Football Freakonomics: Who's Peaking?

(Photo: tyle_r)

The following is a cross-post from, where we’ve recently launched a Football Freakonomics Project.

The preliminaries are finally over. As we head into the first weekend of NFL playoffs, the conversation shifts. No longer are we talking about the long arc of the season – about working out the kinks, getting schemes in place, or jockeying for position. Now, with every game a do-or-die game,  we’re talking about which team is peaking at the right time. Because no matter how good (or bad) your record may be, the final summit is in sight and it’s time to turn on the juice.

In the latest installment of “Football Freakonomics” (see video below), we look at the art and science of peaking. What’s the best way to assess a team’s peak position?

Well, there are a million variables – the kind of arguments the nation is having on sports talk radio 24/7 at the moment – about momentum, injuries, matchups, weather, and so on. And a lot of those are relevant.

But we took a look at one metric that seems to do a good job of describing who’s peaking and who’s not. Kerry Byrne of has identified what he believes is the best peaking indicator: Passer Rating Differential (PRD). It’s nothing more complicated than a team’s offensive passer rating minus its defensive passer rating, but it seems to do a pretty good job of capturing a team’s overall play and momentum.

Here, ranked in order of their season-ending PRD, are the 12 teams still in contention:

  1. Green Bay (42.01)
  2. New Orleans (24.06)
  3. Houston (23.68)
  4. New England (19.58)
  5. Pittsburgh (17.95)
  6. San Francisco (17.89)
  7. Detroit (15.13)
  8. Baltimore (12.83)
  9. Atlanta (7.75)
  10. New York Giants (6.76)
  11. Cincinnati (-5.42)
  12. Denver (-19.67)

To me, Houston is the big surprise there. (What PRD doesn’t account for, however, is playoff experience, of which Houston has zero.) The most remarkable number of course is Denver’s negative 19.67 PRD.

As a Steelers fan, I am hardly comforted by Pittsburgh’s relatively robust 17.95 PRD. Even though the Steelers’ defense has been extraordinary, Rashard Mendenhall’s week 17 knee injury leaves the offense with a game-day running-back corps likely composed entirely of undrafted free agents: Isaac Redman, John Clay, and the newly signed Chad Spann. On the other hand, James Harrison was undrafted too, and that’s worked out pretty well.

If nothing else, this suggests an avenue for future research: overall team success in light of roster-wide draft positions. Just for kicks, here’s a look at one writer’s list of the top 25 undrafted players in NFL history.


The metric regarding passing would only make sense if you used it for the last few games.

The topic of the post is who is peaking NOW.

Big QB numbers from September and October do not show who is peaking going into the playoffs.


my own empirical research shows the saints r peaking and will cruch the nonpeaking lions (by 10.5 i hope)- QED


Two things:

As John B points out, I don't see how this metric, which I assume is an unweighted average of all their performances since early September, indicates which teams are peaking right now as we head into the postseason. If you re-ran the numbers for just the past few weeks, it might be more informative.

But that leads me to my other point, which is that (I believe) passer rating (and its defensive counterpart) is much more explanatory than predictive. If you look at games that have already occurred, you will likely find that the team with the higher passer rating in the game won far more often than not. But it's not very good at predicting who will win games in the future.

Chris Wiley

On a more systems economics approach, I think would like to see the link between the advent of ESPN to Pro Football Revenue disparity instead of a one year analysis. Much more, the revenue from a college football bowl game tied to ESPN coverage. The birth of an internet based College Football media source would be an economic game changer.


Hope it isn't cold, rainy or windy for anyone betting based on passing stats.


2 for 2 on Saturday and 0 for 2 on Sunday of wildcard week.