FORMATION NOTES: UConn did some weird stuff. My lingo on these is probably bad but this was "5-1 nickel split" with a 3-4 front that has two OLBs flanking the line:

And I just gave up when this happened, calling it "5-4 30 front":

There was also a 5-3 30 front that had a deep safety.

This is "shotgun 4-wide tight" for M. You may note the weird tilt of Funchess:

As a rule I count a TE in a two point stance as a WR for purposes of naming a formation.

SUBSTITUTION NOTES: The usual basically everywhere. Save one snap for Derrick Green when Toussaint was momentarily injured, Toussaint got every tailback snap. Butt was preferred to Funchess late when Michigan was running the ball. And it seems like Chesson is slowly absorbing snaps from Reynolds and Jackson.

All else was as before.

[After THE JUMP: points! yards! (none of those things)]

FIRE BORGES

I feel that doing this is neither realistic nor helpful, especially midseason.

FIRE YOU

Awwww, you don't really mean that.

FIRE GARDNER

To properly evaluate this in context we need a—

a ch—

hello?

CHART INTERRUPTION IS ON STRIKE UNTIL WE CAN RUN FOR TWO YARDS ON FIRST DOWN

—chart—

DAMMIT

about these things:

Devin Gardner 2012

Devin Gardner 2013

Shane Morris

That was… okay? I guess? I'm surprised by that number, too. Let's sanity check it. Here are the incompletions on which I did not give Gardner an IN:

1. Gardner blown up on third and long, throws to Gallon, gets PI call.
2. Freeze play gone awry.
3. A slant that went wide was filed an MA as the DB gets a shove on Dileo and busts up the route somewhat.
4. Hitch dropped by Gallon.
5. Pressure up middle, inadvisable spin, low but catchable throw to Chesson.
6. Interception on decently thrown deep ball to Chesson.
7. Chesson misjudges well thrown deep ball.
8. Gardner chucks ball OOB as he gets instant pressure.

You could argue that some of those were bad plays, but I think that's about right. You don't want five balls that are wildly inaccurate in a game but Gardner did not throw into coverage all night and only had one pass on which he had an opportunity to find someone and didn't—a two yards scramble with no pressure on which he had Funchess about to streak open deep. He also missed Funchess open for a touchdown on his first short-yardage boot of the night, but with the corner wide open I'm betting he's told to run the dang ball if he knows he can get a first down.

Unfortunately, Gardner's pocket presence is still an issue. Here he's got a three-man rush on which a cavern is opening up to one side of the nose tackle and still does his spin, this time into trouble:

If he's making that throw to Chesson by stepping into it as he nears the LOS it likely arrives on time and whistling, giving Michigan a shot at a first down. He appears to be working on it, at least.

A few things were mitigated by repeat views. Inexplicable covered fade to Joe Reynolds was a freeze play on which Miller snapped the ball without an opponent in the neutral zone. I think this third and medium pass is less bad than it seemed since Dileo gets knocked off his route:

It's subtle, but Dileo is moved more parallel to the LOS than he wants to go and Gardner seems to throw it to where he's supposed to be instead of where he is. I chalk that up to good defense, mostly.

Gardner didn't get much help from either his line or his receivers in this one, and the happy feet don't really show up in the UFR chart except as that one TA. And it's debatable how happy those feet were anyway. His six scramble rushes averaged 9.5 yards an attempt and turned third and eleven from the 16 into a critical touchdown. Pretend those are completions and his line is… well… slightly less horrific. It goes from 13/25 for 111 yards to 19/31 for 168 yards. Woo?

A lot of slack was cut to Gardner for a reason you might be able to guess if you look at the "PR" section above.

With that number up the offensive line chart must be just…

Well.

That protection number is miserable, barely better than Michigan's performance against Notre Dame. Gardner did not cope as brilliantly as he did against the Irish.

I'm actually surprised by the Toussaint number, which also includes a –3 for his fumble. Leave that aside and that's a really strong day. Despite the 120 yards at 5 yards a pop in my head I still think he had a bleah day. That's what the charting's for, though: to make you reconsider your lying eyes.

The rest of it lines up with my expectations: Lewan and Glasgow were both good, as was Schofield; Kalis had a bit of a struggle, Miller did some good things but got destroyed on way too many plays, the tight ends are huge liabilities blocking, Gardner's legs saved Michigan's ass except when his arms were fumbling.

