Monday, October 21, 2013

2013 CFB Modeling: Week 8 Game by Game Predictor

  • Ouch for the Huskies. What a disastrous performance in the desert. Here's hoping 3 weeks of Cal-bye-Colorado can cure what ails them. 
  • The model had Clemson #12 (AP had them #3). I guess even #12 was way too high; the model has them #24 now.
  • I have an actuarial exam coming up in a week and a half, and I've already gone dark (24 if you don't get the reference). I'll be able to get a Watchability chart up on Thursday as usual, but probably not Predictor update next Monday. 
  • Speaking of the exam...
  • I've gotten plenty of questions on methodology. Once the exam has passed, I'll actually have time to post about methodology. I'm going to do it in a series of Monday morning posts, we'll have Methodology Mondays around here. It's kind of like #fridayfocus or #throwbackthursday except with Actual Math, I hope to share enough details so people can understand what's going on and offer criticism, while still maintaining some level of propriety.

  • OK. Predictor has been updated with current data.
  • As per usual:
    • Type any FBS team into the box and the table will populate with 
      • Games
      • Past outcomes 
      • Future win probabilities 
      • Team's ranking each week (according to the model) 
      • Scenario analysis
      • Graph of expected total regular season wins
Please let me know if you have any questions or find any scheduling/outcomes errors.

No comments:

Post a Comment