Why it’s so hard to get sacks in the NFL

I like to track data and if you are a fan of defensive football, you will probably already know about the trend that I am going to spend the next few paragraphs discussing: Quarterback play in the NFL is getting incrementally better every year. Don’t believe me - look at the data on two of the most important stats that track quarterback play - completion % and TD/INT ratio.

I’ve (crudely in red) circled the first year that league-wide the ratio of passing touchdowns to interceptions was 1.0 (1983) but I also circled (crudely in blue this time) the last time that the ratio was 1:1 (1992).

Before getting clawed back a little in the past two seasons, the league was fast approaching a 2:1 ratio. Keep in mind that is what the AVERAGE NFL QB does in the modern league. To be considered an elite QB you have be in the 3:1 or 4:1 range. The Denver Broncos quarterbacks over the past three seasons have thrown 58 TDs and 56 INTs.

I also find it interesting to note that the same downward trend that we saw in TD/INT we saw league-wide in completion % over the past two seasons.

So what does this have to do with sacks (and getting pressure on QBs)?

Well, generally speaking, the more pressure that you put on quarterbacks the more sacks you are going to get, except that doesn’t hold true anymore. Quarterbacks are being trained to throw the ball away in order to avoid the sack (and the rules allow them to do much more of that than when John Elway played).

The correlation between pressure and sacks is not as strong as it once was, but there is a nice strong tie between the declining sack rate and the declining interception rate. The interception rate is just total interceptions divided by total passing attempts. The sacks rate is sacks divided by the sum of passing attempts plus total sacks. See the graph below:

We can see that the generally trend line is downward and that the correlation is pretty strong (it’s 0.828 if you really want to know). Note that the slopes are almost exactly the same. That correlation of 0.828 is about as strong as a correlation as you are going to find in sports (and I’m sure - now - someone is going to both point out a stronger correlation and that correlation is not causation).

In this case, we can make a very strong argument for causation though. Quarterbacks are not getting sacked anywhere near as often as they were in the 80’s and even the 90’s. It would stand to reason that if you are not getting sacked as often you are going to probably be able to throw the ball more accurately (fewer interceptions) unless you just have a problem with ball placement like the guys throwing the ball for Denver last year.

Another possible reason for the decrease in sacks is that quarterbacks in the NFL are generally getting larger making them harder to bring down than the signal callers from twenty and thirty years ago. Think about quarterbacks like Philip Rivers (who is listed as 6-5, 230 lbs), Ben Roethlisberger (who is listed as 6-5, 240 lbs), Matt Ryan (6-4, 220 lbs), Blake Bortles (6-5, 236 lbs) and Joe Flacco (6-6. 250 lbs).

In fact if you look at the top 10 in lowest sack % in 2017, only Drew Brees (6-0, 210 lbs), Derek Carr (6-3, 215 lbs) and Case Keenum (6-1, 215) are lighter than 220 lbs. It doesn’t hurt to be big if you are trying to avoid sacks, but the one thing that all of the top 10 from last year have in common is that they are mobile within the pocket. In other words, they know how to move in the pocket to buy just enough time to get off the throw that they want to make (and not the throw that the defense is forcing them to make).

In fact, two of the hardest to sack quarterbacks in the league in 2017 played in the division with arguably the best pass rushers in the league - the AFC West (see below).

I will note that the Broncos avoid many of the top 10 “hardest-to-sack” quarterbacks from 2017 in 2018. Keenum is on the team. They face Carr and Rivers twice each, while they take on Big Ben and Joe Flacco once. Six games against difficult to sack quarterbacks and 10 against more sackable guys or guys with little to no data on (Pat Mahomes, Baker Mayfield, maybe Sam Darnold and Josh Rosen).

So if you combine the first two charts you can see that as sacks and interceptions have gone down, TD/INT has gone up (see below).

Note that sacks were not an official NFL stat prior to 1982 so the data from the 70s is suspect (you can find it and all of the rest of the data from this article at pro-football-reference.com). I wonder which of these stats will flatten off first. Are you approaching the limit of TD:INT ration? Will sacks continue to decrease? What about interceptions?

I don’t mean to say that you cannot get pressure on quarterbacks as the modus operendi of a defense. The 2015 Broncos and the 2017 Eagles proved recently that if you can consistently pressure the opposing quarterback with your front four (or five) then you can dictate what that offense can do. That usually means that you dictate that they go three and out and if you can force a ton of three and outs you have a good shot of making the playoffs and winning the Lombardi trophy.

I am loving the fact that the Broncos have added another putative premiere pass rusher to the fold in Bradley Chubb, but I don’t want anyone to have outrageous expectations about sack totals and/or turnovers forced. The best team ever (the 1984 Bears) had 72 sacks in 16 games (sacks were not an official stat when teams last played 14 regular season games). That means that the ‘84 Bears got 4.0 sacks per game. The Steelers led the league last season with 56 total sacks. The Cardinals led the league in 2016 with 48 sacks. The Broncos led the league in 2015 with 52 and we had the no fly zone and VonWareWolfe.

In fact, you have to go back to 2013 to find a team with 60 or more sacks (Carolina had exactly 60 in 2013). You want to know the last time a team even got close to matching or besting the ‘84 Bears? 1989 when the Vikings had 71. No team has broken 70 since then. Think about this, in 2008 there were only 1036 sacks during the regular season meaning the the league average was 32.4 sacks. To have one team that is close to double the league average in any stat is just crazy and that is a reason why even if the 2018 Broncos are historically great at getting pressure on the opposing quarterback, I sincerely doubt that they will even come close to the getting 72 sacks in the regular season.

The other thing to keep in mind is that the level of offensive line play in the league has gotten significantly worse over the past 20 years. In theory this should make it easier for teams to get sacks, but that is not the trend that we see. So the conclusion is that quarterback play has gotten significantly better even as offensive line play has gotten significantly worse.