Play the angles on NFL season win totals

By Scott Kellen  () 

You probably were not paying much attention in that 10th-grade geometry class when we learned all those math formulas. Learning math formulas was far from my mind.

But one formula was drilled into our heads and is hard to forget — the Pythagorean Theorem, or A2 +  B2 = C2. I’ll spare you the details of how it works, but feel free to Google it.

Bill James used the Pythagorean Theorem many years ago to indicate a baseball team’s future performance based on runs scored and allowed as they correlate to winning percentage. From there a formula was created to apply it to the NFL. I am unsure who created the NFL formula, but Football Outsiders has given it some research as a way to predict a team’s outcomes based on its prior-year performance. I have used this for 12 years or so to help gauge NFL season win totals. While it’s only one piece of the puzzle, some lessons can be learned to set us up in the right direction on season win totals. So, for the second year in a row, let’s look at what we can learn from this geometry class formula.

Let’s start by applying the Pythagorean Theorem to the 2019 season. The Green Bay Packers went 13-3 in the regular season, scoring 376 points and allowing 313. The Pythagorean Theorem suggests the Packers should have won only 9.7 games and lost 6.3. In other words, the Packers won 3.3 more games than they should have based on this formula.

Now let’s look at all teams going back to 2007 that won two or more games than they should have and how that correlated to meeting the next season’s win totals. If a team overachieved regular-season wins by two or more games the year before, we look to go Under its season win total the next season. Since 2007, there have been 22 such teams, or just under two a year. By fading those 22 teams to the Under season win total the next season, you would have gone 18-3-1. Of those 22 teams, 18 won at least three fewer games the next season than they won the previous season, including all eight teams that won 13 or more. 

The New Orleans Saints went 13-3 last year, but their numbers suggest they should have gone only 10.7-5.3, meaning they overachieved by 2.3 wins. They also went 6-1 SU in games decided by seven or fewer points in the regular season. Two other teams also overachieved by two or more wins last year. The Seattle Seahawks went 11-5 (8.2-7.8) and the Houston Texans went 10-6 (7.8-8.2). So the Packers, Saints, Seahawks and Texans overachieved by two or more wins last year.

But let’s take this one step further. Teams that typically win more games than their numbers suggest likely also had very good fortune in close games. The more close games you win, the greater likelihood that fortune will regress to the mean the next season. The 2019 Packers not only won 3.3 games more than their numbers suggest they should have, but they also went 6-1 SU in games decided by seven or fewer points. They had an extreme amount of good luck in those close games.

When we look at those 22 teams that overachieved by at least two games and consider only teams that won at least four more close games than they lost, we find those teams went Under their season win totals the next season by a count of 14-1-1. Since the Packers, Saints, Seahawks and Texans won at least four more close games than they lost, all look to be Under plays this year based solely on this formula.

Before we move on, let’s look at a 2018 team based on its 2017 numbers that qualified for this situation. The Pittsburgh Steelers went 13-3 in 2017, while their Pythagorean numbers suggested they should have won 10.5 games, or 2.5 fewer. The Steelers were also a very fortunate 8-2 in games decided by seven or fewer points. The next year they slipped back to 9-6-1 while going a more normal 6-5 in close games. And, true to form, they went Under their season win total of 10 wins. Last year this formula applied to the Rams. They came through by winning only nine games and going Under their 10.5 season win total one year after playing in the Super Bowl.

What about the other side of the coin? What about teams that underachieved the previous season? Since 2007 there have been 26 teams that won at least two games fewer than their Pythagorean numbers suggest they should have won. We would look to go Over their season win totals the next season. Those 26 teams went 16-9-1 the next season versus their season win total. We have four teams that fit this pattern for this year. With Matthew Stafford injured, the Detroit Lions went 3-12-1 last year but should have gone 6-10 based on their numbers. The Dallas Cowboys went 8-8 but should have gone 10.7-5.3. The Los Angeles Chargers went 5-11 but should have gone 7.8-8.2. And the Cincinnati Bengals went 2-14 but should have gone 4.4-11.6. So the Lions, Cowboys, Chargers and Bengals qualify to go Over their season win totals this year.

Let’s look at another number that fit that mold for the 2018 season. In 2017, the Cleveland Browns went 0-16, but their Pythagorean numbers suggested they should have won 3.3 games. Despite being marked with a six-win total for 2018 — six more wins than the previous season — the Browns managed to win seven games and go Over their season win total number.

Teams that won eight or more games and still underachieved by at least two wins the previous season have gone Over their season win total the next year all five times since 2007. The Dallas Cowboys fit that formula this year, having gone 8-8 last year when they should have gone 10.7-5.3. They underachieved by 2.7 wins but still won eight games in 2019.

Between the two situations suggesting to go Over or Under a team’s win total, you would have gone 34-12-2 since 2007 following these Pythagorean plays. Eight such plays apply this year. Here are the best three.


2 overachieving 2019 teams to fade in 2020


Green Bay Packers Under 9 -115

Besides being incredibly lucky last year by going 6-1 SU in games decided by seven points or fewer, Green Bay also lost RT Bryan Bulaga and LB Blake Martinez in free agency. The schedule will do the Packers no favors. Their own division had three teams at .500 or better, and Detroit figures to be better with Matthew Stafford healthy all year. The Packers face the other No. 1 teams from the NFC in San Francisco, New Orleans and Philadelphia. They face an AFC South with three teams capable of finishing .500 or better. And they face improved Tampa Bay.


New Orleans Under 10.5 -110

The Saints have won their division three straight years and have won 13 games each of the last two seasons. But last year they won only three games against teams that finished above .500. This year they will play all but two games against teams with season win totals of seven or higher. And they went 6-1 SU in games decided by seven or fewer points, so they will need to continue to be incredibly fortunate in close games to get over their season win total. Tampa Bay figures to be much tougher with the addition of Tom Brady and Rob Gronkowski. The Saints will also face first-place teams in San Francisco, Green Bay and Philadelphia in addition to defending Super Bowl champion Kansas City.



1 underachieving 2019 team to play in 2020


Dallas Cowboys Over 10 -110

Mike McCarthy takes over for Jason Garrett as Dallas’ coach. While McCarthy faced plenty of criticism at the end of his run in Green Bay, he has always been a particularly good quarterbacks coach and used his one-year hiatus to retool his offensive scheme. The Cowboys won eight games last year despite going 0-5 SU in games decided by seven points or fewer. They play five games against teams predicted to win fewer than seven games. They were just 2-7 SU in games against teams finishing .500 or above last year. That includes going 0-4 SU in games decided by seven or fewer points. Those close losses have a way of evening out from year to year. If they do that, the Cowboys have a chance to go over their season win total.


back to news

Live On Air

Streaming Now: The Lombardi Line

play Watch Live radio Listen Live


Andy MacNeil: NHL - Stanley Cup Finals Most Goals - Sam Bennett (15/1). View more picks.


The Greg Peterson Experience: There is very little value at the top of the College Basketball futures board this time of year, no need to make any bets this early as the numbers won't move during the Summer. View more tips.