Sunday, May 20, 2012

2012 NFL Rankings - First Look

Green Bay Packer hearse
In this recent post, I took a look at the Week 1 opening point spreads for the 2012 NFL season in an effort to get an early read on changes in team strength. In the comments, it was pointed out that Cantor Gaming has published point spreads for the entire 2012 NFL season (except for Week 17). You can find them here (courtesy of

Getting 16 weeks of point spreads is a virtual goldmine for my ranking system, as my rankings developed in-season are based on weekly point spreads that are constantly being re-calibrated to injuries, roster changes, etc.  I think there's a lot of interesting results to tease out of this data set, but for this post, I am focusing on just looking at overall rankings and how they compare to last year.

For reference, here is some background on how the rankings are developed.  The Methodology page also has a simple example.

The Rankings

The table below summarizes the rankings implied by the 16 weeks of point spreads. GPF stands for "Generic Points Favored".  It's what you would expect a team to be favored by against a league average opponent at a neutral site.  The "Last Year" column refers to the Week 16 rankings I published last year at Advanced NFL Stats Community.  "GPF Change" is the change in the team's GPF relative to last year's Week 16 GPF.

  • Green Bay and New England stand atop the rankings, nearly even.  There's a pretty big gap between them and the rest of the league.
  • There is definitely some regression to the mean baked into the point spreads.  You can see that most clearly in the "GPF Change" column.
  • If these rankings hold, look for Houston to have the AFC South locked up by Week 12.  There is a huge gap between the Texans and the other three AFC South teams.
  • The Texans, Broncos, Bears, and Chiefs all show significant improvement, most likely due to better expected quarterback play.  The Texans, Bears, and Chiefs are getting their starting QB back from injury, and we all know what the Broncos did in the offseason.
  • The Saints had the biggest drop in GPF.  Some of the drop is likely due to regression to the mean, but is probably also a reflection of the loss of Williams, Payton, and Vilma.

I did some testing of rankings week to week and could not find any evidence that Cantor "regressed" the point spreads for the later weeks of the season.  In a future post, I hope to take a look at what these rankings imply about team-specific home field advantage.

Thanks again to the anonymous commenter in last week's post who called these point spreads to my attention.


  1. Very interesting! Am surprised the Broncos improved by only 2.7 points going from Tebowtime to Peyton, but then there's obviously a lot of uncertainty surrounding their new QB - aging, one year absence and whether he'll recover fully from his neck injury.

    The Saints - yep, and don't forget Brees hasn't signed a new contract. If the Cantor oddsmaker indeed factored that uncertainty in, I suppose we'll see NO's "power rating" changing per se the latest news of the negotiations with Brees. If not, somebody could make a killing by guessing the outcome correctly and bet the whole season for/against NO before it actually starts.

    So Schaub is worth around 3 points. Seems about right, huh?

    49ers expected to see a significant regression in spite of their talented additions. Makes sense as they were unusually lucky last season, with A.Smith throwing only 5 INTs.

    WAS declining by 2 points with RGIII? A rookie QB thing, I guess.

    Nice work, Michael!

  2. Thanks! Washington's drop confused me as well.

    That being said, these are the team strengths according to Cantor Gaming, not necessarily the betting market at large. It will be interesting to see how these lines recalibrate as the money starts pouring in on each side.