Writers of Pro Football Prospectus 2008

Most Recent FO Features

owie.jpg

» Turf Type and NFL Injuries: Part II

In our second look at turf and injuries, we look at specific stadiums, with some surprising results. It turns out the turf in Washington isn't nearly as dangerous as you may have been led to believe.

22 Jun 2017

Defense and Pass Pressure, 2016

by Scott Kacsmar

If your kid doesn't look like he has a bright future as a professional quarterback, then perhaps he should learn to love attacking quarterbacks. Pass-rushers are the closest anyone gets to making "quarterback money" in the NFL. Von Miller and Ndamukong Suh each make nearly $20 million per season, and players such as Muhammad Wilkerson, Olivier Vernon, Justin Houston, Fletcher Cox, and J.J. Watt aren't far behind.

Every defense understands the importance of generating pressure. When we looked at the impact of pressure on quarterbacks in 2016 earlier this week, we saw once again that every quarterback performs worse when pressure gets to him. Likewise, the presence of pressure makes every defense more successful.

Sacks are often the gold standard of measurement for pressure, but simply hitting a quarterback while he's in his throwing motion can stall a drive just as well, or even lead to an interception. Pressure brings a lot more value than just sacks. Even ignoring sacks, passers under pressure in 2016 lost roughly 2.0 yards per throw, saw their completion rate drop by 23.5 percentage points, and had their touchdown-to-interception ratio cut in half.

As Carl Yedor pointed out in last year's defensive pressure study, we have seen offenses continue to get better under pressure. This makes sense when many of the skilled passers release the ball so quickly these days, and the athletic quarterbacks are very good at extending plays. Over the last three seasons, the average DVOA under pressure has climbed from -76.0% to -67.3%, and then to -60.3% in 2016. Remember, since this is for defense, the more negative the DVOA, the better for the defense.

The following table shows each defense's success with and without pressure in the 2016 regular season. Pressure plays include sacks, hurries, and forced scrambles. Aborted snaps and coverage scrambles/sacks are not included, but defensive pass interference penalties are. This data primarily comes from Sports Info Solutions charting, but we also check it against the ESPN Stats & Information data. The table is sorted by descending pressure rate.

2016 Defenses: With and Without Pass Pressure
Defense Plays Pressure
Plays
Pressure
Rate
Rk DVOA
w/Pressure
Rk DVOA
no Pressure
Rk DVOA
Dif.
Rk
DEN 625 201 32.2% 1 -63.7% 12 -8.8% 1 -54.8% 32
ARI 630 202 32.1% 2 -92.1% 1 29.9% 13 -122.0% 3
PHI 617 195 31.6% 3 -63.8% 11 13.7% 2 -77.4% 24
MIA 662 208 31.4% 4 -40.6% 27 29.9% 12 -70.5% 29
MIN 615 192 31.2% 5 -77.2% 7 35.3% 18 -112.5% 7
CIN 652 201 30.8% 6 -58.7% 18 34.5% 17 -93.2% 18
CAR 682 206 30.2% 7 -77.4% 6 36.0% 19 -113.4% 6
NYG 678 198 29.2% 8 -80.8% 4 23.3% 6 -104.1% 10
BUF 579 168 29.0% 9 -49.9% 25 40.9% 24 -90.8% 19
SEA 597 171 28.6% 10 -47.7% 26 23.5% 7 -71.2% 28
LARM 647 184 28.4% 11 -56.1% 21 40.6% 23 -96.7% 14
WAS 645 183 28.4% 12 -61.1% 16 45.1% 26 -106.2% 9
CHI 602 169 28.1% 13 -29.8% 31 26.0% 10 -55.8% 31
NO 644 176 27.3% 14 -30.3% 30 49.8% 29 -80.1% 22
JAC 614 167 27.2% 15 -61.2% 15 34.4% 16 -95.6% 16
GB 638 172 27.0% 16 -80.3% 5 48.0% 28 -128.4% 1
Defense Plays Pressure
Plays
Pressure
Rate
Rk DVOA
w/Pressure
Rk DVOA
no Pressure
Rk DVOA
Dif.
Rk
OAK 595 160 26.9% 17 -61.7% 14 41.6% 25 -103.3% 11
SD 646 173 26.8% 18 -56.3% 20 20.3% 5 -76.6% 26
BAL 630 168 26.7% 19 -72.1% 8 26.1% 11 -98.2% 13
ATL 720 190 26.4% 20 -53.6% 24 32.5% 14 -86.1% 21
NYJ 598 157 26.3% 21 -39.2% 28 51.4% 30 -90.7% 20
HOU 588 154 26.2% 22 -54.5% 22 15.2% 3 -69.7% 30
TEN 704 175 24.9% 23 -53.8% 23 45.4% 27 -99.2% 12
PIT 651 161 24.7% 24 -56.8% 19 20.3% 4 -77.1% 25
TB 619 153 24.7% 25 -89.2% 2 24.2% 8 -113.4% 5
NE 657 162 24.7% 26 -69.7% 9 39.3% 21 -109.0% 8
KC 659 161 24.4% 27 -89.0% 3 25.1% 9 -114.1% 4
CLE 590 140 23.7% 28 -69.0% 10 57.1% 32 -126.1% 2
DAL 687 159 23.1% 29 -61.7% 13 33.5% 15 -95.2% 17
DET 603 139 23.1% 30 -24.9% 32 54.9% 31 -79.8% 23
SF 579 131 22.6% 31 -35.1% 29 39.3% 22 -74.4% 27
IND 651 124 19.0% 32 -59.4% 17 37.1% 20 -96.5% 15
NFL 635 172 27.1% - -60.3% - 33.5% - -93.8% -

