Writers of Pro Football Prospectus 2008

20 Jun 2011

Lockout Sure to Burn Blitzing, Complex Defenses

Part Two of my miniseries on the simplest and most complex systems in the NFL. Gauging defensive complexity is even harder than it was for offense, but this method provided some results that both meshed with common sense and provided a surprise or two.

Posted by: Mike Tanier on 20 Jun 2011

10 comments, Last at 27 Jun 2011, 12:13pm by Joseph


by Harris :: Mon, 06/20/2011 - 2:29pm

Castillo was an in-house hire, but he wasn't a defensive assistant. All the scuttlebutt suggests he's going to defer to Washburn and the Titans were one of the most conservative teams. But anything that keeps Trent Cole from dropping into coverage is a good thing.

Hail Hydra!

by tuluse :: Mon, 06/20/2011 - 2:44pm

I like this measure of complexity better than your offensive one.

by KB (not verified) :: Mon, 06/20/2011 - 3:40pm

I really expected to see the Steelers or Packers on this list. I kind of thought Capers had elaborate blitzes always going on.

by DisplacedPackerFan :: Mon, 06/20/2011 - 4:36pm

The Packers weren't quite what Capers may have wanted to do last year. They sat in a 2-4-5 around 45% of the time because of injuries to the linemen (and the backers), and it was usually a 4 or 5 man rush from it, only rushing three or rushing 6 or more was not common. The 3-4-4 wasn't used much and the four man fronts that were there at times in 09 pretty much never showed up. The complexity comes from the fact that Capers tries to make sure one of those rushers gets to come clean or at least is in a favorable 1 on 1 battle. So while it may be complex, it doesn't "look" complex, there weren't a lot of different formations and there wasn't a lot of variation on the number of rushers, so this metric would miss it.

The blitzes aren't that elaborate. It's twists and changing who the 4 or 5 guys are out of the same fundamental set, which makes it easier to learn than expected. Sure a linebacker may be a blitzer one play, zone coverage the, next, man the next, and a spy the next, but his responsibilities don't shift that much and are pretty clear.

What this does allow though is for the defense to move a lot pre snap so the offensive doesn't know just who is doing what so it's harder to pinpoint the favorable matches up. You know that the defense is going to be sending 4 or 5 and that the outsides are going to be covered man-up but you don't know if it will be Sam Shields or Charles Woodson or Nick Collins or Charlie Peprah covering the wide-out. You don't know if it will be Clay and Woodson or Clay and Bishop coming in at you with the linemen. But it's a fairly simple premise so the players can easily see how they fit in even if their job changes from play to play, so it's easier to execute and know what you simply can't let happen and where you can try and gamble some.

The fact they had to "Next man up!" so much last year with rookies and street free agents is a bit of testament to that.

by tuluse :: Mon, 06/20/2011 - 5:00pm

Yeah, I think it's more confusing to the offense looking at it then the actual defensive assignments. It's more of a confusion smoke screen.

by Aaron Brooks' Good Twin (not verified) :: Mon, 06/20/2011 - 6:52pm

This still suffers from the same problems as the offensive projection -- this analysis would hold up if you took 11 new players and tried to teach them a given defensive system.

But the NFL doesn't work that way. Will the Jets and Ravens have *any* rookie starters next year? Who cares how hard a defense is to teach if every player has already learned it?

by Podge (not verified) :: Wed, 06/22/2011 - 4:54am

That's what I was thinking. If you have 11 starters returning then you don't really need to teach the defense. On the other hand, it is informative. I think you could probably come up with a decentish metric combining offensive complexity, defensive complexity, expected number of new starters and changes to scheme due to new coordinators (that'd have to be fairly subjective unfortunately) to measure something like scheme consistency.

It is kind of informative though, because you can have a look at teams with complex or simple schemes and use that to inform your expectations for rookies or new signings.

by rageon :: Tue, 06/21/2011 - 4:26pm

I'll state up front that a piece like this is useful in the sense that it is interesting to see which team runs what most often.

That said, I don't believe I see anything to suggest that teams with certain tendancies are "sure to" suffer.

