Picture Pages: Zone Minus Read

Submitted by Brian on November 9th, 2011 at 10:53 AM

This week in spread zealotry we've got an example of something you can't do against the spread without getting a face full of Denard: crash. It's second and three early in the third quarter. Fitzgerald Toussaint has just taken an iso seven yards thanks to Molk and Hopkins making quick work of the NT and MLB.

Michigan will run an inside zone from an ace set. I'm pretty sure that Michigan screwed this up because I've never seen an inside zone play on which a guy who is not the end man on the line scrimmage is let go unless he's getting read. Here the backside DE is let go while Watson flares out to block a guy with a longer path to the ballcarrier.

The setup:


Get used to both halves of this. Iowa is in a 4-3 under. The key guy is the DE at the top of the screen—the guy in a two point stance next to the standing SLB.

Michigan runs an inside zone. They double the NT and the SDE (at the bottom of the screen) as the linebackers flow to the LOS.


By the handoff point a couple things have happened. Both linebackers are at the LOS and engaged; the MLB is actually doubled by Lewan and Schofield. Sometimes a doubled LB means you've blocked a play so well that there's no one else to get. Not so much here.


I don't want to make too much of this because this is clearly a bust by the line (in all likelihood Lewan), but when I saw this I immediately wished Michigan was in the shotgun and Denard was reading the guy they let go. He'd have two choices: remain responsible on Denard and open that cutback up (he'd likely recover in time to tackle but not at the LOS) or do what he's doing now and put Denard one on one with the safety.

Similarly, with the linebackers one and two yards off the LOS, a pass like the one that started off their second hurry-up drive would be open. These things are all possible if you're reading the guy you've let go.


When you're not he just tackles you.


Toussaint does make the guy miss, but only by redirecting into a pile of bodies. He gets a yard. On the next play Michigan runs a QB power with Denard that Iowa is all over until two guys fall down after beating their blockers to the spot. Twenty two yards later they've got a first down. A field goal results.


Items of Interest

This seems strictly less effective than the same thing run from the gun. I'm not sure what the advantage of operating from under center on this can be. You hear a lot about getting downhill quickly as an advantage of playing from under center, but pistol sets and even Michigan's old belly setup where the QB is a yard in front of the tailback get guys going downhill pretty damn fast without giving up the mesh read.

The other advantage suggested by commenters when I tried to answer some guy's question about the advantages of the I-Form over the spread was an ability to keep your eyes on the coverage downfield instead of catching a shotgun snap*. Here Robinson turns his back to the defense and has no idea what's going on behind him until he turns around.

So… yeah. Living in a world without post-snap reads is giving up something when your quarterback is mobile.

This is an example of the "ten man football" Borges was talking about. Even so, the play should still work for a few yards. The blocking's decidedly mediocre—in the last few frames you see a DT chucking Omameh, forcing the cutback—but the nice thing about the zone is it's hard for the defense to be right when you've got an effective cutback runner. Toussaint is that.

If the backside end actually gets blocked, Toussaint looks like he has the cutback for decent yardage. While that safety is probably going to come down and hold it to a modest gain, the first down is well within reach. Lewan busts and Michigan gets zilch. That was a theme on the day: one guy doing something wrong on these run plays and Michigan getting stuffed.

I wonder if spread stuff has a greater failure tolerance. You'd think it would because you are optioning off a potential defender and therefore get a double on someone. The alternative is forcing a safety into the box, which isn't bad.

*[Something that didn't seem particularly convincing since the shotgun is the preferred passing formation for long-yardage situations and hurry-up even in the NFL.]



November 9th, 2011 at 11:11 AM ^

I know many people don't like Jay Cutler, but he's been sacked more than any QB in the NFL recently, and when asked if the shotgun would help with pressure he said it wouldnt because he couldn't see the defense shift the same way.  /just sayin



November 9th, 2011 at 12:31 PM ^

because a) he's already on shaky ground due to a number of things, some of which are crap, and basically saying that his OC doesn't know how to run an offense is not going to help his case, and b) the Bears' offense has been Martzed to the point where there isn't a lot they can do to help with pressure in the first place. (Their best TE is playing in Carolina, not that Martz believes in the position in the first place, and he's still using personnel groups that make it harder to shift protection.)

