Puzzle behind the randomness
By Stephen J. Dubner
It doesn't take a genius to argue that injuries can have a massive impact on an NFL team's fortunes. This season, we might be living through the most heightened example in history of that fact. The Indianapolis Colts, with Peyton Manning sidelined since Week 1 with a neck injury, currently stand winless at 0-12. Over the previous five seasons with Manning in charge, the Colts have gone 61-19 during the regular season.
How can the absence of one player, even a star quarterback, have such an impact? As Aaron Schatz of Football Outsiders contends in the latest episode of Football Freakonomics: "Not only were they built around him offensively, but the defense was generally built around them getting the lead and then having defensive ends just tee off on the opposing QB while the other team has to pass to try to catch up."
The Manning-less Colts are losing off the field too -- attendance is down, Manning jersey sales are down, and some Colts fans have jumped on the "Suck for Luck" campaign, figuring that if the Colts are going to be bad they might as well be bad enough to snare the top pick in the draft.
Schatz makes an empirical argument that you don't have to lose a Peyton Manning to lose games in the NFL. There is a clear correlation, he argues, between a team's injured players and its won-loss records. The key metric he uses is called Adjusted Games Lost, and it's calculated like this:
Adjusted Games Lost =
.05 x Players listed as probable
.38 x Players listed as questionable
.99 x Players listed as out
1.00 x Players on IR or PUP lists
To me, the trickiest part of injuries is their randomness -- or, specifically, the question of how random they are. There is a lot we don't definitively know (yet). How evenly (or unevenly) are injuries distributed among teams? What sort of preventive measures work and which don't? Do better players tend to get injured less? If so, why? Maybe they're in better condition or in better position; maybe they're protected better by teammates and officials?
If I were an NFL owner, GM, or coach, I'd set aside a little pot of money to try to answer some of these questions empirically. There is a lot of advantage to be gained by keeping even a few more players per season off the injured reserve list -- to say nothing of the fact that it's the right thing to do. NFL medical staffs have become astonishingly sophisticated in recent years, but the fact is that the game of football exacts a price on muscle and bone that is constant and huge. The more we can learn about cutting that price, the better off we'll all be.