Computer Rating System Prediction Results for College Football (NCAA IA)
2017 Retrodiction Results
Through
| Rank |
System |
Pct. Correct |
Against Spread |
Absolute Error |
Bias |
Mean Square Error |
games |
suw |
sul |
atsw |
atsl |
| 1 | Sagarin | 0.74903 | 0.00000 | 12.9214 | 0.0742 | 268.260 | 773 | 579 | 194 | 0 | 0 |
| 2 | Sagarin Predictive | 0.56921 | 0.00000 | 16.8393 | -1.7192 | 456.794 | 773 | 440 | 333 | 0 | 0 |
| 3 | Massey Concensus Rank | 0.66192 | 0.00000 | 15.1267 | 1.1300 | 355.343 | 772 | 511 | 261 | 0 | 0 |
| 4 | SuperList | 0.67012 | 0.00000 | 15.2351 | 0.3900 | 363.511 | 773 | 518 | 255 | 0 | 0 |
| 5 | Sonny Moore | 0.70763 | 0.00000 | 14.2927 | 0.2849 | 318.035 | 773 | 547 | 226 | 0 | 0 |
| 6 | Laz Index | 0.71022 | 0.00000 | 14.4727 | 0.1033 | 331.945 | 773 | 549 | 224 | 0 | 0 |
| 7 | Born Power Index | 0.73609 | 0.00000 | 13.5998 | 0.9244 | 299.913 | 773 | 569 | 204 | 0 | 0 |
| 8 | Stat Fox | 0.73609 | 0.00000 | 12.1138 | 1.4968 | 235.481 | 773 | 569 | 204 | 0 | 0 |
| 9 | NutShell Sports | 0.73739 | 0.00000 | 15.4470 | 1.5045 | 383.138 | 773 | 570 | 203 | 0 | 0 |
| 10 | Pigskin Index | 0.73739 | 0.00000 | 12.5602 | 0.6965 | 253.261 | 773 | 570 | 203 | 0 | 0 |
| 11 | Sagarin Golden Mean | 0.74515 | 0.00000 | 12.9126 | 0.0803 | 268.141 | 773 | 576 | 197 | 0 | 0 |
| 12 | Least Squares w/ HFA | 0.79922 | 0.00000 | 9.4763 | 0.1814 | 142.922 | 772 | 617 | 155 | 0 | 0 |
| 13 | Sagarin Recent | 0.75420 | 0.00000 | 12.3486 | 0.1758 | 245.305 | 773 | 583 | 190 | 0 | 0 |
| 14 | Ashby AccuRatings | 0.75518 | 0.00000 | 10.9170 | 1.6325 | 192.564 | 772 | 583 | 189 | 0 | 0 |
| 15 | system Median | 0.75679 | 0.00000 | 11.5088 | 0.0720 | 222.198 | 773 | 585 | 188 | 0 | 0 |
| 16 | System Average | 0.76585 | 0.00000 | 11.1224 | 0.0099 | 208.711 | 773 | 592 | 181 | 0 | 0 |
| 17 | Stephen Kerns | 0.77034 | 0.00000 | 10.8683 | 0.9803 | 197.085 | 762 | 587 | 175 | 0 | 0 |
| 18 | Beck Elo | 0.77232 | 0.00000 | 10.5646 | -0.1582 | 191.351 | 773 | 597 | 176 | 0 | 0 |
| 19 | Logistic Regression | 0.77850 | 0.00000 | 15.2796 | -3.3124 | 386.847 | 772 | 601 | 171 | 0 | 0 |
| 20 | Least Squares | 0.77850 | 0.00000 | 10.4450 | -0.4734 | 172.947 | 772 | 601 | 171 | 0 | 0 |
| 21 | PerformanZ Ratings | 0.78886 | 0.00000 | 11.3575 | 0.1725 | 205.494 | 772 | 609 | 163 | 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