Statistical In-Ferentz, Week 4: Six Defensive Backs in Search of a Receiver

Statferlogo_medium
The Iowa offense got the lion's share of the attention in the game this weekend, and deservedly so, but there was a lot of interesting and sometimes dismaying stuff happening on the other side of the ball, so I thought I'd look that way this week. First, some general facts about the defense:

  • Opponents are averaging 23 points per game against Iowa (18.8 if we ignore the overtime periods against Iowa State), which ranks 60th in the FBS (30th without the ISU overtime scores)
  • Iowa is giving up 385.8 yards per game, 77th in FBS
  • Iowa is giving up 3.2 yards per rush, 36th in FBS
  • Iowa is giving up 10.6 yards per pass completion, 31st in FBS
  • Opposing QBs are averaging a 66% completion percentage, 103rd in FBS

Ranking in the top 40 may seem pretty good, but compared to recent Iowa defenses, it does mark a dropping off: from 2007-10, Iowa ranked 15th, 8th, 10th and 7th in points allowed per game, and 40th,12th, 11th and 15th in opponent's yards per game. Iowa generally gives up a higher completion percentage than you would assume for such a good defense, but this year that number has become particularly awful. For comparison's sake, consider that only three starting quarterbacks in the Big 10 are averaging better than a 66% completion percentage: Russell Wilson, Nathan Scheelhaase and Kirk Cousins. With all due respect to Steele Jantz and Tino Sunseri, Iowa has made some fairly average quarterbacks look very good in their first four games.

So there is some doom and gloom in that area. But what exactly is Iowa doing on defense that is allowing opposing quarterbacks to feast? The University of Louisiana-Monroe game provided a useful case study: prior to the Iowa game, the Warhawk offense averaged 176 yards passing per game and a 53.3% completion percentage (and was held to 95 yards passing by Florida State); against Iowa, ULM went off for 293 yards and a 65.9% completion percentage. What happened?

The formations

I watched the ULM game again and tried to figure out which defensive formations Iowa was using on pass plays. I don't claim to be a savant when it comes to distinguishing defenses from their initial set-ups (TV broadcasts games also makes it hard to see what defenses do after the snap), but I did try to distinguish between cover 1, cover 2 and cover 4 looks, as well as noting when unusual personnel groups were in, such as three linemen or extra defensive backs:*

* If I got the names of formations wrong here, please let me know in the comments. I tried to be consistent, so if what I call "cover 2" is not really cover 2, at least you can be sure that I made the same mistake for every defense that looked like that at the snap. Also note that I ignored the final ULM drive where the second string played.

1) 4-3-4, cover 2 - two safeties covering deep halves of the field, with corners and linebackers in either man or zone underneath

Screenshot4__cover_2_man__medium

2) 4-3-4, cover 4 - corners and safeties covering deep quarters, with linebackers in man or zone underneath

Screenshot3__cover4__medium

3) 3-4-4 cover 2 or cover 4 (i.e. three down linemen with zone behind)

Screenshot4_3linemen_medium

4) 4-1-6 cover 1 or cover 2 (i.e. four down linemen, six defensive backs, one or two deep safeties)

Note on this screenshot: the linebacker doesn't necessarily blitz on every play, but he did on this one (this play will come up later).

Screenshot5-_6dbs__medium

So those are the defenses Iowa used as far as I could tell.  There were a couple other formations -- a goal-line defense and a nickel package -- but these four defenses accounted for 41 of the 43 pass plays I looked at. And here is how the different set-ups compared in terms of usage and yards allowed on pass plays:

Formations_medium

As it turns out, Iowa relied on cover 2 about half of the time, with a good mix of cover 4 (23%) and opportunistic use of 3-4-4 and 4-1-6 schemes. Most of the damage came against the cover 2. The three-linemen look worked well, but it was used mainly in situations that greatly favored the defense (e.g. 3rd and 16, 3rd and 9), so  it did have some inherent advantages. The cover 2 defense was used in more neutral situations (e.g. 1st and 10, 2nd and 7), so it didn't have quite the same advantages, but at the same time, that defense did allow an 81% completion percentage, 7.9 yards per pass play, and 9.7 yards per completion. That's ... that's not good.

