Computer Rating System Prediction Results for College Football (NCAA IA)
2009 Last Week
Through
| Rank |
System |
Pct. Correct |
Against Spread |
Absolute Error |
Bias |
Mean Square Error |
games |
suw |
sul |
atsw |
atsl |
| 1 | Massey * | 0.79412 | 0.81818 | 9.5426 | 3.5397 | 156.537 | 34 | 27 | 7 | 27 | 6 |
| 2 | Hank Trexler | 0.67647 | 0.53125 | 12.9412 | 4.2353 | 274.943 | 34 | 23 | 11 | 17 | 15 |
| 3 | CPA Retro | 0.64706 | 0.60606 | 12.2321 | 4.9915 | 235.369 | 34 | 22 | 12 | 20 | 13 |
| 4 | Edward Kambour | 0.64706 | 0.60606 | 12.4276 | 4.6224 | 245.812 | 34 | 22 | 12 | 20 | 13 |
| 5 | PerformanZ Ratings | 0.64706 | 0.57576 | 13.3356 | 5.4809 | 271.902 | 34 | 22 | 12 | 19 | 14 |
| 6 | ARGH Power Ratings | 0.61765 | 0.58065 | 12.6691 | 5.1985 | 258.005 | 34 | 21 | 13 | 18 | 13 |
| 7 | Nutshell Girl | 0.61765 | 0.60606 | 11.8676 | 3.6406 | 240.044 | 34 | 21 | 13 | 20 | 13 |
| 8 | Tom Benson | 0.60606 | 0.65625 | 13.3982 | 5.6012 | 273.036 | 33 | 20 | 13 | 21 | 11 |
| 9 | NutShell Combo | 0.58824 | 0.63636 | 12.7156 | 4.5479 | 274.975 | 34 | 20 | 14 | 21 | 12 |
| 10 | Least Squares w/ HFA | 0.58824 | 0.45455 | 15.3100 | 7.9982 | 387.760 | 34 | 20 | 14 | 15 | 18 |
| 11 | Marsee | 0.57576 | 0.51515 | 13.6364 | 4.4848 | 306.481 | 33 | 19 | 14 | 17 | 16 |
| 12 | Sagarin Predictive | 0.55882 | 0.60606 | 13.0376 | 5.3565 | 272.720 | 34 | 19 | 15 | 20 | 13 |
| 13 | System Average | 0.55882 | 0.51515 | 12.8350 | 5.0597 | 269.424 | 34 | 19 | 15 | 17 | 16 |
| 14 | Dave Congrove | 0.55882 | 0.60606 | 12.8162 | 5.0744 | 282.453 | 34 | 19 | 15 | 20 | 13 |
| 15 | Billingsley | 0.55882 | 0.69697 | 12.5265 | 4.0265 | 257.252 | 34 | 19 | 15 | 23 | 10 |
| 16 | Howell | 0.55882 | 0.62500 | 12.3094 | 4.5153 | 248.448 | 34 | 19 | 15 | 20 | 12 |
| 17 | Moore Power Ratings | 0.52941 | 0.57576 | 13.5482 | 5.3788 | 304.799 | 34 | 18 | 16 | 19 | 14 |
| 18 | Pigskin Index | 0.52941 | 0.53125 | 13.5882 | 4.5882 | 301.028 | 34 | 18 | 16 | 17 | 15 |
| 19 | NationalSportsRankings | 0.52941 | 0.51515 | 13.6171 | 5.2653 | 304.380 | 34 | 18 | 16 | 17 | 16 |
| 20 | NutShell Sports | 0.52941 | 0.51515 | 14.0056 | 5.4250 | 321.499 | 34 | 18 | 16 | 17 | 16 |
| 21 | Lee Burdorf | 0.52941 | 0.51515 | 14.2941 | 4.7824 | 335.417 | 34 | 18 | 16 | 17 | 16 |
| 22 | Keeper | 0.52941 | 0.36364 | 14.3859 | 6.0547 | 362.625 | 34 | 18 | 16 | 12 | 21 |
| 23 | Born Power Index | 0.52941 | 0.38710 | 14.3915 | 5.5562 | 342.801 | 34 | 18 | 16 | 12 | 19 |
| 24 | Beck Elo | 0.52941 | 0.54545 | 13.4900 | 4.6135 | 279.185 | 34 | 18 | 16 | 18 | 15 |
| 25 | Atomic Football | 0.52941 | 0.48485 | 13.1229 | 5.4918 | 279.812 | 34 | 18 | 16 | 16 | 17 |
| 26 | Line (opening) | 0.52941 | 0.53125 | 13.1176 | 5.6471 | 292.150 | 34 | 18 | 16 | 17 | 15 |
| 27 | Billingsley+ | 0.52941 | 0.66667 | 12.7100 | 6.1471 | 280.866 | 34 | 18 | 16 | 22 | 11 |
| 28 | Covers.com | 0.52941 | 0.60606 | 12.7150 | 3.4879 | 266.437 | 34 | 18 | 16 | 20 | 13 |
