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

2010 Last Week

Through
Rank System Pct. Correct Against Spread Absolute Error Bias Mean Square Error games suw sul atsw atsl
1Lee Burdorf0.742860.5882411.8689-2.7369221.304352692014
2NationalSportsRankings0.714290.6764711.9823-2.6286231.3833525102311
3Laffaye XWP0.714290.5454513.6769-4.2711293.9033525101815
4Dunkel Index0.685710.5588215.0151-1.1317397.2903524111915
5ARGH Power Ratings0.685710.5757613.4000-2.7143284.0753524111914
6Catherwood Ratings0.685710.5882413.0857-2.5143265.4863524112014
7Keeper0.685710.5588212.93541.3554270.6973524111915
8Moore Power Ratings0.685710.4705912.8737-3.0869255.4413524111618
9Pigskin Index0.685710.5294112.8000-3.7126265.0263524111816
10Brent Craig 20.685710.6176512.6000-2.2000244.1433524112113
11Stat Fox0.685710.5882412.1711-2.5700230.1393524112014
12Least Squares w/ HFA0.657140.5000014.2331-0.0274301.5353523121717
13Warren Claassen0.657140.5294114.0697-3.2783307.4833523121816
14Covers.com0.657140.5454514.0100-4.6911313.7053523121815
15Howell0.657140.5882413.6143-2.2714291.4503523122014
16Bihl System0.657140.6470613.3703-3.2286266.4813523122212
17Edward Kambour0.657140.5882413.3020-3.2289275.8093523122014
18System Median0.657140.6764713.1840-3.0143262.7233523122311
19Sagarin Predictive0.657140.5882413.1383-3.2137263.6363523122014
20Linear Regression0.657140.6470612.9671-2.9174258.1183523122212
21Payne Power Ratings0.628570.5757613.8534-3.2637293.5643522131914
22Tempo Free Gridiron0.628570.5151513.9714-4.8286298.5433522131716
23Dokter Entropy0.628570.5151514.0666-3.2837291.0953522131716
24Billingsley0.628570.4705914.2286-3.6857304.7323522131618
25Atomic Football0.628570.4117614.2446-3.0023302.7323522131420
26Computer Adjusted Line0.628570.4545513.5286-2.3286286.92135221356
27Laz Index0.628570.5588213.4929-3.5963280.2093522131915
28Stephen Kerns0.628570.4705913.3200-2.4457275.5753522131618
29Born Power Index0.628570.6176511.8826-2.3911228.5433522132113
30CPA Rankings0.628570.6176513.1009-2.4734258.6243522132113
31BG Sports0.628570.6060613.12863.7571277.8643522132013
32Austin Sports0.628570.5882413.2674-2.9817261.5293522132014
33Marsee0.628570.6060613.2857-3.1143269.6293522132013
34Regression-Based Analys0.600000.4687514.2571-1.7429309.1713521141517
35Sportrends0.600000.5000014.0000-3.6571293.3573521141515
36Massey Consensus0.600000.5588213.9729-3.2980300.3043521141915
37Hank Trexler0.600000.4848513.8857-4.2286301.4293521141617
38Sagarin0.600000.5882413.8169-3.1580286.3903521142014
39Dave Congrove0.600000.5294113.7889-2.6626279.8873521141816
40Line (updated)0.6000013.4571-2.2000287.471352114
41PerformanZ Ratings0.600000.5294113.3783-2.9697280.5943521141816
42System Average0.600000.6176513.3197-2.8540267.0913521142113
43NutShell Combo0.600000.6764713.1703-3.1063250.2203521142311
44Anderson/Hester *0.571430.5000014.6220-3.6763331.8783520151717
45Sagarin Elo0.571430.5294114.9189-2.9029328.7973520151816
46Massey *0.571430.5294114.9674-2.5463328.6603520151816
47Martien Maas0.571430.5000016.4829-0.0771394.2553520151717
48Ashby AccuRatings0.571430.4375014.1143-3.3143293.4863520151418
49Schmidt Comp. Ratings0.571430.4411814.1000-3.9263292.5603520151519
50Beck Elo0.571430.4117613.7823-4.3989281.3703520151420
51Brent Craig0.571430.5151513.5714-3.8857311.2143520151716
52CF By the Numbers0.571430.5454513.3429-1.5143271.0573520151815
53NutShell Sports0.571430.6176512.0786-1.3831217.6403520152113
54Massey Ratings0.542860.5588213.5434-2.8331270.0463519161915
55Logistic Regression0.542860.4411815.2091-6.3983337.5493519161519
56CPA Retro0.542860.4117615.1217-4.1486354.4193519161420
57Wolfe *0.542860.5294114.6777-2.8269323.8793519161816
58Line (opening)0.542860.4000014.0571-2.2000296.8863519161218
59Nutshell Girl0.514290.4411814.6617-4.8326314.9333518171519
60Billingsley+0.514290.4411813.5906-3.1797266.3083518171519
61Super List0.514290.5294114.2900-1.1643298.1193518171816
62Bassett Model0.619050.6500013.8262-2.1414307.92321138137

* 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