This article stole the sponsor:  Starting this season, Neck Sharpies is sponsored by HomeSure Lending and Matt Demorest. Mortgage rates are LOW right now, so if you've been waiting to see if football is actually played as a sign that a refinance can save you a bunch of money, now's a good time to strike. It usually takes Matt and his people about 5 minutes to know if a refi makes sense, and if it doesn't they'll tell you. You know the guy, and you know from a bunch of other readers that he does a better job and charges less than the mills that mistreat Beileins.

-----------------------------------------

If I were tell you an offense is running the same handful of plays and throwing to the same four spots most every standard down, how would you defend it? Why not cover those four spots and react to those same plays, training every player in exactly what he has to do to defend them?

Well, Minnesota has ways to stretch your needs. They have good receivers—gotta leave an extra linebacker to keep leverage on them. They have big offensive linemen and a great RB—gotta activate your linebackers before your DL crumble. And they stretch your linebackers between them. They also change up their formations, personnel, and shifts to turn your linebackers into linemen and your safeties into linebackers. And they'll get to the line quickly after a play so they have a good 35 seconds to survey your defense, check into what would work against it, check again if they don't like your reaction, and then run another RPO.

That's a recipe for last year's #7 offense to SP+, one which put up 7.67, 6.25, 6.82, 5.55, and 6.59 yards per play against Penn State, Iowa, Northwestern, Wisconsin, and Auburn to close out 2019. Yet starting mostly the same lineup—though absent a few important lineman and a graduated NFL receiver—Michigan held the Gophers to their least efficient passing outing since 2017. Yes, the year they started Demry Croft and let Khaleke Hudson roam free in the backfield.

That 2017 Minnesota offense got 4.88 yards per play. Demry Croft's career yards per play was also 4.88. Minnesota's offense on Saturday: 4.53. For every snap they took, Minnesota was half a yard per play less efficient than Demry Croft.

It wasn't for lack of trying. I want you to watch this sequence. It's not a play, but all the pre-snap stuff as Brown and Fleck played the game within the game.

Let's unpack it.

[After THE JUMP: How to be in two places at once]

Last week I wrote about how PJ Fleck's RPO offense likes to work. To summarize:

1. Get the key unblocked linebacker trapped away from the run game using a run-pass option.
2. Provide easy reads on the other side of the field to the same four spots the QB is comfortable throwing to.

A fifth read is the QB runs into the alley

They can run all kinds of route combinations to get the same reads and timing and formations. They can also turn one of these routes off and remove another defender by loading up the other side with bulky dudes. Regardless, one of them should always be open if he keeps it because that means there are only three defenders, or at best a fourth defender being carried from the backside, since the extra LB has gone to defend the run. But if you do the same thing all the time it grows stale. Defenses will start blitzing into running lanes, or have unseen defenders pop up in the four throw spots for easy picks.

So much of Minnesota's RPO offense takes place before the snap. They'll see how the defense is aligned to decide if they've got a defender they can pick on, and often audible if they sense there's something fishy going on. Audibles are where you often see the counters to RPO: a quick toss outside to screw over a slant, or a true play-action to mess with linebackers and safeties hammering down on those crossers.

Michigan's gameplan was to leave the linebacker back on the passing side and trust the rest of their front to deal with the consequences. Minnesota has seen that gameplan before, and has ways to punish it. If you're hanging back on the pass not only will they run, they're going to create matchups with your worst run defenders against their best run blockers, and go right after them.

That brings us to our first slide.

Minnesota here is loading up their run personnel on the left side kind of like a single-wing offense. Why go heavy like that? Because it changes everybody's jobs. Michigan's already been putting their safeties on the edges and their cornerbacks deep all game when there's an "open" (no receiver) side. Two tight ends however turns Michael Barrett into Aidan Hutchinson, Brad Hawkins into a viper, and Gemon Green into your last line of defense.

But all's not right in Gopherland. They may have some matchups, but do they have numbers? To this point Michigan had been rolling a high safety down into the box on these formations, and when Minnesota didn't check out of it they met a safety at two yards.

That is an eight-man box. It is not safe to run into.

So back to our main play. When they notice a safety—Makari Paige—lining up as a hybrid linebacker with one DB deep, it says Michigan wants to play eight in the box again.

If Paige is out there that means Paige is likely to be sitting on the intermediate (#1) route while Josh Ross—who's been converted into the Mike by the alignment—can activate aggressively against anything that resembles a run. This 4-4 look has one too many guys in it for Minnesota's standard RPO. They check to the sideline and get the call, which I'm going to guess is some kind of deep route that puts Paige on Rashod Bateman in a huge amount of space.

I can't be 100% positive but I'm taking clues that the play they get is a variation of this:

Makes sense right? All that space for a good receiver against a new defensive back.

They could also have called a run flip, running a high-low RPO on Makari Paige since they guess he's the important gap's defender on the backside. They did the same to good effect against Daxton Hill earlier, as he banjoed between getting depth on a pass drop and covering the gap they run into. In that case they got Dax for being too anxious about the pass.

watch #30 at the bottom.

