Upon Further Review still has a sponsor.

I've been informed to emphasize the rates and accessibility of your loan guy and not so much the pants. I do feel that the pants are an excellent metaphor for those aspects of going with Homesure Lending. But I do as I am told. Also, Matt tells his kids to go to bed at 6 PM so you know he's not to be messed around with.

Matt's got a ticket offer going for a Michigan football or basketball game. If you're buying a home or refinancing, he's the right guy to call. (No pants required.)

FORMATION NOTES: This is the "diamond" formation referenced below:

Michigan showed this more than they ran it, often motioning a TE to the side the FB was on.

Meanwhile your weird thing of the week was Tom Strobel, OL:

He is outside of "right tackle" Patrick Kugler with Cole lined up outside of him. This was a failed fourth down conversion that in retrospect probably would have been a touchdown if Smith hadn't fallen over untouched.

As for UNLV was pretty typical:

They spent the day with between 8 and 10 men in the box. Plays on which a safety was at least 10 yards deep were conservative ones.

PERSONNEL NOTES: Starting line was as expected. From left to right he second team line was Bushell-Beatty, Dawson, Kugler, walk-on Ben Pliska, and Bars. Kugler got a couple of snaps with the first team when Michigan went to a goofy seven-man line in the third Q. Tom Strobel, wearing 50, also got in on those plays.

The rest of the rotation was pretty much as before. Smith was the lead back backed up by Isaac and Johnson in that order; Green did not get in until the final drive. It was mostly Kerridge at FB until he got hurt; not much AJ Williams at TE, almost all Butt and Poggi.

Moe Ways got more playing time at WR, but there were not a ton of WR snaps to go around.

[After THE JUMP: selling out on the interior]

I feel so meh about this.

I felt better after watching the film. Michigan was content to grind this out without showing their hand for BYU. UNLV was content to get gashed on the outside as long as they made it look good on the inside. There was no better example of this than that time on third and one where Michigan put Kerridge in the slot and nobody from UNLV bothered to acknowledge his existence:

Michigan converted on a run up the gut anyway. So it goes.

So we are not that worried about an anemic run game?

We are not. Given the way UNLV was playing the long runs that busted to the outside were less a happy fluke and more an inevitable consequence. Literally every time Michigan tested the edge against UNLV they found very little resistance. The jet sweep touchdown needed two blocks, and I'm not entirely sure it needed the first:

Meanwhile Michigan messes up the Isaac pitch sweep and still rips off a 76 yard touchdown because UNLV just cannot defend the edge given the way they are playing.

De'Veon Smith's part in the run game was anemic, but large chunks of that was Michigan running into the teeth of stacked fronts that were overplaying anything between the tackles.

Michigan countered that with the two big runs and got reasonable production despite the overplays to assemble 6.5 YPC. That's good.

I mean, the

CHART

chart is pretty good.

I don't want to take too much away from the individual numbers since this is still UNLV we are talking about here. I will say that Braden is way better than he was a year ago. In week three against a similarly dire opponent he was 8-4=4.

Also of note was Cole's play in space in this game. That was a weak point through about a game and a half; he was very good in this one. This is the kind of cut block that cannot be leapt or recovered from:

He sealed ends off when required and got very good second level blocks.

The line didn't get a ton of great blocks but they lost very, very little. The overall result were numbers lower in amplitude than usual but way over our 66% benchmark for a good outing. Cole took –2 for a clip; interior guys got occasional dings for picking the wrong second level guy; I had almost no negatives for lost blocks.

But Smith was not so hot in this game?

He didn't have many opportunities to bust through guys. This was UNLV's design and his presence created a gameplan that was exploitable. It was even more exploitable than we saw, as Michigan was largely content to butt its head against a wall in the second half.

That the gameplan designed to limit him limited him isn't a huge surprise. I did think he missed a couple more cuts. Contrast this pitch sweep with Ty Isaac's:

That TFL is 100% on Smith. I also thought he had a lane here to be mean ol' De'Veon Smith in but instead exposed himself to the unblocked guy outside:

That's a play on which a quicker guy with the same awareness scrapes out some yards; Smith has a much smaller margin for error with his vision because he's a nuclear-powered icebreaker.

