The 49ers and Positional AGL

Guest column by Arjun Menon

Injuries play a big part in a team’s success throughout the course of the year. Teams that can stay the healthiest tend to perform better and win more games.

Football Outsiders’ adjusted games lost metric (AGL) helps to analyze how hard the injury bug hit certain teams. Today I will take a look into positional AGL and its relationship to variables such as team wins, offensive/defensive efficiency, and run/pass splits.

Image 1

As a whole, AGL has shown that there is an obvious negative correlation with team wins. The more a team gets injured, the less likely they are to win more games. For example, the 49ers made the Super Bowl in the 2019 season, but in this past year, they were the second-most injured team since 2006 (161.6 AGL), which contributed to them underperforming and winning only six games.

Image 2

If we break down AGL by offense and defense since 2006, it’s clear that offensive injuries contribute more to losing than defensive injuries do. This should be expected given that offensive success is vital to winning and generally more stable than defensive success.

Image 3

To dive a bit deeper into this idea, I looked at the three most valuable positions on the field: quarterback, wide receiver, and defensive back. These three position groups have shown to produce the highest WAR of any position group in football, thus being the most valuable. The data confirms priors, that losing your starting quarterback to injury can drastically affect a team’s chances of winning games.

Now, let’s move on to offensive and defensive efficiency. It’s safe to say that quarterback play is the most vital to a team’s offensive success. However, is there a significant relationship between suffering injuries to the offensive line or wide receiver room and a team’s offensive success?

Position QB OL WR TE RB
Correlation -0.33 -0.08 -0.06 -0.07 -0.04

There really seems to be only a small correlation between offensive success and injuries to any position outside of the quarterback. I believe scarcity is really the answer to this finding, as there are only 20 to 30 starting-caliber quarterbacks in the league, so when your starter goes down, it hurts more than any other position. Most teams carry eight to 10 offensive linemen, ive or six wide receivers, three or four tight ends, and three or four running backs, so there is likely to be more quality depth in those positions than at quarterback.

Image 4

On the defensive side of the ball, it is clear that the secondary and its AGL has the most significant relationship with defensive EPA/Play. It’s also important to note that positive EPA/Play for defenses is not good, which is why the correlation here is positive. On the other hand, negative EPA/Play is bad for offenses, which is why the correlations there were negative.

Year-by-Year Changes

Image 5

When looking at the Total AGL accumulated by season for each offensive positional group, there doesn’t seem to be a particular trend anywhere. I would also like to point out that the AGL data used from 2020 does not include COVID absences or opt-outs, in order to keep the data consistent from year to year. There are a couple things to note:

  • The offensive line AGL in 2020 was the highest it has been all decade. This could be due to COVID and the lack of training reps that the linemen got. Additionally, linemen are among the few players on a football team that play close to 100% of the snaps if they’re healthy. Thus, conditioning is likely more important for them, and the lack of OTAs/minicamps might have hurt them last year.
  • Another note on the offensive line was that there seemed to be a gradual rise in injuries starting in 2013, which could be due to the NFL shifting towards more of a passing league. Most linemen like running the ball because they can impose their strength on a defensive lineman and move forward. However, when passing the ball, linemen have to absorb the impact of the hit they take from defensive linemen, which potentially could lead to more injuries.
  • This same idea applies to the running back AGL since 2014. Since 2014, there has been a slight decrease in running back AGL every year, which again can be attributed to running the ball less often. Additionally, there have been more “running back by committee” situations across the NFL. Teams would much rather divide up the carries between two or three running backs than give 25-plus carries to one player (with a handful of exceptions such as Derrick Henry and Christian McCaffrey).

Image 6

The defensive side of the ball does appear to show a pattern for each position group. The data could suggest a couple things:

  • The linebacker group was the most interesting case to me. Since 2016, the number has declined pretty significantly. There were 107 more AGL by linebackers in 2016 than in 2020. I would say this is due to the rise of nickel and dime defenses, which have proven to be much more successful in stopping the pass. As a result, linebackers have seen the field less and less, which results in them getting injured less often.
  • The effect of nickel and dime defenses can also be seen when looking at defensive backs. As more defenses used five or six defensive backs on the field at around a 70% or 80% clip, the rate at which defensive backs got injured rose dramatically from 2017 to 2018, and stayed around the same level in 2020. This position group will likely always be the leader in AGL on the defensive side of the ball.
  • The defensive line group followed a similar pattern to the linebackers, which is interesting. The point about offensive linemen needing conditioning could be applied here. However, starting defensive linemen don’t usually play more than 75% or 80% of the snaps, and coaches like to rotate their players to keep them fresh. This is likely why there wasn’t a dramatic AGL increase in 2020 like there was for offensive linemen.

Run/Pass Splits

The final topic I want to cover was looking at AGL and run/pass splits. When teams run the ball, there is a lot more contact between players. Usually, every offensive player is blocking a defensive player, and there are more defenders around the ball ready to make a tackle. I hypothesized that teams who run the ball more could be susceptible to more injuries.

Image 7

However, the data doesn’t fit with this idea. It seems that teams that pass the ball more suffer more injuries. This can also be misleading. Instead of thinking about running or passing first, it’s more important to think of injuries as the independent variable, and run/pass splits as the dependent variable.

  • Teams who suffer injuries are less likely to win games. This means they will be playing from behind more often than not. As a result, they are probably going to pass the ball a majority of the time, which causes a higher pass rate.
  • Additionally, teams that are playing from behind play with more pace and tempo than teams that are winning. This puts a lot of stress on the strength and conditioning of the offensive players as they try to score as quickly as possible to get back into the game. On the other hand, teams that are winning like to slow the pace down and eat up clock, which allows them to rest a little bit more on average in the huddle.

Long-Term Totals

Finally, here is a look at the mean Total AGL for every team dating back to 2006.

Image 8

The most healthy offenses of the past decade have been the Falcons, Saints, Titans, Cowboys, and Ravens. The most injured offenses of the past decade have been Washington, the Colts, Patriots, Giants, and Chargers.

Image 9

On the other side of the ball, the Colts, Browns, Giants, Washington, and Bears have been among the most injured defensive teams. The Titans, Steelers, Vikings, Eagles, and Rams have been the most healthy since 2006.

Image 10

That will wrap up this article. I hope it gave everyone some insight into some of the data and trends regarding positional AGL and how it relates to offensive and defensive success.

Arjun Menon is a sophomore at the University of Michigan. He is a lifelong Chargers fan and has gotten into football analytics in the last two years. Arjun can be followed @arjunmenon100 on Twitter.


Leave A Reply

Your email address will not be published.