Contrast that approach with the Lions: a team with a line that might or might not be adequate is running more plays from the gun than any other NFL team (69.3% through 8 games, which would top their 2010 percentage of 64.5%) and also has a relatively low Adjusted Sack Rate. Part of that is Stafford learning to get rid of the ball more quickly, but a good bit of it is giving him time to find someone in the first place.


November 9th, 2011 at 12:36 PM ^

Isnt this assuming Denard makes the right read?  How often did that happen last year?  Seems he's been a little better at it this year in limited opportunities, but there's a reason RR went to mostly called runs at the end of last year.


November 9th, 2011 at 11:44 AM ^

the first play of the game where Borges calls a Zone Read from the gun.  The backside stays honest and forces Denard into a race to the sidelines where he gets help from other defenders. (this may not be exactly how it happened, just how I remember) 

The play above is obviously a blown assignment by Lewan.  I just don't get why assuming the zone read would have worked here makes more sense.  The zone read is not very effective very discipline defenses. The backside DE has to stay home, thats it. That forces the QB to handoff everytime or take a chance on making him miss or outrunning him. So at the point it basically becomes a zone give, so why not run it from under center. 

Questioning whether zone is effective or not from under center is kind of silly.  Look at just about every NFL offense and many of the better College offenses like Alabama. They seem to be pretty effective running from under center.


November 9th, 2011 at 12:37 PM ^

Straight from the 2011 FO Almanac:

Shotgun formations are generally more efficient than formations with the quarterback under center.

Over the past three seasons, offenses have averaged 5.9 yards per play from Shotgun, but just 5.2 yards per play with the quarterback under center. This wide split exists even if you analyze the data to try to weed out biases like teams using Shotgun more often on third-and-long, or against prevent defenses in the fourth quarter. Shotgun offense is more efficient if you only look at the first half, on every down, and even if you only look at running back carries rather than passes and scrambles.

That does not mean that every team is more efficient running from the gun, and a contributing factor may be that not enough teams run enough plays from the gun to get DCs to focus on stopping those plays, but right now, it's not silly at all to question running plays from under center. In fact, it's fair to suggest that what's silly is not running more plays from the gun.


November 9th, 2011 at 12:53 PM ^

A couple things-I'd like to see data on fumbles/interceptions lost.  Also, just weeding out third and longs isnt enough - its possible that even on first down, teams run longer passing routes out of shotgun.  That will result in more average yards, but not necessarily more "successful" plays.  Not to say the overall conclusion is wrong, just noting its just hard to boil football down to any one statistic other than wins and losses.


November 9th, 2011 at 12:37 PM ^

If the backside DE stays home every time, then there is 5 on 5 blocking for the Oline.  Thats the whole point of running the play.  It gives you a numbers advantage regardless of how the defense defends it.  You want to bring a safety into the box?  I'll hit a seam behind him.  You want to run the scrape exchange?  I'll run the zone read with a bubble or a triple option.  When an offense like that is executing at a high level, it can make the defense wrong on almost every play.


November 9th, 2011 at 12:54 PM ^

Any offense executing at a high level is going to make things tough on a defense.

From the double tight look like this one you can use the backside TE to block the DE or you can just let him go and have the TE pickup the 8th man in the box. If the backside DE crashes down and makes the play (which unless your Julius Peppers you will only make it 1 of 10 times) the next time you hit him with a bootleg to keep him honest.

The zone read play has run its course.  Your not seeing the same impact Pat White had 5-6 years ago because defenses have figured out how to stop it.

You can list all the options you have off that look, but they still rely on someone on the defense not doing their job. Which Iowa, MSU, and Ohio are going to do a lot less than UMASS. I'd much rather depend on being technically sound and tough that quick and deceptive.


November 9th, 2011 at 2:11 PM ^

I've got no problem with pro-syle offenses.  To expand on what you said, with good talent and good execution a pro-style offense is hard to stop.  Michigan should win 9+ games a year now that they have competent defensive coaches.

However, saying that spread offenses have been figured out is silly.  In my opinion LSU is the best defense this year (Alabama is 2nd -- slightly better numbers but worse schedule).  The only teams to score more than 11 points or gain more than 300 yards on LSU were Oregon (zone-read based spread and shred) and WVU (Holgorsen's run-heavy airraid spread). 

