College Football Polls
Jeff Sagarin NCAA football ratings

The Jeff Sagarin NCAA football ratings were developed and are released by Jeff Sagarin, an American sports statistician. Sagarin is well-known for his development of a methodology for ranking and rating football teams and other sports teams.  Sagarin ratings have been have been a part of the Bowl Championship Series since its inception in 1998 and have been a regular feature in the USA Today sports section since 1985

Sagarin Methodology

For obvious reasons, does not divulge the exact methodology behind his system. This is not uncommon. Developers of almost all the other sports rating systems don't tell us how they do it. Sagarin has two rating systems, each of which gives each team a certain number of points. One system, Elo chess, is presumably based on the Elo rating system used internationally to rank chess players.  In the system ELO , only winning and losing matters; the score margin is of no consequence, which makes it very politically correct. However it is less accurate in its predictions for upcoming games than is the "Pure Points," or "Predictor," in which the score margin is the only thing that matters. In the Predictor, the difference in two teams' rating scores is meant to predict the margin of victory for the stronger team at a neutral venue.

For both systems, teams gain higher ratings within the Sagarin system by winning games against stronger opponents, factoring in such things as home-venue advantage. For the Predictor system, margin of victory (or defeat) factors in also, but a law of diminishing returns is applied. Therefore, a football team that wins a game by a margin of 7-6 is rewarded less than a team that defeats the same opponent under the same circumstances 21-7, but a team that wins a game by a margin of 35-0 receives similar ratings to a team that defeats the same opponent 70-0. This characteristic has the effect of recognizing "comfortable" victories, while limiting the reward for running up the score.

At the beginning of a season, when only a few games have been played,  the starting ratings have weight in the process(BAYESIAN), but once the teams are all CONNECTED, then the starting ratings are no longer used and all teams are started equal and the ELO CHESS is then done in an UNBIASED manner from that point on. RATING and PREDICTOR are now unbiased and the ELO-CHESS is now unbiased.

FINAL College Football 2009 through games of 2010 January 7 Thursday  the BCS uses the ELO_CHESS from here

HOME ADVANTAGE=  3.31           RATING    W   L  SCHEDL(RANK) VS top 10 | VS top 30 |  ELO_CHESS   |  PREDICTOR  

   1  Alabama              A  = 100.25   14   0   77.28(   2)    3   0  |    7   0  |  103.11    1 |   97.87    1

   2  Florida              A  =  95.75   13   1   75.39(  15)    1   1  |    4   1  |   96.44    2 |   94.68    2

   3  Texas                A  =  92.39   13   1   73.09(  38)    0   1  |    3   1  |   91.81    4 |   92.51    4

   4  TCU                  A  =  90.16   12   1   70.59(  60)    0   1  |    3   1  |   90.29    6 |   89.56    6

   5  Boise State          A  =  89.35   14   0   64.97(  96)    1   0  |    2   0  |   94.89    3 |   85.57   11

   6  Ohio State           A  =  88.35   11   2   72.05(  50)    2   0  |    4   1  |   87.22    8 |   89.08    7

   7  Virginia Tech        A  =  87.69   10   3   75.77(  13)    0   1  |    2   2  |   84.03   13 |   92.71    3

   8  Cincinnati           A  =  86.84   12   1   72.51(  44)    0   1  |    2   1  |   90.57    5 |   83.74   15

   9  Iowa                 A  =  85.82   11   2   72.64(  41)    1   1  |    3   1  |   88.17    7 |   83.50   17

  10  Penn State           A  =  85.43   11   2   69.21(  73)    0   2  |    1   2  |   84.79   10 |   85.60   10

FINAL College Football 2009 through games of 2010 January 7 Thursday  the BCS uses the ELO_CHESS from here

