National Hockey League Computer Ratings


Introduction

This is my rough attempt at a computer rating that can, among other things, come somewhat close to predicting final game scores. 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 "Points" column (just subtract the teams in question) and add the home "field" advantage, which is on the right below the predictions.

All of the predictions on the right already have the home "field" added in, and also show a predicted total. Shootouts and overtimes make this tricky, and are a lot of why the average error is 1.5 goals or so.

Early-season predictions are based almost entirely on last year's games and the preseason, so use those with extreme caution. New important note for 2021-22: games that are still tied after overtime will be recorded as ties. This will, hopefully, improve the accuracy of the predictions.


Ratings last updated Tuesday 10/26/21, 11:18 AM ET Rank Team W L T Rating Points BCS 1 Florida 6 1 0 3.06 51.13 1 2 Pittsburgh 4 1 1 2.97 51.09 2 3 Carolina 5 1 0 2.85 51.23 7 4 Tampa Bay 3 3 1 2.75 51.16 6 5 NY Islanders 3 4 0 2.63 50.99 3 6 Boston 3 1 0 2.52 50.94 4 7 Washington 4 2 0 2.37 50.83 5 8 NY Rangers 5 3 0 1.97 50.68 8 9 Nashville 3 4 0 1.65 50.61 9 10 Dallas 2 4 1 0.94 50.46 12 Rank Team W L T Rating Points BCS 11 Philadelphia 2 1 1 0.69 50.00 11 12 New Jersey 3 1 0 0.61 49.92 10 13 Colorado 2 3 1 0.17 50.32 18 14 Edmonton 5 0 1 0.09 50.14 17 15 Chicago 1 6 0 0.04 50.00 16 16 Columbus 4 3 0 0.02 49.86 13 17 Detroit 3 4 0 -0.09 49.86 15 18 Toronto 3 5 0 -0.12 50.20 20 19 Las Vegas 1 5 0 -0.30 50.11 21 20 Buffalo 4 2 1 -0.35 49.65 14 Rank Team W L T Rating Points BCS 21 Minnesota 4 2 0 -0.87 49.58 19 22 Winnipeg 2 3 0 -1.02 49.85 24 23 Calgary 3 2 0 -1.17 49.68 23 24 St Louis 5 0 0 -1.32 49.51 22 25 Montreal 1 5 0 -1.46 49.55 25 26 Ottawa 2 5 0 -2.14 49.20 27 27 Vancouver 2 3 2 -2.26 49.13 29 28 Arizona 0 5 1 -2.43 48.99 28 29 San Jose 5 1 0 -2.48 48.92 26 30 Seattle 1 5 0 -2.80 48.90 30 31 Los Angeles 1 5 1 -2.87 48.96 31 32 Anaheim 2 4 1 -3.67 48.56 32
(The "BCS style" ranking is one based entirely on wins and losses, similar to what is used in college football.)

When I get them working, divisional rankings will go here:

Rank Conference Mean Rating ---- ------------------------- ----------- 1 North 1.76 2 East 0.52 3 Central -0.35 4 West -1.93

