Before I start, let me just note that I am pretty much positive there will be negative backlash towards this post, and am ready to be flamed.
Recently there was a post/comment, don't remember by whom, about the concept of "hot streaks" and how statistical analysis seems to show that hot streaks are really just a product of a) small samples and b) cherrypicking. The more I think about it, the more that that makes sense to me. First, how often do you see on the TV screen "in his last 8 games at DH, hitting .367" or something of the like. Doesn't that sound a helluva lot like statistical cherrypicking? Why isn't it fair to say "in his last 12 games, hitting <exactly the same as season/career lines>"? Of course, because it's not interesting television. But it's not limited to player analysis either. We always hear "this guy is 5-2 in his last 7 day games" or "they've won 5 of their last 7" or other obviously cherrypicked numbers.
My main point: is it possible that "hot streaks" are called such based on reverse causation? The thinking seems to be "this guy is hitting .400 this week because he's hot" when isn't it just as likely (and, as I suggest, more likely) that "this guy is hot because he's hitting .400 this week"? Basically, is it possible that hot streaks are made up just as a result of a series of good outcomes, and a player does not truly "create" a series of good outcomes because he's hot?
Now I'm sure the most common response would be "but that ignores the human element - every athlete ever can tell you that 'in the zone' exists in some form." And this was my biggest hurdle when I was thinking about this. I've played sports at a high enough level to know that "hot" most certainly exists. But then, as I think about it more, I really think it's just mental acrobatics. When I hit 10 straight shots in basketball, of course I feel hot. But I really think that this is the causation: I hit a bunch of shots - I feel hot, so maybe I hit some more shots. The causation is such that good outcomes lead to hotness, not the other way around. And, of course, there is plenty of random variation in all sports at every level (a gust of wind, another player slipping on a slick floor, an act of Superman, etc) so there can be random variation positively too (I would most certainly not expect to hit 10 shots in a row very often). Why isn't random variation the cause of hot streaks?




There are 29 Comments. Load Now.
Shortcuts to mastering the comment thread. Use wisely.
C - Next Comment
X - Mark as Read
R - Reply
Z - Mark Read & Next
Shift + C - Previous
Shift + A - Mark All Read
Comment Settings
Live comment alert: Hide it!
Comments for this post are closed.