Computer Rating System Prediction Results for College Football (NCAA IA)
2017 Last Week
Through 2018-01-09
Rank |
System |
Pct. Correct |
Against Spread |
Absolute Error |
Bias |
Mean Square Error |
games |
suw |
sul |
atsw |
atsl |
1 | Sagarin Points | 0.67500 | 0.69231 | 12.9743 | 1.4812 | 299.521 | 40 | 27 | 13 | 27 | 12 |
2 | ThePowerRank.com | 0.60000 | 0.69231 | 12.7175 | 1.2825 | 288.079 | 40 | 24 | 16 | 27 | 12 |
3 | TeamRankings.com | 0.65000 | 0.69231 | 12.9775 | 1.7325 | 298.291 | 40 | 26 | 14 | 27 | 12 |
4 | Linear Regression | 0.60000 | 0.66667 | 13.5300 | 2.2980 | 319.335 | 40 | 24 | 16 | 26 | 13 |
5 | ComPughter Ratings | 0.63158 | 0.64865 | 13.9800 | 2.4611 | 348.728 | 38 | 24 | 14 | 24 | 13 |
6 | Talisman Red | 0.65625 | 0.64516 | 12.9741 | 0.5578 | 300.058 | 32 | 21 | 11 | 20 | 11 |
7 | The Sports Cruncher | 0.63636 | 0.62500 | 12.9300 | -1.3730 | 284.107 | 33 | 21 | 12 | 20 | 12 |
8 | Ashby AccuRatings | 0.67500 | 0.62162 | 13.1750 | 1.1755 | 302.887 | 40 | 27 | 13 | 23 | 14 |
9 | Sagarin Ratings | 0.72500 | 0.61538 | 12.9975 | 1.2450 | 304.097 | 40 | 29 | 11 | 24 | 15 |
10 | Edward Kambour | 0.67500 | 0.61538 | 13.2163 | 0.9543 | 296.423 | 40 | 27 | 13 | 24 | 15 |
11 | System Average | 0.62500 | 0.61538 | 13.3943 | 1.4342 | 309.432 | 40 | 25 | 15 | 24 | 15 |
12 | Stat Fox | 0.57500 | 0.61111 | 13.8748 | 1.2258 | 300.837 | 40 | 23 | 17 | 22 | 14 |
13 | System Median | 0.62500 | 0.60526 | 13.2865 | 1.3195 | 308.260 | 40 | 25 | 15 | 23 | 15 |
14 | Bihl System | 0.60526 | 0.59459 | 13.8158 | 1.2221 | 334.578 | 38 | 23 | 15 | 22 | 15 |
15 | Computer Adjusted Line | 0.60000 | 0.59259 | 13.5125 | 0.8625 | 295.425 | 40 | 24 | 16 | 16 | 11 |
16 | Born Power Index | 0.50000 | 0.58974 | 14.1448 | 1.4303 | 334.187 | 40 | 20 | 20 | 23 | 16 |
17 | PerformanZ Ratings | 0.60000 | 0.58974 | 14.0363 | 0.7233 | 339.135 | 40 | 24 | 16 | 23 | 16 |
18 | Atomic Football | 0.65000 | 0.56757 | 13.0000 | 1.4500 | 294.147 | 40 | 26 | 14 | 21 | 16 |
19 | Howell | 0.60000 | 0.56757 | 13.8123 | 1.7627 | 356.027 | 40 | 24 | 16 | 21 | 16 |
20 | Line (updated) | 0.60000 | 0.56522 | 13.5000 | 0.8000 | 293.459 | 40 | 24 | 16 | 13 | 10 |
21 | Donchess Inference | 0.65000 | 0.56410 | 13.2233 | 1.3682 | 308.508 | 40 | 26 | 14 | 22 | 17 |
22 | Laz Index | 0.57500 | 0.56410 | 13.6638 | 1.2807 | 318.624 | 40 | 23 | 17 | 22 | 17 |
23 | Sagarin Golden Mean | 0.67500 | 0.56410 | 13.1620 | 1.4570 | 304.601 | 40 | 27 | 13 | 22 | 17 |
24 | Moore Power Ratings | 0.52500 | 0.56410 | 14.1400 | 0.6920 | 323.252 | 40 | 21 | 19 | 22 | 17 |
25 | Loudsound.org | 0.64103 | 0.55556 | 14.0769 | 0.5385 | 340.993 | 39 | 25 | 14 | 20 | 16 |
26 | Stephen Kerns | 0.57500 | 0.55263 | 14.3275 | 2.8775 | 350.243 | 40 | 23 | 17 | 21 | 17 |
27 | FEI Projections | 0.65000 | 0.55263 | 13.1500 | 2.2500 | 299.673 | 40 | 26 | 14 | 21 | 17 |
28 | DP Dwiggins | 0.55000 | 0.55263 | 14.2498 | 2.3498 | 368.360 | 40 | 22 | 18 | 21 | 17 |
29 | Liam Bressler | 0.47368 | 0.54054 | 13.5866 | 1.4608 | 307.963 | 38 | 18 | 20 | 20 | 17 |
