2013 Adjusted Games Lost
by Scott Kacsmar
Wide receiver Michael Crabtree tore his Achilles tendon last May, but returned in time to help San Francisco peak. Denver left tackle Ryan Clady suffered a Lisfranc injury in Week 2 and missed the remainder of the season, including the run to the Super Bowl. New England tight end Rob Gronkowski started the season on the mend, played seven games, and ended 2013 on injured reserve after tearing his ACL. Percy Harvin played just 19 snaps on offense in the regular season for Seattle, but he was healthy enough to make his biggest contribution in the Super Bowl victory.
Injuries are an unfortunate part of the game, but those teams overcame some big ones to make it to the final four in 2013. There were of course many more injuries suffered along the way, and the severity and duration can vary greatly. Sometimes the player makes it back in time, while some teams have to prep a third starter after the backup gets injured as well.
With Football Outsiders' Adjusted Games Lost (AGL) metric, we are able to quantify how much teams were affected by injuries based on two principles: (1) Injuries to starters, injury replacements and important situational reserves matter more than injuries to bench warmers; and (2) Injured players who do take the field are usually playing with reduced ability, which is why Adjusted Games Lost is based not strictly on whether the player is active for the game or not, but instead is based on the player's listed status that week (IR/PUP, out, doubtful, questionable or probable).
As long as NFL teams are solely responsible for producing weekly injury reports, we cannot say that every single injury has been accounted for, but secrecy is an unavoidable aspect of this side of the game.
Obviously every team would prefer a low AGL, but some teams will inevitably have a tough year of injuries. The following table lists the AGL totals and rankings along with the results from 2012 for comparison. This data is only for the regular season.
|Team||2013 AGL||2013 Rank||2012 AGL||2012 Rank|
|Team||2013 AGL||2013 Rank||2012 AGL||2012 Rank|
Perhaps this is the one argument for teams doing a better job of reporting injuries, because once again the average AGL increased. After a league average of 64.5 last season, the 2013 average is 70.2, making it three years in a row that we had a new high in our AGL database (complete for 2002-13).
The correlation for AGL between 2012 and 2013 was 0.10, which is more in line with results from 2002 to 2008 than the higher correlations found in the previous four seasons (2009-12). The correlation between 2013 AGL and 2013 team DVOA was -0.03, which is the smallest since 2002 based on the table Danny Tuccitto prepared in last season's analysis.
Similarly, the correlation between 2013 wins and 2013 AGL was 0.10, which is the second-weakest correlation since 2002 and the first time the relationship was actually positive. Yes, as AGL went up, so did wins, in a bizarre 2013 where seemingly every AFC contender had several crippling injuries to some of its best players. Recall some of the names who finished the season on injured reserve from the AFC playoff teams alone: Reggie Wayne, Dwayne Allen, Geno Atkins, Leon Hall, Vince Wilfork, Rob Gronkowski, Jerod Mayo, Sebastian Vollmer, Dwight Freeney, Von Miller and Ryan Clady.
One AFC contender avoiding the big injuries was Kansas City, which had a league-low 27.6 AGL in 2013 after ranking 22nd a season ago. The Chiefs had big changes at coach and quarterback, but better health is another plus. The irony is that many will just remember the injuries the Chiefs experienced in the AFC Wild Card Game in Indianapolis, such as losing star running back Jamaal Charles on the opening drive. The Colts erased a 38-10 deficit for the win to join the 2012 Packers as the only teams to win a playoff game despite having an AGL above 100.
The Colts ranked 31st (110.3) in AGL for the third time in five years and have ranked 24th or worse in eight straight seasons. Jon Torine was the strength and conditioning coach from 1998-11, but he was replaced in 2012 by Roger Marandino. Despite three head coaches since 2002, GM Ryan Grigson's annual roster purge and new philosophies on both sides of the ball, the Colts continue to be one of the NFL's most injured teams each season.
Seven of the bottom 10 teams in 2013 AGL made the postseason compared to only three of the 10 healthiest teams. In fact, none of the teams ranked 4-10 even had a winning record. It was just an unusual season in how the injuries impacted teams. While a team like the Falcons fell apart, veteran teams used to overcoming injuries like the Packers and Patriots were able to do so again. By some stroke of luck, Aaron Rodgers and Randall Cobb returned just in time for Week 17 to down the Bears and win the NFC North.
