- Joined
- Sep 18, 2011
- Messages
- 5,344
- Reaction Score
- 21,804
Look closer at the ratings. On a simple average basis, Sagarin rates the SEC-West the top group with the B12 second. Sagarin also has a central mean rating which gives a higher weight to the schools in the middle of the conference and progressively less weighting to the higher and lower schools and the central mean rates the B12 1st. Makes sense since the lowest rated B12 team is WVU at #50. B12 is strong in the middle and it is not top or bottom heavy.As someone who likes Sagarin, I think he does a better job with college basketball than college football. He has the Big 12 as the #1 league right now. Yeah.
If you use the Sagarin ratings now and use them for UConn's games, they are fairly accurate:
Utah St.: Predicts 6.7 point loss and it was 11 point loss.
CCSU: Predicts 33.8 point win and it was a 25 point win.
Syracuse: Predicts 10.5 point loss and it was a 34 point loss.
Michigan: Predicts 37.5 point loss and it was a 59 point loss.
NC St.: Predicts 20.1 point loss and it was a 31 point loss.
Fresno St.: Predicts 9.4 point loss and it was a 5 point win.
FIU: Predicts a 10.8 point win and it was a 21 point win.
Ball St: Predicts a 6.6 point loss and it was a 4 point loss.
BC: Predicts a 2 point loss and it was a 10 point win.
UMass: Predicts a 18 point win and it was a 17 point win.
Liberty: Predicts a 7.3 point loss and it was a 3 point win.
You have to remember, college football has fewer datapoints than college basketball so the statistical models take more time to normalize. And, the Syracuse, Michigan, and NC St., losses were pretty negative for the model and you can see the turn starting with the Fresno St. game.