30 | Dokter Entropy | 0.57500 | 0.53846 | 13.5815 | 1.3425 | 316.082 | 40 | 23 | 17 | 21 | 18 |
31 | Least Squares w/ HFA | 0.60000 | 0.53846 | 15.7933 | 4.5727 | 416.573 | 40 | 24 | 16 | 21 | 18 |
32 | ARGH Power Ratings | 0.70000 | 0.53846 | 13.7000 | 1.4125 | 338.006 | 40 | 28 | 12 | 21 | 18 |
33 | Payne Power Ratings | 0.67500 | 0.53846 | 13.6138 | 1.0353 | 337.626 | 40 | 27 | 13 | 21 | 18 |
34 | Dave Congrove | 0.60000 | 0.53846 | 13.2933 | 1.9327 | 284.087 | 40 | 24 | 16 | 21 | 18 |
35 | Super List | 0.67500 | 0.53846 | 13.9545 | 2.5890 | 366.486 | 40 | 27 | 13 | 21 | 18 |
36 | Laffaye RWP | 0.62500 | 0.53846 | 13.7765 | 0.9945 | 323.429 | 40 | 25 | 15 | 21 | 18 |
37 | ESPN FPI | 0.62500 | 0.53846 | 13.3940 | 1.1685 | 299.311 | 40 | 25 | 15 | 21 | 18 |
38 | Massey Ratings | 0.65000 | 0.53846 | 13.5320 | 1.5345 | 315.007 | 40 | 26 | 14 | 21 | 18 |
39 | NutShell Sports | 0.65000 | 0.53846 | 14.6210 | 1.6585 | 376.613 | 40 | 26 | 14 | 21 | 18 |
40 | Sagarin Recent | 0.60000 | 0.53846 | 13.5785 | 0.7330 | 324.817 | 40 | 24 | 16 | 21 | 18 |
41 | Roundtable | 0.61538 | 0.52632 | 14.1026 | 1.0256 | 343.204 | 39 | 24 | 15 | 20 | 18 |
42 | Brent Craig | 0.56410 | 0.52632 | 13.9582 | 2.5382 | 335.512 | 39 | 22 | 17 | 20 | 18 |
43 | Lee Burdorf | 0.48718 | 0.51351 | 14.8949 | 0.8077 | 346.899 | 39 | 19 | 20 | 19 | 18 |
44 | Logistic Regression | 0.60000 | 0.51282 | 14.3945 | -0.9840 | 359.797 | 40 | 24 | 16 | 20 | 19 |
45 | Billingsley | 0.52500 | 0.51282 | 14.7808 | 0.3247 | 365.236 | 40 | 21 | 19 | 20 | 19 |
46 | Pi-Ratings Mean | 0.55000 | 0.51282 | 13.3575 | 1.8775 | 310.816 | 40 | 22 | 18 | 20 | 19 |
47 | Keeper | 0.62500 | 0.51282 | 13.9915 | 1.5740 | 324.742 | 40 | 25 | 15 | 20 | 19 |
48 | Marsee | 0.55000 | 0.47368 | 14.4500 | 1.5500 | 339.990 | 40 | 22 | 18 | 18 | 20 |
49 | Pi-Rate Ratings | 0.52500 | 0.46154 | 13.7425 | 2.0375 | 324.805 | 40 | 21 | 19 | 18 | 21 |
50 | Beck Elo | 0.57500 | 0.46154 | 13.6770 | 0.8750 | 322.358 | 40 | 23 | 17 | 18 | 21 |
51 | Cleanup Hitter | 0.52500 | 0.46154 | 15.4500 | 0.5750 | 335.942 | 40 | 21 | 19 | 18 | 21 |
52 | Billingsley+ | 0.47500 | 0.46154 | 14.9725 | 0.5230 | 383.969 | 40 | 19 | 21 | 18 | 21 |
53 | Daniel Curry Index | 0.50000 | 0.46154 | 15.4815 | 1.7450 | 370.910 | 40 | 20 | 20 | 18 | 21 |
54 | Catherwood Ratings | 0.52500 | 0.44737 | 14.3250 | 1.2750 | 321.434 | 40 | 21 | 19 | 17 | 21 |
55 | Massey Consensus | 0.52500 | 0.41026 | 16.3817 | 1.5152 | 397.649 | 40 | 21 | 19 | 16 | 23 |
56 | Dunkel Index | 0.60000 | 0.41026 | 14.7320 | 1.6530 | 358.535 | 40 | 24 | 16 | 16 | 23 |
57 | Line (opening) | 0.45000 | 0.40741 | 14.0750 | 1.0750 | 323.871 | 40 | 18 | 22 | 11 | 16 |
58 | Pigskin Index | 0.47500 | 0.39474 | 14.7005 | 0.5505 | 356.405 | 40 | 19 | 21 | 15 | 23 |
59 | PI-Rate Bias | 0.50000 | 0.38462 | 13.7675 | 2.0475 | 325.636 | 40 | 20 | 20 | 15 | 24 |
60 | Line (Midweek) | 0.52500 | | 13.7500 | 0.8000 | 309.047 | 40 | 21 | 19 | | |
* 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