Writers of Pro Football Prospectus 2008

Most Recent FO Features

CooperAma15.jpg

» Futures: Alabama WR Amari Cooper

Cooper's game has elements of Michael Crabtree and Roddy White. See what he must learn to fit into either one of these starters' development trajectory.

28 Oct 2011

Under Pressure: Coming Unblocked

by J.J. Cooper

Gregg Williams loves to take risks.

When Saints defensive coordinator Gregg Williams goes shark hunting, he leaves his shark cage at home. Williams spends his vacations climbing Mount Everest, and he doesn't need any supplemental oxygen. Whenever he's in Japan, he has to order the Fugu.

He's even been known to go to an Eagles' game wearing an opposing team's jersey.

Actually, I don't know anything about Williams' personal habits. But on the field, he coaches with a bravado and a go-for-broke insanity that is unlike anyone else in the game. There are risk takers, and then there is Williams.

For most teams, a seven-man rush is a fine play call to use every couple of weeks (or once or twice a year depending on the coach). Teams use it as the occasional change-up, an all-out assault to try to catch the offense unaware, or to force a quarterback into a quick checkdown pass on third-and-long.

Last year, the Saints sent six or more rushers 50 percent more often than any other team in the league. The Packers sent seven men twice last year, the Redskins sent seven once. The Saints called a seven-man blitz 42 times.

It hasn't changed this year. We don't have the complete data yet on the number of times teams have sent seven, but the Saints have seven sacks on seven-man rushes this season. The rest of the league has nine. Give Williams credit, he knows how to get a blitzer a free run at the quarterback. In logging sacks where the rusher comes untouched, the Saints have eight unblocked sacks. No other team has more than five.

Unblocked Sacks by Team, 2011
Team Unblocked Sacks Team Unblocked Sacks Team Unblocked Sacks
NO 8 ARI 2 SD 1
WAS 5 CAR 2 SF 1
PIT 4 CLE 2 STL 1
SEA 4 MIA 2 TEN 1
BAL 3 NYG 2 BUF 0
DAL 3 NYJ 2 CHI 0
DEN 3 PHI 2 CIN 0
GB 3 TB 2 DET 0
JAC 3 ATL 1 IND 0
MIN 3 HOU 1 KC 0
OAK 3 NE 1 NFL 2

Williams' frequent blitz calls also cover up something else -- when New Orleans doesn't send a lot of men, they generally don't rack up sacks. The Saints have as many sacks on seven-man rushes as they do on four and five-man rushes combined. New Orleans lacks pass rushers who can consistently generate pressure on their own, so Williams has figured out a way to compensate for it: sending everyone. That also explains why of the five defenders who have picked up two unblocked sacks this year, New Orleans has two of them, including the NFL leader, Jonathan Casillas, who has three.

Unblocked Sack Leaders, 2011
Player Team Quick Sacks
Jonathan Casillas NO 3
Roman Harper NO 2
Jason Babin PHI 2
Jared Allen MIN 2
James Farrior PIT 2
Jonathan Wilhite DEN 2

New Orleans isn't a particularly talented defense, and Williams' blitzing isn't a perfect panacea -- they are 25th in the league in DVOA and 20th in pass defense -- but Williams has figured out how to generate pressure with few true pass rushers.

When it comes to teams who are victimized by unblocked defenders, is anyone surprised to see the Bears at the top of this list? Combine Mike Martz's aversion to hot routes and max protect and Jay Cutler's tendency to hold the ball and you get some unblocked sacks. When the Bears had the misfortune to face the Saints, New Orleans racked up four unblocked sacks came against the pair.

With Seattle, it would be easy to blame the quarterback for some of those, but they are divided rather evenly between Tavaris Jackson and Charlie Whitehurst, and on only one of them did a defensive back come unblocked (those are usually plays where a quarterback has to check down to a hot receiver). The other four all appear to be a case of miscommunication or other confusion among the offensive lineman and the backs.