But Glasgow's a walk-on?

I'm about a month away from giving him knighthood in the Order of St. Kovacs, whereupon the fact that he is a walk-on becomes a quaint historical fact relevant only to Tom Rinaldi.

He's got to show it against Big Ten foes, but Glasgow is good. He is just good. Even when Michigan is not picking up yards on run plays he is firing off the ball and getting push, consistently.

A few times in this game he'd club some guy only to see another player mess something up such that Michigan could not take advantage of it. Maybe he won't perform as well against Big Ten teams—he struggled against Nix—but he is definitely Michigan's best interior lineman.

I won't say fire, but… like… there seems to be kind of a glaring hole in the above chart. Other than the tight ends.

Yes. You, like the rest of the internet, clamors for the insertion of Chris Bryant and benching of Jack Miller. It seems like you may get your wish in two weeks against Minnesota, as buzz around says Michigan is going to make that switch as long as Bryant stays healthy. Which he's never been, but whatever.

Miller did struggle. He was in large part responsible for three busted zone plays and gave up two pressures up the middle that resulted in lost plays for M, and while UConn's DL isn't terrible it's not Notre Dame where you can just about write a bad performance off. He also messed up that freeze play. He was run over twice for stretch TFLs.

This was just… are you seriously leaving this guy?

In what world is that guy not lined up over you? In what world does Kyle Kalis have the faintest prayer of blocking that guy without a scoop? This is game four and if you're still making enormous, obvious mistakes like that on the stretch while simultaneously being the smallest dude on the line, you're nearing your expiration date.

Meanwhile in pass protection:

"That can't happen" is the worst analysis, but man… that can't happen. (Craig Ross did make a great point on WTKA today, though: Why is Glasgow helping Lewan? Lewan is fine. Lewan is doing sudoku he is so bored. Help Miller.)

Miller did do some positive things. One of them was covered in the picture pages; here's a second-half-opening inside zone on which he's the primary block that gets Toussaint a nice gain:

He and Toussaint did pick up a twist blitz from UConn that gave Gardner a lane for Michigan's first touchdown, as well.

He's inconsistent and when he is on the bad side of that things tend to go really badly because of his size—290 is probably an exaggeration. He can't get away with any slip in technique at all. Really makes you appreciate David Molk. I know I just argued that benching Miller is not an easy choice, and I still think it's not a slam dunk because Chris Bryant may shatter at any moment… but once you add it up it's stark. Michigan has to try something else here after consecutive days of –5.5 and –7 against Akron and UConn, the latter with seven protection minuses besides.

Unfortunately and as detailed extensively earlier this week, Miller's failings are far from the only issue with the run game and benching him will not be a magic wand.

Sigh. Well, then, what are the issues with the run game?

A dossier of failed (< 4 yards except short yardage type plays) Michigan runs broken down by responsibility (I split some plays 50/50 between two categories):

• Rock-paper-scissors loss: 6
• Toussaint issues: 1
• Missed reads: 1
• TEs: 3
• Line biffs: 5

Note: left out the end around and a two yard run on second and three.

Miller was responsible for over half the the line issues, but those RPS hits are a big part of the problem. Again and again Michigan got caught in crappy situations like zone read against a corner blitz, and just sucked it up. If Gardner has a check maybe that's on him.

So you give Toussaint a super grade and think that he's not responsible for short runs in this game and you're still down on him?

I can't argue. I will have to resolve this inconsistency going forward.

My feelingsball take: I still think he's deeply uncomfortable with the idea of following his blocks. At the slightest hint of adversity he wants to bounce, bounce, bounce. Sometimes that's because he's got an unblocked guy he's trying to deal with, but especially on power plays even minor issues cause him to lose faith, think about a bounce, and blow up the precise timing required to run the ball. Canonical example:

Butt gets rocked back there by the end. Toussaint wants to bail, figures out he can't bail, and then gets swallowed up for zero yards. If he follows his lead blocker that's the worst case scenario. More likely he grinds out four or something.

When faced with a hole occupied by a defensive player he almost invariably tries to bounce it outside, finds that a Michigan player is in the way since he's kicking out a defender, and costs Michigan two or three yards:

That adds up when it happens on six to eight plays a game.

On a rather heavy other hand, he did have a couple of nice downshift-through-the-hole moments on zone plays, most prominently the inside zone embedded above and the 14-yarder that kicked off Michigan's winning field goal drive:

He also showed his skill in the open field on the 35-yard speed option touchdown, setting up the last man with an upfield cut and then bouncing to the sideline for a critical 20 bonus yards. In this game the good outweighed the bad. I guess.

