« Blueprint To Beat LeBeau Written In 2002 | Home | 41 Year Old Gary Roberts Schools Flyers Enforcer »
Random Playcalling in Sports
By i-Maque | December 11, 2007
I’m sure that many of you are familiar with the book, Freakonomics. What you may not know is that the NY Times hosts a Freakonomics blog to, in their words, “keep the conversation going.”
They have a very interesting guest post today from Ian Ayres entitled “Why Don’t Sports Teams Use Randomization?”
He wonders,
But an even bigger puzzle is why teams don’t exploit the other powerful use of randomization. To my knowledge, no sports team in the history of humankind has ever run a random control trial to figure out which strategies work the best. (I make this extravagant claim in hopes of provoking you all into providing some counterexamples.) Randomized studies are the gold standard of medical testing, and they’re now the hottest thing in Internet ads.
He suggests that football coaches use a random-number generator to determine what play a football coach should call, for instance. To my knowledge (I may be wrong), but I don’t think teams are allowed to use computers during games. But it is still a very interesting read.
—
And in the “If you’re bored” category of things, check out this robot that Toyota unveiled. It plays the violin. I think it’s pretty darn cool.
Topics: Sports | 5 Comments »
December 11th, 2007 at 4:26 pm
Sure, they may not be able to use a computer during the game, but there’s any number of pseudo random events that you can see without needing a computer.
Direction of a wind sock in swirling wind
A die
The number of people in a row of stands who are wearing a black hat vs gold
The first letter in the last word you just heard
A sheet with a list of computer created pseudorandom numbers printed out!
December 11th, 2007 at 4:48 pm
DrObviousSo – you’re absolutely right. But one has to be careful. If it’s an event that’s easily observable, then one has to assume that the opponent will eventually figure it out. One must be random in how they choose to be random. Can’t let the key be public, as it were.
December 11th, 2007 at 5:58 pm
The problem with that randomization is on 3 or 4 and 10 in a crucial situation, you might get a dive play from a 0 WR set. What are you gonna do then?
December 11th, 2007 at 8:16 pm
i-Maque: Absolutely. However, there is so much going on in a football game that it shouldn’t be to hard to find something seemingly insignificant from game to game. I personally like keying off a letter in the last spoken word over a headset.
gr8one626: Think of it more like this. On first down at mid field, a coordinator probably has a handful of run plays and a handful of pass plays that are all good options. He would randomly only pick from them, instead of the whole gameplan. The idea isn’t to use new plays in new situations, but to keep people from using their internal bias to pick a predictable play.
If for example we know that an OC likes to run the ball up the middle on 2nd and short plays, but has at his disposal a reverse, a bomb, a screen, and a short yardage passing play that all work on that down against that defensive scheme, this would keep him honest. If the OC is kept honest, the defense has to stay honest, too.
Currently, OC’s must monitor themselves to keep themselves honest, which is a very, very hard thing for a single human to do himself.
At least, that’s where I see value in random play selection.
December 11th, 2007 at 8:24 pm
that video freak anybody else out??