Parent DirectoryI am using a least squares algorithm implemented in Visual Basic using an Access Database to store all my data. The factors I am using for the pre-season rating are last years final ratings, the number of returning starters, schedule strength, coaching experience, and a key returnee factor (e.g., Peyton Manning).
For the season I am going to add the results of the current week, winning margin weighted with the team strength and possibly a weather scaling factor, but I'll have do a little analysis to see how it come out.
I would like to include recruiting, but it's hard to figure when this really becomes a factor, probably the 3rd or 4th games. I have numerical ratings, but I really don't want to publish them at this time.
Michael DeSimon / desimone@algorithms.com