After five straight seasons of ranking in the top 10 in AGL, the 49ers suffered 2013's biggest gain of injuries, going from a league-leading 16.2 AGL in 2012 to a 23rd-ranked 84.2. The 68.0 AGL increase is the third highest on record. Crabtree's injury was the big one, but Mario Manningham's struggle to return from his own serious injury was another problem for the wide receivers. The defensive line was also banged up, with nose tackle Ian Williams going down early and standout Justin Smith playing despite a questionable label several times. Despite the increase in injuries and a drop of 12.1 percentage points in DVOA, the 49ers did win 12 games and reached their third-straight NFC Championship. Had the Crabtree injury occurred in July instead of May, he likely would have missed all of 2013.
One team I focused on after Week 1 was Philadelphia, as I asked if the Eagles' fast pace would have any impact on offensive injuries. The Eagles did lead the league by averaging a play every 23.38 seconds. However, despite losing Jeremy Maclin in the summer and another Michael Vick injury early in the year, the Eagles actually ranked eighth in offensive AGL (21.4) after being dead last in 2012. Chip Kelly's team ranked second overall in AGL (33.6), so perhaps his hiring of a sports science coordinator is having a desired impact.
This ESPN Insider piece has more on the top and bottom teams in AGL and what their ratings may mean for 2014.
2013 AGL by Team Units
The New York Giants (7-9) had their first losing season since 2004, but setting a new standard with 144.6 AGL can help explain that. The previous high belonged to the 2009 Bills (6-10) at 122.8 AGL. We'll break down the injuries by units next, but the Giants were banged up on both sides of the ball.
Here are the 2013 AGL splits for offense and defense:
|Top 8||Bottom 8|
The Giants almost pulled off the deflating feat of leading the league in AGL on both offense and defense, but the Saints just edged them out with the worst defensive AGL (74.4) in our database. The Giants rank third behind the 2012 Packers (62.8). However, the Giants, Buccaneers and Colts did all beat out the 2011 Rams (66.6) for the worst offensive AGL in our database. New York's 56.0 increase in offensive AGL is the largest for one season.
Interestingly enough, the first three groups produced three or four playoff teams, with the group of the bottom eight defenses producing five playoff teams. That does not include Chicago and Dallas, both of which hosted a win-or-go-home situation in Week 17. The Cowboys had the sixth-healthiest offense, but Tony Romo's back injury came at the most inopportune time, which I guess is better than a crushing interception. Kyle Orton had his back for that one.
[ad placeholder 3]
Seattle's defense was great, but it sure helps to be healthy. Ranking sixth in defensive AGL works. By the time 27th-ranked Denver reached Super Bowl XLVIII, the Broncos were missing starters such as Von Miller, Kevin Vickerson, Derek Wolfe, Rahim Moore and Chris Harris. Denver has loaded up in free agency on the defensive side of the ball this offseason, though the departure of other players means there won't necessarily be more depth in 2014.
Tampa Bay's personnel (and MRSA outbreak) would suggest a defense-oriented team, but health also played a factor in Greg Schiano's swansong. The Buccaneers paired the healthiest defense with the second-least healthiest offense. New Orleans made the playoffs with the opposite approach: third-healthiest offense and least-healthy defense.
Since there were 13 teams with 0.0 AGL at quarterback (up from eight such teams in 2012), we listed every team.
Thanks to Mark Sanchez, the Jets bring up the rear. No further comment may be necessary, but this is a tough one since most people may just think of Geno Smith as the team's 2013 starter, but do recall the situation in which Sanchez was injured. There was an open competition for the job between Sanchez and Smith. In the third preseason game against the Giants, Smith got the start, but struggled with three interceptions. With backup offensive linemen in and just six minutes to play in the fourth quarter, Rex Ryan had Sanchez in the game. He suffered a serious shoulder injury and missed the season while Smith took over as the starter. Given his release, the injury goes down as the final play of Sanchez's Jets' career.
The surprising team here has to be Green Bay. You could sum up the Packers' quarterback AGL for 1993-12 and it likely would not come close to the 14.0 from 2013 alone. Four starting quarterbacks in one month (November) is a stat you would expect to be associated with a 1987 team (replacement games) or one of Mike Ditka's Chicago teams. The Packers have had so many injuries in recent years, but as long as Aaron Rodgers wasn't one of them, they would be okay. That was greatly tested on a fairly innocent-looking hit against Chicago that broke Rodgers' collarbone. Fortunately, the Packers won some games in crazy comeback fashion with Matt Flynn and an 8-7-1 record was good enough to claim the NFC North. Any other year and a quarterback-dependent team like Green Bay likely would have been toast for the playoffs.
