MBB: Michigan Intercollegiate Athletic Association

[Gregory Sager]

Looks like FDF is saying the same thing I am about to, here goes anyway:

Based on this broadening of the scope, I agree.

If a player get's "hot" the opposing coach has to determine whether he is subjected to normal random variation or if there is something his defense is doing (more appropriately, not doing) to aid the shooter's success.

Ah, but what is "normal random variation"? By strict definition, variation could mean anything that makes one shot different from another. Shooter A may be a great shooter, but he may not be as good from one spot on the floor as he is from another, even if he's equally open in two sequences in which he gets the ball in first one spot and then the other. Or he may get the ball in the same spot two possessions in a row, but the second time the defender might be a foot closer to him, or a foot further away ... or the shot clock might be down to the point where he has to rush his motion the second time around. See what I mean? There's no way you can "cleanly" analyze a given sequence statistically, in the sense that game factors don't come into play -- not unless you can somehow recreate exactly identical scenarios from one shot to the next.

In other words, the variation inherent to basketball isn't random -- it's predicated upon conscious action and reaction.

This is why good coaches are always more likely to use the word "tendency" than the word "probability" -- they may trust statistics to a point, but they trust their ability to read specific game situations and the individual quirks of a player even more.

[ziggy]

The discussion started on the macro level and then was dragged to the micro. Sure, there are plenty of things a coach needs to take into account to make a particular discussion, that is what they are there to do. It all boils down to this at the most basic level:

1. If you subscribe to the statistical analysis that says there is no such thing as a "hot" hand then you believe that players of equal ability in an equal situation will perform equally. Your proof is the studies referenced in the past few pages.

2. If you take the "yeah but I don't think that's true" side of the argument and hold to the theory of the "hot" hand then you believe that players of equal ability in equal situations will not necessarily perform equally. Your "proof" is based on confirmation bias (Thanks to Happy Calvin Guy)

[Happy Calvin Guy]

And while all of the arguemnets that have been made here are sound - I still go back to the fact that a coach is likely to make decisions (offensively and defensively) based on who he /she perceives to be hot or cold.

Actually, KVS has been criticized on these boards for making decisions based on a predetermined rotation and NOT based on who is "hot" or "cold".  So I think that these discussions are in fact very relevant to the coaching style of a team near and dear to (some of) our hearts, and can't just be dismissed as theoretical banter.

[Flying Dutch Fan]

At this point I can't wait for the games to begin.  I'm really anxious to see who starts off the season with a hot hand  

[ziggy]

At this point I can't wait for the games to begin.  I'm really anxious to see who starts off the season with a hot hand  

me too.

[HopeConvert]

We might construe the distinction this way: being "hot" or "cold" is a subjective state, while streaks are (statistically) objective. HCG is right in that, objectively speaking, how a person has shot in his prior five attempts has little to no predictive value in telling us whether he is going to make his next shot. For that, long-term statistical aggregates are better predictors (in which case you would go with the 40% shooter over the 25% one). But in the context of the game, the player who is actually the better shooter might be having confidence problems, or an excellent defender on him, in which case you might want to run the play for someone else. The coach has to balance these concerns in a short time span.

Subjectively speaking, I think there pretty clearly is such a thing as "feeling it." Anyone who has played sports know there are elusive times where everything seems to be clicking: you're balanced, have excellent tempo, and everything just feels right. Other times, you don't have that. Golfers go out one day and shoot a 62 and a 77 the next, and part of that is simply because they do or do not have the feel of their swing. Over the course of a season, however, such groupings are apparently only random in the sense there is no discernible pattern.

[KnightSlappy]

Subjectively speaking, I think there pretty clearly is such a thing as "feeling it." Anyone who has played sports know there are elusive times where everything seems to be clicking: you're balanced, have excellent tempo, and everything just feels right.

Again, there are examples of empirical data that show that "feeling it" does not always improve performance.

Chapter two of The Book (a very good book if you like baseball) is on streaks. Here's the intro to the chapter http://www.insidethebook.com/c02.shtml.

The chapter concludes that a hot or cold streak has no predictive ability for hitters, but it does have predictive ability for pitchers.

The basketball studies linked in the previous pages has a conclusion similar to that of baseball hitters: the data shows no predicitive power to a shooter's hot streak.

One would have to conduct a similar study for golf, but my gut feeling is that 'feeling it' doesn't predict golf scores well either (at least at the PGA level).

Unless you're saying that truly 'feeling it' is a significantly rarer (is that a word?) event than a 'hot streak'.

[realist]

Sorry!  I was just "feeling" it.

[ziggy]

Sorry!  I was just "feeling" it.

The horse must be continually beaten because we can not be sure that just because it is dead now that it will be dead in the future.

[Gregory Sager]

You're right, Ziggy. In the picture posted by Realist, our top-hatted friend is merely poking at ol' Dobbin rather than continually beating him. Therefore:

[Civic Minded]

And still a month or more to go until an actual basketball game is played.  Heaven help us in the interim!!!      

[Gregory Sager]

Actually, we have less than a month to go. The first day of the regular season is Monday, November 15, which is 27 days from now.

[Civic Minded]

Actually, we have less than a month to go. The first day of the regular season is Monday, November 15, which is 27 days from now.

I suppose "we" is subjective.  Hope's first game isn't until the 23rd; not sure if other MIAA schools have games before that.  Thanks for proving my point though...quibbling over the calendar now.   

[sac]

[Mr. Ypsi]

Wow, am I noticing the party late - poor Dobbin has already been beaten!

Nontheless, as a retired statistics and probability instructor (but also a social psychologist, and former bball player - though my lack of stature kept me from even hs ball), I wish to come down firmly on both sides of the debate!

I have never seen a study that indicated performance outside the realm of probability for a coin - even an honest coin will come up heads seven times in a row occasionally, and only a soon-to-be-bankrupt gambler would think a coin is "hot".

On the other hand, coins don't play bball (or have memories or emotions); I clearly recall days I felt the basket was 3 feet wide, and others I thought it was slightly narrower than the bball.  The perception of the 'hot hand' obviously exists, and I cannot help thinking it affects performance.

On the other other hand, as Greg observed (and probably others, but I've lost track ), bball is not a static coin-flip situation - the opposition will make adjustments.  You get 'hot' and the defense stiffens.

I think the question is probably objectively unanswerable.  I believe the "hot hand" does exist; I also think that opposition adjustments make it impossible to prove either way.  (I saw Draayer at Wheaton in 2005(?) [don't recall if the opponent was Wheaton or Carthage] - no one in the crowd would doubt that he had the "hot hand".)