Unblocked Sacks Allowed by Team, 2011
Team Unblocked Sacks Team Unblocked Sacks Team Unblocked Sacks
CHI 6 ARI 2 OAK 1
SEA 5 BAL 2 SD 1
CAR 4 CIN 2 TB 1
CLE 4 DEN 2 WAS 1
NYJ 4 MIA 2 ATL 0
DET 3 MIN 2 BUF 0
GB 3 NYG 2 DAL 0
JAC 3 STL 2 KC 0
PIT 3 HOU 1 NE 0
SF 3 IND 1 PHI 0
TEN 3 NO 1 NFL 2

On to this week's sack notes, which leads off with a note on an unblocked sack.

QUICKEST SACK OF THE YEAR

In doing this project, you see a lot of sacks -- nearly 3,000 at last count. But in two-and-a-half years of logging sacks, I’d never been a sack like the one Ronde Barber pulled off against the Bears last Sunday.

Normally, timing a sack is pretty easy: you hit the stopwatch’s plunger at the snap and then again when the sacker hits the quarterback. But in this case, it’s hard to promise that my nearly 40-year-old reflexes are up to the task. Barber hit Cutler so quickly that the start and stop clicks are nearly instantaneous.

By my best estimate, Barber first hit Bears quarterback Jay Cutler 0.3 seconds after the snap. In the previous 2,701 sacks, none had been recorded in less than 0.8 seconds, and only two had been recorded in less than a second -- the next fastest sack this year took 1.5 seconds. Barber didn’t really hide his blitz. He came up to the line, stood in the A gap and timed the snap to hit Cutler before Cutler had even gotten his first step into his dropback. It’s impossible to know exactly what the Bears’ blocking scheme on the play was, but in many schemes, a blitzing defensive back is the responsibility of a running back. If that was the case, there was absolutely no chance to stop Barber.

POOR SCHEME?

The Redskins sacked Panthers’ quarterback Cam Newton four times, but what was notable is that only one of those four sacks involved a Panthers’ linemen actually being beaten by a defender.

One of those sacks came when Newton rolled out of the pocket, found no one open, and eventually ran out of bounds. Another came when LaRon Landry was unblocked on a safety blitz from one side while Adam Carriker beat Geoff Hangartner on the other to combine for a sack.

But the two puzzling ones were both the results of what appeared to be either significant mental breakdowns or simply poor line calls. Carriker was left unblocked on a simple four-man rush for the easiest sack of his career on one play, while on another, the Panthers somehow left Brian Orakpo unblocked on a short edge with no tight end on his side. No matter how athletic Newton is, he’s not going to avoid an unblocked Orkapo, especially when he’s coming from Newton’s blind side. The result was what you would expect: a crushing hit only 1.7 seconds after the snap.

INVENTIVE

In that same Panthers-Redskins game, Panthers’ defensive coordinator Sean McDermott came up with a very inventive zone blitz call to force a sack on a key fourth-and-two early in the second half, with the Panthers up by just three.

With the Redskins lined up with three receivers to the right of the formation and another split out to the left, Carolina could feel confident that Washington was going to throw for the first down. The Panthers lined up with linebacker James Anderson and safety Charles Godfrey standing in the A gaps, flanking defensive tackle Ronald Fields. Linebacker Antwan Applewhite also showed blitz as he was lined up just outside of right defensive end Charles Johnson.

At the snap, Anderson and Godfrey came on blitzes, just as they had telegraphed before the play. But Fields, the defensive tackle, and defensive end Greg Hardy both fired off for one step, then peeled back into short zones, as did Applewhite. What had looked like an all-out blitz ended up being a four-man rush.

The Redskins had six blockers to handle the four Panthers’ pass rushers, but since center Erik Cook was focused on blocking Fields, he let Anderson and Godfrey run right by him. Running back Roy Helu picked up Godfrey, but he couldn’t block two men, and Anderson came flying in nearly untouched for an easy sack. The key to the play was sending both rushers up the middle. Neither Fields or Hardy was going to stick with any of the Redskins’ receivers for long, but since Anderson got there 2.2 seconds after the snap, they didn’t need to. John Beck had an open receiver over the middle he could have connected with for the first down, but before he could get rid of the ball, Anderson was already there.

LONGEST SACK OF THE WEEK

Newton's rollout sack was timed at 6.1 seconds, but that was only the second longest sack of the week. Aaron Rodgers had a similar sack that took 6.7 seconds against the Vikings, as he rolled out, found no one open, tried to run for it, and was caught by Marcus Sherels before he could get past the line of scrimmage.

