The phrase, on any given Sunday, embodies the idea that the outcome of any NFL game depends not only on the intrinsic strength of each team, but also on how well each team plays on that particular day and which team benefits more from the breaks. This random component is so pervasive that, according to a fascinating study published a few years ago by Doug Drinen, the strongest team in any particular NFL season is likely to win the Super Bowl only about 24% of the time.
We can consider the role of stochasticity in the just-completed regular season by assuming that the team efficiency ratings at the end of the season represent the true strength of each team. In order to avoid the effect of some teams giving less than a full effort in the final week of the season, Ill use the week-17 team efficiency ratings, as recommended by Brian Burke. These ratings should be reasonable estimates of the “true” team strength because they are calculated from the statistics that are most highly correlated with winning and they are based on a large enough data set to have reasonable significance.
From the efficiency ratings, we can estimate the true odds of each game of the season, and then use these odds to simulate the entire NFL season 5,000 times. The results are given below:
Percent probability of team finishing in each place within the division
AFC EAST
Team | 1st | 2nd | 3rd | 4th |
New England | 68.96 | 27.66 | 3.06 | 0.32 |
NY Jets | 28.98 | 58.08 | 10.72 | 2.22 |
Buffalo | 1.3 | 8.78 | 50.28 | 39.64 |
Miami | 0.76 | 5.48 | 35.94 | 57.82 |
AFC NORTH
Team | 1st | 2nd | 3rd | 4th |
Pittsburgh | 54.94 | 30.0 | 14.94 | 0.12 |
Baltimore | 29.72 | 39.74 | 30.1 | 0.44 |
Cincinnati | 15.34 | 30.16 | 53.82 | 0.68 |
Cleveland | 0.0 | 0.1 | 1.14 | 98.76 |
AFC SOUTH
Team | 1st | 2nd | 3rd | 4th |
Indianapolis | 78.46 | 17.02 | 3.94 | 0.58 |
Houston | 15.62 | 48.02 | 24.2 | 12.16 |
Tennessee | 4.02 | 20.66 | 39.32 | 36.0 |
Jacksonville | 1.9 | 14.3 | 32.54 | 51.26 |
AFC WEST
Team | 1st | 2nd | 3rd | 4th |
San Diego | 78.88 | 21.08 | 0.04 | 0.0 |
Denver | 21.08 | 77.58 | 1.24 | 0.1 |
Kansas City | 0.02 | 1.0 | 62.74 | 36.24 |
Oakland | 0.02 | 0.34 | 35.98 | 63.66 |
NFC EAST
Team | 1st | 2nd | 3rd | 4th |
Philadelphia | 47.86 | 31.6 | 16.52 | 4.02 |
Dallas | 35.68 | 36.9 | 21.82 | 5.6 |
NY Giants | 15.02 | 25.08 | 41.34 | 18.56 |
Washington | 1.44 | 6.42 | 20.32 | 71.82 |
NFC NORTH
Team | 1st | 2nd | 3rd | 4th |
Green Bay | 66.86 | 30.2 | 2.92 | 0.02 |
Minnesota | 31.46 | 59.6 | 8.72 | 0.22 |
Chicago | 1.66 | 10.04 | 81.82 | 6.48 |
Detroit | 0.02 | 0.16 | 6.54 | 93.28 |
NFC SOUTH
Team | 1st | 2nd | 3rd | 4th |
New Orleans | 94.9 | 4.32 | 0.74 | 0.04 |
Carolina | 2.74 | 45.64 | 38.52 | 13.1 |
Atlanta | 2.22 | 42.04 | 40.3 | 15.44 |
Tampa Bay | 0.14 | 8.0 | 20.44 | 71.42 |
NFC WEST
Team | 1st | 2nd | 3rd | 4th |
Arizona | 65.38 | 28.28 | 5.76 | 0.58 |
San Francisco | 29.96 | 50.6 | 16.94 | 2.5 |
Seattle | 4.48 | 18.34 | 57.46 | 19.72 |
St Louis | 0.18 | 2.78 | 19.84 | 77.2 |
Generated: Wed Jan 06 22:42:25 EST 2010
AFC Percent Probability Playoff Seeding
Team | 1st | 2nd | 3rd | 4th | 5th | 6th | Total |
Indianapolis | 34.56 | 21.82 | 14.4 | 7.68 | 8.62 | 5.18 | 92.26 |
San Diego | 29.32 | 23.68 | 15.58 | 10.3 | 8.82 | 4.92 | 92.62 |
New England | 10.92 | 14.56 | 20.0 | 23.48 | 5.76 | 5.88 | 80.6 |
Pittsburgh | 9.64 | 13.62 | 16.2 | 15.48 | 7.96 | 8.72 | 71.62 |
Denver | 4.06 | 6.02 | 5.74 | 5.26 | 17.56 | 15.32 | 53.96 |
Houston | 3.76 | 5.0 | 4.08 | 2.78 | 18.3 | 15.06 | 48.98 |
Baltimore | 3.32 | 5.56 | 9.0 | 11.84 | 8.56 | 11.0 | 49.28 |
NY Jets | 2.72 | 5.38 | 8.62 | 12.26 | 9.52 | 10.62 | 49.12 |
Cincinnati | 1.0 | 2.74 | 4.28 | 7.32 | 6.12 | 9.14 | 30.6 |
Tennessee | 0.58 | 1.24 | 1.06 | 1.14 | 4.68 | 7.54 | 16.24 |
Jacksonville | 0.1 | 0.32 | 0.72 | 0.76 | 3.46 | 5.08 | 10.44 |
Buffalo | 0.02 | 0.06 | 0.2 | 1.02 | 0.48 | 0.92 | 2.7 |
Miami | 0.0 | 0.0 | 0.12 | 0.64 | 0.16 | 0.54 | 1.46 |
Kansas City | 0.0 | 0.0 | 0.0 | 0.02 | 0.0 | 0.08 | 0.1 |
Oakland | 0.0 | 0.0 | 0.0 | 0.02 | 0.0 | 0.0 | 0.02 |
Cleveland | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 |
NFC Percent Probability Playoff Seeding
Team | 1st | 2nd | 3rd | 4th | 5th | 6th | Total |
New Orleans | 33.96 | 26.3 | 26.9 | 7.74 | 0.74 | 0.76 | 96.4 |
Green Bay | 20.46 | 21.58 | 19.52 | 5.3 | 11.12 | 9.9 | 87.88 |
Philadelphia | 19.3 | 15.52 | 11.0 | 2.04 | 21.6 | 14.02 | 83.48 |
Dallas | 12.32 | 13.08 | 8.72 | 1.56 | 24.46 | 17.6 | 77.74 |
Minnesota | 7.04 | 10.18 | 10.82 | 3.42 | 18.2 | 16.92 | 66.58 |
NY Giants | 4.2 | 5.6 | 4.6 | 0.62 | 15.54 | 19.64 | 50.2 |
Arizona | 1.88 | 4.94 | 11.12 | 47.44 | 0.62 | 1.46 | 67.46 |
San Francisco | 0.22 | 1.3 | 3.42 | 25.02 | 0.72 | 2.66 | 33.34 |
Carolina | 0.18 | 0.48 | 1.2 | 0.88 | 1.7 | 4.16 | 8.6 |
Washington | 0.18 | 0.38 | 0.64 | 0.24 | 2.98 | 6.08 | 10.5 |
Atlanta | 0.14 | 0.28 | 0.94 | 0.86 | 1.44 | 3.18 | 6.84 |
Chicago | 0.12 | 0.34 | 0.8 | 0.4 | 0.78 | 2.92 | 5.36 |
Seattle | 0.0 | 0.0 | 0.22 | 4.26 | 0.06 | 0.36 | 4.9 |
Tampa Bay | 0.0 | 0.02 | 0.1 | 0.02 | 0.04 | 0.28 | 0.46 |
St Louis | 0.0 | 0.0 | 0.0 | 0.18 | 0.0 | 0.02 | 0.2 |
Detroit | 0.0 | 0.0 | 0.0 | 0.02 | 0.0 | 0.04 | 0.06 |
Generated: Wed Jan 06 22:42:25 EST 2010
As NFL seasons go, this one had relatively few rare outcomes. The favored team won 5 of the 8 divisions. Of the three favorites that did not win their division, only Pittsburgh failed to qualify as a wild card. With an intrinsic 70% chance of making the playoffs, this disappointing follow up to their championship season of a year ago was a bitter pill for Steelers fans to swallow. Cincinnatis intrinsic chances of winning their division were only 15%. They should consider themselves fortunate. Likewise, Minnesota not only managed to win their division over favored Green Bay, but also captured the 2 seed in the NFC. Their season ranked in the top 15% outcomes available to them. Miami, with only a 1% chance of making the playoffs, was competing for a playoff spot in the final week of the season. Another interesting observation is that once every 100 years or so, some truly bottom dwelling team like the Chiefs, Raiders, Browns. Bucs, Rams, or Lions would actually qualify for the playoffs. Any given Sunday, indeed.
One final curiosity. After week 6, I promoted in some circles the fact that the Broncos had essentially won the AFC West, since my software projected that they had a 99.6% chance of winning their division. These odds were based on the team efficiencies at the time. Today, I re-simulated weeks 7 through 17 again using the true end-of-season efficiency ratings. Their actual odds were only about 96%. So in reality their odds of missing the playoffs were about 10 times larger than I was estimating at the time. It might be worthwhile for me to incorporate some kind of uncertainty estimate in the early-season efficiency ratings in order for me to avoid similar future embarrassments. On the other hand, Denver’s improbable collapse provided a great deal of comic relief to my fellow Chief fans and I was credited with unique powers of “reverse mojo”.