HOME ADVANTAGE=  3.31           RATING    W   L  SCHEDL(RANK) VS top 10 | VS top 30 |  ELO_CHESS   |  PREDICTOR  

  11  Oregon               A  =  85.27   10   3   76.29(   6)    0   2  |    2   3  |   84.35   11 |   85.77    9

  12  Georgia Tech         A  =  84.60   11   3   76.17(  10)    1   1  |    3   3  |   84.27   12 |   84.45   12

  13  LSU                  A  =  84.22    9   4   76.64(   5)    0   3  |    3   4  |   83.74   14 |   84.21   13

  14  Nebraska             A  =  84.06   10   4   71.83(  52)    0   2  |    1   3  |   80.01   25 |   89.64    5

  15  BYU                  A  =  83.60   11   2   70.34(  61)    0   1  |    2   1  |   84.94    9 |   81.97   23

  16  Pittsburgh           A  =  83.50   10   3   72.90(  40)    0   1  |    1   1  |   82.81   16 |   83.73   16

  17  Oklahoma             A  =  83.15    8   5   74.59(  22)    0   1  |    1   5  |   79.13   29 |   88.52    8

  18  Arkansas             A  =  82.49    8   5   75.93(  12)    0   2  |    1   5  |   81.60   19 |   82.93   19

  19  Mississippi          A  =  82.38    9   4   72.47(  46)    0   1  |    2   2  |   81.63   18 |   82.68   20

  20  Southern California  A  =  81.95    9   4   74.96(  19)    1   0  |    1   2  |   81.04   20 |   82.41   22

FINAL College Football 2009 through games of 2010 January 7 Thursday  the BCS uses the ELO_CHESS from here

HOME ADVANTAGE=  3.31           RATING    W   L  SCHEDL(RANK) VS top 10 | VS top 30 |  ELO_CHESS   |  PREDICTOR  

  21  Miami-Florida        A  =  81.72    9   4   76.18(   8)    0   1  |    2   3  |   80.42   21 |   82.63   21

  22  Clemson              A  =  81.54    9   5   74.42(  24)    0   1  |    1   3  |   79.19   28 |   83.83   14

  23  Wisconsin            A  =  81.21   10   3   70.18(  64)    0   2  |    1   2  |   81.97   17 |   80.04   25

  24  Utah                 A  =  80.59   10   3   69.06(  75)    0   1  |    0   3  |   82.95   15 |   78.18   32

  25  Texas Tech           A  =  80.25    9   4   70.25(  62)    0   1  |    2   1  |   77.53   36 |   83.05   18

  26  Georgia              A  =  80.08    8   5   76.22(   7)    0   1  |    3   2  |   80.32   24 |   79.38   29

  27  Auburn               A  =  79.43    8   5   75.22(  17)    0   1  |    1   4  |   80.38   22 |   78.09   33

  28  Connecticut          A  =  79.42    8   5   73.45(  34)    0   1  |    0   2  |   78.76   30 |   79.61   28

  29  Stanford             A  =  78.89    8   5   73.85(  29)    0   0  |    2   1  |   76.19   43 |   81.61   24

  30  Villanova            AA =  78.74   14   1   60.49( 123)    0   0  |    0   0  |   80.36   23 |   76.84   38

FINAL College Football 2009 through games of 2010 January 7 Thursday  the BCS uses the ELO_CHESS from here

HOME ADVANTAGE=  3.31           RATING    W   L  SCHEDL(RANK) VS top 10 | VS top 30 |  ELO_CHESS   |  PREDICTOR  

  31  Florida State        A  =  78.56    7   6   77.11(   3)    0   1  |    1   4  |   79.20   27 |   77.48   35

  32  West Virginia        A  =  78.55    9   4   73.71(  31)    0   1  |    2   2  |   79.81   26 |   76.95   37

  33  North Carolina       A  =  78.47    8   5   73.66(  32)    1   0  |    3   2  |   77.22   38 |   79.31   30

  34  Oregon State         A  =  78.35    8   5   74.56(  23)    0   1  |    1   4  |   78.63   31 |   77.61   34

  35  Air Force            A  =  78.21    8   5   68.39(  79)    0   1  |    0   3  |   76.37   42 |   79.76   26

  36  Tennessee            A  =  77.96    7   6   74.80(  21)    0   3  |    1   5  |   76.03   44 |   79.62   27

  37  Arizona              A  =  77.64    8   5   76.16(  11)    0   1  |    2   3  |   78.59   32 |   76.29   42

  38  Navy                 A  =  77.51   10   4   68.19(  82)    0   1  |    0   2  |   77.14   40 |   77.41   36

  39  South Florida        A  =  77.28    8   5   70.83(  59)    0   1  |    0   4  |   77.53   37 |   76.57   40

  40  Oklahoma State       A  =  77.28    9   4   72.25(  49)    0   1  |    2   3  |   78.07   33 |   76.07   43