These totals do include fullbacks.
|Top 8||AGL||Bottom 8||AGL|
The 2013 season was the pass-happiest in NFL history, but six of the top eight teams with the healthiest running backs did make the postseason. Chip Kelly turned the Eagles into the best rushing offense in the league and kept rushing champion LeSean McCoy healthy in the process. On the other side of Pennsylvania, the Steelers had the worst running back AGL in 2012 (25.4) and things did not get much better in 2013, but that's nothing compared to the Giants and Colts.
New York's 42.3 AGL is the highest for running backs in our database, beating out the 2007 Buccaneers (37.5). Second-year back David Wilson could never get it going on the ground before he suffered a herniated disc in his neck. Andre Brown missed half the season with a broken leg (second year in a row he's done that). Things got so bad the Giants brought back Brandon Jacobs, and even once he looked to be in vintage shape, he too went down with an injury. Peyton Hillis received 86 touches thanks to these injuries. The Giants even led the league in fullback-only AGL (13.0) due to Henry Hynoski's serious shoulder injury three games into the season.
When can an injury have long-lasting ramifications for a team? After losing one game following a Week 1 season-ending injury to starter Vick Ballard, the Colts kind of panicked and sent a first-round pick to Cleveland for Trent Richardson in a shocking trade. Ahmad Bradshaw would have been a fine replacement for Ballard, but he too went down with a season-ending injury in September. The Colts had Richardson, but he ran like he was hurt too, averaging a pathetic 2.9 yards per carry while Donald Brown ranked second in DVOA and Success Rate behind the same offensive line. Now the Colts enter 2014 with Trent 2.9, Ballard and Bradshaw all returning, but no first-round pick. Things may have looked bleak in September, but there was no reason to sacrifice resources for future building to a position that has been devalued.
|Top 8||AGL||Bottom 8||AGL|
Breakout sophomore seasons by Alshon Jeffery and Josh Gordon headline the two teams with the best AGL for wide receivers. Pittsburgh and Carolina fared well too before letting six of their top seven receivers leave in free agency after the season.
The NFC West powerhouses had their big-name receivers missing for most of the season, but those teams could survive by other means. The Chargers were without their two leading wide receivers from 2012, Danario Alexander and Malcom Floyd, for most of the season and posted the highest AGL (36.0) in our database. So how did Philip Rivers and the Chargers survive and still have one of the league's most efficient offenses? Eddie Royal took a time machine back to his 2008 debut and started the season strong (five touchdowns in Weeks 1-2), but a toe injury slowed him down. The real savior was rookie Keenan Allen, the FO readers' choice for Offensive Rookie of the Year. He fell to the 76th pick in the draft, but Allen looked like the real deal with 1,046 yards and eight touchdowns. He also caught 68.3 percent of his targets -- a rate practically unheard of for a high-volume, outside receiver, especially in his rookie season.
|Top 8||AGL||Bottom 8||AGL|
Hey, look, the Giants actually weren't that bad off at a position. Unfortunately, Brandon Myers was not much of a difference maker and he's now off to his third team in three years. Tampa Bay may be glad to have him given that the Buccaneers set the new benchmark for worst tight end AGL (30.8) in our database. Tim Wright came on late in the season, but the cupboard really was bare in Week 1 when the Buccaneers were trying to start Luke Stocker and Nate Byham (more of a H-back). Both quickly wound up on injured reserve and even reserve Tom Crabtree joined them by season's end.
Consider the Buccaneers a case study in why AGL does not have stronger correlation with performance. The numbers suggest their tight end injuries were catastrophic, but given the caliber of players involved and how they were used, all of Tampa Bay's injuries combined were not worse than the teams missing players like Dennis Pitta, Dwayne Allen, Rob Gronkowski and Dustin Keller (horrific injury in the preseason) for large periods of time in 2013. Tampa Bay was starting guys who would not even register as important reserves for most teams in the league. That was a bigger problem than the injuries themselves.
|Top 8||AGL||Bottom 8||AGL|
Philadelphia's balanced offense was helped by the biggest yearly improvement in AGL (-41.8) on record. Last year Danny mentioned the Chicago Bears having a healthier offensive line for the 2012 season, but 2013 saw dramatic improvement in part because every starter but rookie right tackle Jordan Mills played all 1,057 offensive snaps, and Mills still managed 1,010 snaps (95.6 percent).
On the other side, the 2013 Raiders and Giants both rank among the five worst AGL's for offensive lines in our database. Oakland's macabre game of musical chairs is too much to recap, but let's just say letting Jared Veldheer go in free agency and trying to sign an injured (allegedly, after the fact) Rodger Saffold from St. Louis to a huge contract tells us all we need to know about how the Raiders are evaluating these positions.
