Hot Hand

Home for all your discussion of basketball statistical analysis.
Post Reply
rlee
Posts: 2105
Joined: Tue Sep 10, 2013 3:58 pm

Hot Hand

Post by rlee » Sat Jun 10, 2017 7:24 pm


permaximum
Posts: 413
Joined: Tue Nov 27, 2012 7:04 pm

Re: Hot Hand

Post by permaximum » Sun Jun 11, 2017 7:55 pm

If hot hand is not proven by analytics it's only because analytics are not there yet. No one can convince me there's no such thing as hot hand with the usual very poor statistical arguments. I'm a former basketball player that played as an SG at school-level and I know the whole deal. If these so-called analysts continue to pursue these kind of things, they are gonna stay irrelevant to mainstream. I suggest starting with hormone level/body control correlation first. I'm sure they'll see the beginning of hot hand there.

We're in the age of artificial intelligence that creates artificial intelligence (reinforcement learning algorithms that create/tweak deep learning neural network architectures for optimum quality in tasks such as image and speech recognition, language modelling) and some people try to "prove" things with such arguments. I appreciate the effort but that's about it.

Edit: I'm talking about those people who defend the idea of not having a state called hot hand. In the article apperantly some people found evidence of it.
Last edited by permaximum on Sun Jun 11, 2017 8:04 pm, edited 2 times in total.

Dr Positivity
Posts: 298
Joined: Thu Sep 20, 2012 6:44 pm

Re: Hot Hand

Post by Dr Positivity » Sun Jun 11, 2017 7:56 pm

I believe the hot hand is real, the players think it exist and if the game didn't have emotional factors or importance of feeling good rhythm wise, stuff like home court advantage wouldn't exist...

Mike G
Posts: 4425
Joined: Fri Apr 15, 2011 12:02 am
Location: Asheville, NC

Re: Hot Hand

Post by Mike G » Sun Jun 11, 2017 9:47 pm

Every person, animal, or machine has times when they work more efficiently than at other times. Why would anyone suppose "basketball shooters" are the one exception in the world?

A bunch of people spent a lot of time proving only that they could not prove anything.

mystic
Posts: 470
Joined: Mon Apr 18, 2011 10:09 am
Contact:

Re: Hot Hand

Post by mystic » Mon Jun 12, 2017 9:22 am

Well, that is a discussion we had 2 years ago. Miller and Sanjurjo explained why the "hot hand" was not found in reallife experiments, because the statistical analysis applied was inappropiate. To be more specific: When selecting only shots after a certain sequence of made or missed shots, we are changing the expected probability similar to the Monty-Hall problem (events go from independent to be dependent upon a() previous event(s)). In case of the "hot hand" we can't expect the probability of a shot being made after a sequence of made shots to be the same as the general expected probability (which is based on the assumption that events are independent), but the real expected probability has to be lower. We can correct that by applying the Bayes' theorem here (conditional probability). And that is not just related to basketball, but is true for all such experiments (even random coin flips, as Miller&Sanjurjo show in the introduction of their working paper, which can be found here: https://papers.ssrn.com/sol3/papers.cfm ... id=2627354).

That's said, I feel like I have to emphasize here, that the "hot hand fallacy" does not deny the existence of variance, nor did the research claim that no streaks of made/missed shots would exist. Players have better and worse games, just that those can be explained by normal variance of their performance level and not some mysterious hot/cold hand. Or in other words: A coin can show head/tail a couple of times in row, but that doesn't mean that the coin is loaded. The research, which concluded that hot hand does not exists, found that the sequences of made/missed shots do not differ from random head/tail sequences of coin flips. That is/was actually a pretty strong argument against the hot hand. Just that, as Miller&Sanjurjo pointed out, the results were based on the wrongly applied statistical analysis.

