If a blogger with no readers has his home internet connection disrupted while on sabbatical, will anyone miss his posts? I believe so, I really do believe so.
Black Sunday, 2007. The football season is officially over. There is nothing to watch, nothing - ok, except three weekends of college basketball in a couple of weeks - until next September. But
1. Did you notice that the sun has been shining just a little brighter, the birds chirping just a little bit more merrily the last couple of days? That would be because Brett Favre, greatest quarterback ever, has decided to strap on the ol' helmet for one more year. Look for your Green Bay Packers in the playoffs next January.
2. The seven month hiatus until next season gives you time to bone up on your statistical skills with the aim of outperforming the Vegas bookies in next season's games (naturally this ought to be an intellectual exercise only!). Take a look at the article by Mark Glickman and Hal Stern in the March 1998 Journal of the American Statistical Association, "A State-Space Model for National Football League Scores." The abstract:
"This article develops a predictive model for National Footbal League (NFL) game scores using data from the period 1988-1993. The parameters of primary interest – measures of team strength – are expected to vary over time. Our model accounts for this source of variability by modeling football outcomes using a state-space model that assumes team strength parameters follow a first-order autoregressive process. Two sources of variation in team strengths are addressed in our model; week-to-week changes in tea strength due to injuries and other random factors, and season-to-season changes resulting from changes in personnel and other longer-term factors. Our model also incorporates a home-field advantage while allowing for the possibility that the magnitude of the advantage may vary across teams. The aim of the analysis is to obtain plausible inferences concerning team strengths and other model parameters, and to predict future game outcomes. Iterative simulation is used to obtain samples from the joint posterior distribution of all model parameters. Our model apears to outperform the Las Vegas “betting line” on a small test set consisting of the last 110 games of the 1993 NFL season."
Honors in economics to any student who replicates their analysis and applies it profitably in the 2007 football season. (Note: if Glickman and Sterm are still in academia seven years after publication of this article, chances are their algorithm has not worked out great since the 93 season. You may need to do some fine-tuning.)
Posts
0 comments:
Post a Comment