Which Defenses Depended on "Non-Pressure Sacks" in 2017?
by Aaron Schatz
Data dump time! I was putting together some data for the Patriots chapter in Football Outsiders Almanac 2018 and I decided why not share with the class a little early. I'm not sure we've ever looked at sacks in exactly this way. Not much commentary here, but readers can feel free to respond with their own thoughts and point out what they think is interesting.
I've counted up all the sacks that fit into one of the following three categories:
- Coverage Sacks
- Failed Scramble, which refers to sacks where the quarterback just takes off to run and doesn't make it past the line of scrimmage
- QB Fault, which refers to sacks where the quarterback just drops the ball or trips over his own feet without being touched by the defense
This helps us get to the bottom of the question, "How did the Patriots rank tenth in adjusted sack rate but just 30th in pressure rate?" They were getting a lot of their sacks with great coverage, plus a couple where they didn't do anything at all and just watched the quarterbacks screw up. The Patriots led the league with 16 "non-pressure sacks." No other team had more than 13. Yet the Patriots didn't lead the league in the rate of such sacks. Oh man, was the Tampa Bay pass rush non-existent last year. On the flip side, the Washington pass rush led the league in pressure rate and was massively underrated.
| Rate of "Non-Pressure" Sacks, 2017 Defenses | |||
| Defense | Non-Pressure Sacks |
Total Sacks |
Rate |
| TB | 9 | 22 | 41% |
| NE | 16 | 42 | 38% |
| HOU | 11 | 32 | 34% |
| NYG | 9 | 27 | 33% |
| LAC | 13 | 43 | 30% |
| KC | 9 | 31 | 29% |
| NYJ | 8 | 28 | 29% |
| GB | 10 | 37 | 27% |
| MIN | 10 | 37 | 27% |
| NO | 11 | 42 | 26% |
| CAR | 13 | 50 | 26% |
| OAK | 8 | 31 | 26% |
| DET | 9 | 35 | 26% |
| CIN | 10 | 41 | 24% |
| MIA | 7 | 30 | 23% |
| PIT | 13 | 56 | 23% |
| BUF | 6 | 27 | 22% |
| JAX | 12 | 55 | 22% |
| CHI | 9 | 42 | 21% |
| DEN | 7 | 33 | 21% |
| DAL | 8 | 38 | 21% |
| PHI | 8 | 38 | 21% |
| TEN | 9 | 43 | 21% |
| CLE | 6 | 34 | 18% |
| LAR | 8 | 48 | 17% |
| ARI | 6 | 37 | 16% |
| IND | 4 | 25 | 16% |
| BAL | 6 | 41 | 15% |
| SF | 4 | 30 | 13% |
| ATL | 5 | 39 | 13% |
| SEA | 5 | 39 | 13% |
| WAS | 4 | 42 | 10% |
Here's the flip side, looking from the offensive perspective. Tyrod Taylor sure likes to run around a lot and try to make something happen, and sometimes he just runs out of time.
| Rate of "Non-Pressure" Sacks, 2017 Offenses | |||
| Offense | Non-Pressure Sacks |
Total Sacks |
Rate |
| BUF | 20 | 47 | 43% |
| MIN | 10 | 27 | 37% |
| LAC | 6 | 18 | 33% |
| TEN | 11 | 35 | 31% |
| GB | 16 | 51 | 31% |
| IND | 16 | 56 | 29% |
| DET | 13 | 47 | 28% |
| KC | 10 | 37 | 27% |
| ATL | 6 | 24 | 25% |
| CIN | 10 | 40 | 25% |
| DEN | 13 | 52 | 25% |
| LAR | 7 | 28 | 25% |
| PHI | 9 | 36 | 25% |
| CAR | 8 | 35 | 23% |
| BAL | 6 | 27 | 22% |
| CLE | 11 | 50 | 22% |
| DAL | 7 | 32 | 22% |
| SEA | 9 | 43 | 21% |
| SF | 9 | 43 | 21% |
| OAK | 5 | 24 | 21% |
| NYG | 7 | 34 | 21% |
| NO | 4 | 20 | 20% |
| ARI | 10 | 52 | 19% |
| CHI | 7 | 39 | 18% |
| NE | 6 | 35 | 17% |
| HOU | 9 | 54 | 17% |
| MIA | 5 | 33 | 15% |
| TB | 6 | 40 | 15% |
| NYJ | 7 | 47 | 15% |
| JAX | 3 | 24 | 13% |
| PIT | 3 | 24 | 13% |
| WAS | 4 | 41 | 10% |
Data comes from Sports Info Solutions charting, of course.
Comments
24 comments, Last at 13 Jun 2018, 2:38pm
#1 by Aaron Brooks G… // May 31, 2018 - 5:45pm
When did Roethlisberger stop Roethlisbergering?
3 non-pressure sacks seems like a huge decline for him.
Also -- Jesus, Houston. 45 pressure sacks? No wonder Watson never has his feet set during the film review.
Judging from Indy, Luck may want to be "hurt" for a while longer.
#12 by jtr // Jun 04, 2018 - 10:13am
>When did Roethlisberger stop Roethlisbergering?
Around 2012, when they ditched Bruce Arians. Todd Haley focused on getting Ben to take less hits by working more short and intermediate timing routes rather than forcing Ben to buy time in Arians' bombs away offense. Rothlisberger still breaks some tackles in the pocket since he's so freaking huge, but he's long since stopped with the crazy 12-second scrambles ending in a 30 yard completion.