Where did the damage come?

The chart below shows  where ULM's passes were completed in terms of areas of the field (imagine this chart as a football field from the defensive perspective, so the top left is the short left zone, the top right is the short right zone, etc.):

Zones_medium

As you can see in the chart, the defense by and large accomplished the "don't break" part of "bend but don't break" against ULM. The Warhawks went 3/11 for 78 yards in the deep zones of the field, and Micah Hyde was untouchable, allowing zero completions on six deep targets. In addition, two of the long completions were fairly fluky: one came on a play where the ULM quarterback picked up a bad snap and alertly threw the ball over Shaun Prater's head for a 25-yard gain, and the other was a fourth-quarter throw into triple coverage that was threaded between Tanner Miller and Anthony Hitchens.

On the short routes, though, Iowa got pretty much carved up: ULM went 11-13 for 72 yards in the short zones and 13-18 for 129 yards in the medium zones. These weren't complex routes, mostly short hitches, outs and crossing routes. The most common route ULM used was a quick hitch to the outsider receiver, taking advantage of the soft zone coverage by the cornerbacks. But ULM also had success sending receivers in motion across the interior of the defense and bringing running backs out of the backfield to work against linebackers.

Who was targeted?

For every passing play I also looked at which defender was either in man coverage on the receiver or in the zone of the targeted receiver:

Targets_medium

Micah Hyde was targeted the most, but also did the best at limiting the oppositions' completion percentage. His numbers are a little misleading, though, as four of those targets were throw-aways into the stands due to good overall team coverage. The linebackers were picked on pretty heavily, James Morris most of all. Just to be clear, I don't mean to criticize Morris too much here. He plays the very demanding middle linebacker spot and is asked to cover a very large zone, and the "targets" here merely mean that he was the nearest defender to the receiver on these plays, not that he was necessarily locked up in man coverage and got beat. The outside linebackers (Tyler Nielsen and Christian Kirksey) benefit somewhat from corner help on the outside, but Morris was frequently left as the only man in a very wide middle zone against speedy wide receivers. His zone was targeted a lot, usually on short slants, crossing routes or passes to running backs coming out of the backfield. In other words, ULM used the spread to do exactly what it's designed to do: they spread out Iowa's defenders and forced slower players (linebackers) into bad match-ups with faster players (receivers and backs). 

I was surprised to see how successful ULM was targeting Prater. It's hard to tell how much of that success was due to Prater's play and how much to design (I would wager on design), but the deep cushion he gave receivers allowed a number of hitches and short out patterns to gain 5-10 yards.

The Weird Stuff

There were some interesting wrinkles that Norm Parker threw in to counter the spread formations ULM ran. On third and long, the defense frequently went to a three-linemen, four linebacker set with Hitchens in as the fourth linebacker, and that seemed to work very well (the only completion against this defense came on that thread-the-needle-into-triple-coverage throw mentioned earlier). The Warhawk quarterbacks had trouble finding anyone downfield against this defense, while the three linemen got surprisingly good push.  

The really fascinating defense was the six defensive back look Iowa employed a few times in the red zone. Castillo and Sleeper came in as the extra DBs, and Morris was left as the sole linebacker. On consecutive plays from this formation in the second quarter, Iowa used a cover 1 version of this defense (i.e. Miller was the only deep safety and the other five DBs were locked on receivers in man coverage) and had Morris blitz from the outside. Both times it worked like gangbusters and forced a hurried throw into coverage (Iowa also used a cover 2 version of the same personnel group, which wasn't quite as effective). It will be interesting to see if Iowa goes back to these extra LB/DB looks in the future. In any event, it seems clear that Parker and company are working on ways to combat the spread.