The longest non-scramble sack of the week also went to Rodgers, as he held the ball for five seconds in the pocket before Everson Griffen caught up to him.

Posted by: J.J. Cooper on 28 Oct 2011

32 comments, Last at 01 Nov 2011, 8:53am by nat

Comments

1
by trill :: Fri, 10/28/2011 - 12:21pm

Williams loves the big blitz, but he also likes to show the threat of a 6/7 man rush only to drop 8 into coverage. Seems like they've backed off of this tactic a little since the start of the season, but the Saints D definitely lives on the margins. Hopefully as rookies Cam Jordan and Martez Wilson grow into their roles, the pass rush talent will be there to support some less extreme strategies.

2
by JasonG (not verified) :: Fri, 10/28/2011 - 12:37pm

"New Orleans isn't a particularly talented defense, and Williams' blitzing isn't a perfect panacea -- they are 25th in the league in DVOA and 20th in pass defense"

Is the superaggressiveness making a would-be league-worst defense better or a would-be average defense worse? For example, and not to channel TMQ, but sure the big blitz gets them sacks, but what kind of plays does it give up? And how do those plays compare to plays given up when the D is more straight-up?

15
by trill :: Fri, 10/28/2011 - 3:12pm

I wish Brian Fremeau and Bill Connelly would take the drive-based, efficiency, and explosiveness measures they use for CFB and crunch the numbers on NFL teams. Comparing the Saints' passing S&P+ and Adjusted Sack Rate to the league average would essentially answer your question.

3
by nat :: Fri, 10/28/2011 - 12:49pm

The Saints have as many sacks on seven-man rushes as they do on four and five-man rushes combined. New Orleans lacks pass rushers who can consistently generate pressure on their own, so Williams has figured out a way to compensate for it: sending everyone.

Arg! Once again you're using the wrong ratio, and thus reaching unwarranted conclusions. You need to use rates (e.g sacks per 4-man rush), and compare to league average rates. Instead you use cumulative sacks and compare sacks to other sacks.

Is New Orleans bad at getting sacks with just 4 or 5 rushers? Comparing to their successes with 7 rushers tells you exactly nothing.

If you don't have the data yet, just say so. Just stop making unsupported claims. Please. This whole series has been plagued by wild claims that are unsupported by the data they purport to be derived from.

9
by nat :: Fri, 10/28/2011 - 1:24pm

Why is using the wrong ratios so bad? Here's an example:

Team A has as many completions on screen passes as they do on deep posts and fades combined. Team A lacks a QB who can throw post patterns and the fade, so Coach Jones has figured out a way to compensate for it: throw screens.

How could you possibly draw these conclusions without knowing the number of each type of pass thrown and completed, and typical completion rates for each type of pass?

14
by trill :: Fri, 10/28/2011 - 3:07pm

You're totally right that sack rate is a much more descriptive/useful stat that raw, unadjusted sacks. However, despite the fact that the Saints blitz 6+ more than anyone else in the league, I doubt they're big blitzing with greater frequency than they are rushing <6. So the article's point is still valid.

But yeah, using sack rates rather than raw sacks will help adjust for the number of passes a team sees and the frequency at which they blitz a given number of guys.

18
by nat :: Fri, 10/28/2011 - 4:02pm

My issue is that we can't tell if this phenomenon is due to (a) unusually high sack rates on seven-man blitzes (b) unusually large numbers of seven-man blitzes, (c) unusually low sack rates with 4 or 5 rushers, or (d) unusually low numbers of plays with 4 or 5 rushers.

The point he was trying to make may end up being true, or not. But the data he claims to be using is useless to answer the question, unless you make a lot of assumptions.

I agree that it is unlikely that they are sending seven more than they send less than six. But I don't know the expected sack rates for those two types of rushes. I'm pretty sure that the point of the seven man rush is that it is more likely to get a sack than other rush packages.

