9/22/2004

Optimization in Football

On Sunday I happened to catch a short blip of a story on 60 minutes about the key success in football: the coach. They explained a little bit about how coaches today use stats to not only find the probability of what an opponent will do in a certain situation but other strategic moves involved in contracts and trades.

Bill Belichick, the head coach of the Super Bowl Champion- New England Patriots, was actually an economics major in college. I found that very interesting.

Anyone who follows sports knows about the salary cap. Teams cannot exceed this limit in aggregate salary paid to its players. Belichick, in particular, uses optimization to decide when it is time to release a high paid star and find a new, emerging player with star potential. This way, his team stays under the salary cap, while having a consistently successful team.

3 comments:

Ernie said...

It may be true that there is actually a lot of thought that goes into being a coach in a high stakes league such as the NFL, but I think that most of a persons success in coaching comes from having a knack for doing it. In other words the ability to see a players talent and put together a group of individuals that come out of the mix as a team of winners. Economics may help in areas such as opportunity costs and other certain areas. A coach may evaluate what may have happened differently in a game had he or she started someone else in a certain position. I also find it sort of hard to believe that a coach actually has that much bearing on whether a team is successful or not. You can have the best coach in the world as far as play calling and figuring out strategies of other teams, but without the talent on the team the franchise will go nowhwere.

Kristin and Scottie said...

Coaching does play an important part in a team's success. Success from a coach's view is not always a superbowl ring. You can say a team is successful even if they didn't get the ring. A coach must know his players and Belichick does.

I think it is not completely economics but yes it does play a part. You cannot have a player on your team that is making twice as much as other players but is not working for you. Usually these players are the ones who cause the problems. As a coach you want the emerging player because you will be able to have them longer than your current players. The NFL is like a business. Belichick is right as far as using economics but there is more to it than just economics.

Dr. Tufte said...

The post and the comments are all good ... but would be better if you all discussed game/management scenarios where knowing some economics might help. I offer some tips on further reading below, but if you are interested I have literally dozens of articles and many books about this topic that you are free to borrow.

There actually have been a lot of advances from a number of decision making fields (as opposed to phycisal sciences which have made many more advances) that have contributed to impovements in various aspects of sports. And there are some big name coaches who have listened. Bellichek is the one that everyone talks about because there were actually some articles discussing this published in the major media over the last year. One of the big (new) results in football is that coaches should be much more willing to go for it on 4th down, even from longer distances and even from further back towards their own end zone. I have a copy of that paper in my office if anyone would like to get a copy of it.

Baseball has actually moved further in this direction - both because there are more statistics to analyse, and because it is easier to associate a single player's performance with measurable statistics about that player (problem: how do you measure the contribution of a right guard in football)? There is even a bestseller called Moneyball about how the Oakland Atheletics have been managed better than other teams.

And, yours truly has even written several papers (and published two) about economic aspects of sports. On football, one looked at whether there are ways to make abnormal profits from betting on football (there are, but the odds are minute), while the other used techniques that are common in economics to argue that Super Bowl XXV (arguably the closest Super Bowl ever) was not actually that close. On baseball, we came up with a model predicting which hitters will get voted into the Hall of Fame after they retire (we found that there are no magic numbers - like 300 home runs, or 3000 hits - that make a difference to the voters).

And ... our library has a book called Curve Ball about the statistical analysis of baseball. It is pretty basic stuff, and good reading for someone who wants to understand statistics better without reading dry textbooks.