NCAA Football Bowl Results

I am posting my bowl season results with only the championship game remaining tonight. Here is how the model performed by game.

Matchup
63% BYU 52 24 UTEP 37%
54% Northern Illinois 40 17 Fresno State 46%
52% Ohio 21 48 Troy 48%
51% Southern Miss 28 31 Louisville 49%
32% Utah 3 26 Boise State 68%
45% Navy 14 35 San Diego State 55%
32% Tulsa 62 35 Hawaii 68%
46% Florida Intl. 34 32 Toledo 54%
59% Air Force 14 7 Georgia Tech 41%
55% West Virginia 7 23 NC State 45%
62% Missouri 24 27 Iowa 38%
32% East Carolina 20 51 Maryland 68%
51% Illinois 38 14 Baylor 49%
62% Oklahoma State 36 10 Arizona 38%
41% Army 16 14 SMU 59%
59% Kansas State 34 36 Syracuse 41%
38% North Carolina 30 27 Tennessee 62%
60% Nebraska 7 19 Washington 40%
46% South Florida 31 26 Clemson 54%
57% Notre Dame 33 17 Miami (FL) 43%
51% Georgia 6 10 UCF 49%
51% South Carolina 17 26 Florida State 49%
44% Northwestern 38 45 Texas Tech 56%
56% Florida 37 24 Penn State 44%
57% Alabama 49 7 Michigan State 43%
64% Mississippi State 52 14 Michigan 36%
46% Wisconsin 19 21 TCU 54%
31% Connecticut 20 48 Oklahoma 69%
57% Stanford 40 12 Virginia Tech 43%
52% Ohio State 31 26 Arkansas 48%
39% Mid Tennessee 21 35 Miami (OH) 61%
47% LSU 41 24 Texas A&M 53%
58% Pittsburgh 27 10 Kentucky 42%
65% Nevada 20 13 Boston College 35%
51% Oregon Auburn 49%

The expected model accuracy was 58%. Actual model performance was 59%. If we look at the distribution of expected results, we get the following. Blue bars represent relative probabilities of the model predicted the correct results for the number of games on the x-axis. 34 games have been played. The dark line represents the expected probability from guessing each game (50%). The model is clearly a little better than guessing.
Distribution of expected results
The model was not completely uniform. It predicted better at the higher probabilities and not as well near 50%.
Expected probability vs actual win/loss
If we consider the results as the percentage of points score by the predicited winner, then the model correlates well to the result.
Expected probability vs percentage of game points scored by expected winner
So, going into the championship game, I don’t think I could do any better than a coin toss on picking the winner. I would expect the game to go right down to the last couple of minutes of the 4th quarter.
Still: GO DUCKS!
O