With the exception of the quarterback, there’s no position the NFL values more than the pass rusher. Over the past few years, teams have consistently spent high draft capital on defensive linemen who can pressure and sack the quarterback. In this year’s draft, Ohio State defensive end Chase Young was taken with the second overall pick by Washington. In 2019, defensive linemen Nick Bosa, Quinnen Williams and Clelin Ferrell went 2-3-4 in the first round. And in 2017, Texas A&M defensive end Myles Garrett was picked first overall, while Stanford’s Solomon Thomas went third.
These players are prized because they possess the ability to end drives by taking away an offense’s most dangerous weapon: the passing game. But exactly how much of an effect can a sack or pressure on the QB have on the outcome of a drive? What are the chances that a team can overcome a sack for a productive series? To find out, we looked at 126,205 regular-season plays from the past three seasons and grouped them by drive. We removed series that ended at halftime or the end of the game itself, and we took out rare events like safeties and blocked field goals or punts. Then we tallied up the outcomes of drives that included at least one sack or quarterback pressure and compared them with drives with no sacks or quarterback pressure.
It’s no surprise that getting pressure on the opposing QB is good for the defense. Over the past three seasons, NFL teams forced the opposing offense to punt 5.7 percentage points more often when they pressured the QB at least once on a drive.
Pressure has a modest effect on drive outcomes
Drive results on series with and without pressure on the quarterback, regular-season games 2017-19
| Defensive pressure | No pressure | |||||
|---|---|---|---|---|---|---|
| Drives | Frequency | Drives | Frequency | Diff | ||
| Punt | 3,907 | 41.4% | 2,867 | 35.7% | +5.7 | |
| Made field goal | 1,570 | 16.6 | 900 | 11.2 | +5.4 | |
| Turnover on downs | 518 | 5.5 | 206 | 2.6 | +2.9 | |
| Missed field goal | 274 | 2.9 | 163 | 2.0 | +0.9 | |
| Fumble | 432 | 4.6 | 350 | 4.4 | +0.2 | |
| Interception | 664 | 7.0 | 599 | 7.5 | -0.5 | |
| Touchdown pass | 1,257 | 13.3 | 1,125 | 14.0 | -0.7 | |
| Touchdown | 551 | 5.8 | 745 | 9.3 | -3.5 | |
But pressure doesn’t appear to have a massive impact on interceptions and fumbles, two key measures of defensive success and plays that are often pivotal in winning or losing football games. Overall, these pressure numbers don’t fully explain why teams would spring for elite pass rushers to the extent they do in the draft. More punts and fewer touchdowns are certainly great results for a defense, but the absolute effect of pressure is rather modest: pressure plays in 2019 were worth 0.41 expected points added for the defense.
The types of pressure are fairly diverse, so perhaps this shouldn’t be surprising. A play where a QB takes a hard hit in the pocket while throwing downfield is likely to lead to a better result for the defense than a play that causes a mobile QB to scramble, yet both are counted as pressures.
The same, however, can’t be said for sacks. Sacks are always bad for the offense — they only vary in degree. In fact, when a drive includes at least one sack of the opposing QB, defenses went on to force a punt over 18 percentage points more often than on drives with no sacks.
Sacks have a much more dramatic effect
Drive results on series with and without sacks of the quarterback, regular-season games 2017-19
| Sack | No sack | |||||
|---|---|---|---|---|---|---|
| Drives | Frequency | Drives | Frequency | Diff | ||
| Punt | 1,843 | 53.4% | 4,931 | 35.2% | +18.2 | |
| Fumble | 304 | 8.8 | 478 | 3.4 | +5.4 | |
| Turnover on downs | 202 | 5.9 | 522 | 3.7 | +2.2 | |
| Made field goal | 520 | 15.1 | 1,950 | 13.9 | +1.2 | |
| Missed field goal | 91 | 2.6 | 346 | 2.5 | +0.1 | |
| Interception | 107 | 3.1 | 1,156 | 8.2 | -5.1 | |
| Touchdown | 78 | 2.3 | 1,218 | 8.7 | -6.4 | |
| Touchdown pass | 170 | 4.9 | 2,212 | 15.8 | -10.9 | |
If offenses are punting more often on series when they take a sack, it makes sense that scoring drives on those series are also less frequent. From 2017 to 2019, drives culminating in touchdown passes occurred 10.9 percentage points less frequently when the QB was sacked, and fumbles occurred 5.4 percentage points more frequently.
There is a saying in the NFL that disruption is production, meaning that even if a defensive pressure play doesn’t end in a sack, that player’s contribution is still beneficial. We can see from the data that this is certainly true — pressure plays have value. But pressures are worth only a little over a quarter of the value of a sack. Overall, sacks are strongly positive plays for the defense, accounting for 1.47 EPA per play — over a full point better than the average pressure play. This helps explain why NFL teams value elite pass rushers as highly as they do. A player capable of averaging a sack a game over the course of a season is going to substantially impact a team’s fortunes.
Optimizing for sacks makes sense in this light, but they’re fairly rare outcomes. Across the NFL, teams averaged a pressure about once every nine defensive plays in 2019. Meanwhile, they mustered a sack just about 3 percent of the time, or once every 34 plays. To justify spending top-tier draft capital on a pass rusher, that player needs to do more than just pressure the QB. He needs to get home, wrap up and take the QB to the ground. And he needs to do it frequently.