tag:blogger.com,1999:blog-38600807.post3324901003753462052..comments2022-10-29T04:20:46.650-04:00Comments on Advanced Football Analytics (formerly Advanced NFL Stats): NFL Prediction TrackingUnknownnoreply@blogger.comBlogger3125tag:blogger.com,1999:blog-38600807.post-20948118144480520972007-11-04T11:00:00.000-05:002007-11-04T11:00:00.000-05:00I do have a linear model for prediction point diff...I do have a linear model for prediction point differentials already that I ran for most of 2006. I think it works as well as any other, which doesn't say very much. Scoring in the NFL is so irregular, non-normal, and non-linear. Unlike baseball the points are not in increments of one. There are a lot more 3 point wins or 7 point wins than 1 point wins, which blows the assumptions required for linear regression. Plus, there is always the "trash-time" effect that would be so hard to quantify. I do have some ideas for approaches to overcome these problems, but I'll probably have to work on it in the off-season.Brian Burkehttps://www.blogger.com/profile/12371470711365236987noreply@blogger.comtag:blogger.com,1999:blog-38600807.post-2378111100965608562007-11-04T00:32:00.000-04:002007-11-04T00:32:00.000-04:00You could always run your stats through a linear r...You could always run your stats through a linear regression model to predict the result.Derekhttps://www.blogger.com/profile/17941314072950152029noreply@blogger.comtag:blogger.com,1999:blog-38600807.post-4686182893559227972007-11-03T20:01:00.000-04:002007-11-03T20:01:00.000-04:00I wrote an e-mail but was told that you could not ...I wrote an e-mail but was told that you could not be included on the list because you only predict winners and not results. :(Anonymousnoreply@blogger.com