Alabama has great offensive linemen and a great running back but their pro-style attack only scored 6 points on 295 yards compared to 27 points on 335 yards for Oregon and 21 points on 533 yards for West Virginia.  In LSU's other 6 games they gave up an average of only 7 points and 190 yards.

Does this mean the pro-style offense has been figured out? No.  It means that LSU has a great Defense which can stop any offense.  Those points are season lows for both WVU and Oregon.  The yardage is probably a season low for Oregon.

The beauty of college football is that any offensive system can work.  It doesn't matter if it's pro-style, the spread and shred, the airraid, the flexbone or June Jone's run 'n shoot modifications.  The key is that all these offenses are built together so there are answers to what defenses do to stop you.


November 9th, 2011 at 2:53 PM ^

First of all I said the ZONE READ has been figured out not the SPREAD OFFENSE in all its forms…

All the yards Oregon and WVU put up on LSU are meaningless when they LOST and they are out of the game by the 4thqtr.  You fail to mention the Oregon turned the ball of 4 times to LSU’s 1. And Oregon lost the time of possession battle.  WVU turned the ball over 4 times to LSU’s 0.

Bama didn’t turn the ball over more than LSU and matched them in time of possession.

Regardless I agree with you when you say any offense works.  Any offenses works when you believe in it.  I just think when your Michigan and you can get the biggest, fastest and toughest guys around why try to fool people?  DeBord’s offense might have been bland and predictable but Michigan won a whole lot of games and sent many players to the NFL.  Rich Rod’s offense was exciting but didn’t win many games and didn’t send many players to the next level. 


November 9th, 2011 at 4:43 PM ^

Using your idea, the fact that Alabama didn't turn the ball over more than LSU and matched them in time of possession is meaningless when they LOST .  Their biggest, fastest, toughest guys in the WHOLE COUNTRY (their past 4 recruiting classes have been ranked 1,1,5,1) weren't able to score a single touchdown.

Time of possession is a defensive stat.  If you force a 3 and out everytime the other team has the ball you will win TOP no matter what style your offense is.

Turnovers are a valid comparison though.  Alabama still turned the ball over twice, which admittedly is less than 4.  Maybe as defenses get better at defending the zone read and spread offenses in general they have gotten better at forcing turnovers.  When Rodriguez was still at Michigan the hope was that the turnover problem was due to playing freshmen quarterbacks since WVU was always positive in TO margin while he was the coach there.  That sounded like a reasonable explanation but it is possible that it is a newer problem with spread offenses as defenses adapt.  I don't have any real knowledge either way on that question.

Are you honestly telling me that a combined team of Oregon's offense and Alabama's defense  would have done worse against LSU than Alabama did?  I think that the Bama Ducks would have won by a comfortable margin (although probably not a blowout).

I'm a believer that a great defense makes for a good team on it's own.  A great offense needs at least an average defense to be a great team (compare Michigan and Auburn last year).  Maybe Hoke is right that you can't be a great defense when you are playing a spread offense.  I don't really believe it but Rodriguez unfortunately didn't give us any counterexamples.


Is Michigan allowed to use play-action passes?  Isn't that "fooling people"?  I think a team should build a great defense and use any advantage they can get on offense.  If that means fooling people, why wouldn't you want to fool people?


November 9th, 2011 at 6:35 PM ^

My point was that, yes Oregon and WVU put up yards on LSU (your point) BUT turned the ball over, which is stereotypical of a spread/option team, and lost the game.  Yes Bama lost as well, but they lost by 3 points in OT, not down by 20+ going into the 4th qtr.

Time of possession is only a defensive stat? Really?  All the talk about offensives changing tempo's has no effect of time of possession?  Michigan's defense isnt better this year because the are on the field less? That doesn't make much sense. Yes the defense stopping people affects time of possession but the type of plays (i.e.pass vs. run) and tempo the offense uses also affects time of possession.

I am with you on the spread offense helping or hurting its D as far as practice mainly because of the Urban Meyer Florida teams.  Now those Florida's offenses were about as physical as you can get in the spread too.  Maybe its a good thing we have Mattison as he went throught that at UF. 

