Wednesday, October 17, 2012

College Football Rankings - Week 8

Here are the betting market rankings for College Football for week 8.  See my first post for background and my Methodology page for a simple example of how the rankings work.

For this version, I have added the offensive and defensive components of each team's ranking (oGPF and dGPF).  I can derive these by looking at the over/under in combination with the point spread.

I modified the methodology slightly relative to last week.  I switched to a weighting based on weeks elapsed as opposed to days.  I also modified the weights to favor recent results more heavily.  The modified weights were based on back testing against prior seasons.  The model still has a somewhat large prediction error of about 3 points.  I suspect that with the wide dispersion of talent among Division 1A teams, my simple linear approach to team strength may be breaking down at the extreme ends. In contrast, for the NFL, I've been able to get the error down to about 1.8 points on average.


  1. The charts are really cool looking too. So the vertical axis is the overall successful value, basically? It's a great idea and one I wish we'd see more of - visualization of advanced data is often a much clearer way to get an idea of interesting trends and information.

  2. I think the rankings look a bit out of whack this week, particularly Georgia. Part of the reason is that I currently don't have an adjustment for game outcomes (as mentioned at the bottom of my Methodology page).

    These college football rankings are still a work in progress. Adding a game outcome adjustment looks to improve the rankings significantly. I just need to find the time to optimize the adjustment and flow that into the rankings here.

  3. Am I blind, or is the link for the rankings not on this page?

    1. The table is embedded in the post (there is no link). Are you not seeing the table?

    2. I wasn't seeing at work but am seeing it at home. My office must have a filter of some kind.