It is also an administrative nightmare to keep on top of the vast number of games. Big shouts out to FBSchedules and Sports-Reference for their fantastic resources for this. ESPN are also pretty good for scores updates, as you might expect.
The college model is much more volatile than the NFL one, in that teams gain and lose far more ground with a victory. Home advantage also appears to be much stronger in college football than the professional leagues. However, there have been a touch over 14,300 games since the year 2000 and my model can correctly predict the outcome 74% of the time (compared to 64% for the NFL). So far this year it is sitting at 77% correct over 166 games.
Unlike the NFL and CFL, I am not placing bets on the NCAAF this year. I am tracking the odds and simulating what bets I would place, but the betting aspects of my model is not mature enough for this championship yet for me to be ready to actually risk money. My 52 fake bets so far have made an ROI of 5% though. For comparison, I probably will not even be making 52 total bets on the NFL this year and will be totaling far less than that on the CFL (only 6 from 50 games so far).
I have the top 25 FBS college football teams as we head into week 3 as:
- Alabama
- Clemson
- OhioState
- Georgia
- Oklahoma
- Wisconsin
- PennState
- Auburn
- LouisianaState
- SouthernCalifornia
- Stanford
- NotreDame
- TexasChristian
- OklahomaState
- MississippiState
- CentralFlorida
- VirginiaTech
- Washington
- BoiseState
- NorthCarolinaState
- WashingtonState
- Miami(FL)
- Northwestern
- MichiganState
- Iowa
No comments:
Post a Comment