Computer Rating System Prediction Results for College Football (NCAA IA)

2009 Season Totals

Through
Rank System Pct. Correct Against Spread Absolute Error Bias Mean Square Error games suw sul atsw atsl
1Line (updated)0.7338911.26541.1814206.152714524190
2Computer Adjusted Line0.738100.5166111.26121.0917206.163714527187140131
3Line (opening)0.732490.4974211.41740.8249212.248714523191289292
4Dokter Entropy0.750700.5165511.44440.7752212.719714536178359336
5System Median0.747900.4919711.54950.4387216.008714534180337348
6Pigskin Index0.756300.4947511.54050.6110216.060714540174330337
7Atomic Football0.752100.4971311.51200.8409216.111714537177346350
8System Average0.752100.5014411.57230.4370217.049714537177349347
9Edward Kambour0.746500.5245011.64920.7182219.689714533181364330
10Stat Fox0.739500.5038211.66251.6683221.267714528186330325
11NutShell Combo0.736690.5245011.70030.1868221.595714526188364330
12Laz Index0.749300.5014411.66670.1511221.865714535179348346
13Massey *0.776880.5110511.80201.2431223.55037228983185177
14Sagarin Predictive0.738100.5043111.75290.6497224.315714527187351345
15Ashby AccuRatings0.736690.4916311.75630.1601224.631714526188323334
16NationalSportsRankings0.748480.4846011.75920.3506225.182656491165299318
17Moore Power Ratings0.749300.5172911.71120.5256225.807714535179359335
18Dunkel Index0.739130.4971211.84541.1091227.489713527186345349
19Bihl System0.727060.4878611.84860.8618227.504425309116201211
20Billingsley+0.749300.5093511.75860.4208228.385714535179354341
21CF By the Numbers0.699680.4390211.9457-0.5272228.92931321994126161
22Linear Regression0.733870.5276211.89220.9067228.93037227399191171
23Hank Trexler0.750350.4923811.8850-0.0084231.237713535178323333
24Howell0.733520.5174012.00220.0065232.849713523190342319
25Sagarin0.726890.5000011.95330.4593234.340714519195348348
26Keeper0.736690.4964012.07171.5916234.506714526188345350
27Dave Congrove0.726890.5100612.0814-0.1209235.237714519195355341
28Covers.com0.732490.5348812.09120.0708236.518714523191368320
29Catherwood Ratings0.722690.5000011.92721.2661236.606714516198326326
30Massey Consensus0.725490.5057512.05170.6875237.293714518196352344
31Lee Burdorf0.732490.5036312.05810.4940238.303714523191347342
32Schmidt Comp. Ratings0.731090.4971312.05360.1328238.557714522192346350
33ARGH Power Ratings0.742300.5104512.20170.6674239.155714530184342328
34Beck Elo0.726890.5079412.08900.2810239.203714519195352341
35Born Power Index0.736690.4899112.08340.4529240.313714526188340354
36NutShell Sports0.728290.5143712.21800.5398241.651714520194358338
37CPA Rankings0.732040.5053912.26320.1163244.168668489179328321
38Nutshell Girl0.736690.5258612.2668-0.1733244.838714526188366330
39Harmon Forcast0.723210.4896712.2321-1.4048247.691672486186308321
40Bassett Model0.729960.4891812.46071.8703248.630711519192339354
41Anderson/Hester *0.731180.4930712.51990.7481249.200372272100178183
42Imes Comprank0.723450.4847312.4674-0.0253250.253499361138238253
43DP Dwiggins0.711960.5159312.4536-3.3892250.857677482195340319
44Billingsley0.732490.5287412.5430-0.0059253.321714523191368328
45Payne Power Ratings0.729170.4928612.4982-0.1746255.153576420156276284
46Stephen Kerns0.725490.5258612.52910.7475256.894714518196366330
47Sportrends0.705550.4931912.7319-0.1486259.673703496207326335
48Frank Alder0.725490.4985612.6280-0.4457260.100714518196347349
49Marsee0.737730.4762612.68581.9677262.714713526187321353
50Sagarin Elo0.707280.5071812.74870.2918263.889714505209353343
51PerformanZ Ratings0.724090.4827612.6587-0.0329264.096714517197336360
52Warren Claassen0.708680.5043112.9084-0.4963270.781714506208351345
53Colley Rankings *0.698920.5027612.92310.8179270.791372260112182180
54Martien Maas0.688210.4921613.07660.4696276.537526362164251259
55TW Rankings0.701520.5023613.0547-0.1404283.666660463197319316
56CPA Retro0.724550.5046213.3387-0.0054286.861668484184328322
57Wolfe *0.706990.4944813.46981.5001288.803372263109179183
58Super List0.725490.5050413.58150.9087292.143714518196351344
59Least Squares w/ HFA0.725810.5221013.78620.7620311.808372270102189173
60Tom Benson0.675560.4870314.95173.2272341.119712481231338356
61Logistic Regression0.717740.5359115.0370-2.4933369.468372267105194168

* 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