"A great offense needs at least an average defense to be a great team (compare Michigan and Auburn)."

Come on man... You're going to say 2010 Michigan offense was great? 7 points vs. OSU, 14 points vs. Mississippi State.  In what book is that a great offense?

Okay I wasn't referring to play action passing or counters (if thats your next question) when I said fooling people.  I meant deception based offenses, such as Rich Rod's.  You have to rely on fooling a defender to be successful from play to play.  I don't have problem using a bit of deception while still being a fundamentally sound football team, such as Boise State. Not only does your success rely on basically playing bad defenses you also spend a large percentage of your practice time perfecting reads, meshes, pitches, and pitch relationships. This forces other fundamentals to be neglected. As I said if I am bigger, faster, and stronger why do I want to spend my time perfecting those things only to rely on playing a bad defense as well.  If I am bigger, faster, stronger I am going to line up and knock you off the ball even if you know that's what I'm going to do.  Teams like WVU, Northwestern, and Oregon needed deception offenses to compete with the power programs.  The power programs do not.


November 9th, 2011 at 8:51 PM ^

I don't really know if this is worth much of an argument.  You don't have to convince me that pro-style offenses are good.  They certainly can be.  I'm not going to convince you that a spread-n-shred offense would work at an elite level.  I'll focus on three points that I thought were most interesting.

The first point is that you are always going to play a team that is just as big, fast and strong as you are.  Michigan will see that every year in OSU and one or two of Wisconsin, Nebraska and Penn State (not to mention the possibility of Michigan State and/or Iowa depending on the year).  Alabama saw that in LSU last weekend and their offense didn't score anywhere near their average of 40 points.  Congratulations to them on being in it in the fourth quarter.  It's still a loss in which their offense was shut down.

Secondly, yes I think last year's offense was great.  It wasn't the best in the country and it had room to improve this year, but I would say it was in the top 15.  That Mississippi State game was terrible for sure, no excuses, but that same defense also held Auburn to a season-low of only 17 points.

Finally, I consider TOP a defensive stat, if it has any meaning.  The offense definitely can affect TOP but I primarily view it in terms of whether the defense is getting themselves off the field quickly.  The tempo of the offense definitely affects the defense but the primary reason why I think a slower tempo is making our defense better this year is that the other team has fewer chances to score.  Last year the opponents averaged 12 possessions per game while this year they're only averaging 8 possessions per game.  That right there has probably improved the defensive rankings by about 20-30 spots.

Before I get accused of maligning Mattison or the amazing job he's doing let me say that I think he's great.  This years defense is holding opponents to 1.96 points per possession compared to last year's 2.89 points per possesion.  That is a big improvement made to look even bigger on the scoreboard because they're facing 4 fewer possessions a game.


November 9th, 2011 at 3:40 PM ^

I think part of the issue with UM running the zone read is either A) they don't really have a dynamic tailback to complement Denard, and /or 2) the oline doesn't block that great on the play. 

The second half of last year teams coached the defensive end to stay home on the zone read, and Denard was handing the ball off a huge percentage of the time.  Rarely did it seem like the play was very successful when the running back got the ball.


November 9th, 2011 at 1:18 PM ^

If the backside end stayed honest then Denard should not have kept the ball if it was in fact a Zone Read.  If the backside end doesn't crash down then the tailback should get past the LOS untouched and everyone in front should be blocked into the secondary.

I don't recall the play you're talking about from memory but it was either a bad read by Denard or not actually a read, just a fake handoff that Denard was supposed to keep and run outside.


Except for in the case of talent mismatches, no offense is very effective against a very disciplined defense.  If the talent and execution is even, the offense should be able to get small yards on every play.  But they have to play mistake-free or they'll kill their own drive.  If the defense gets aggressive they can stop the offense but there is always a vulnerability that the offense could find a way to exploit.

The zone-read hands it off to the tailback for 5 yards until the defensive end finally crashes down on the tailback and the qb is on the edge running free.  Pro-style offenses can use the run to get the safeties in the box and then throw it over top.  Or they can use the passing game to clear out the middle of the field and then hand it off for 10-20 yards.

