This article and all of the research was written by 18to88 reader Dan Svirsky and edited by me.
For several seasons the Colts have been ravaged by injuries. The endless wave of big names showing up on the weekly injury sheet has frustrated fans as the Colts have gone into several recent postseason games missing indispensable players like Dwight Freeney, Gary Brackett, and Jerraud Powers. Moreover, on a week to week basis, the team has had to adopt the well-worn "Next Man Up" mantra. The following study seeks to determine the following:
- Do the Colts have more injuries than other NFL teams?
- Can the Colts' injuries simply be the result of bad luck?
- What other causes might explain the injuries?
Measurement of the cost of injuries, both in terms of missed games and games where players were not able to play to their full potential. Estimates a number of games based on whether players are listed as Probable, Questionable, Doubtful, or Out. Introduced in Pro Football Prospectus 2008 essay, "The Injury Effect.
Below are the top five best and worst AGL rates among all NFL teams for the past five years. The tables list the team, the team’s average AGL over the past five years, and the p-value for team’s AGL, which I'll explain below.
NFL Average AGL
There's a lot of interesting stuff here. Just looking at the lists, Indy is in some awful company. We have the second worst overall injury rates in the league along with arguably the worst teams of the past five years (though maybe you'd exclude the Bengals and add a team like the Raiders). Moreover, the Colts have been shown to under-report injuries, which actually depresses their AGL. So, right off the bat, it is incredible that the Colts have had such success given their injury rates.
Where it gets truly crazy though is when you break it down by units. Here are the top five best and worst 5-year average AGL ratings broken down by unit:
The Indy D is the most injured unit in the entire league over the past five years, and it's not even close. To truly understand just how crazy that Colts' D's AGL number is, though, it's best to turn to some stats.
So what's this p-value? Basically, the p-value measures the probability that you'd get an average AGL that high just by luck. Let's start by looking at the Titans. Tennessee's average overall injury rate over the past five years is 13--way below the league average of 22. If AGL rates were completely random, and teams’ injury rates were just luck of the draw, you'd get a number like Tennessee’s only 2.94% of the time, or about 1 in 33 times. Now, this is pretty low, and it's pretty good evidence that Tennessee does a great job with injuries. But remember, we're dealing with a league of 32 teams. So, you actually might expect one team in the NFL to have an injury rate that far below average, since it'll happen 1 in 33 times.
But now look at the Colts D: the AGL for Indy's defense is so bad--our players get injured so much on D--that you'd get an AGL that high by luck only .3% of the time--1 in 333 times. And considering how bad 2010 is shaping up, the Colts D's number is only going to get worse.
The bottom line is that over the past five years, the Colts defense has been the most injured unit in the league, and it is unlikely to be due to bad luck.
So, what might be causing this wave of injuries?
Theory 1: The Tampa 2 Factor: Undersized defenses suffer more injuries
I compiled the average weight and height of every team's defense over the last 5 years, and sure enough, if a team is in the bottom 5% for average weight on defense, they will have an AGL about 7 games higher.
Ave. Wt. on Defense (Lbs)
Remember that Colts' injury rate is unlikely due to just luck. If you don't take their average weight into account, the Colts D's AGL is so bad that you'd expect it to happen by chance only 1 in 333 times. When you do take their average weight into account, however, you get an AGL so bad that it'd happen by chance 1 in 20 times. This suggests that the Colts' insanely high injury rate on D is what happens when bad luck hits an undersized defense.
Theory 2: The Bend-But-Don't-Break Factor
I've heard the argument that because the Colts' D has a 'bend but don't break' style, they'll be on the field more and therefore risk more injury. To test this, I compiled each team's average defensive Time of Possession (measured roughly: defensive time of possession = 60 minutes - offensive time of possession - a couple minutes for special teams plays). Being on the field more as a defense was associated with more injuries, and the Colts do tend to have higher TOP averages (though not crazily so), but the result was not statistically significant.
Theory 3: The 'Training Staff' Effect
I consider this the most interesting result so far: for every point you increase offensive AGL, defensive AGL increases roughly .2 points. For every 10 games that starters on the offense miss, starters on the defense miss 2 games. So what's going on here? On the surface, offensive AGL shouldn't really have an effect on defensive AGL. If Pierre Garcon pulls a hamstring, you wouldn't expect it to be more likely for Melvin Bullitt to mess up his shoulder. You could argue that a bad offense keeps the defense on the field more, thus leading to more injuries, but as we saw above, there doesn't seem to be a relation between time of possession and AGL.
The best interpretation of the offensive AGL/defensive AGL link, in my opinion, is that there's some factor causing both offensive and defensive AGL to rise together--maybe some sort of intrinsic team characteristic. For example: maybe it's a bad training staff; maybe it's a team that cares more about the health of its players; maybe teams have different policies on how long to keep a player out. All of these could be factors because they have an effect on injury rates and are also correlated with offensive and defensive AGL. Maybe it's just the 'end of the season effect' where the Colts rest players heavily down the stretch. So while it seems that higher offensive AGL is causing higher defensive AGL, what's actually happening is that something is causing both.
Here's the bottom line:
The Colts' injury woes on defense have two plausible explanations.
It's either the result of: 1) undersized players and a whole lot of bad luck, or 2) undersized players, some bad luck, and something about the way the team itself handles injuries that causes more missed games.