Computer Rating System Prediction Results for College Football (NCAA IA)
2008 Retrodiction Results
| System |
Pct. Correct |
Against Spread |
Absolute Error |
Bias |
Mean Square Error |
games |
suw |
sul |
atsw |
atsl |
| Least Squares w/ HFA | 0.82238 | 0.00000 | 9.4249 | 0.1467 | 138.649 | 715 | 588 | 127 | 0 | 0 |
| PerformanZ Ratings | 0.79860 | 0.00000 | 11.2465 | -0.6441 | 201.534 | 715 | 571 | 144 | 0 | 0 |
| Sagarin Predictive | 0.79441 | 0.00000 | 10.7132 | -0.6661 | 183.515 | 715 | 568 | 147 | 0 | 0 |
| Beck Elo | 0.79441 | 0.00000 | 10.7993 | -0.6846 | 202.062 | 715 | 568 | 147 | 0 | 0 |
| Born Power Index | 0.79441 | 0.00000 | 11.0206 | 0.3013 | 193.192 | 715 | 568 | 147 | 0 | 0 |
| CPA Retro | 0.79301 | 0.00000 | 11.4122 | 0.8321 | 224.229 | 715 | 567 | 148 | 0 | 0 |
| Sonny Moore | 0.79161 | 0.00000 | 10.9238 | -0.0045 | 194.398 | 715 | 566 | 149 | 0 | 0 |
| CPA Rankings | 0.78741 | 0.00000 | 10.5493 | 0.9923 | 178.431 | 715 | 563 | 152 | 0 | 0 |
| Stat Fox | 0.78462 | 0.00000 | 10.8112 | 0.8056 | 188.265 | 715 | 561 | 154 | 0 | 0 |
| NutShell Sports | 0.78182 | 0.00000 | 11.1838 | 0.0708 | 200.418 | 715 | 559 | 156 | 0 | 0 |
| Pigskin Index | 0.77902 | 0.00000 | 11.0741 | -0.2238 | 201.776 | 715 | 557 | 158 | 0 | 0 |
| NutShell Retro | 0.77622 | 0.00000 | 11.1928 | 0.1162 | 202.749 | 715 | 555 | 160 | 0 | 0 |
| Wolfe * | 0.80000 | 0.00000 | 11.4045 | 0.5148 | 215.352 | 715 | 572 | 143 | 0 | 0 |
| Laz Index | 0.80280 | 0.00000 | 10.7286 | -0.3335 | 194.155 | 715 | 574 | 141 | 0 | 0 |
| Least Squares | 0.80280 | 0.00000 | 10.5116 | -0.2513 | 175.773 | 715 | 574 | 141 | 0 | 0 |
| Sagarin | 0.81818 | 0.00000 | 10.9679 | -0.7726 | 199.406 | 715 | 585 | 130 | 0 | 0 |
| SuperList | 0.81678 | 0.00000 | 12.3963 | 0.2819 | 242.266 | 715 | 584 | 131 | 0 | 0 |
| Massey Concensus Rank | 0.81399 | 0.00000 | 11.2810 | 0.0750 | 210.918 | 715 | 582 | 133 | 0 | 0 |
| Sagarin Elo | 0.80979 | 0.00000 | 11.5516 | -0.8605 | 227.889 | 715 | 579 | 136 | 0 | 0 |
| Massey BCS * | 0.80979 | 0.00000 | 11.7891 | -0.1096 | 240.385 | 715 | 579 | 136 | 0 | 0 |
| system Median | 0.80839 | 0.00000 | 10.5103 | -0.1635 | 185.653 | 715 | 578 | 137 | 0 | 0 |
| Logistic Regression | 0.80839 | 0.00000 | 14.1795 | -4.2595 | 345.511 | 715 | 578 | 137 | 0 | 0 |
| System Average | 0.80559 | 0.00000 | 10.4871 | -0.1082 | 183.715 | 715 | 576 | 139 | 0 | 0 |
| Colley Rankings * | 0.80559 | 0.00000 | 11.5409 | 0.1077 | 224.218 | 715 | 576 | 139 | 0 | 0 |
| Anderson/Hester * | 0.80426 | 0.00000 | 11.5328 | -0.0564 | 226.212 | 705 | 567 | 138 | 0 | 0 |
| Martien Maas | 0.80420 | 0.00000 | 11.2757 | 0.3033 | 213.626 | 715 | 575 | 140 | 0 | 0 |
| Ashby AccuRatings | 0.75664 | 0.00000 | 11.6615 | 0.3133 | 225.496 | 715 | 541 | 174 | 0 | 0 |
* This system does not make predictions. I make predictions for this
system by translating it to a new scale that allows for making predictions.
Retrodictive records are found by taking the ratings from the current week
and applying them to the entire season to date.
The ideal system would be one that has the highest correct game decisions,
has the smallest mean error(deviation from the actual game result), and has
a bias of zero.
Mean Error = average[abs(prediction-actual)]
Bias = agerage(prediction - actual)
Std. = Standard Deviation of individual game biases