NFL: Winning with Power Ratings
September 14, 2005
Can Power Ratings predict which team will cover? If you use mine they can. In College Football, lack of parity and motivational swings can make statistically-based handicapping almost meaningless at times. In the NFL however, there is enough parity and motivational consistency that stats often tell the story. The question is which ones? I've got the bankroll-increasing answer.
I keep three sets of "Point Spread Power Ratings" for each team and call them:
2. HFA- Home Field Advantage
3. AFD- Away Field Disadvantage
The most important is the overall rating. I have perfected it over my 20 years of handicapping the NFL. I track specific statistics for every team that predict their chances of covering the spread. They include rushing, passing and overall offensive and defensive efficiency. I also incorporate third down efficiency, sacks for and against, and turnovers. Every team gets a score on each of these measurements and when I add them up I have a numerical rating.
The other two ratings are used to adjust for the site of the game. Rather than assign what seems to be the standard three point edge to the home team, I look at things from both team's perspective. The HFA can vary tremendously based on outdoors vs. in, weather, crowd, record and time of year. Indy and New England currently have HFA's of 4 or 4½ points; while New Orleans' "home" FA is two. The same goes for the AFD; some teams travel very well, while others perform much worse on the road than at home.
How do I make a line with these numbers? Let's use an example from Week #2 of the 2005 season:
Minnesota +3 @ Cincinnati
Minnesota: Overall -3, HFA 3.5, AFD -3.5
Cincinnati: Overall +8, HFA 3.5, AFD -3
To make a line, we first take the absolute difference between the overall ratings and divide by two. This would be 11 (i.e. +8 to -3) / 2 = 5.5; in favor of the Bengals. To adjust for the site, we take the absolute difference between the HFA of the home team and the AFD of the away team and divide by two; 3.5 to -3.5 is 7 / 2 = 3.5. To get the predicted line, simply add these two numbers together; 5.5 + 3.5 = 9. The Bengals should be a 9 point favorite in this game based on my Point Spread Power Ratings.
The actual posted line on this game of Cincy -3 made this an easy bet and was a winner as the final was 37-8.
Once we are at least three games into the season, there is a final factor to consider. That is strength of schedule. I use The Gold Sheet's ratings to keep a last 5 games moving average of the rating of each team's opponents. We would then adjust the PSPR line up or down by the difference. So if our example game had been in week 6 and Cincy's 1st 5 opponents had a mean rating* of 3 more than Minny's, that would have made the line -12.
Do not blindly follow these ratings; you must include motivation, specific match-ups and other things like injuries in your handicapping. But all things being equal, they are highly predictive of point spread success.
* - I use another source of ratings here and in their case, lower is better (i.e. a more difficult schedule).
SHOW ME the CURRENT RATINGS