#2 by Jonrd // Jun 01, 2018 - 5:13am
Great stat. I would love to see individual leaders (possibly in the FO Almanac appendix).
As bad as the TB defense was at rushing the QB last year, I always felt that it was understated just HOW bad they were. Yes, 22 sacks is a terrible number, but watching the games, even that felt like it was high. The secondary is not a good unit, by any means, but if there were any pressure, they could have at least been average. I guess we'll (hopefully) see this year with the new d-line investments.
#7 by nat // Jun 01, 2018 - 1:43pm
Well, teams got sacked an average of 2.3 times a game last season. Buffalo got sacked an average of 2.6 times per (non-NE) game. NE played BUF 2 times. So the "inflation" you speak of is about 2 x 0.3 = 0.6 sacks, well within the noise of a stat like sacks.
That is to say, you're full of Buffalo chips.
#8 by Aaron Brooks G… // Jun 01, 2018 - 3:43pm
Coverage sacks.
Buffalo had 20, or 4 more than anyone else, and 7 more than 4th place.
The other coverage-sacky team Buffalo played was the Chargers. But the Chargers got the human interception for a half instead, which likely depressed their sack numbers.
#13 by nat // Jun 04, 2018 - 3:45pm
Buffalo gave up 20 non-pressure sacks, for an average of 1.25 per game.
League average was 0.53 per game. So the inflation in coverage sacks (due to Buffalo's tendency, not due to game plan or match ups or the Patriots just playing a particularly good game, etc) was 0.72 x 2 = 1.44.
That is, New England's so-called non-pressure sacks were inflated by around one such sack by getting to play Buffalo twice.
Of course, we'd have to know how many of those ten sacks by NE were categorized as non-pressure to see if they actually got that kind of benefit. And we really should judge Buffalo's contribution to that "inflation" based on their non-NE games only.
If all ten of New England's sacks against Buffalo were called non-pressure sacks, then Buffalo is nearly average outside of the New England games, and there was no "inflation" to be had. If none were call non-pressure, then obviously that zero wasn't inflated. If four were called non-pressure, then the inflation is about one and a half. I'm not sure exactly where the maximum is, but it's less than 2.
All in all, we're talking 0-2 New England non-pressure sacks that can be attributed to the Bills being soft about those.
Noise.
#15 by jtr // Jun 05, 2018 - 9:43am
I'd be curious to see how many of the non-pressure sacks for each team came in those two games. Seems like it would be a perfect storm to generate non-pressure sacks. You have a QB who's not very comfortable in the pocket going against a coach who likes to back off on pressure and just play contain against that type of player. I could easily see 6 or 8 of the 10 sacks in those games being non-pressure, where Taylor just ran right into a defender who was standing there waiting for him.
#16 by nat // Jun 05, 2018 - 10:01am
Here are Patriots D highlights from one of the games.
https://www.youtube.com/watch?v=rerTuWYSsIs
Most of these sacks are pressure sacks. Although you could chart some of them as non-pressure because Taylor extended the play with a scramble only to be chased down a second time. That would be wrong, but you could do it.
YMMV.
#17 by sbond101 // Jun 05, 2018 - 10:20am
Yeah; watching week 16 BB (& staff) vs. McDermott (& staff) was an epic mis-match of coaching talent along the lines. On both sides of the ball Buff has notably more individual talent along the lines, and on both sides of the ball they got badly beaten along the lines.
For the raw numbers on the film study, it's a bit of a challenge as a number of the sacks in that game are borderline coverage/pressure to my eye.
#18 by nat // Jun 05, 2018 - 10:59am
...a number of the sacks in that game are borderline coverage/pressure to my eye
Thus the variable mileage in YMMV. It's a judgment call.
Add to that having Brady on the other team. Brady: famed for his ability to move within the pocket to buy extra time despite pressure. So we can look at Taylor bailing and say to ourselves "Hey, Brady would have sidestepped once or twice and completed a pass. No need to scramble there."
Never mind that 90% of the QBs who tried it would have been sacked or hit while throwing or forced to scramble a split second later anyway.
It makes me wonder about these stats when applied to the Patriots D... does seeing Brady in the same game make everyone else look like they abandon the pocket before they need to?
Are the Patriots' non-pressure sack totals a consequence of this natural bias in the charting rather than an actual difference in play?
#20 by Hoodie_Sleeves // Jun 05, 2018 - 2:15pm
" Buff has notably more individual talent along the lines, and on both sides of the ball they got badly beaten along the lines."
This is a funny trope I've been hearing for nigh on 15 years now. For some reason, the Patriots never have any talent, but always seem to physically dominate teams along the lines.
#21 by sbond101 // Jun 05, 2018 - 3:45pm
The trope comes from the physical stature of the players in question - but don't mistake me. The Bills on paper had some of the best groups of linemen on both sides of the ball in football - the Pats game was a subset oh watching McDermott really butcher a great group of man-blockers on the o-line (the DL is a bit more complicated because of the interaction with the LB group).
#24 by cdubya // Jun 13, 2018 - 2:38pm
The Bills offensive line in that game consisted of rookie LT Dion Dawkins, Richie Incognito,
Eric Wood, Vlad Ducasse and Jordan Mills. While it's arguably not a terrible line, it's quite far from "one of the best groups of linemen in football." On the defensive side, outside of Jerry Hughes and what's left of Kyle Williams, there isn't too much worth mentioning on that side of the ball either.