The past couple years I've been working at developing a statistical model to predict NFL wins against the spread (using sportsbook.ag) for the line. I ran 3 models this year and I'd appreciate it if ya'll could give it a look and let you know what you think.
The blog post has a more in depth description of what methods I used.
---------- Post added at 09:42 PM ---------- Previous post was at 09:41 PM ----------
Oh I should also point out that I was comparing my model's predictions against the initial lines put out on 5/30