How do we fix this?

This is never happening but Michigan would be wise to seriously de-emphasize their tight ends. Funchess's issues have been discussed in depth, but Williams is hardly better. He consistently gives ground against players smaller than him. Toussaint's fumble on the last drive is probably a touchdown if Williams doesn't get blasted back three yards by an OLB:

Don't confuse that with kickout blocks on which offering upfield penetration is acceptable. Michigan cracks down with Chesson and then flares Butt to the corner—one of those tweaks I complain about not seeing enough of in Michigan's run game—on a play designed to get the corner all the way. All Williams has to do is stalemate a 230 pound OLB (with help from a chip!) and Toussaint has the ability to cut back. Instead he has to run to the corner.

This is a disappointment. If they're improved over last year it's incrementally; they have not taken a leap. Putting those  guys in the box is asking for one or both to whiff a block and blow your play up. Michigan's best formation is probably a shotgun 2-back system with Dileo or Funchess in the slot and Kerridge your superback. I'd even flip Funchess out to WR and run some bubbles on the assumption he can handle a cornerback. Get guys out of the box so you have fewer possible blown blocks that end up with you eating grass in the backfield. Lining up with two tight ends is a terrible idea right now.

It is going to keep happening all year.

Receivers?

[Passes are rated by how tough they are to catch. 0 == impossible. 1 == wow he caught that, 2 == moderate difficulty, 3 == routine. The 0/X in all passes marked zero is implied.]

There is the migration from the two veterans to Chesson, who got five targets to Reynolds/Jackson's two. Unfortunately, Chesson showed his inexperience on his two deep targets; he should have been able to shield the corner off the ball on the first one and misjudged the second, possibly because he'd just been chewed out about shielding the ball earlier in the game. But he did get another nasty crackback block in. Chesson for tight end!

Reluctantly tsk-tsk Borges?

Yeah man you've got to be nuts to run out an I-Form big formation on first and twenty and run a stretch with these guys. And the Norfleet stuff…

…is unbelievably predictable. (As was lining up Funchess in the backfield for a wheel route, but I guess it worked.) He was in for one play in this game where he was not provided the ball. His first jet sweep was a success largely because UConn was badly misaligned, and now that it's in the books teams will be prepared for that. I'm hoping that Borges assumed these last two games would be blowouts in which he wouldn't have to show anything he hasn't put on film already.

I'm also getting peevish about Michigan running into stacked boxes. On Michigan's first play from scrimmage they come out in an I-Form big and shoot Funchess into a cornerback, leaving Michigan 7 on 8.5 in the box with predictable results despite the blocking being pretty much fine. No one accounts for the quarterback. It seems like we're back to the DeBord style in which Michigan does a bunch of low-upside stuff in which they lull you into a false sense of security before hitting the big waggle play; I prefer forcing the defense to account for a QB like Gardner consistently. It sucks to have to guess whether UConn will check your QB or ignore him.

At least they ran an identifiable counter play that worked. For five yards, but, hey, five yards.

Heroes?

Toussaint, says logical me. Gardner's legs, when not fumbling. Lewan and Glasgow and Schofield.

Maybe not so heroic.

Tight ends. Miller. And Chesson had a rough all around game.

What does it mean for Minnesota and the future?

It's time to make the change on the OL. Maybe it won't work but with such a clear hole you've got to try it. Time for ol' Bubblewrap Bryant to show out.

Michigan's tight ends still can't block. Not even a little bit. Maybe Butt, but I don't want to test that extensively.

Gardner wasn't as bad as it appeared, turnovers excluded. He got a ton of pressure and clearly lost confidence in his line, for good reason.

You can't exclude turnovers. The first fumble was loose handling of the ball on a QB sneak. Cumong man. Gardner's accuracy was the issue here; he needs to calm down and keep his mechanics going.

Toussaint does need to learn that he can only get what he can get when he's penned in. Hop cut in the hole: fine. Bounce: no bounces.

They want Chesson to emerge, but he's still raw. Losing Darboh is hurting right now.

Michigan needs to decide who they are instead of running the entire world of plays. I think they should be a spread offense that uses half a tight end instead of two, but I think we're stuck watching Michigan waddle aimlessly through every play in the book.