Quarterbacks in the NFL today get rid of the ball in 2.7 seconds on average. If you’re not in the quarterback’s face by that time, chances are he’s just completed a pass on your secondary. Give him more time and he's just lobbed a 60-yard bomb past your safety and into the arms of a streaking receiver.
For many football fans and observers, the success of a team's pass rush has always been measured in sacks. But considering the limited time the ball is in the quarterback's hands in modern offenses these days, just penetrating the pocket and pressuring the QB to throw early must be considered a success for the pass rush. So to get a true feel for the effectiveness of the pass rush, you’ve got to measure a defense’s entire ability to pressure the quarterback, and not just the number of sacks.
We know that the Cowboys track sacks, hits and pressures as a way to evaluate their pass rush. Linebackers coach Mat Eberflus mentioned this in the 2012 offseason when he talked about Anthony Spencer:
Linebackers coach Matt Eberflus points out that Spencer was fourth in the league in sacks (6), pressures (31) and hits (9), though pressures and hits are unofficial stats.
One source that also records sacks, pressures, and hits is Pro Football Focus (PFF), and in Spencer's case, they saw 6 sacks, 35 QB pressures, and 9 QB Hits in 2012. That's only four pressures more than the team recorded, but close enough to the numbers the Cowboys use to make me feel very confident about the PFF numbers.
But while these numbers are a good metric for evaluating an individual's pass rush contribution, they can be a bit blurry when measuring team performance, because there's bound to be some double counting going on (for example, more than one player can be credited with a pressure or even a sack on a single defensive snap).
So what PFF offers is a look at the total number of dropbacks and the number of times a QB was pressured (via a sack, hit, or pressure) on those dropbacks. Here's what that looks like for the Cowboys defense over the first seven games of the year.
|Passing Under Pressure
|Pressure in %||28%||39%||40%||33%||38%||45%||26%|
Over the first seven games, the Cowboys sacked or pressured opposing QBs on 36% of their dropbacks. That's right in line with the league average for the top 27 QBs, who are also sacked or pressured on 36% of their dropbacks. And it paints a slightly different picture of the Cowboys defense than if you were to look only at sacks.
With seven sacks in seven games, the 6-1 Cowboys are tied with the 2-5 Falcons and the 0-6 Raiders at 1.0 sacks per game, all ranked second-to-last in the league. The only team worse at collecting sacks so far this season are the 2-4 Rams, who have four sacks in six games - despite a starting defensive line with three first-round picks and one third-round pick.
There are many reasons why the Cowboys have a much better record than the other teams at the bottom of the sack totals ranking, and many of those reasons have to do with the offense, but one reason also has to do with getting pressure on the opposing QB - even without getting a boatload of sacks.
Here's the same table as the one above, except this time I've added the passer rating of the opposing QB in each game.
|Pressure percentage vs. Passer rating
|Pressure in %||28%
|Opp. Passer Rating
Notice a trend? The lower the pressure percentage, the higher the opposing QB's passer rating. You don't see it? Here's a helpful visualization that makes the correlation a lot more visible.
You can see visually that there is a strong correlation between passer rating and the amount of pressure a QB is under - at least in this admittedly small sample. But you can also see it mathematically. R² (the correlation coefficient that measures the strength of the relationship between two variables) for the data above is 0.77, a very high number. The closer R² is to 1 or -1, the stronger the relationship between the two variables. Statistically speaking, the amount of pressure on the defense explains 77% of the opposing team's passer rating in our sample.
But we can top that. If you look closely at the graph above, you'll see that St. Louis is a bit of an outlier in our seven-game sample. If we were to exclude the Rams from our sample, R² jumps to a staggering 0.95. In six of the seven games played so far, the Cowboys pass rushing performance explains 95% of the opposing team's passer rating. Of course, those types of correlations are quite unlikely to hold up over an entire season, but they do show that over the first seven games, the key driver of the Cowboys' defensive performance has been the ability, or inability, to generate pressure, regardless of the sack count.
In yesterday's press conference, Jason Garrett was asked how the Cowboys could get a more consistent pass rush.
A lot of different ways. You have to win one-on-one rushes. You've got to get more pressure with your down guys. Maybe it's one-on-one rushers, maybe it's rush games, maybe it's bringing people.
I don't think we did a good enough job affecting the quarterback in yesterday's game.
You can talk all you want about not getting any sacks, but you also have to affect the guy. They did a pretty good job in their passing game, getting the ball out quickly, and so we didn't get a chance to get to them and disrupt that passing game as much as we'd like.
Now there were some good individual efforts, and at times we did a good job, but we have to improve there. Because it's a really important part of playing good defense.
When we evaluated the game yesterday, that's one of the things that stood out. [Manning] seemed too comfortable, he was getting through some progressions. He did get the ball out quickly, so sometimes it's hard to get there, but we have to continue to improve in that area. Individual rushes, how we're running our games, how we're bringing our pressures. Some how, some way, we've got to affect that guy.
The key to slowing down modern offenses is to get pressure on the passer. If the quarterback is pressured or hurried, he's much more likely to make a mistake and turn the ball over or make enough bad passes to make the offense one-dimensional. If you give any NFL quarterback enough time, he’s going to pick you apart. Pressure him, force him into making a decision and the chances of the QB making a mistake go way up.
Over the first seven games, the Cowboys have shown that they can get pressure on the passer, but not consistently. And that consistency is what the Cowboys will need to work on. The sack totals will follow eventually.