Michigan State has a pretty good defense, as you may have heard, and this deep into Michigan State having pretty good defenses there's nothing you can say about it other than "I wish that was not the case." There is some Bud Foster action going on in East Lansing.

Let's review what went down the last two years, in an effort to figure out what Michigan's looking at and what they might do in response.

2011

The trash tornado game. With winds howling around Spartan Stadium, Borges put the game on the arms of Robinson and Gardner and got little in return.

Shotgun or under center? This was almost entirely a shotgun game, and when Michigan went under center it was to throw. Setting aside the disastrous fourth-and-inch from the nine on which Michigan went goal line play action and Brandon Moore didn't block his guy, Michigan saw five snaps from the I-form (two more were penalties, one on each team). One of these was a pitch to Toussaint for four yards. The others were throws. The first was a wild, jinking 15-yard touchdown scramble for Robinson on Michigan's first drive; the others were a sack, a one-yard throwback screen, and an incompletion to Hopkins in the flat off play action.

First down approach? 16 passes, 11 runs. Not that it really mattered. Michigan got a 34-yard touchdown when Roy Roundtree broke a tackle on a slant. They had one other good gain, a 29-yarder acquired when Robinson ignored two open guys, threw at Hemingway as a linebacker was undercutting him, and threw it high enough to get over the LB but short enough for Hemingway to grab it—a lucky fluke. Their other first-down passes acquired a total of –3 yards.

Running was no better, with 36 yards on their 11 attempts. Four of these were Denard's (18 yard total), which boggles. Michigan had 27 opportunities to run Denard Robinson on first down and did it four times. But that's not really relevant with Gardner, who's much more of a dual-threat.

What worked? Virtually nothing. That one long completion where Roundtree beat a tackle in cover zero was the longest play, Denard's should-have-been-pick-six was #2, and #3 was a double A gap blitz on which Vincent Smith was hit in the backfield but managed to pop through a tackle. Breaking tackles to get more than five yards is no way to live.

The rest of it was Robinson running around: a scramble, a jet sweep, a run-around improv throw, a jet stretch, and a QB draw were Michigan's most successful plays outside of the aforementioned.

Was it really as bad as all that? Yes and no. Borges did rip open the MSU defense for several plays that should have been big gains only for his quarterbacks to throw it at covered guys.

Spielman's faith that Robinson would have found the open guy is probably optimistic, but inserting a clearly overwhelmed Gardner was a huge tactical misstep, as he blew multiple opportunities to gash MSU. This in turn may have led to the WR move, which in turn led to the Nebraska game, which… let's stop the counterfactual history of the Michigan program before our heads go numb.

On the downside: Michigan had no answer for MSU timing their snaps, both with double A blitzes and plain old running at the quarterback. This was maddening since it had just happened the year before; there were no adjustments. And they relied on Brandon Moore, who had seen virtually no meaningful snaps in his career, on that deadly fourth down. When that guy doesn't execute, you share in the blame for putting a guy who'd never seen live bullets in a stressful situation. But he's a tight end, and all tight ends must tight end even if they obviously can't tight end.

Wurrfle furffle torghern furfen. Michigan abdicated without really trying here. See: 4 first down Denard runs. The constant snap issues were a coaching issue, not a toughness one. Ditto the thing where Molk put his head up a nanosecond before the snap and had to figure out which of the two linebackers he was going to block and which one he was going to let scream up the middle unmolested.

[2012 and 2013 after the jump]

2012

Shotgun or under center? With Denard returning for his senior year, Michigan entirely abandoned the notion of going under center. Their four snaps from under center last year were two throwback screens for 22 and –1 yards, a goal-line fade to Funchess (incomplete) and a play action hitch to Gallon that was batted at the line of scrimmage.

This sort of helped. Both Toussaint and Robinson squeezed out respectable days on the ground, combining for 148 yards on 30 carries. However, half of these came on a 44 yard Robinson QB draw and that Toussaint run where Joe Reynolds sliced down a safety. Eleven of Michigan's 32 runs were TFLs or zero yard gains; ten more were 1-4 yards. In that environment it's difficult to do anything consistent since eventually you end up behind the chains, and Denard Robinson behind the chains against MSU is not a recipe for great success. Michigan's drives into MSU territory petered out into field goals as they acquired 326 yards.

First down approach? Run run run. Michigan attempted six passes on first down, one the 35-yard quick post to Dileo that would have been a touchdown if Dileo was fast…

…the other five incompletions or a TFL on Toussaint. Michigan ran 18 times for 108 yards, with five failures (<2 yards), six meh(2-5 yards), and 7 successes.

What worked? The inverted veer. Aside from the two big plays, Michigan's five most successful runs were inverted veers (two gives, three keeps), at the cost of two TFLs. Michigan successfully blew MSU DT James Kittredge off the ball all game. Unfortunately, Kittredge is now third-string.

Also: Drew Dileo, who racked up a number of catches in the seam.

Was it really as bad as all that? Well, they won, and few teams really did better against MSU than Michigan did last year in a yardage sense. The snap timing and jumping receded despite Michigan going from the 'gun the whole game. Michigan had a productive day running the ball, all things considered, and avoided the fatal turnover. Michigan should have scored the touchdown that would have put them ahead, but Denard threw an easy pass hard and behind Gallon in the endzone.

On the downside, Michigan's general incoherency hurt them in big spots. Kerridge ran by a DE on a first-and goal inverted veer  that Denard gave on:

Either someone's got to crack back on the safety there or Kerridge needs to pop the DE, because once he gets out there there's no one for him to block. Michigan has been terrible at executing their runs mentally for going on three years now, and when things got tight in the redzone the costs were steep. On the ensuing third down from the seven, a double-A gap blitz saw Toussaint whiff and Denard threw it away. MSU is still not getting burned at all on that.

Also: the two minute drill was an abomination, taking Michigan down to 18 seconds before the Dileo bacon-saving completion when a relatively competent offense still has 41.

Wurrfle furffle torghern furfen. LeVeon Bell ran for 2.8 YPA, Michigan actually ran productively, and Andrew Maxwell threw it 34 times. Toughness was not an issue.

2013?

Shotgun or under center? Even after the Penn State game, Michigan stuck much closer to their under center dreams, splitting plays almost down the middle. Since it was Indiana, it's tough to draw conclusions, but even against the Hoosiers, running from under center was iffy.

One thing that's over, or at least should be: tackle over. Michigan ran seven times from it and acquired 10 yards against the worst defense known to man; the payoff was one comeback route and a throwback screen that went for 70 that should have gone for six. Short yardage, fine, but if it comes out elsewhere it'll probably be to run super tricky play action.

As for regular under-center business, the rest of Michigan's under-center runs (goal line and the Toussaint pitch fumble excluded) included 5 for 45 on their final drive, and 20 carries for 47 yards the rest of the day. Add in the debatably-useful data from the final drive if you like; either way under four yards a carry against Indiana corresponds to down-burning against MSU. Michigan may be able to get compensation for that with the Jimmy Clausen/Golden Tate/Michael Floyd gameplan wherein you go max protect and go deep, hoping to win one on one downfield.

Any semblance of an offense that moves the ball on the ground will be from the shotgun.

First down approach? Probably more 2011 than 2012. Gardner is proficient deep and has two excellent threats; the run offense has been one of the worst in the nation at avoiding TFLs. With Funchess and Gallon available to test MSU defensive backs in multifarious and sundry ways, unleashing the dragon seems like the way to go.

Wurffle furffle torghern furfen. 27 for 27 pretty much says it all, especially after Penn State got annihilated on the ground by OSU. Evasive action required.