Let's say hypothetically that a defenese has less time to prepare their blitz packages and they won't be as effective. If that's true, wouldn't the same be true for an offense's ability to prepare for the -- albeit less effective -- complex defense? So that even a complex defense, ran at 90%, is still going to confuse an offense that itself is running at only a 90% abilty to recognize and react to such a defense? Or taken the other way, perhaps a team runs a simple defense well -- but the offense is only prepared to effectively stop a simple defense.

I love the work at this site, but I think the main heading of these articles assumes something that hasn't been proven. It's almost as if the study on complexity was done, and just sort-of fit into the context of the lockout. Perhaps one could tie these complexity scores into prior years following lockouts/strikes and look for a coorelation? I suppose game charting is probably not available for such a task, but I think that would likely be the best way to support it. Or perhaps complexity tied to the effects of a shortened week to prepare (MNF, Thursday games, etc...). Or complexity versus the effects of losing multiple players to injury at a time. Something.

(And to echo what others have said, my (equally without evidence to support it) theory is that teams with the most continuity will probably get the most benefit from the lockout.)

by Podge (not verified) :: Wed, 06/22/2011 - 5:10am

Regarding your second paragraph, I disagree, assuming both teams' players are at 100%.

I think a "complex" defense is one thats more likely to see a blitzer come in from an unexpected angle, or at an unexpected time, which gives them either a free path to the QB or an advantage over the person trying to block them. The other plays are basically leaving your guys in one on one match ups and leaving them to win it.

I'm just going to make up the figures, but lets say the most complex defense has a blitz that tries to do that on 50% of plays (I'm sure that's way out of whack with reality, but I'm just saying that so its easy to do maths in my head). But say the defense has only been able to install 90% of its plays that means those complex blitzes will only be on 45% of plays, with the remaining 55% "simple" plays. I would assume that an offense doesn't know when the D is going to run the complex plays and when its going to run the simple ones, so has to plan to block (or pass before) a complex blitz on every play, meaning that the offense should be kind of overprepared (schematically) on the simple plays. To my mind that means that the offense would be overprepared on a higher number of plays, and therefore at a bit of an advantage.

I am of course assuming that pass blocking schemes are relatively simpler than pass rushing schemes, which probably isn't fair.

I think where the advantage is likely to lie is with players who are simply better, as its more likely that the offense wouldn't be able to scheme as well to get a double team on a great pass rusher, or the D wouldn't be able to scheme as well to get their pass rushers in situations where they've naturally got an advantage over their blockers.

Yes, my conclusion is that teams with more good players are likely to be better than teams with fewer good players, and better players will be likely to be better than not as good players. I hope you haven't collapsed from the shock.

by Joseph :: Mon, 06/27/2011 - 12:13pm

Without having read either article, I will mention this:
All articles that I have seen regarding player-organized activities talk about the offensive players doing drills, running 7-on-7 drills, etc. I have seen very little of defensive players doing much more than working out. Since the QB's seem to be doing most of the organizing, it seems to me that a) the offense is getting more benefit from these informal workouts than the defense; and b) that in general, the offenses will be ahead of the defenses.
IMO, this will work to the advantage of teams with great offenses (IND, NE, NO, GB) and to the teams who must play against teams with great defenses EARLY (Jets, PIT, CHI) before said defenses really get everybody on the same page.
Another note: since offenses have 6 guys who are on the field for every play (OL & QB--except for a few Wildcat teams), while defenses might have 2 or 3 (maybe the MLB and 1 or 2 secondary players), more defensive players have to really know the playbook. IMO, this means that an offense can install more of its scheme than a defense in the same (shortened) time period.
If my premises are correct, teams with high player & coach continuity (IND, NE, NO, GB, PIT--oh, look the same four from before plus PIT--NFL owners, this might not be a coincidence) are going to be well ahead of those who aren't (SF, CAR, DEN).

In fact, (I know this isn't exactly jumping out on a limb), I will say that at least two of (IND, NE, NO, GB & PIT) will get a bye and be in their conference championship game.