A big part of the struggles for the Giants in the running game and Eli Manning's bad season was the lack of continuity on the line (especially the interior), which is something the Giants have often enjoyed. Veteran starters Chris Snee and David Baas each missed 13 games. David Diehl missed five games and had to move to guard while rookie right tackle Justin Pugh was really the only constant at his position all season.
Only two teams in the top eight for offensive line injuries actually made the playoffs. Half the teams in the bottom eight made the playoffs, but it helps to have mobile quarterbacks or someone who gets rid of the ball as quickly as Peyton Manning.
|Top 8||AGL||Bottom 8||AGL|
While the Bears had good luck on one side of the trenches, the defense was not as fortunate. Perhaps it's no coincidence the Bears ranked dead last in DVOA for rush defense. Defensive tackle Henry Melton was a big loss, and he left in free agency to Dallas, which obviously had its own difficulties at keeping the defensive line healthy (and stopping offenses). Chicago and Dallas were the worst defenses at getting the opponent off the field last season.
Danny's fears about the 49ers' lack of depth at defensive line did somewhat catch up in 2013. The 21.2 AGL was more than what the 49ers had in 2005-12 combined (16.8). Justin Smith rarely misses a game, but he will be 35 this season and the 49ers have to start thinking about a replacement. At least Glenn Dorsey did not look like a total bust, and Tank Carradine will also have his opportunity in 2014.
|Top 8||AGL||Bottom 8||AGL|
This position did not see teams making any huge jumps from 2012 or setting a new benchmark in AGL. Seven of the top eight teams failed to finish with a .500 record. At the bottom, remember this list is only for injuries, so the early-season suspensions served by Arizona's Daryl Washington (four games) and Denver's Von Miller (six games) do not count here.
Green Bay followed up the worst linebacker AGL ever in 2012 with the 13th-worst season in 2013.
|Top 8||AGL||Bottom 8||AGL|
If we look at the eight healthiest defenses for all three units (linemen, linebacker and secondary), we find just four playoff teams -- that's counting both the Eagles and Chiefs twice. No one's going to dispute that it's better to have some ailing good players than to have healthy scrubs. The Steelers had the best AGL in the secondary, but failed again to generate many big plays while Ike Taylor allowed quite a few.
Fittingly, we end with the Giants setting another benchmark for worst AGL, beating out the 2006 Browns (40.4). Safety Stevie Brown had a big 2012 with eight interceptions, but missed the entire 2013 season. Veteran Giants like Corey Webster and Aaron Ross both missed 12 games. When Prince Amukamara is the healthiest cornerback on the roster, you know something's wrong in New York.
To recap, the 2013 Giants had the worst single-season AGL for running backs, defensive backs, total offense and overall team. They were also the only team in 2013 to play a player who was listed as "Doubtful" for the game.
Injury Reporting Tactics
Would a team falsify the injury report to try to gain a competitive advantage? If you have followed along at all with injury reports over the years, you probably suspect the New England Patriots to be the team most likely to engage in such deception. Brandon Spikes and Aqib Talib, two ex-Patriots who have moved on to new teams this offseason, recently spoke publicly about their 2013 experiences. Spikes was placed on injured reserve, but said it was a false report. He was late to a team meeting prior to the move. The Patriots listed Talib with a hip injury, but the cornerback has said it was actually his quadriceps.
[ad placeholder 4]
Fabricating an injury to put a player on IR or to put the wrong body part are likely tactics the NFL would frown upon, but can the league really do anything about it? Hiring 32 independent medical staffs to rotate around the country to examine teams on a weekly basis would be an extreme, cost-heavy approach, but that might be the only way to get a fully accurate injury report.
We can use the given data to figure out which teams may be trying to deceive the opponent with the likelihood of a player playing. Generally, probable is nearly a sure bet the player will play, questionable is a 50/50 proposition and doubtful might as well mean out these days.
But the shadiest territory is naturally listing a player as questionable. In 2013, 61.3 percent of questionable players played that week. That's down a bit from 69.3 percent in 2012, but it's still not the theoretical 50 percent. The ranges are quite wide too. Denver (21.7 percent) and Seattle (25.0 percent) actually had the lowest rate of questionable players active while the Falcons let 48-of-57 questionable players (84.2 percent) take the field that week. The Patriots were above average with 69.9 percent active.