For the second year in a row, Denver had the NFL's highest pressure rate. The 32.2 percent figure almost matched 2015's 32.7 percent, so it was another stellar year for Wade Phillips' unit in the coordinator's swansong with the team. We'll see how Joe Woods fares as a first-time coordinator, because Phillips has a long history of creating excellent pass-rushing defenses. Woods' background is coaching defensive backs, but he can feel good about his results with Denver's "No Fly Zone" secondary over the last two seasons. Perhaps the most impressive number in this table is the fact that Denver still produced negative DVOA (-8.8%) when there wasn't any pressure. The league average was 33.5%. That is a testament to how well the Broncos can cover with Aqib Talib, Chris Harris, Bradley Roby, Darian Stewart, and T.J. Ward in the secondary. The only other defenses to produce negative DVOA without pressure since 2010 were the 2012 Bears (-1.6%) and 2013 Bills (-0.3%), so Denver's pressure-less mark is the best over the last seven seasons.

Arizona also had a strong pass rush again, finishing second in pressure rate and first in DVOA with pressure. The trade for Chandler Jones worked out, but we didn't expect to see the team's offense regress so much after Carson Palmer's MVP-caliber season in 2015. Of course, Palmer was the most knocked-down quarterback of 2016, so there was a lot of pressure against the Cardinals as well last year.

A couple of NFC East teams really improved last year. Philadelphia climbed from 19th in pressure rate in 2015 to third in 2016. Brandon Graham is a case in point that sacks aren't everything. He only had 5.5 sacks, but he was second in the NFL with 52 hurries. The Giants were also a big climber on their way back to the playoffs, and the acquisition of Olivier Vernon sure paid off. He led the NFL with 66 hurries. Imagine if Miami would have paid Vernon instead of bringing in Mario Williams, who was benched and then cut after one disappointing season. Still, the Dolphins had the fourth-highest pressure rate thanks to a combined 71 hurries from Ndamukong Suh (35) and Cameron Wake (36), but only ranked 27th in DVOA with pressure, the lowest for any of the top 12 teams in pressure rate. The coverage needs to get better in Miami.

Carolina (seventh) and Seattle (10th) were still top 10 defenses in pressure rate, but the quality of their secondaries was also on display. The Legion of Boom, which might not have been as strong overall with Earl Thomas and Richard Sherman fighting injuries in 2016, helped Seattle finished seventh in DVOA without pressure. Meanwhile, the Panthers struggled a bit after losing Josh Norman, and only finished 19th when pressure didn't get there.

Jacksonville was a model of mediocrity on this front: 15th in pressure rate, 15th in DVOA with pressure, and 16th in DVOA without pressure. Dante Fowler (31) and Malik Jackson (29) combined for 60 hurries, but that talented-on-paper defense is still a work-in-progress on the field. Gus Bradley's time on the job is done, and he is the new defensive coordinator for the Chargers, an interesting defense to watch in 2017 with Joey Bosa and Melvin Ingram leading the rush. The Chargers finished fifth in DVOA without pressure, and top cover corner Jason Verrett only played four games due to injury. Cornerback Casey Hayward was a big surprise for the Chargers, leading the NFL with seven interceptions.

Last year was Hayward's first with the Chargers after four years in Green Bay, and the Packers would have loved to have had him back in 2016, or maybe just a few healthy corners period. These numbers summarize the 2016 Packers well. The pass rush didn't get home often, but was effective when it did (fifth in DVOA), and the banged-up secondary fell apart when it didn't (28th in DVOA). Hence, the Packers had the biggest decline in DVOA (-128.4%) between when their pass-rush failed to generate pressure. Cleveland finished second with a similar effect, including the worst DVOA without pressure, but will hope that No. 1 overall pick Myles Garrett can bolster the pass rush.