Speaking of combating the spread, a quick note on the topic of quarterback contain. It's become commonplace in the past few weeks to blame Iowa's problems with scrambling quarterbacks on the failure of the defensive ends to "contain" the quarterback in the backfield, but this strikes me as a little unfair to the linemen. Think about a simple play where the offense sends four or five men downfield 10-20 yards deep against the cover 2 or cover 4 defense. In order to cover the receivers in space, the linebackers drop back into deeper-than-usual zones, leaving a 10-yard void between the defensive line and the next wave of defenders. If the zone works as designed, no receiver will be open, but that won't end the play -- the quarterback will still have the opportunity to run into the enormous emptiness in the middle of our defense. So for the D-line, It's not so much a matter of "outside contain" as "total contain." Even if the defensive line does a perfect job keeping the quarterback from reaching the edge, all it takes is one gap anywhere on the interior of the line and Dan Persa/Terrelle Pryor/Denard Robinson/[insert name of fast quarterback here] is off to the races.

And because opposing teams know that our zone defense has this property, they can and do design plays to exploit it: To take a particularly painful recent example, look back to the crucial 4th and 10 Ohio State had against Iowa in the fourth quarter of their game last year (skip to 2:51 in the video -- also, you'll probably want to mute this, as it has a loud, obnoxious soundtrack):

Looking at this play schematically, it seems clear that Ohio State purposely cleared out our zone with three deep routes by the wide receivers in order to generate room for Pryor to run:

Step1_medium

Once the zone was stretched deep down the field, it was up to Pryor to escape contain and run into  space created in the middle of the zone:

Step2_medium

You can say this is just a great player making a great play and that Iowa's linemen failed to contain him, but consider what was being demanded of Adrian Clayborn et al. here: the linemen were asked to a) get up-field to rush the passer and b) stop a very speedy quarterback from escaping in basically any direction, all with no linebacker help. It's a very tall task, and we shouldn't be too surprised when things break down.  It's not impossible for the linebackers to drop back in zone and then come back and make plays on the quarterback (on this play in particular, they probably allowed themselves to get too deep), but it is a very difficult play, especially when the offense sends receivers long.

Which is one reason the formations that employ extra linebackers/defensive backs are so promising against spread looks and running quarterbacks: they allow the defense to maintain its zone look downfield while leaving an extra speedy defender (either a linebacker or defensive back) to spy on the quarterback. I don't know if Iowa actually did that kind spying against ULM (TV broadcasts don't really show the entire defensive backfield on most plays), but that capability is there. 

Notes on strategy

Thankfully there was not too much need of strategery in this game, but Kirk Ferentz did make two smart fourth down calls on the first two drives, converting a fourth and inches from the ULM one yard-line for a touchdown on the first drive and another fourth and inches from the Iowa 49 to keep the second drive alive. When you combine those plays with the no-huddle that KF employed to start the game, I was starting to suspect some kind of Invasion of the Body Snatchers-type phenomenon had taken place in the Iowa locker room prior to the game. If you see giant green pods lying around Kinnick, please report them to the authorities. Or don't! All of those developments were very welcome.

X
Log In Sign Up

forgot?
Log In Sign Up

Please choose a new SB Nation username and password

As part of the new SB Nation launch, prior users will need to choose a permanent username, along with a new password.

Your username will be used to login to SB Nation going forward.

I already have a Vox Media account!

Verify Vox Media account

Please login to your Vox Media account. This account will be linked to your previously existing Eater account.

Please choose a new SB Nation username and password

As part of the new SB Nation launch, prior MT authors will need to choose a new username and password.

Your username will be used to login to SB Nation going forward.

Forgot password?

We'll email you a reset link.

If you signed up using a 3rd party account like Facebook or Twitter, please login with it instead.

Forgot password?

Try another email?

Almost done,

By becoming a registered user, you are also agreeing to our Terms and confirming that you have read our Privacy Policy.

Join Black Heart Gold Pants

You must be a member of Black Heart Gold Pants to participate.

We have our own Community Guidelines at Black Heart Gold Pants. You should read them.

Join Black Heart Gold Pants

You must be a member of Black Heart Gold Pants to participate.

We have our own Community Guidelines at Black Heart Gold Pants. You should read them.

Spinner.vc97ec6e

Authenticating

Great!

Choose an available username to complete sign up.

In order to provide our users with a better overall experience, we ask for more information from Facebook when using it to login so that we can learn more about our audience and provide you with the best possible experience. We do not store specific user data and the sharing of it is not required to login with Facebook.

tracking_pixel_9347_tracker