| 29 | System Median | 0.52941 | 0.50000 | 12.8079 | 5.0891 | 266.100 | 34 | 18 | 16 | 16 | 16 |
| 30 | Anderson/Hester * | 0.52941 | 0.60606 | 12.8485 | 4.2038 | 270.931 | 34 | 18 | 16 | 20 | 13 |
| 31 | Schmidt Comp. Ratings | 0.52941 | 0.57576 | 12.9650 | 4.4297 | 260.545 | 34 | 18 | 16 | 19 | 14 |
| 32 | Warren Claassen | 0.52941 | 0.48485 | 12.9771 | 4.8035 | 270.412 | 34 | 18 | 16 | 16 | 17 |
| 33 | Frank Alder | 0.52941 | 0.54545 | 13.1153 | 4.6588 | 275.601 | 34 | 18 | 16 | 18 | 15 |
| 34 | Stephen Kerns | 0.50000 | 0.57576 | 14.6212 | 7.2194 | 338.948 | 34 | 17 | 17 | 19 | 14 |
| 35 | Sportrends | 0.50000 | 0.54839 | 14.0735 | 4.3676 | 319.843 | 34 | 17 | 17 | 17 | 14 |
| 36 | Stat Fox | 0.50000 | 0.48276 | 14.0597 | 6.0597 | 321.118 | 34 | 17 | 17 | 14 | 15 |
| 37 | Linear Regression | 0.50000 | 0.51515 | 14.0194 | 5.7500 | 299.180 | 34 | 17 | 17 | 17 | 16 |
| 38 | CPA Rankings | 0.50000 | 0.50000 | 13.6641 | 5.5806 | 293.364 | 34 | 17 | 17 | 16 | 16 |
| 39 | CF By the Numbers | 0.50000 | 0.50000 | 13.5000 | 5.1471 | 279.721 | 34 | 17 | 17 | 15 | 15 |
| 40 | Bihl System | 0.50000 | 0.51515 | 13.3994 | 5.3406 | 277.549 | 34 | 17 | 17 | 17 | 16 |
| 41 | Line (updated) | 0.50000 | | 13.1176 | 6.2941 | 290.717 | 34 | 17 | 17 | | |
| 42 | Payne Power Ratings | 0.50000 | 0.54545 | 13.1091 | 4.8526 | 271.686 | 34 | 17 | 17 | 18 | 15 |
| 43 | Computer Adjusted Line | 0.50000 | 0.64286 | 13.0441 | 6.1029 | 287.059 | 34 | 17 | 17 | 9 | 5 |
| 44 | Sagarin Elo | 0.50000 | 0.57576 | 12.9212 | 5.0300 | 263.856 | 34 | 17 | 17 | 19 | 14 |
| 45 | Super List | 0.47059 | 0.51515 | 13.8665 | 6.6818 | 306.290 | 34 | 16 | 18 | 17 | 16 |
| 46 | Laz Index | 0.47059 | 0.45455 | 13.4824 | 4.7929 | 289.480 | 34 | 16 | 18 | 15 | 18 |
| 47 | Dokter Entropy | 0.47059 | 0.54545 | 13.2968 | 4.9015 | 276.828 | 34 | 16 | 18 | 18 | 15 |
| 48 | Massey Consensus | 0.47059 | 0.57576 | 13.1544 | 5.1021 | 280.704 | 34 | 16 | 18 | 19 | 14 |
| 49 | Wolfe * | 0.47059 | 0.57576 | 13.0665 | 5.1682 | 274.854 | 34 | 16 | 18 | 19 | 14 |
| 50 | Martien Maas | 0.47059 | 0.57576 | 12.9306 | 4.4629 | 271.795 | 34 | 16 | 18 | 19 | 14 |
| 51 | Dunkel Index | 0.44118 | 0.33333 | 14.6071 | 6.1347 | 348.705 | 34 | 15 | 19 | 11 | 22 |
| 52 | Bassett Model | 0.44118 | 0.51515 | 14.1688 | 6.0806 | 307.953 | 34 | 15 | 19 | 17 | 16 |
| 53 | Harmon Forcast | 0.44118 | 0.45161 | 13.9412 | 5.3529 | 300.157 | 34 | 15 | 19 | 14 | 17 |
| 54 | Ashby AccuRatings | 0.44118 | 0.51613 | 13.4118 | 5.6471 | 300.591 | 34 | 15 | 19 | 16 | 15 |
| 55 | Sagarin | 0.44118 | 0.54545 | 13.0294 | 5.1782 | 262.816 | 34 | 15 | 19 | 18 | 15 |
| 56 | Logistic Regression | 0.44118 | 0.66667 | 12.5526 | 2.7774 | 248.845 | 34 | 15 | 19 | 22 | 11 |
| 57 | Catherwood Ratings | 0.41176 | 0.48387 | 14.5882 | 5.5882 | 336.670 | 34 | 14 | 20 | 15 | 16 |
| 58 | Colley Rankings * | 0.38235 | 0.57576 | 13.0550 | 4.6485 | 273.241 | 34 | 13 | 21 | 19 | 14 |
* 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