before too

First of all the argument is that a player will always regress to the mean. If a player shoots at a 50% average, then its the exact same as a coin flip. And basic statistics tells us that no matter how many times you get heads, the next flip will always be 50%.

There’s actually a couple different logical fallacies here, but the biggest problem I have with these arguments is that stat geeks generally don’t play basketball a whole lot. It might be a definition problem though, I believe the “hot hand” is when you feel you can’t miss, and you just don’t get that from one shot. I believe the second shot (the only shot they are analyzing) is called the “Heat Check” which is usually a pretty low percentage shot.

And I know there is such thing as the hot hand, because I’ve had it before. It just doesn’t happen as often as people think. Sometimes you just get to that point that you have to heave up the most ridiculous fall away deep 3 pointer just to get yourself to miss just once. And even then you aren’t sure you are going to miss.

by Batumshakalaka on Dec 8, 2011 1:27 PM PST reply actions

Excellent point

The study should have tracked the next two shots after two (or more) consecutive makes. This might help explain the occasional 45 point game (or 15+ in a quarter) for a guy averaging 20 points a game, especially if his usage doesn’t double.

"What we have here, is a failure to communicate."

by with_a_z on Dec 8, 2011 1:54 PM PST up reply actions

No the problem is people who play basketball a lot don't understand statistics.

You have never had a hot hand. You’ve had strings of good luck accompanied by positive emotional responses. There’s no way to tell the difference by anecdotal evidence or by doing it. But there are ways to tell by statistics, and they all agree.

by pualo on Dec 8, 2011 9:23 PM PST up reply actions

Sounds good, and I've never actually done the math

But I’m guessing my good luck streaks would end up something like 9 shots in a row, 10, 12, 8, maybe 15 I guess, but the day’s I would have the “hot hand” I would shoot bout 95% no matter how long I shot. And yes, technically you could flip a million heads in a row, but I really just can’t explain the feeling that get from days like this. It’s not a magical thing, you’re muscle memory and coordination is just so in-tune that everything flows perfectly. That 5% that you do miss, you can usually feel it before it even leaves your hand.

Mind you, I’m talking about maybe a half dozen times in my life. I understand that emotional high you get after making a couple shots in row, especially in a game. It’s the same as the “clutch” factor. The fact that there is less time left in the game doesn’t make you more likely to make the basket. Although even this can be argued by the calm and confidence of certain mindsets when taking the big shot.

Statistics is important and all, but you could explain anything away with it. What I consider the hot hand would be called an outlier and be dismissed anyway. Either way, I think Blazerblaze hit it right on the head below.

by Batumshakalaka on Dec 9, 2011 9:02 AM PST up reply actions