23
by J.J. Cooper :: Fri, 10/28/2011 - 5:03pm

I agree that sack rate is a more descriptive sack, but as mentioned in the article, we don't have the numbers on how many times the Saints have sent 3, 4, 5, 6 or 7 in 2011 yet. We have the 2010 numbers, so I used them as a notation that the Saints send 6 or 7 way more than anyone else. Maybe I should have made it more clear, but that observation came both from statistics and also from watching the Saints play. If you watch the Saints on defense, one of the things that becomes clear quite quickly is they do not have many pass rushers who are able to win one-on-one battles. Yes, sack rate would help buttress that observation, but does anyone who has watched the Saints this year disagree with it?

25
by nat :: Fri, 10/28/2011 - 11:05pm

The observation didn't come from the stats. It couldn't have. Because you don't have the right stats.

It's okay that you don't have the right stats yet. My objection is that you use the incomplete stats you do have to "observe" something that you cannot possibly observe or even hint at in those stats. You are doing good work collecting all that raw data. Thank you for that. But please, please stop saying the stats show things they cannot possibly show.

30
by Joseph :: Mon, 10/31/2011 - 1:37pm

Yo, nat, take a chill pill, dude.

I watch all the Saints games--he is absolutely right. First, under GW, the Saints send 6+ rushers more than ANYONE else--and it's not even close (for 2009, it was 22%--glancing through my almanac, nobody else was above 12.5%). Second, the Saints send more than 4 about 50% of the time (in 2009, it was 47.3% of the time; I don't have the 2010 stats, but it's about the same). They absolutely do not get enough pass rush with 4 rushers.
Third, if you want the "real" numbers, run them yourself. Sophandros & mm can confirm what I have said here. The Saints rarely get sacks from just beating their man--they come from unblocked (or barely blocked) rushers.

32
by nat :: Tue, 11/01/2011 - 8:53am

So, to summarize: We all agree that the Saints send six or more rushers a lot more than other teams. We have stats to back that up. We have anecdotal or subjective opinions that they don't have a good sack rate with 4 or 5 rushers, but we don't have any stats (at least here) to back that up. We have another stat (sending four 50% of the time) that doesn't have any bearing on how effective they are when sending 4 or 5, since it only addresses how often they do it, not the results they get.

I don't doubt (or believe, either) your opinion about the Saints' pass rush with 4 or 5 rushers. I'm just saying that J.J.'s stats, and now your stats, don't address the issue at all.

Claiming that the stats quoted here show a weak 4-5 man pass rush was just bad statistics, at a really basic level. J.J.'s articles would be better, and more in keeping with FO traditions, if he didn't make wild claims about what his stats show. FO built its reputation as much on debunking bad statistical "analysis" as it did on DVOA. It's a shame to see that reputation tarnished.

4
by hansolocup :: Fri, 10/28/2011 - 12:52pm

does anyone have a link to the video of the Barber - Cutler sack? I haven't seen it and would be very entertained.

thank you

13
by turbohappy (not verified) :: Fri, 10/28/2011 - 2:51pm
20
by ok you know who I am (not verified) :: Fri, 10/28/2011 - 4:46pm

It is pretty interesting to watch. Talk about knowing the snap count .....
http://www.nfl.com/videos/nfl-fantasy/09000d5d823651c4

24
by Marko :: Fri, 10/28/2011 - 6:15pm

While that was a huge play, its impact thankfully was negated by the personl foul called on Aqib Talib after the play was over.

5
by IAmJoe (not verified) :: Fri, 10/28/2011 - 12:57pm

That Barber sack was pretty quick, but I'm surprised at the margin by which it takes the fastest sack title. If I had to guess off the top of my head, I would think Troy Polamalu, with his "Superman" blitz as the ball is snapped, would've held the fastest sack mark, and at quite a bit lower than 0.8 seconds.

11
by trill :: Fri, 10/28/2011 - 1:50pm

Not to take anything away from that very impressive play by Polamalu, but it looks like the Titans were going for a QB sneak in that situation. Didn't work, obviously, and maybe the stat sheet says it's a sack rather than a TFL, but I think for the purposes of this column we're sticking to pass plays.

