Writers of Pro Football Prospectus 2008

21 Oct 2009

AGL Through Week 6

With regards to the DVOA Discussion in this week's Ratings thread, I must admit that I'm chalking up the Chargers' struggles this year to an absolutely atrocious run defense, thanks primarily to the absence of the injured Jamal Williams.

Of course, the run defense even with Williams hasn't been great, so perhaps there's more to it than that. Regardless, it got me thinking that I should run our injury metrics through the first six weeks of this year's season.

The data below includes the first six weeks of the season, but not players who were injured in Week 6 and then went on IR, so the likes of Antwan Odom are not included. It also does not include the AGL we know players will gather later on in the season, so while we know that Brian Urlacher will accrue 15.0 AGL by the end of the year, he's only counted in the table below for 4.0 AGL, having missed games in Weeks 2, 3, 4, and 6. Since not all teams have played the same amount of games, the AGL totals for each team are adjusted for the number of games played.

Finally, please note that this is position-adjusted AGL and not the new "HGL" metric discussed in our New York Times article from earlier this year.

Team AGL
SEA 27.5
PHI 24.7
CIN 20.5
NYG 20.4
NE 19.4
TB 18.8
BUF 18.7
IND 17.4
DET 15.4
STL 15.2
JAC 13.9
PIT 13.3
NO 12.7
CLE 12.4
SD 12.3
CAR 11.6
HOU 10.4
OAK 10.4
GB 9.1
SF 9.1
CHI 8.1
WAS 6.6
ATL 6.1
ARI 5.9
MIA 5.9
DEN 5.8
DAL 5.7
NYJ 5.3
TEN 4.5
KC 4.1
BAL 3.8
MIN 3.6

Some interesting things to note:

  • Baltimore has gone from the second-most injured team in the league a year ago to one of the league's healthiest.
  • Teams like Cincinnati and Seattle remain among the most-injured teams in the league, although they aren't as injured through six weeks as they were a year ago. Meanwhile, Kansas City and Tennessee are among the league's healthiest teams for the third consecutive year, despite their poor records.
  • Although Denver has a decent amount of players listed as Probable and Questionable, the only players of note who I believe have missed games are Correll Buckhalter and Ben Hamilton. (The data used above doesn't factor in whether the players in question actually missed the game or not.)

    That surprising Denver defense has been buoyed by the fact that their starters have made it through six consecutive games without, from what I can tell, a single missed start.

Just because a team is suffering through injuries now doesn't mean that they'll be doing so at the end of the year, necessarily. While we know that injuries rise at a roughly linear rate throughout the season, there were six teams that actually averaged more AGL per week through the first six weeks of last season than they did in the final 11 (after adjusting for bye weeks). They included playoff teams in Arizona, Carolina, Philadelphia, and San Diego, along with non-playoff teams in Jacksonville and Seattle.

The four teams that saw the biggest rise in injuries over the last 11 weeks a year ago did not fare very well. The Lions and Rams combined for a 1-21 record, while the Broncos went 4-6 and fell out of the playoffs, and the Bengals went 4-5-1.

While we can't predict who will get injured between now and the end of the year, we can take a look at the players who were placed on IR in the first six weeks (again, not including Week 7 guys like Odom or Kris Jenkins) who we know are going to be missing time.

The team that stands out as the likely Champions of AGL (an award the 2007 Dolphins and, so far, the 2008 Bengals rode to success in the following season) are the Buffalo Bills. Although they're only seventh in AGL at the moment, the presence of Brad Butler, Leodis McKelvin, Kawika Mitchell, and Derek Schouman on IR mean that they're going to accrue 40 AGL between now and the end of the 2009 season. Only one other team -- Philadelphia -- is above 20 AGL the rest of the way, at 22. (The Bengals will join them when Odom is factored in, starting next week.)

The teams with 20 AGL on the way include Cincinnati, Jacksonville, Pittsburgh, St. Louis and Tampa Bay. The teams without any impending AGL are Arizona, Baltimore, Dallas, Denver, Green Bay, Indianapolis, Kansas City, Minnesota, Oakland, Seattle, and Tennessee.

Posted by: Bill Barnwell on 21 Oct 2009

20 comments, Last at 22 Oct 2009, 12:00pm by Wicker

Comments

1
by Key19 :: Wed, 10/21/2009 - 2:33am

Yet more data to confirm that the Titans are pulling a definite "WTF?" season.

Nice to see Dallas back near the bottom after last year's injury issues (which if I remember correctly, resembled an average season injury-wise, but the difference between that and the previous seasons was immense).

6
by Temo :: Wed, 10/21/2009 - 10:15am

I love AGL, I think it's one of the better researched topics on this site, but the obvious deficiency in it right now is that it doesn't adjust for the quality of the injured player. (HGL, obviously, fixes many of these concerns)

Certainly, the difference between Kyle Kosier and Cory Proctor last year was immense and significantly affected the team as a whole. (Obviously Romo to Brad Johnson qualifies as huge too, but that's true for most teams to some degree)

This year, the difference between Gerald Sensabaugh and Pat Watkins has been immense for the pass defense. If it had been Ken Hamlin getting injured, I'm not sure the drop off matters as much. And while we're on the topic... holy hell, please cut Pat Watkins already. He sucks on defense, he sucks on special teams, he just SUCKS.

10
by chemical burn :: Wed, 10/21/2009 - 11:44am

Yeah, I was thinking the same thing because the Eagles have lost three of their most crucial players in Shawn Andrews, Stewart Bradley and Jason Peters. But at the same time, losing Omar Gaither hurt even though he's not very good: losing a back-up on top of a starter creates a gigantic black hole of suck (named Jeremiah Trotter's decrepit knees) that's in some ways more damaging than even the initial loss... Going from back-up the 3rd stringer is higher order of loss, even though the quality of player involved in the new injury isn't as significant...