PREDICTIONS FOR UPCOMING GAMES Date Away Team Home Team Total Prediction ----------- -------------------- -------------------- ----- -------------- 26-Oct-2021 Calgary New Jersey 5.0 HOME by 0.57 26-Oct-2021 Tampa Bay Pittsburgh 5.7 HOME by 0.26 26-Oct-2021 Las Vegas Colorado 5.8 HOME by 0.54 26-Oct-2021 San Jose Nashville 5.7 HOME by 2.02 26-Oct-2021 Winnipeg Anaheim 5.3 AWAY by -0.96 26-Oct-2021 Montreal Seattle 5.2 AWAY by -0.32 26-Oct-2021 Minnesota Vancouver 6.0 AWAY by -0.12 27-Oct-2021 Boston Florida 5.4 HOME by 0.51 27-Oct-2021 Detroit Washington 5.8 HOME by 1.29 27-Oct-2021 Toronto Chicago 6.4 HOME by 0.12 27-Oct-2021 Las Vegas Dallas 5.0 HOME by 0.68 27-Oct-2021 Philadelphia Edmonton 7.2 HOME by 0.46 28-Oct-2021 Boston Carolina 4.6 HOME by 0.62 28-Oct-2021 Calgary Pittsburgh 5.2 HOME by 1.74 28-Oct-2021 Arizona Tampa Bay 6.2 HOME by 2.50 28-Oct-2021 Colorado St Louis 6.4 AWAY by -0.48 28-Oct-2021 Buffalo Anaheim 6.1 AWAY by -0.77 28-Oct-2021 Minnesota Seattle 5.7 AWAY by -0.35 28-Oct-2021 Philadelphia Vancouver 6.8 AWAY by -0.54 28-Oct-2021 Winnipeg Los Angeles 5.3 AWAY by -0.57 28-Oct-2021 Montreal San Jose 6.1 AWAY by -0.30 29-Oct-2021 Chicago Carolina 5.5 HOME by 1.57 29-Oct-2021 Florida Detroit 6.0 AWAY by -0.93 29-Oct-2021 Columbus NY Rangers 5.6 HOME by 1.15 29-Oct-2021 Arizona Washington 6.1 HOME by 2.17 29-Oct-2021 Anaheim Las Vegas 5.2 HOME by 1.88 29-Oct-2021 Ottawa Dallas 6.2 HOME by 1.59 30-Oct-2021 NY Islanders Nashville 4.5 AWAY by -0.04 30-Oct-2021 Montreal Los Angeles 5.2 AWAY by -0.26 30-Oct-2021 Florida Boston 6.5 HOME by 0.14 30-Oct-2021 New Jersey Pittsburgh 6.3 HOME by 1.50 30-Oct-2021 Winnipeg San Jose 6.3 AWAY by -0.60 30-Oct-2021 Detroit Toronto 5.8 HOME by 0.67 30-Oct-2021 Chicago St Louis 6.6 AWAY by -0.16 30-Oct-2021 Philadelphia Calgary 6.8 HOME by 0.01 30-Oct-2021 Minnesota Colorado 6.4 HOME by 1.07 30-Oct-2021 Edmonton Vancouver 5.8 AWAY by -0.68 31-Oct-2021 Arizona Carolina 5.5 HOME by 2.58 31-Oct-2021 Montreal Anaheim 5.1 AWAY by -0.66 31-Oct-2021 Buffalo Los Angeles 6.2 AWAY by -0.37 31-Oct-2021 Columbus New Jersey 5.2 HOME by 0.39 31-Oct-2021 NY Rangers Seattle 5.6 AWAY by -1.44 01-Nov-2021 Washington Tampa Bay 6.9 HOME by 0.66 01-Nov-2021 Ottawa Chicago 7.2 HOME by 1.13 01-Nov-2021 Seattle Edmonton 6.2 HOME by 1.56 02-Nov-2021 Detroit Montreal 5.5 HOME by 0.01 02-Nov-2021 Arizona Philadelphia 6.4 HOME by 1.34 02-Nov-2021 Las Vegas Toronto 5.4 HOME by 0.42 02-Nov-2021 Ottawa Minnesota 6.8 HOME by 0.71 02-Nov-2021 Dallas Winnipeg 5.4 AWAY by -0.28 02-Nov-2021 Nashville Calgary 5.9 AWAY by -0.60 02-Nov-2021 New Jersey Anaheim 6.1 AWAY by -1.03 02-Nov-2021 NY Rangers Vancouver 6.0 AWAY by -1.22 02-Nov-2021 Buffalo San Jose 7.1 AWAY by -0.40 03-Nov-2021 Carolina Chicago 7.0 AWAY by -0.91 03-Nov-2021 Nashville Edmonton 6.4 AWAY by -0.15 03-Nov-2021 Columbus Colorado 6.0 HOME by 0.79 03-Nov-2021 St Louis Los Angeles 5.8 AWAY by -0.23

Current home field advantage is: 0.33 MAE for games to date: 1.99 These ratings fit to produce 0.63 of the correct winners. Pct when predicted MOV is above 0.66: 0.68 A favored away team rarely loses when favored by more than -0.78. A favored home team rarely loses when favored by more than 0.95.

Above are some statistics about the ratings model. Each team has its own home "field" advantage, but the average all of them is shown here. The 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 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 am now a faculty member at Indiana University in Bloomington, teaching courses in the broad areas of weather and climate. Don't hesitate to contact me using this email form if you have questions or non-hateful comments. shopify stats