Oakland did get a Defensive Player of the Year effort from Khalil Mack, who finished in a third-place tie with Cameron Jordan with 50 hurries. Bruce Irvin (38 hurries) was also productive in coming over from Seattle, but the Raiders were not good without pressure (25th in DVOA). Cornerbacks Sean Smith and David Amerson failed to repeat their 2015 successes, so the Raiders wound up spending a first-round pick on controversial rookie Gareon Conley from Ohio State. Oakland has to play much better defense if it is to be taken seriously as a Super Bowl contender.

How about the teams in Super Bowl LI, Atlanta and New England? Neither defense produced a good pressure rate, but the Patriots were effective (ninth in DVOA) with pressure, while the Falcons (14th in DVOA) fared relatively better without it. Cornerback Desmond Trufant was a big loss for the Falcons last year, but even if he had been there in February, the whole defense looked exhausted in the later stages of the championship game when Brady began to lead scoring drive after scoring drive. Meanwhile, those five sacks of Matt Ryan really made the comeback possible for the Super Bowl champs. Bill Belichick does a great job of confusing offenses and making the quarterback hold the ball. You don't have to always get a lot of pressure to be a Super Bowl team, but you definitely need to get timely pressure. The Patriots did that last year, while the Falcons needed just one more stop that never came.

Houston's fall to 22nd in pressure rate makes perfect sense when J.J. Watt missing the majority of the season. The coverage was still very good. Pittsburgh and Tennessee had very similar results, except when it came to DVOA without pressure. Dick LeBeau's Titans defense finished 27th, while the Steelers were a surprise at No. 4. The Steelers certainly got a schedule boost from the quarterbacks they faced, but also did a good job of keeping the play in front of them. Pittsburgh faced the second-shortest throws on average at just under 7.0 air yards per attempt. In the secondary, Sean Davis was a big improvement at safety over Robert Golden, and Ross Cockrell charted better than expected in coverage (46th in success rate, 31st in adjusted yards per pass).

Kansas City's No. 27 ranking in pressure rate is concerning, but Justin Houston did miss 11 games last season. If he can stay healthy, and along with Dee Ford's progression, then the Chiefs should have a strong duo, plus Tamba Hali off the bench. We know the Chiefs can cover well with two All-Pros (Marcus Peters and Eric Berry) in the secondary, and the defense was ninth without pressure.

Finally, we have the bottom four teams in pressure rate. The Lions and Cowboys allowed a lot of short, easy completions last season, but Dallas was better at squeezing out a few more sacks without having to blitz as often. Sacks aren't everything, but the Lions sure expect a lot more than 2.0 of them out of Ezekiel Ansah. The 49ers have been building a defense for the future with several first-round picks used on the front seven, but we knew 2016 was going to be a struggle. We just didn't know the team was only capable of beating that horrific Rams offense.

Finally, there were the Colts. For a defense that ranked 27th in pass defense DVOA, the pressure splits look more mediocre than terrible here, ranked 17th with pressure and 20th without it. Unfortunately, no defense generated pressure less often than the Colts, who only bothered quarterbacks 19.0 percent of the time, the lowest rate of any defense in the last four seasons.

This was the final straw in general manager Ryan Grigson's tenure. Grigson notoriously ignored defense in the draft for years, opting for old free agents that rarely performed. Erik Walden had the oddest pass-rushing seasons in our data this year. Walden was one of 16 players charted with double-digit sacks, but he only had 13 additional hurries. The other 15 ten-sack players averaged 31.5 hurries, and Minnesota's Danielle Hunter (19) was the only other player with fewer than 20. Out of 44 players with at least 7.0 sacks, only Walden had fewer than 18 hurries. When Walden, now a free agent, actually got pressure, he finished it with a lot of sacks, but he generated very little pass rush otherwise for the Colts. In fact, two of his 13 hurries still led to touchdown passes for the opponent. The Colts project to only start three defensive players they drafted, including rookie safety Malik Hooker.

We conclude with a brief look at something readers have mentioned in the comments and on Twitter that they wanted to see: pressure splits by the number of pass rushers. While it would also be helpful to show the frequency of how often teams used each strategy, and the accompanying DVOA, that all starts to look like a mess in one big table, so we'll stick with the pressure rates for now.

The following table splits frequency of pressure for each defense when using two or three rushers; four-man rushes; and anything with five-plus rushers (blitzes). The table is sorted by descending pressure rate on blitzes.