2
by Danish Denver-Fan :: Wed, 10/21/2009 - 4:55am

How is Tennesees score kept? They've lost the primary starter, the backup and the backup's backup at corner. Are the Titans picking up AGL for all those guys, or only the original starter?

And is there some sort of adjustment for the fact that all their AGL has come on the same position? I would say that it's worse to loose 5 players at the same position than to loose 5 players spread out on the team. Right?

Or is that what "position adjusted" means?

3
by Theo :: Wed, 10/21/2009 - 6:27am

Maybe a quick link to the explanation of the used stats would be nice, I agree.

4
by narticus :: Wed, 10/21/2009 - 7:26am

The Titans have been without both starting corners and their nickle back for several weeks. That's healthy?

9
by dedkrikit (not verified) :: Wed, 10/21/2009 - 11:21am

I second (or third) this question. Does this take into account backups that are injured? For example, Titans TE Cook hasn't been healthy this year (though he has played in at least one game).

5
by moe :: Wed, 10/21/2009 - 9:18am

It might be interesting to have a second column that included the future GL from players on IR

7
by Chip :: Wed, 10/21/2009 - 10:29am

Chicago's AGL seems light.

They've been without 4 starters for at least 2 weeks each, not to mention a number of starters appearing questionable / doubtful at times:
Urlacher missed 4 games (weeks 2-6)
Tinoisamoa 3 games (2-4)
D.Clark 2 games (2-3)
Hillenmeyer 2 games (4-6)

These starters appeared as doubtful / questionable for 1-2 games
A.Adams
A.Brown
L.Briggs
C.Tillman

8
by Matt M (not verified) :: Wed, 10/21/2009 - 10:48am

Has any study been done that looked at medical staff or team medical policy and compared it to AGL or HGL? It seems that the Pats are always one of the most injured teams in the NFL - I'm wondering if it's linked to their medical care. Also, I heard anecdotally that Arizona has one of the best medical staffs in the league, and that's contributing to their continued health.

Anyone else noticing these trends?

11
by chemical burn :: Wed, 10/21/2009 - 11:44am

FO did a study of this once (in the 2007 Prospectus, I think) and the only team that stood out in one way or the other was that Dallas had consistently low injury numbers. The article went on to explain that they used specialized training methods that focused specifically on flexibility (as opposed to muscle-building)... Not sure if other teams have caught on in the meantime or if the study has otherwise been updated... (this is all culled from shaky memories, for the record)

12
by Matt M (not verified) :: Wed, 10/21/2009 - 11:49am

Good note - I'll definitely look it up, I don't remember that article.

Thanks!

16
by Danish Denver-Fan :: Wed, 10/21/2009 - 2:16pm

I don't have any numbers, but I'd suspect that NE has been one of the oldest teams in the league.

On that note: Does age and AGL corrolate?

17
by Boston Dan :: Wed, 10/21/2009 - 2:33pm

You sure about that?

Look at the ages of the guys on the offensive and defensive lines. Are the Pats an old team? Do they have more old players at the positions that are more likely to be injured if there is such a thing?

19
by theshadowj :: Wed, 10/21/2009 - 5:06pm

In the article on HGL it mentions that age and HGL are not correlated because of the survivor effect; that is, older players in the NFL have stuck around in part because they have been able to stay relatively healthy.

13
by Bobman :: Wed, 10/21/2009 - 12:36pm

Well, not that they are a total surprise, but this helps explain Minny's uber-dominance, and the future (lack of IRed guys) bodes well for them, as well as Den and Indy, two of the other three undefeateds. (as a Colts fan, all I can say is will wonders never cease?)

Looking at the four undefeateds and maybe four 1-2 loss teams, I suspect that the team that is healthiest now and down the stretch has a huge advantage. Yes teams get hot, get lucky bounces, and improve their cohesion that may be lacking early on, but winning it all without Antoine Winfield, Dwight Freeney, or Brandon Marshall would be pretty tough for those three teams teams.

Balt is kinda scary--healthy so far, few IRs for the future, and considerably better than their record, IMO. Could be one of those "teams nobody wants to play" come playoff time.

14
by BroncosGuy (not verified) :: Wed, 10/21/2009 - 1:38pm

Bill, I haven't seen every Broncos game, but I believe you are correct that Hamilton and Buck are the only notables to miss a game (and Buckhalter is a committee-member), and the defense has been clean (despite the secondary having an average age of 103).

I think the Chargers certainly miss Williams and, more subjectively, a guy like Merriman doesn't look like the same player post-injury, even if he doesn't show up on the report.

15
by theshadowj :: Wed, 10/21/2009 - 1:40pm

I missed the article on HGL when it was put up, that definitely addresses some of my concerns with AGL. I think it would also be cool if there could be a way to adjust DVOA for injuries in specific games. If a team loses a bunch of their good starters for one game, shouldn't the other team be penalized for not having to play the team at full strength?

For example, say Tom Brady goes out for a game or two this season and Brian Hoyer has to come in. The other team will probably do much better against Brian Hoyer than they would have against Brady. Shouldn't that be adjusted for in DVOA?

Also, unless I'm not understanding this right (which very well might be the case) shouldn't Houston be on the list of teams with 20 AGL on the way? Both of their starting guards are on IR.

18
by bravehoptoad :: Wed, 10/21/2009 - 3:08pm

I'd like to see more about HGL, period. It seems way cool.

20
by Wicker (not verified) :: Thu, 10/22/2009 - 12:00pm

I like the idea of adding a column for future HGL but also a column for this week's HGL -- an estimate of how much HGL the team is expected to accrue just for this week's games. It would interesting, and maybe could be in tandem with the black and blue report.