Group Think in Football

I have written a couple of articles about predictive markets (Gambling, Predictive Markets, and Product Development, and Predictive Markets), and recently I've started reading the book, The Wisdom of Crowds by James Surowiecki.  More years ago than I care to think about, a friend and I used to talk about strategies we would use as football coaches in the NFL.  Neither of us had or has any background or ability in football coaching, other than being big fans.  This is the kind of thing engineers talk about even in recreation.  One of our key strategies was to go for it on fourth down very frequently.  We just intuitively figured that you might be better off going for it much more often than is done now, and also that if you built your offense around this strategy you effectively give yourself 4 downs to get 10 yards for a first down instead of only 3.  In the book, there is an example starting on page 44 (paperback version) that describes the work of David Romer [1].  He performed a statistical analysis of going for it on fourth down versus kicking a field goal using data from previous NFL seasons.  He used the probabilities of making a field goal versus the probabilities of making a first down and the eventual points scored starting with a first down at every point on the field.  He was able to assign a point value of kicking a field goal versus going for it for every position on the field.  At a high level, the results indicate that it almost never makes sense to kick a field goal when you are inside the 10 yard line or at a position where the field goal is very long such as the 30 yard line.  The results further show that it makes sense to go for it on fourth down at many points in many situations on the field [2].  The point of the example is that even though it would be a very good strategy to go for it on fourth down much more often, NFL coaches are stuck being much more conservative because of imitation, social proof (if everyone else is doing it this way it must be the best way), and group think [1].  Romer is an economics professor at UC Berkeley and another another article about this work can be found on the UC Berkeley web site.

[1] Surowiecki, James. 2004. The Wisdom of Crowds. Anchor Books: New York.

[2] Romer, David. July 2005. "Do Firms Maximize? Evidence from Professional Football."  [Accessed January 27, 2008: WWW at http://elsa.berkeley.edu/~dromer/papers/PAPER_NFL_JULY05_FORWEB_CORRECTED.pdf].