Computer Rating System Prediction Results for College Football (NCAA IA)
2009 Retrodiction Results
Through
| Rank |
System |
Pct. Correct |
Against Spread |
Absolute Error |
Bias |
Mean Square Error |
games |
suw |
sul |
atsw |
atsl |
| 1 | Least Squares w/ HFA | 0.84314 | 0.00000 | 8.3487 | 0.2736 | 120.486 | 714 | 602 | 112 | 0 | 0 |
| 2 | Logistic Regression | 0.80952 | 0.00000 | 13.1234 | -3.6743 | 285.559 | 714 | 578 | 136 | 0 | 0 |
| 3 | NutShell Combo | 0.80812 | 0.00000 | 10.1149 | 0.3401 | 168.472 | 714 | 577 | 137 | 0 | 0 |
| 4 | Least Squares | 0.80532 | 0.00000 | 9.4972 | 0.2247 | 148.888 | 714 | 575 | 139 | 0 | 0 |
| 5 | Sagarin Predictive | 0.80392 | 0.00000 | 9.7150 | 0.7855 | 153.902 | 714 | 574 | 140 | 0 | 0 |
| 6 | Laz Index | 0.80252 | 0.00000 | 9.7953 | 0.3511 | 158.911 | 714 | 573 | 141 | 0 | 0 |
| 7 | Billingsley+ | 0.80112 | 0.00000 | 9.8054 | 0.7180 | 166.083 | 714 | 572 | 142 | 0 | 0 |
| 8 | Sonny Moore | 0.79972 | 0.00000 | 9.8386 | 0.7398 | 159.226 | 714 | 571 | 143 | 0 | 0 |
| 9 | Covers.com | 0.79832 | 0.00000 | 10.6341 | -0.2744 | 186.175 | 714 | 570 | 144 | 0 | 0 |
| 10 | Stephen Kerns | 0.79552 | 0.00000 | 9.8010 | 0.9677 | 157.736 | 714 | 568 | 146 | 0 | 0 |
| 11 | NutShell Sports | 0.79552 | 0.00000 | 9.9998 | 0.8335 | 164.251 | 714 | 568 | 146 | 0 | 0 |
| 12 | CPA Rankings | 0.79132 | 0.00000 | 9.5698 | -0.2067 | 150.610 | 714 | 565 | 149 | 0 | 0 |
| 13 | Born Power Index | 0.78992 | 0.00000 | 9.9041 | 1.0303 | 158.078 | 714 | 564 | 150 | 0 | 0 |
| 14 | Pigskin Index | 0.78992 | 0.00000 | 9.9132 | 0.5382 | 161.535 | 714 | 564 | 150 | 0 | 0 |
| 15 | Nutshell Girl | 0.78711 | 0.00000 | 11.4523 | -0.1535 | 220.037 | 714 | 562 | 152 | 0 | 0 |
| 16 | NationalSportsRankings | 0.77591 | 0.00000 | 9.8039 | 0.4618 | 165.187 | 714 | 554 | 160 | 0 | 0 |
| 17 | Ashby AccuRatings | 0.77171 | 0.00000 | 10.6569 | 1.0623 | 183.613 | 714 | 551 | 163 | 0 | 0 |
| 18 | Stortrends | 0.76814 | 0.00000 | 10.1856 | -0.0655 | 166.544 | 703 | 540 | 163 | 0 | 0 |
| 19 | PerformanZ Ratings | 0.81092 | 0.00000 | 10.2000 | 0.3289 | 168.966 | 714 | 579 | 135 | 0 | 0 |
| 20 | Beck Elo | 0.81092 | 0.00000 | 9.8873 | 0.1811 | 166.104 | 714 | 579 | 135 | 0 | 0 |
| 21 | Billingsly | 0.81653 | 0.00000 | 11.0919 | 0.2098 | 204.488 | 714 | 583 | 131 | 0 | 0 |
| 22 | Sagarin Elo | 0.84034 | 0.00000 | 10.5056 | 0.5339 | 186.933 | 714 | 600 | 114 | 0 | 0 |
| 23 | Wolfe * | 0.83894 | 0.00000 | 11.4750 | 0.9808 | 219.604 | 714 | 599 | 115 | 0 | 0 |
| 24 | Sagarin | 0.83473 | 0.00000 | 9.9181 | 0.6455 | 163.121 | 714 | 596 | 118 | 0 | 0 |
| 25 | Martien Maas | 0.83193 | 0.00000 | 10.5164 | 0.6175 | 184.620 | 714 | 594 | 120 | 0 | 0 |
| 26 | SuperList | 0.83193 | 0.00000 | 11.2859 | 1.1261 | 208.702 | 714 | 594 | 120 | 0 | 0 |
| 27 | Massey BCS * | 0.83053 | 0.00000 | 9.8853 | 0.8813 | 163.786 | 714 | 593 | 121 | 0 | 0 |
| 28 | Edward Kambour | 0.82913 | 0.00000 | 8.7623 | 0.5477 | 129.109 | 714 | 592 | 122 | 0 | 0 |
| 29 | Anderson/Hester * | 0.82913 | 0.00000 | 10.4552 | 0.7338 | 183.230 | 714 | 592 | 122 | 0 | 0 |
| 30 | CPA Retro | 0.82913 | 0.00000 | 10.6822 | -0.2083 | 193.188 | 714 | 592 | 122 | 0 | 0 |
| 31 | Payne Power Ratings | 0.82773 | 0.00000 | 9.9365 | 0.0928 | 162.893 | 714 | 591 | 123 | 0 | 0 |
| 32 | Colley Rankings * | 0.82773 | 0.00000 | 10.6053 | 0.6938 | 184.955 | 714 | 591 | 123 | 0 | 0 |
| 33 | System Average | 0.82633 | 0.00000 | 9.6524 | 0.3864 | 154.135 | 714 | 590 | 124 | 0 | 0 |
| 34 | Warren Claassen | 0.82633 | 0.00000 | 10.0728 | 0.3127 | 169.502 | 714 | 590 | 124 | 0 | 0 |
| 35 | system Median | 0.82493 | 0.00000 | 9.6589 | 0.3472 | 154.646 | 714 | 589 | 125 | 0 | 0 |
| 36 | Massey Concensus Rank | 0.82353 | 0.00000 | 10.1665 | 0.8555 | 170.974 | 714 | 588 | 126 | 0 | 0 |
| 37 | Frank Alder | 0.82073 | 0.00000 | 9.9358 | 0.3487 | 162.454 | 714 | 586 | 128 | 0 | 0 |
| 38 | Schmidt Comp. Ratings | 0.81793 | 0.00000 | 10.0645 | 0.1443 | 171.262 | 714 | 584 | 130 | 0 | 0 |
| 39 | Stat Fox | 0.75910 | 0.00000 | 10.0126 | 1.6586 | 159.284 | 714 | 542 | 172 | 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