Either way Minnesota can waste Michigan's extra man in the box. But then Michigan changes their look:

QB Tanner Morgan is looking right at this, since Paige is the guy they're planning to attack, and Paige is backing out with Ross sliding over. In other words, Michigan is going back to a base two-high look. Whatever was meant to attack Paige now has Ross under it. That also gives Michigan enough defenders to cover the areas Morgan likes to throw to.

On the other hand it pulls 2.5 defenders significantly further from the spot where Minnesota likes to run:

Morgan checks the clock to see if he's got time for an audible. He does.

They get another play in from the sideline:

And share it with the committee:

And Michigan doesn't move. By the time Minnesota snaps it, Michigan still hasn't changed their alignment. The center calls McGrone the "Mike", which means Ross (linebacker on the bottom) is the guy being read again now. The ball is snapped with a second left.

Let's pause here and see if you can guess what play they checked into. I guessed they'd checked into some variant of their standard RPO that runs left or passes to the four spots on the right.

It may have been something like that, but Morgan doesn't wait to see if Ross activates against the run very long—just a quick glance out of the corner of his eye—before making the pitch read. He's sure they've caught Michigan in another passive response to their RPO game, and this play is all about making that read pre-snap and getting the ball outside where the offense has numbers. Do they?

Yes and no. Minnesota got what they wanted, as Michigan's DL are all reached. The key blocks are tight ends versus Michigan's safety and safety-like outside linebacker, and with Ross sitting out on the backside they've got a hat for a hat/an extra gap. However things don't work out for them. Why?

This is also what Michigan wanted. Having the backside DL take backside gaps—essentially a slant without having to slant much—they've realigned themselves so that the DL are again taking DL jobs—Kemp has the backside A gap so McGrone can travel over to that big playside C gap. Hutchinson murdered that frontside B gap, where he hung up both the tackle (#70) who had to seal him and the guard (#65), who appeared in a lot of Michigan defensive highlights. That keeps McGrone free to make the tackle.

Technically they do have Ross caught between things: the sit route by Bateman and that frontside A gap that technically exists between the center tripping Kemp (cheater!) and the left guard who's going to need therapy before he sees another winged helmet.

That gap is of minimal concern because of where Hutchinson is and the direction of the play that was trying to get outside. Had Michigan not been slanting, the linebackers would both have backside gaps, and those gaps would be widening as the mass of people sauntered left together.

With the slant however the edge of the defense is set by Hawkins and then an upfield move by Barrett, barely to the hash. Ross is staying put on the hash so he can break on Bateman or the QB in case of a throwback, and McGrone doesn't have as far to go before he turns vertical.

Also since that side has no receivers, Gemon Green gets to come help clean up. No gain, 3rd and long.

Here's the whole sequence:

LESSONS:

This was just one play-calling battle and both teams had more to show. In fact Michigan prepared a really cool blitz for this game specifically because it messes with Tanner Morgan's reads, and it did. See if you can go through his progressions for him here:

1. Cam McGrone crosses Josh Ross, so "re-Mike"—flip the guy you're reading to Ross.
2. Ross has stepped too far with the run to be in the way of the TE's route to spot #1.
3. Crap, it's already too late because I was watching those linebackers. That's okay, TE will go to spot #2.
4. Hey, why isn't that safety (ahem, cornerback) attacking the run?

Oh right, he doesn't need to because Michigan's got this.

Josh Ross also finally learned how to drop into those spots. Morgan's effectiveness over the years has been tied to his system because it rarely betrays him and he knows it well enough. But if you speed him up enough he's like every other quarterback; he won't think before doing the next thing he was taught to do.

Let's read with Tanner again through this RPO/bluff arc.

1. Josh Ross is my read man. He's stepping with the direction of the run: keep.
2. Paige is going with the flat route so #1 should come open if the safety isn't hammering down on him.
3. No safety, I'm clear, which is good because this Feldman Freak is about to kill me again.
4. LOL JOSH ROSS!

That's just a Cover 2 look converting to Cover 3 but Tanner Morgan isn't thinking in terms of coverage because his system is about finding your guy at spot 1, 2, 3, 4. He's not expecting Ross to reverse course from past the middle of the field, get depth, and get underneath. But the back-foot throw gives him time, and he knows exactly which spot to go, and presto! Game over!

Minnesota gained their yards mostly on two long throws and a few good runs when Michigan lost the edge for one reason or another. The RPO game also got a few slants when they caught Ross out in the Viper role (like he is here). The other thing that worked early was play-action, because that opens up a more diverse passing game. Minnesota wasn't completely shut down. And there was a drive when the running game rolled over some 3-3-5.

However Michigan's strategy against RPOs largely worked because they had the players to do it. Minnesota wanted to do more than hand it off and hope their OL could get some victories against Michigan's DTs, or their tight ends could push around Barrett and the safeties. But Michigan was content to let them move their pieces around where they may, counter with Aidan Hutchinson or Kwity Paye popping up in B gaps, and trust their linebackers to cover a lot of ground.

Even with 40 seconds to change their call twice, nothing could be done about the fact that Michigan's front was better than Minnesota's, and Michigan knew it, so they didn't have to commit any more material than necessary to stopping it.