16
by IAmJoe (not verified) :: Fri, 10/28/2011 - 3:28pm

Well, I know I've personally seen him do it at least twice (did it against Gabbert recently, I think he ended up breaking up a WR screen), and the one against the Titans, but I don't even remember the situation against the Titans, if it was somethere where they definitely were going for a QB sneak, or if it was a situation where Polamalu was on top of the QB before we could even find out. But even if that was a QB sneak, I wouldn't be surprised if he's done it more than the two times I remember seeing it, and its pretty effective when he does, so I figured I'd throw it out there.

21
by J.J. Cooper :: Fri, 10/28/2011 - 4:57pm

As is mentioned below, the Polamalu hit, while great, wasn't a sack. Collins was clearly running the ball on a quarterback sneak. If a quarterback is clearly running the ball (not scrambling after dropping back to pass), no sack is recorded.

26
by Intropy :: Sat, 10/29/2011 - 1:24am

Before reading the comments I was going to mention that same play. I took a look at the official stats for the game, and they do not record a sack for Polamalu. Watching the video, I agree with you it's extremely difficult to time with a stopwatch. As near as I can tell contact occurs .3 seconds after the snap (8 frames of video at 27 FPS). As to whether it's a "sack" in the actual sense if not the official sense there's just no way to know. Collins is flinching away from a midair Polamalu before he even receives the ball from the center. The flinching causes him to fail to receive the snap cleanly, and he just collapses directly to the ground. It brings up an interesting point about a lower bound on how quickly a sack can be completed. The only reason we can tell Barber's tackle is a sack is that Cutler is strong enough to keep going through the dropping back motion for a couple of steps despite being in the process of being tackled.

Oh, BTW, again counting frames, I get 10 for Barber's sack, or .37 seconds. In both cases the first frame where the ball is not on the spot from which is was snapped is frame #0, and the first frame showing any contact at all with the QB is the stopping frame.

27
by tuluse :: Sat, 10/29/2011 - 11:20am

Shouldn't footage of football be 29.97 or 30 FPS?

28
by Intropy :: Sat, 10/29/2011 - 9:21pm

Do'h. You are right, of course. I knew that, but my brain decided the "9.9" in the middle was unimportant.

Make that .27 for Polamalu and .33 for Barber, plus or minus 0.07.

29
by Aaron Brooks' Good Twin (not verified) :: Mon, 10/31/2011 - 10:46am

Also, it's the Bears, who have Mike Martz and a RB who can't convert goal-line opportunities. You can safely assume it was going to be a pass.

6
by Mr. X (not verified) :: Fri, 10/28/2011 - 1:00pm

Separate of older issuers where he used to hold the ball too long, Rodgers has learned to take a sack rather than try a desperation pass that will more often fail than not. Who cares if he takes a few more sacks if his interception rate is significantly lower?

8
by andrew :: Fri, 10/28/2011 - 1:17pm

well in theory if he could throw the ball away... and he did several times sunday as well....

At least one of those sacks knocked them out of normal field goal range, but Crosby hit the resulting 58-yarder anyway so all was good.

7
by JMM* (not verified) :: Fri, 10/28/2011 - 1:12pm
12
by chemical burn :: Fri, 10/28/2011 - 1:52pm

Am I wrong or is that not really a sack? Collins seems to be ducking down for a QB sneak (or draw?) I guess the league scores them the same way, but in Barber's sack, Cutler is attempting to drop back and Barber catches him. In that sack, Collins ducks forward into the oncoming blitzer...

10
by Nate :: Fri, 10/28/2011 - 1:36pm

It has to be a good number.

17
by IAmJoe (not verified) :: Fri, 10/28/2011 - 3:30pm

When it comes to teams who are victimized by unblocked defenders, is anyone surprised to see the Bears at the top of this list? Combine Mike Martz's aversion to hot routes and max protect and Jay Cutler's tendency to hold the ball and you get some unblocked sacks. When the Bears had the misfortune to face the Saints, New Orleans racked up four unblocked sacks came against the pair.

22
by J.J. Cooper :: Fri, 10/28/2011 - 4:58pm

Yeah, four of them.

19
by Tepid Coffee (not verified) :: Fri, 10/28/2011 - 4:24pm

"In doing this project, you see a lot of sacks -- nearly 3,000 at last count."

Who knew you had something in common with Bibi Jones?

/shows self out

31
by akn :: Tue, 11/01/2011 - 12:31am

You owe me a keyboard.