Football-o-Genetics is an application for "evolving" near-optimal offensive play calling strategies. The application incorporates ideas from both artificial intelligence (specifically, genetic algorithms) and advanced statistics (in the form of a Markov model of an offensive drive) to accomplish this end. For an in-depth (i.e., academic) discussion of the methods used in Football-o-Genetics (in addition to an example analysis), please read the white paper.
Addendum: I wrote this application when I was first teaching myself computer science. I've since realized that a genetic algorithm is not the best solution for this type of reinforcement learning problem. I've also implemented a Q-learning solution for play calling strategy, but even this approach is inadequate because it fails to model the adversarial aspect of the defense. For more information on the Q-learning approach, see my blog post here.
2016/12/09:
2013/08/24:
2013/07/14: