Part 2: NFL Pythagorean Expectation
In sports, we often hear phrases such as “the final score was not indicative of how close these two teams played”. Such rhetoric is grounded in the belief that scoring margin is an imperfect measure of a team’s true ability. Randomness and variations of luck over the course of a game or season can result in a team’s performance on the scoreboard deviating from its actual talent level.
This understanding served as the primary motive in last week’s In case you missed Part 1 of the story click here column which discussed my creation of a second-order Pythagorean expectation for NFL teams. A team’s yards per play and yards per point metrics are regressed to predict what its net scoring margin should have been. From there, we can calculate how many games the team should have won based on its key performance statistics.
Listed below is each team ranked by its adjusted Pythagorean wins along with its actual and standard Pythagorean wins. It is important to remember that it is a retrodictive evaluation of last season that documents which teams overachieved or underachieved relative to their Pythagorean expectations. Over the past four seasons, teams that have outperformed their adjusted Pythagorean expectation by greater than one win have experienced, on average, a 1.77 decrease in wins the following season. In contrast, those who underperformed by greater than one game have seen their win totals increase by 2.75 games, on average.
Below is a chart of how many games each team won in 2012 vs what would have been expected by their advanced metrics.
Team |
Wins |
Pythag. Wins |
Adjusted Pythag. Wins |
Seahawks |
11.00 |
12.39 |
12.26 |
49ers |
11.50 |
11.33 |
11.53 |
Broncos |
13.00 |
12.32 |
11.53 |
Bears |
10.00 |
10.75 |
11.26 |
Patriots |
12.00 |
12.39 |
10.44 |
Falcons |
13.00 |
11.04 |
10.39 |
Ravens |
10.00 |
9.37 |
10.17 |
Bengals |
10.00 |
9.86 |
10.14 |
Giants |
9.00 |
10.05 |
10.13 |
Packers |
11.00 |
10.33 |
10.10 |
Redskins |
10.00 |
9.10 |
9.28 |
Vikings |
10.00 |
8.81 |
9.18 |
Texans |
12.00 |
10.12 |
9.08 |
Dolphins |
7.00 |
7.09 |
8.41 |
Steelers |
8.00 |
8.64 |
8.34 |
Panthers |
7.00 |
7.84 |
8.33 |
Saints |
7.00 |
8.15 |
8.15 |
Buccaneers |
7.00 |
7.88 |
8.15 |
Chargers |
7.00 |
8.00 |
7.92 |
Rams |
7.50 |
6.58 |
7.25 |
Browns |
5.00 |
6.16 |
7.12 |
Cowboys |
8.00 |
7.41 |
7.12 |
Bills |
6.00 |
5.83 |
6.77 |
Titans |
6.00 |
4.81 |
6.25 |
Colts |
11.00 |
7.24 |
5.96 |
Lions |
4.00 |
6.49 |
5.80 |
Jets |
6.00 |
5.37 |
5.24 |
Cardinals |
5.00 |
4.81 |
4.64 |
Raiders |
4.00 |
4.29 |
4.39 |
Jaguars |
2.00 |
3.39 |
4.07 |
Eagles |
4.00 |
4.02 |
3.92 |
Chiefs |
2.00 |
2.56 |
1.57 |
The data is provided for you to draw your own conclusions but let me share my analysis on some of the most fortunate and unfortunate teams of 2012:
Overachievers
Indianapolis Colts (O/U 8.5 wins):
The 2012 Colts serve as a textbook example for the disciples of Pythagorean win expectation. The team managed to win 11 games despite being outscored by 30 points over the course of the season. And for Colts backers, their arguments only get weaker as you dive deeper into the numbers. Based on its key performance statistics, the team should have only won 5.96 games—marking the Colts as the 25th worst team in football behind the likes of the Buffalo Bills and Tennessee Titans. Lastly, consider this: the Jacksonville Jaguars performed better than the Colts in net yards per play last season.
Houston Texans (O/U 10.5 wins):
According to the data, the Houston Texans, whose record was tied for 2nd-best overall in the NFL last season, were not as dominant as their wins and losses suggest. The Texans yards per play and yards per point metrics were just slightly above league average; and this is precisely who the 2012 Texans were—a slightly above average football team. The 12-win Texans should have won just 9.08 games which was good for only 13th overall in the league. The market-makers, however, continue to disagree as just five teams are priced ahead of the Texans in terms of 2013 expected win totals.
Underachievers
Cleveland Browns (O/U 6.5 wins):
Given the team’s recent struggles and placement in the always competitive AFC North, the Cleveland Browns could really use some good fortune on their side, but this was certainly not the case in 2012. Winners of just 5 games, my model suggests the Browns should have won 7.12 games. For a team that is already inherently underrated due to their recent ineptitude and unsexy location, give them some better luck combined with the progression of young offensive talent under Norv Turner and the Cleveland Browns offer value headed into 2013.
Carolina Panthers (O/U 7 wins):
The Panthers were fifth overall in net yards per play in 2012, ranking strikingly ahead of team such as the New England Patriots, Green Bay Packers, and Super Bowl Champion Baltimore Ravens. Despite this impressive statistical performance, the Panthers managed to win only 7 games last season. The lack of success can be partly attributed to Cam Newton’s disappointing sophomore campaign. A large degree of the team’s struggles, however, can be attributed to bad luck. With the public’s focus now shifted to other young, star quarterbacks, the Panthers are flying under the radar as a team that could play up to its true potential in 2013.