Last but not least: The best way to learn new things is to be open minded and be able to change your opinion about a subject based on new evidence. Just believing in something is not knowledge (even though in our current "Trumpian age" it seems like there are a lot of people pretty much convinced that they are smarter than others, because they believe that this is the case); and the personal belief is a really poor argument especially combined with personal incredulity.

permaximum
Posts: 413
Joined: Tue Nov 27, 2012 7:04 pm

Re: Hot Hand

Post by permaximum » Mon Jun 12, 2017 10:21 am

If we had to call upon maths to prove everything and advance that way as the human race, our race would already be extinct.

Even now you can see some completely conflicting findings on most of the new research areas and both sides claim they have statistical evidence for it. You just need to look for those papers. I promise, you will find those conflicting papers where they think they have proven their claim. Which one is right? Time will answer that eventually. Just like a physics problem. Classical physics were the law, then quantum physics appeared with a stronger case.

Hot hand exists. Simply because you feel it. If you think there's evidence for its non-existence, you're in denial of your nature AND the history of maths. It means you're conservative of the current state of maths. You're an obstacle to advancement. You're not fit for any kind of research. You're biased by the rules of now.

Trying to prove hot hand's existence or non-existence is stupid.

xkonk
Posts: 294
Joined: Fri Apr 15, 2011 12:37 am

Re: Hot Hand

Post by xkonk » Tue Jun 13, 2017 1:40 am

The first step in the scientific enterprise, certainly in academia, it to show that you can measure something. If you can't show that something exists, how would you go about studying why or when it happens?

permaximum suggests that hormones are involved. Great! What exactly would researchers compare hormone levels to if they can't measure the hot hand? Dr. Positivity says that emotion is important. I like that hypothesis too! So we'll have players rate their confidence at various times and then compare ratings from.... well, we can't divide them up because we can't even tell if players get hot.
Trying to prove hot hand's existence or non-existence is stupid.
If all you were interested in was showing that the hot hand exists, and you were already certain it exists, I guess that's true. But if you want to know who gets hot and who doesn't, or when players get hot, or anything else (can you induce a hot hand? I bet someone would pay for that), then first you have to show you can measure it.

watto84
Posts: 81
Joined: Tue Feb 23, 2016 9:45 am
Contact:

Re: Hot Hand

Post by watto84 » Tue Jun 13, 2017 12:01 pm

I once did a training course mid-season with my team that was called "introduction to mindfulness", where one of the topics we spoke about and explored was what it felt like to be "in the zone" and how to try and get there more often. This could be intrepreted the same as "hot hand".

We spoke about how Mindfulness is a particular way of paying attention to yourself and your surroundings on the court and enables you to more skillfully respond, rather than just react to events that happen on the court. When you respond to an event you do so in a way that is controlled and measured.

When you are in the zone, you feel like you are 2 steps ahead of everyone and the game just comes to you.

Look at someone like khwai Leonard. Controlled and measured in his approach to all events on the court wether they are good or bad. It's no doubt a key reason for his consistency from game to game.

Mindfulness can be taught and players can become more Self aware and it brings in elements of meditation to help achieve it.

would be interesting to see if there is a correlation between game to game consistency of a player and if those players have more "hot hands". Could then perhaps correlate if a players who's overall performance fluctuates from game to game is more "streaky".

larrycoon
Posts: 1
Joined: Wed Jun 14, 2017 10:19 pm

Re: Hot Hand

Post by larrycoon » Wed Jun 14, 2017 10:54 pm

Hi all,

People who are saying things like "emotional factors or importance of feeling good rhythm wise," or "Hot hand exists. Simply because you feel it," or " hormones are involved" are misunderstanding the question. Is there PSYCHOLOGICAL momentum? The answer to that is a clear yes, and everything I quoted above refers to psychological momentum.

But psychology is subject to all of the cognitive biases and flaws in thinking that are inherent in the human brain. One of these flaws is hindsight bias -- we interpret and retro-dict the significance of events once we know what the outcome is.