Where the Patriots blow away the average is just how often they list a player as questionable. It happened 113 times last season, or 36 more than runner-up New Orleans (77). The other 31-team average was 32.2 questionable players. That's actually down from 2012 when Bill Belichick's squad had 138 questionable designations compared to an average of 21.6. In case you were wondering, Danny Amendola was listed as questionable four times, though he was on the injury report every single week of the season.
Sticking with the AFC East, Rex Ryan and the Jets continued their probable fetish with 189 designations after a league-high 164 in 2012. However, Houston led everyone with 217 probable listings in 2013. The league average was 80.0.
Overall, 94.4 percent of probable players played that week. The Patriots were one of three teams to have 100 percent participation. In terms of percentage played, the Broncos again bring up the rear with just 85.2 percent. So it seems like Denver listed too many players as probable when they should have been questionable. Given only five of Denver's 23 questionable players played, maybe some of those should have been doubtful.
Of the league's 211 players listed as doubtful, just one played that week. Giants rookie defensive end Damontre Moore was doubtful for Week 6 against the Bears, but he played four snaps on special teams. Perhaps helping his cause was the fact it was a Thursday game, so on a short week, the Giants may have needed any bodies they can find to play special teams. But generally, do not expect a doubtful player to play anymore. The Ravens led the league with 21 doubtful players. Three teams (Atlanta, Houston and Tennessee) did not use the designation once.
13 comments, Last at 18 Nov 2014, 12:36pm
#1 by dansvirsky // Mar 28, 2014 - 12:49pm
Interesting stuff. This is a great stat.
At some point, don't the Colts have to do something? It could still be dumb luck (Wyatt Earp stuff)...but if you're willing to pay 20 million a year or whatever for a Peyton Manning level player, why not spend at least that much to try to lower your AGL? Maybe it's not that easy, but yikes, Indy's AGL is always ugly.
#2 by Ryan D. // Mar 28, 2014 - 1:56pm
"New Orleans won the NFC South with the opposite approach: third-healthiest offense and least-healthy defense."
The Saints tied for the best record, but the Panthers actually won the NFC South division title.
#3 by Scott Kacsmar // Mar 28, 2014 - 2:04pm
#4 by ammek // Mar 28, 2014 - 6:19pm
If Sanchez was considered the Jets' starting QB for 2013, should DuJuan Harris have been given the same honor as Green Bay's running back? Mike McCarthy said he "viewed" Harris as the starter before a preseason injury sent him to IR. Was that just spin? It's a difficult one, because nobody believed that Harris would still be getting the majority of the carries by the end of the season, given that the Packers had just spent second-round and fourth-round picks on tailbacks. But then, could we say the same for Smith and Sanchez?
I don't envy you FO writers having to pick through this kind of quandary. But if you included Harris as even a situational player, I'm sure the Packers would show up on the RB list, as Lacy missed a game, his replacement Franklin was knocked out of the same game and missed four others, and James Starks missed three.
#5 by wiesengrund // Mar 29, 2014 - 3:58am
How does AGL deal with Training Camp injuries to probable starters, for instance Mike Johnson in Atlanta? Are they incorporated as being the losss of a full season, or is there an offset because there was still a small chance that they would lose their starter job at some point during the year due to performce?
#6 by Insancipitory // Mar 29, 2014 - 7:25am
How do suspensions impact AGL? Right now it looks like that's just a measure of health, not who's on the field. Now I kind of wonder if suspensions are included, does that improve the correlation.
#7 by Scott Kacsmar // Mar 29, 2014 - 12:53pm
Suspensions are not included. Guys who show up on the injury report with something that's listed as "Not Injury Related" are not included either. In that case, a lot of time it's just rest or maybe a personal matter came up they had to attend to.
#8 by thok // Mar 29, 2014 - 9:14pm
It's sort of impressive that even if you give New England an additional 16 games for Aaron Hernandez, Tampa Bay still lost most player games at tight end.
#12 by eggwasp // Mar 31, 2014 - 4:56am
Oakland had their expected starter - Ausberry, out for the entire year, yet for some reason they were counted as 0.6 AGL for this position.
#9 by MarkV // Mar 30, 2014 - 1:16am
I am always curious when I see this list about injury density and if that correlates to anything. Things such as teams that lose their top 4 cornerbacks for a few weeks stretch, or x players from specified unit at the same time vs spread out over the season.
#10 by Jimmy // Mar 30, 2014 - 2:01pm
I agree with this. Losing starters is bad, losing their backups means you end up playing street scrubs, that's worse.
#11 by Jimmy // Mar 30, 2014 - 2:02pm
#13 by JohnnyZ // Nov 18, 2014 - 12:36pm
Any chance on updating this for 2014 YTD? or posting the list?