2016 Defenses: Pass Pressure Rate Splits by No. of Pass Rushers
Defense 2-3 Rk 4 Rk 5+ Rk
CAR 27.8% 6 28.1% 9 46.5% 1
NYG 25.0% 12 27.5% 14 45.3% 2
JAC 37.5% 4 22.7% 26 45.2% 3
CIN 40.9% 2 29.3% 7 44.7% 4
PHI 77.8% 1 29.9% 6 43.1% 5
SEA 37.5% 3 24.9% 22 41.9% 6
MIN 8.3% 31 31.8% 2 40.8% 7
DEN 17.9% 23 30.4% 4 39.9% 8
ARI 18.5% 22 30.1% 5 39.7% 9
MIA 6.7% 32 32.3% 1 39.0% 10
CHI 22.0% 16 29.2% 8 38.1% 11
BUF 20.2% 19 31.7% 3 37.7% 12
SD 25.4% 9 25.6% 20 36.9% 13
WAS 32.1% 5 26.2% 17 36.3% 14
TB 27.3% 7 22.7% 27 36.1% 15
DET 21.1% 18 19.9% 30 36.1% 16
Defense 2-3 Rk 4 Rk 5+ Rk
ATL 14.8% 28 27.9% 11 36.0% 17
DAL 23.7% 14 19.6% 31 35.6% 18
LARM 24.6% 13 28.0% 10 35.5% 19
PIT 15.9% 26 24.7% 23 35.5% 20
TEN 22.5% 15 21.1% 29 35.0% 21
NO 21.4% 17 25.5% 21 35.0% 22
HOU 14.1% 30 27.2% 15 34.7% 23
BAL 18.8% 21 26.8% 16 34.0% 24
OAK 14.3% 29 27.6% 12 33.6% 25
NYJ 25.4% 10 24.4% 24 33.5% 26
KC 14.8% 27 26.1% 18 32.4% 27
GB 25.0% 11 27.5% 13 31.7% 28
CLE 16.2% 24 22.4% 28 31.2% 29
SF 25.9% 8 22.7% 25 28.7% 30
NE 19.0% 20 26.1% 19 28.2% 31
IND 16.0% 25 18.0% 32 24.2% 32
NFL 23.7% - 37.5% - 36.6% -

The Panthers, Bengals, and Eagles were the only defenses to rank in the top 10 for each split. The Eagles and Bengals were the only defenses to rush four more than 75 percent of the time, doing so on 79 percent of passes. Carolina's blitzing produced the highest pressure rate (46.5 percent), while the Colts brought up the rear in both blitzing (24.2 percent) and a standard four-man rush (18.0 percent).

As we observed leading up to Super Bowl LI, the Patriots used a three-man rush more than any defense (24.0 percent of pass plays). Part of this was playing with the lead so often, as Belichick liked to keep things in front of his defense instead of being too aggressive. It did not help that the Patriots had the second-lowest pressure rate when blitzing last season (28.2 percent). New England's blitzing frequency rate was the sixth-lowest in 2016 at 21.0 percent. The Saints and Jets both blitzed 41 percent of the time, but both pass defenses were awful, and neither produced a high rate of pressure with that strategy. No defense blitzed less frequently than Chicago (15.0 percent), but maybe the Bears could have used some more aggression. They ranked 11th in pressure rate on the blitz.

The Eagles had an absurd pressure rate of 77.8 percent with the three-man rush, but we're talking about a league-low nine plays there, so don't get too excited. Miami had a poor pressure rate from a three-man rush but the highest pressure rate with a four-man rush, led again by Suh and Wake.

Posted by: Scott Kacsmar on 22 Jun 2017

4 comments, Last at 24 Jun 2017, 8:49am by Whidbey Bronco

Comments

1
by killwer :: Fri, 06/23/2017 - 4:56am

How on earth was the Eagles 2nd in DVOA with no pressure? We had ZERO above average CBs last year and we could still cover better than 30 teams when the DL wasnt getting pressure?

2
by Hoodie_Sleeves :: Fri, 06/23/2017 - 12:38pm

This whole article would be a lot more useful if you posted the table of percentage of snaps by number of rushers that you mention several times in the article.

The last table is pretty much useless without knowing how much of the variance is just sample size issues.

3
by gofastjoey :: Fri, 06/23/2017 - 1:04pm

agreed. it basically only helps confirm the colts and browns had terrible defenses.

4
by Whidbey Bronco :: Sat, 06/24/2017 - 8:49am

What an excellent article. Thanks for the research. If I'd been on a desert island for three years, this would catch me up.

BTW Go Broncos NFZ.