...a companion blog to "Math-Frolic," specifically for interviews, book reviews, weekly-linkfests, and longer posts or commentary than usually found at the Math-Frolic site.

"Mathematics, rightly viewed, possesses not only truth, but supreme beauty – a beauty cold and austere, like that of sculpture, without appeal to any part of our weaker nature, without the gorgeous trappings of painting or music, yet sublimely pure, and capable of a stern perfection such as only the greatest art can show." ---Bertrand Russell (1907) Rob Gluck

"I have come to believe, though very reluctantly, that it [mathematics] consists of tautologies. I fear that, to a mind of sufficient intellectual power, the whole of mathematics would appear trivial, as trivial as the statement that a four-legged animal is an animal." ---Bertrand Russell (1957)

******************************************************************** Rob Gluck

Sunday, October 18, 2015

Hot Hand... You Betcha!

via Reisio/WikimediaCommons
"Gödel's Lost Letter" tackled the "hot-hand fallacy" recently:


 I have to confess to tiring a bit of this whole debate: there IS such a thing as being 'in the groove' or 'in the zone' or 'on your game' or 'HAVING A HOT HAND' (IMHO) and everyone who has ever played basketball or tennis or golf or bowling or any number of other sports KNOWS it (there are times we ought not downplay people's personal experience in favor of slapping dry stats and randomness onto situations that are exceedingly difficult to analyze, and where uncontrolled variables abound -- reminds me of what is routinely done in epidemiology... don't get me started).

A lot depends on simply how you define "hot hand" and what units of time are considered... i.e., does someone have a hot-hand for a game, or for a 13-min. stretch of a game. And in the case of the basketball "hot-hand" the stats often look at 2-4 shots in a row to predict the next shot, when larger samples, probably 5-10 shots minimum, need to be considered, because the variables are so-o-o many -- also, if you make 4-5 layups in a row it probably means nothing; but if you repeatedly put in shots from the far corner, the 3-point-range, and while being double-teamed (i.e, lower-percentage shots) that begins to mean something, yet I've never seen "shot-type" or shot-circumstances taken into consideration.

Any athlete will have experienced that rare feeling when their health/nutrition/sleep/physiology/physical prowess/movement/mood/psychology/whatever all seem to coalesce to yield an excellent performance, where they can be depended upon, more than other teammates, for crucial plays. Not every instance that looks like a "hot-hand" of course, to the outside observer, may be one, but I'm a believer ;-) that it does exist on occasion (and am old enough to recall Wilt Chamberlain's 100-point effort in 1962, including a phenomenal 28 out of 32 free throws! And a 'hot' Michael Jordan's winning shot for the 1982 NCAA championship, or Christian Laettner's 1992 championship shot, and on and on).
I s'pose next the statisticians will try to tell me that Reggie Jackson was NOT really ever "Mr. October" for the NY Yankees! ;-)  DON'T even go there!!

Sports performance is clearly in part a function of skill and experience (and psychology), which can vary from day-to-day (even moment-to-moment) for a given individual. A 'hot-hand-like fallacy' is more likely to hold sway in something like gambling where outcomes are more strictly governed by "chance," not skill, and a perceived "streak" may not be real (even in gambling though, it is possible that tiny, almost imperceptible clues, trends, properties, signals, are picked up by the experienced gambler at times that raise his/her performance on certain games).

Anyway, while I'm merely banking on common sense here (admittedly, a dangerous medium), there are statisticians who have also found technical flaws with the 'hot-hand fallacy' argument (see Gelman, for example, here), and therefore speak of the 'fallacy of the hot-hand fallacy' to which, of course, their detractors can respond with the fallacy of the fallacy of the hot-hand fallacy... but then, I find their arguments fallacious.

Now, excuse me while I go shoot some baskets, while I'm feeling kinda hot (...under the collar).


ADDENDUM ==> the above post was written a few days back and pre-scheduled for Sunday-posting. Lo-and-behold, just yesterday, science writer George Johnson had a piece in the NY Times on, of all things, the hot-hand fallacy!:


As indicated above I consider the "hot-hand fallacy" and "gambler's fallacy" two very different subjects and levels of complexity; referencing them together is mixing apples and oranges a bit (though I understand why both show up in such discussion).
Like most articles, this one fails to take into account the intrinsic oversimplifications of hot-hand analyses, and again treats the hot-hand as something spectators observe, rather than something an athlete 'feels' or experiences.
I wish this whole area would just move along now as not worthy of further exploration (...but am sure it won't).

No comments:

Post a Comment