To the best of my knowledge — which is limited, as Josh Miller has pointed out — scientific hot hand research has focused on detection: to show that the hot hand exists researchers have collected information and developed statistical methods to show, with statistical significance, that athletes perform better some of the time. From a sociology of science perspective, this was crucial.
But there’s a much more valuable direction to take the hot hand. In my paper on NBA free throw shooting I included sections on ‘Some Implications for Basketball Strategy’ and ‘Speculative Strategy Implications.’ Start with implication #5:
Heat Up Your Team
Thanks to my 1989 investigation, I became aware of repetition as a cause of ‘heating up.’ This gave me a perhaps unique fan perspective following Phil Jackson’s championship Chicago Bulls team in the ‘90s. In particular, I saw how they frequently went right back to a player after a miss, provided that he had a similar opportunity on the next possession.
This flies in the face of all basketball tradition! — which is why I’ve ascribed this feature to Jackson rather than Michael Jordan. If you’re playing pick-up basketball and brick your first attempt, good luck ever getting a pass from your teammates the rest of the game.🤪 It also contradicts the conventional hot hand logic; one consequence is that missing your shot means the other team pays less attention to you thereafter.1
It appeared to me that Jackson knew that repetition improves his players’ chance of making the next shot, further enhanced due to reduced defensive focus after missing.
The more general lesson: don’t try to figure out who’s got the hot hand; instead, create hot hands and then position your team or yourself to reap the heat.
Never Sabotage Your Team
I have determined that conventional hot hand strategies are particularly damaging in amateur softball.2 In the short run, nearly all softball hitters alternate: that is, they are more likely — compared to their long run average — to follow a hit with an out, and to follow an out with a hit.3 Furthermore, if you as a manager move ‘hot’ hitters up in the lineup and ‘cold’ batters down, you increase your players’ anxiety: instead of focusing on technique in the moment, you divert their attention to results.4
So riding the (conventional) hot hand not only immediately harms your team’s performance, it also poisons team chemistry, pitting teammates against each other.
Cool Down Your Opponents
In my free throw research, I found evidence that interruption, fatigue and stress degrade performance. Interruption and stress motivate the well-known icing strategy (implications #2 and #3 in my paper): call a timeout so that the opposing free throw shooter or field goal kicker has more time to build anxiety. This is offset to the degree that the opponent benefits from recovery time, reducing fatigue.
My BrewDogs softball team found that producing stress in our opponents by interrupting them was inordinately effective in state, regional and national tournaments. At this level of play, softball hitters typically get on base over 60% of the time. Subjectively — unfortunately, I don’t have concrete data for this — the timeouts we called at key moments reduced this below 20%.
How Large Are Hot Hand Effects?
On Andrew Gelman’s blog, James B. Shearer asks:
By the way how large is the hot hand effect believed to be? How does the difference between a player having a good day or a bad day compare to the differences in ability among players?
My research on NBA free throws shows that individual players (1) systematically improve by, on average, 5 percentage points from their first to second shot in each trip to the free throw line, (2) have intra-game standard deviations in their shooting-percentage-expectation (‘SPE’) of about 5 percentage points, and (3) inter-game standard deviations of single-game mean SPE about 4 percentage points. The standard deviation across players for career mean SPE is about 8 percentage points.
So, for example, a poor free throw shooter during a hot streak can be as successful as a good shooter at a typical time.5
I anticipate that comparable field goal analysis will reveal comparable magnitudes.
There’s also a wide degree of intra-career variation in SPE. In particular LeBron James exhibited the greatest intra-career variation within the 2000 through 2014 data. It appears there were often times where his free throw SPE was as low as 60% or as high as 85%!
I haven’t followed the NBA lately, so I’m curious if anyone has noticed teams realizing they need to continue closely guarding players at their next attempt after a miss. Or do most teams still avoid giving the ball back to a player who just missed?
My guess is that this generalizes to many other sports, but I’ve got solid data for amateur softball. I’ll leave the analytical report for a later date.
The main reason hitters alternate is because when they feel ‘hot’, they swing at more pitches — this appears to be true even for the most sensible players — while when they feel cold, most hitters, and especially the more astute, become more selective.
I realize there have been recent criticism of growth mindset claims, but I’ve found in every arena I’ve encountered that results-orientation is counterproductive.
Technical details for this example: define a poor free throw shooter as one standard deviation below average — “-1 sigma” — a good shooter as +1 sigma, and “hot streak” as +1.6 sigma, i.e., the best 5% of the time.