Computer Rating System Prediction Results for College Football (NCAA IA)
2021 Season Totals
Through 2022-01-11
Rank |
System |
Pct. Correct |
Against Spread |
Absolute Error |
Bias |
Mean Square Error |
games |
suw |
sul |
atsw |
atsl |
1 | Brent Craig | 0.70843 | 0.51942 | 13.0799 | 0.5354 | 263.541 | 415 | 294 | 121 | 214 | 198 |
2 | Keeper | 0.71035 | 0.51531 | 13.4245 | 1.2305 | 283.635 | 763 | 542 | 221 | 387 | 364 |
3 | Least Squares w/ HFA | 0.66584 | 0.51515 | 15.6582 | 0.6268 | 372.986 | 401 | 267 | 134 | 204 | 192 |
4 | Stephen Kerns | 0.68831 | 0.50725 | 13.9070 | 1.0216 | 301.969 | 770 | 530 | 240 | 385 | 374 |
5 | Pi-Rate Ratings | 0.69870 | 0.50531 | 13.2049 | 1.0277 | 264.988 | 770 | 538 | 232 | 381 | 373 |
6 | Dave Congrove | 0.68270 | 0.50528 | 13.7253 | 0.9936 | 288.525 | 769 | 525 | 244 | 383 | 375 |
7 | FEI Projections | 0.68277 | 0.50412 | 13.0317 | 0.3315 | 257.875 | 621 | 424 | 197 | 306 | 301 |
8 | Laffaye RWP | 0.70141 | 0.50359 | 13.8664 | 0.2000 | 304.175 | 710 | 498 | 212 | 351 | 346 |
9 | Pigskin Index | 0.68312 | 0.50211 | 13.1947 | 0.9822 | 262.218 | 770 | 526 | 244 | 357 | 354 |
10 | Roundtable | 0.69231 | 0.50198 | 13.6266 | 0.7186 | 293.881 | 533 | 369 | 164 | 253 | 251 |
11 | Line (updated) | 0.71818 | 0.50000 | 12.6247 | 0.4247 | 245.711 | 770 | 553 | 217 | 165 | 165 |
12 | Catherwood Ratings | 0.77027 | 0.50000 | 13.7432 | 0.6351 | 276.443 | 148 | 114 | 34 | 70 | 70 |
13 | Line (opening) | 0.70779 | 0.50000 | 12.7526 | 0.2760 | 251.696 | 770 | 545 | 225 | 337 | 337 |
14 | ARGH Power Ratings | 0.71688 | 0.50000 | 13.4558 | -0.1104 | 281.555 | 770 | 552 | 218 | 369 | 369 |
15 | Super List | 0.71299 | 0.49934 | 15.2931 | 1.5619 | 355.876 | 770 | 549 | 221 | 378 | 379 |
16 | Dunkel Index | 0.67630 | 0.49780 | 13.5055 | 1.3240 | 279.919 | 692 | 468 | 224 | 340 | 343 |
17 | Linear Regression | 0.69077 | 0.49747 | 13.1928 | 0.4825 | 267.460 | 401 | 277 | 124 | 197 | 199 |
18 | DP Dwiggins | 0.69191 | 0.49589 | 13.9295 | 0.3133 | 304.002 | 766 | 530 | 236 | 362 | 368 |
19 | PI-Rate Bias | 0.71039 | 0.49538 | 13.2086 | 1.1340 | 264.793 | 770 | 547 | 223 | 375 | 382 |
20 | Beck Elo | 0.69610 | 0.49474 | 13.4583 | 0.7619 | 284.370 | 770 | 536 | 234 | 376 | 384 |
21 | Pi-Ratings Mean | 0.70909 | 0.49398 | 13.1366 | 1.0153 | 261.875 | 770 | 546 | 224 | 369 | 378 |
22 | Bihl System | 0.69324 | 0.49392 | 13.0753 | 0.2528 | 258.480 | 414 | 287 | 127 | 203 | 208 |
23 | Howell | 0.70833 | 0.49389 | 13.5013 | -0.1457 | 282.844 | 768 | 544 | 224 | 364 | 373 |
24 | Talisman Red | 0.68994 | 0.49081 | 13.3432 | 0.1340 | 271.022 | 716 | 494 | 222 | 347 | 360 |
25 | Stat Fox | 0.68701 | 0.49038 | 13.4117 | 2.1963 | 274.644 | 770 | 529 | 241 | 357 | 371 |
26 | Moore Power Ratings | 0.69221 | 0.48816 | 13.4232 | 0.6346 | 279.246 | 770 | 533 | 237 | 371 | 389 |
