Chris Graythen/Getty Images
In my site bio, I wrote:
As a self-proclaimed “numbers guy,” I have always been fascinated by the way mathematics and statistics, if used properly, can thoroughly explain seemingly complex phenomena. Like the motion of the planets or the path of an ant, I truly believe football can be perfectly represented by numbers.
One formula that has always intrigued me is Pythagorean Win Expectation. Like many of the stats I use, it originated in baseball. When adapted to football, the formula predicts the number of wins a team “should” have given their points scored and allowed. Pythagorean Win Expectation is a far superior tool in forecasting a team’s future record than even their past record. This is because it takes “luck” out of the equation.
In football, the formula is PF^2.37/(PF^2.37 + PA^2.37). Why an exponent of 2.37? I’ll take the pragmatic stance and say “because it works.” There’s nothing “magical” about 2.37–it simply has been proven more effective at predicting future records than 2.00 or 2.50, that’s all.
When we calculate the 2010 Cowboys’ Pythagorean Win Expectation based on their points scored and yielded, we see that they “should have” had a win percentage of .440—equivalent to 7.04 wins. This isn’t significantly superior to the six wins they recorded, but it’s still interesting to know the team’s six total wins isn’t perfectly representative of how they played.