Again, there is no quesiton of psychological momentum exists -- what happens in a competition affects the psychology of the athlete, and even the physiology. But again, that's not the real question here. The question is, does psychological momentum have a measurable impact on the outcome of the game? And here the answer is NO. When you look at the patterns of made vs. missed shots, there is no statistically signficant difference between what we observe and normal, random variance. There is no empirical EFFECT to explain -- the effects are all psychological.

Of course, Miller & Sanjuro claim that the statistical tools we've been using to analyze the effect are flawed, and these flaws are masking an actual, subtle hot hand effect. While the statstical flaws may be true, I disagree with their interpretation. One big issue that exists is with the magnitude of the claimed effect. In the ESPN piece Tom used a "room with clocks" analogy, so I'll use an analogy of my own -- ESP research (sorry if I'm pissing off anyone who believes ESP is real, but for the purpose of my argument, let's just assume for a moment that we live in a world in which it doesn't exist). The null hypothesis is that it doesn’t exist (even though there are lots of people who will swear up and down that it DOES exist -- sound familiar?), but there has been a lot of research done to see if it can be identified. The study results are typical of what can be found when the thing they’re looking for really isn’t there – the smaller, less controlled and more methodologically flawed studies have positive results; once you start improving the methodologies, controlling for weaknesses in the process and analysis the effect size diminishes; and the best studies are effectively all negative. In other words, it’s the pattern you find when you’re trying to tease out a signal from what is effectively just noise. (And you can look up things like P-Hacking to see just how easily and subtly a false signal can creep into your results.)

So here’s the important part – believers in ESP swear it exists because they’ve seen it. (Whereas a skeptic would say that what they’ve seen is easily explainable through the normal parameters of human psychology and doesn’t require the invention of new fields of physics to explain.) But when you compare what the proponents claim to see with what a study’s statistics claim to show, there’s a difference in magnitude. If people notice something (and therefore commission studies to ferret out) you’d expect the effect size to be huge when you look at it statistically – otherwise we never would have noticed it without statistics. The face claim is that our brains – as blunt as they are and as easily misled as they are and as biased as they are – were able to pick out a signal in a phenomenon that a carefully conducted study – which seeks to isolate variables, remove bias, carefully analyze, replicate, etc. – can barely, barely detect, and which – if it does exists – is a signal that’s barely distinguishable from noise. Our brains require a huge signal to reliably detect something, and what we found is the tiniest of signals -- maybe. That’s incongruent. That’s not the mark of there being something there, it’s the mark of the phenomenon you’re searching for NOT being there.

Back to Miller & Sanjuro and the hot hand – they’re claiming there’s a subtle, subtle error that’s eluded us for decades, and when you fix the data, a small hot-hand effect emerges. Even if we assume they’re correct, the effect is so small that there’s a massive disconnect between the perception of the hot hand and the reality. Even if it If you look at all the data, it still shows that there’s the tiniest – or no – hot hand effect. At best, Miller & Sanjuro have correctly identified a subtle error that shows a small effect that does nothing to explain the phenomenon that proponents claim to observe. They’re simply not describing what people are claiming to see.

And the people arguing for a hot hand are really just arguing for psychological momentum, not an empirical effect on the outcome of the game.

Mike G
Posts: 4425
Joined: Fri Apr 15, 2011 12:02 am
Location: Asheville, NC

Re: Hot Hand

Post by Mike G » Thu Jun 15, 2017 12:10 am

Have you ever noticed that when a player starts to get hot, the opponent will put a better defender on him, or even double team him?
The effect should be a reduced likelihood of his making more shots; yet there's no effect to be seen?

Golfers and tennis players get hot, and there's no help on the way defensively. Does anyone doubt these cases of 'hotness'?

Perhaps a bigger dampening effect, in some cases, is that a player may attempt tougher shots, once he's in a groove. Steph Curry likes to shoot from beyond 30 feet, just to show that he can. These are lower-%age shots.

Post Reply