27 | Massey Consensus | 0.72857 | 0.48748 | 13.5983 | 1.2433 | 283.608 | 770 | 561 | 209 | 370 | 389 |
28 | Laz Index | 0.71429 | 0.48681 | 13.1609 | 0.6398 | 263.527 | 770 | 550 | 220 | 369 | 389 |
29 | System Average | 0.70649 | 0.48617 | 12.9646 | 0.5539 | 255.949 | 770 | 544 | 226 | 369 | 390 |
30 | Born Power Index | 0.68442 | 0.48553 | 13.5410 | 1.2933 | 278.308 | 770 | 527 | 243 | 369 | 391 |
31 | Massey Ratings | 0.69091 | 0.48289 | 13.4370 | -0.0081 | 276.138 | 770 | 532 | 238 | 367 | 393 |
32 | TeamRankings.com | 0.70390 | 0.48271 | 13.1130 | 0.7473 | 263.278 | 770 | 542 | 228 | 363 | 389 |
33 | Billingsley | 0.71169 | 0.48221 | 13.8826 | 0.1543 | 304.154 | 770 | 548 | 222 | 366 | 393 |
34 | Edward Kambour | 0.70909 | 0.48158 | 13.2806 | 0.6883 | 265.528 | 770 | 546 | 224 | 366 | 394 |
35 | Loudsound.org | 0.65395 | 0.47962 | 14.7039 | -2.9618 | 340.061 | 760 | 497 | 263 | 353 | 383 |
36 | PerformanZ Ratings | 0.70909 | 0.47958 | 14.0244 | 0.5273 | 308.273 | 770 | 546 | 224 | 364 | 395 |
37 | Sagarin Recent | 0.69221 | 0.47895 | 13.5677 | 0.4317 | 292.251 | 770 | 533 | 237 | 364 | 396 |
38 | Computer Adjusted Line | 0.71299 | 0.47674 | 12.6591 | 0.4734 | 246.306 | 770 | 549 | 221 | 205 | 225 |
39 | Payne W/L | 0.71299 | 0.47625 | 14.3145 | -0.7173 | 318.431 | 770 | 549 | 221 | 361 | 397 |
40 | System Median | 0.70130 | 0.47410 | 12.9598 | 0.5655 | 255.160 | 770 | 540 | 230 | 357 | 396 |
41 | Donchess Inference | 0.71688 | 0.47383 | 13.2899 | 0.2138 | 267.461 | 770 | 552 | 218 | 353 | 392 |
42 | Cleanup Hitter | 0.66883 | 0.47312 | 14.3228 | 0.8333 | 319.305 | 770 | 515 | 255 | 352 | 392 |
43 | Logistic Regression | 0.67830 | 0.47222 | 15.4058 | -2.1711 | 369.469 | 401 | 272 | 129 | 187 | 209 |
44 | ESPN FPI | 0.69870 | 0.46772 | 13.0218 | 1.1573 | 260.627 | 770 | 538 | 232 | 355 | 404 |
45 | Versus Sports Simulator | 0.70143 | 0.46764 | 13.3260 | 0.6634 | 271.270 | 767 | 538 | 229 | 354 | 403 |
46 | Payne Power Ratings | 0.70390 | 0.46711 | 13.6009 | -0.1149 | 286.661 | 770 | 542 | 228 | 355 | 405 |
47 | Sagarin Golden Mean | 0.68182 | 0.46711 | 13.5186 | 0.3441 | 283.960 | 770 | 525 | 245 | 355 | 405 |
48 | Daniel Curry Index | 0.68961 | 0.46711 | 13.8791 | 0.1010 | 298.036 | 770 | 531 | 239 | 355 | 405 |
49 | Payne Predict | 0.69091 | 0.46316 | 13.8672 | 0.3339 | 301.146 | 770 | 532 | 238 | 352 | 408 |
50 | Dokter Entropy | 0.71651 | 0.45971 | 13.1405 | 0.5041 | 262.208 | 769 | 551 | 218 | 348 | 409 |
51 | David Harville | 0.70390 | 0.45910 | 13.1467 | 0.2825 | 262.413 | 770 | 542 | 228 | 348 | 410 |
52 | Sagarin Points | 0.69221 | 0.45850 | 13.3517 | 0.2749 | 270.378 | 770 | 533 | 237 | 348 | 411 |
53 | Sagarin Ratings | 0.69091 | 0.44737 | 13.2541 | 0.2903 | 266.691 | 770 | 532 | 238 | 340 | 420 |
54 | Line (Midweek) | 0.71169 | | 12.6429 | 0.4740 | 246.300 | 770 | 548 | 222 | | |
* 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