I think that's what people are frustrated with.  What is Michigan's base offense?  What are their constraints to keep the defense from cheating to stop their base offense?

Blue in Seattle

November 9th, 2011 at 2:27 PM ^

I think you did a great job of describing how each scheme has ways of setting up the defense to start cheating, and then having an option to exploit the cheating.  On this specific play I think the next option to punish the defense is that Denard keeps it and continues his motion in a roll out, at which point Roundtree would break off from his blocking, and the defender either needs to stay in coverage or come up to tackle the incredibly dangerous running threat of Denard's legs.  But what happens, is the offense blows an assignment when Borges is trying to establish if they are cheatingt, and since the base play isn't successful, there is less chance the option off of this base play will have the huge exploitation expected.  But if this play had gone for 5+ yards a couple times, then the next time Iowa sees it there is a chance the Safety or the ends break discipline and then Denard keeping will be a huge success.

Now, on this specific play, assuming that it was in shotgun and Denard gets to make a choice, I think it's clear that the WDE would have come in under control, like he did in this play, and then he would have help contain on Denard, forcing the handoff to Fitz, who would have had no where else to go but up into the hole that the rest of the offensive line was not really creating.  If you watch the video, the unblocked WDE only gets a hand on Fitz, and yes it slows him down a little, but there is no where to go, the hole never really forms.  BUT if Lewan had blocked the WDE and NOT doubled, then the cutback would have been there for Fitz.

The bottom line though, is not that this is a RPS fail, but that it is an execution fail.  And maybe Lewan screwed up because they are transitioning away from the Read Option Zone to the Zone No Option, but I don't see where the extra Lewan body enabled the hole to open up for the RB.  And I think this was the problem for all of last year against quality defenses.  I did a quick tally of the Wisconsin game plays from the UFR and on well over 75% of the plays Brian had designated as Read Option, Denard handed off.  In all of those Brian added a comment that it was the correct Read by Denard.  But the RB on average gained less than Denard did on the QB ISO's against Wisconsin.

The bottom line is that Denard is the best RB we have.  Unfortunately when you are tackled like a RB your arm gets pounded on over and over, which means throwing the ball turns into shotputting the ball pretty quickly.

Now, don't get me wrong.  I think last year's offense against Wisonsin did move the ball and it did run a lot.  And if we had this year's defense with last year's offense then the result may have been a close win of many points against many points.  But we know that we just can't mix last year's offense with this year's defense.  We have a different OC, and what he knows really well is how to develope passing QB's.  He loves passing QB's that run, but he's still the best at calling plays and coaching players to be Passing QB's that may or may not run good.  And while the team is learning the new schemes, they are probably going to continue to have failures like they did in this specific play.  But that's not on the coaches, it's on the players to execute the play called.

Overall I don't like the run heavy Rodriguez system over the more pass oriented offenses, like they run in the Pros.  And it's because over the long haul you aren't going to get the best talent at QB and RB if you aren't preparing those guys to play on Sunday eventually.  And I think Denard wants to play on Sundays, and his dream is doing it as QB.  I don't think he would have gotten that from the Rich Rodriguez system.  He might not get it from two years in the Borges system, but at least the framework is there, and the rest is on him to do what it takes to execute.



November 9th, 2011 at 12:45 PM ^

The existence of other countermeasures does not negate the efficacy of the one in question.  The fact is that during our last road game, as the game went on, we became less adept at using many of those other countermeasures.  I think it would have been reasonable to decide to run some amount less shotgun to deal with that.

Anyway, I didnt say that was THE reason we ran this play from under center, I said it was one advantage, which is certainly true.  It may not have factored in to the decision at all for all we know.  I'm merely commenting on the disconnect for a guy who complained that our coaches didnt do ONE THING to stop teams jumping the snap count against our previous two opponents, and that is UNACCEPTABLE, now being unable to see ONE REASON for running this play under center.  And thats not directed at you


November 9th, 2011 at 1:05 PM ^

Am I completely wrong, but I would expect the backside TE to cover/help with the DE in that instance off a zone run to the right. Lewan briefly chips the DE and can then get upfield to stop backside pursuit and open up a cut back? I mean yea Lewan has got to get into DE because he's lined up in front of him but shouldn't the TE have helped so he could bail.