2011-12 English Premier League
Computer Ratings
Introduction

This is my first rough attempt at a computer rating that can, among other things, come somewhat close to predicting the margin of victory of a match. The actual computer rating, which describes team performance based on games played to date, is found under the "Rating" column. To determine a hypothetical margin of victory, use the "Pts" column. You can the calculation for some upcoming games on the right. Don't forget to add in home "field" advantage, which is listed just below the current predictions on the right. 

Ratings last updated Sunday 02/19/12, 10:05 AM ET

 -------------------------------------------------------------------
                                                          BCS-Style 
 Rank  Team                     W  L  T   Rating   Pts      Rank    
 -------------------------------------------------------------------
   1  Manchester City          19  3  3    4.54    1.43       1
   2  Manchester United        18  3  4    4.00    1.18       2
   3  Tottenham Hotspur        16  4  5    3.08    0.79       3
   4  Chelsea                  12  6  7    1.92    0.60       4
   5  Liverpool                10  6  9    1.58    0.53       5
   6  Arsenal                  13  8  4    1.06    0.30       6
   7  Everton                   9 10  6    0.23    0.13      12
   8  Newcastle United         12  7  6    0.14   -0.15       7
   9  Norwich City              9  8  8   -0.11   -0.19       8
  10  Fulham                    7  9  9   -0.16   -0.06      14
  11  Sunderland                9 10  6   -0.35   -0.16      13
  12  Swansea City              7  9  9   -0.54   -0.16      15
  13  West Bromwich Albion      8 12  5   -0.65   -0.18      16
  14  Stoke City                8 11  6   -0.88   -0.26      17
  15  Aston Villa               6  9 10   -0.89   -0.16      18
  16  Blackburn Rovers          5 14  6   -2.23   -0.67      19
  17  Wigan Athletic            4 14  7   -2.47   -0.79      20
  18  Wolverhampton             5 14  6   -2.52   -0.75      21
  19  Bolton Wanderers          6 17  2   -2.58   -0.68      22
  20  QPR                       5 14  6   -3.18   -0.74      23
(The "BCS style" ranking is one based entirely on wins and losses, similar to what is used in college football.)

 

                     PREDICTIONS FOR UPCOMING GAMES

   Date       Home Team                  Away Team                    Prediction
----------    ------------------------   ------------------------   --------------
Sat 25 Feb    Liverpool                  Everton                    HOME by   0.76  
Sat 25 Feb    Chelsea                    Bolton Wanderers           HOME by   1.64  
Sat 25 Feb    Newcastle United           Wolverhampton              HOME by   0.95  
Sat 25 Feb    QPR                        Fulham                     AWAY by  -0.33  
Sat 25 Feb    West Bromwich Albion       Sunderland                 HOME by   0.34  
Sat 25 Feb    Wigan Athletic             Aston Villa                AWAY by  -0.27  
Sat 25 Feb    Manchester City            Blackburn Rovers           HOME by   2.46  

Sun 26 Feb    Arsenal                    Tottenham Hotspur          AWAY by  -0.13  
Sun 26 Feb    Norwich City               Manchester United          AWAY by  -1.01  
Sun 26 Feb    Stoke City                 Swansea City               HOME by   0.26  


Current home field advantage is:  0.36

MAE for games to date:  1.25

These ratings fit to produce 0.52 of the correct winners.
Pct when predicted MOV is above 0.72:  0.66

A favored away team rarely loses when favored by more than -0.55.

A favored home team rarely loses when favored by more than 0.64.
Above are some statistics about the ratings model. Home "field" advantage is shown first. MAE is the mean absolute error of the ratings fit to all the games played to date. This number is usually larger than you think it should be, but to me it's a good measure of how variable (or maybe "predictable") game outcomes can be.

Immediately below that, you can see how this best fit does in retro-predicting (there's a better word I'm sure) just the game winners. My favorite stats are the last two--when the home or away team is favored by the given margin, they only lose 30 percent of the time. This is the kind of information that people in the sports wagering world might find useful.

About the author

I have a Ph.D. in Atmospheric Science from the University of Alabama in Huntsville. I now work in Boulder, Colorado, in the broad areas of science education and numerical weather prediction (that is, forecasting the weather using computer models). Contact me using this email form if you have questions or non-hateful comments.