And then there was the real bad thing:

Michigan did not convert a fourth and one because Smith fell over of his own volition; the weird formation had flanked UNLV and without that trip the play goes one of two ways: Smith gets caught by a guy shooting a gap inside of Braden, or he scores. That is a huge swing in our perception of this run game.

Aside from the pitch above, the cuts Smith missed were mostly ones he doesn't like to make and has probably been told to avoid at all cost: bounces. Smith knows his strengths and generally goes straight upfield. (Or you could say he rarely sees opportunities to deviate from the play design.) If he bounces it's often because he has no other choice. So UNLV spent the day diving their EMLOS (end man on the line of scrimmage) inside Michigan tight ends.

That constricted a ton of lanes and left huge tracts of land to the outside. Smith did not test that.

In general Michigan did not look like they were particularly focused on beating this schematically.

Wait, what?

Against Oregon State they ran a bunch of power plays where they had two guys pulling, one to either side of a block on a DE.

If they'd done that in this game and their RBs bounced it out they would have slaughtered this D. But for whatever reason this game saw both those guys go inside:

This clip was the inside run game in a nutshell: Michigan hammers UNLV off the ball and is thwarted by the Rebels' willingness to put everyone at the LOS. Even so Michigan gets a solid outcome for first and goal from the nine; Kerridge arcing outside like he did towards the end of the Oregon State game is a walk-in TD. I started minusing Kerridge for this but after a while it became clear this was what they were supposed to do and I shifted those over to RPS.

It was getting a bit frustrating watching this happen repeatedly. This one is a monster gain if anyone arcs outside:

Ideally Isaac would bust this himself; 1 out of 2 on the day ain't bad.

But the RPS came out positive anyway?

Yeah. Michigan got a free jet sweep TD, four quick screens that succeeded, and a bunch of easy pitch sweeps. Michigan also had sufficient changeups to prevent the extra guys in the box from blowing too much up. I usually offer an RPS minus if I don't think it's reasonable for the offense to block the guy who made the tackle (if the tackle is near the LOS) and that didn't happen too much in this game.

If this is weird to you, it is also weird to me.

AJ Williams kind of disappeared after his best day as a Wolverine. What's the deal?

In this game at least, Poggi was the top inline TE. This is a situation where Poggi has a major advantage if the two guys are at all close: Poggi is relatively new to the position and Michigan has him for another two years. He has much more upside and reason to develop that upside.

His performance is justifying the playing time. He got the (possibly) critical block on the jet sweep TD, and this is a play-long whuppin';

I bet a dollar Poggi got a sticker for that. Poggi also crushed his guy down the line on the ill-fated fourth down attempt.

The trajectory here is highly encouraging. He was bad in the opener, better but still erratic in game two, and very good in game three. Tough to tell quite yet but it looks like he's a real player who had some first-game jitters in his debut at a new spot.

Also cutting into Williams's PT was Jake Butt. Butt had another good day as a blocker. On this counter run at the weak side of a tackle over formation he wins the key block at the POA:

That is their WDE, sure. Here's Butt rocking their SDE back with whiplash:

I'm still excited when he does anything that says ACTUAL TE NOT DEVIN FUNCHESS. (Not that I wouldn't stab someone to get Funchess on this team. It's just that M needs a TE like Butt for maximum Harbaughs.)

Over the course of the day he got a lot of guys diving inside of him. In that situation the move is to shove 'em, get some depth, and open up the cutback; Michigan didn't take advantage but that had little to do with how well the TE blocking was or was not going.

And Hill?

We didn't see him until late and then only briefly. He ended up a slot receiver on another play that looks like a pitch sweep; he got a shove in but then lost the guy. Poggi is the man right now.

AAARGH Jake Rudock /throws trident

There is no way around it: Rudock was very bad. If Jake Rudock is not going to complete a five yard swing route that is at least 30 yards of YAC, I'm going to write a column about how I'm freakin' out, man:

This was the worst I'd ever seen him play by about a million miles.

Jake Rudock 2014

Jake Rudock 2015

Good news: that's probably not going to happen again. Bad news: that is a downgrade in expectation anyway.

Rudock had a number of instances in this game where he ended up sitting in the pocket forever and then running around. That is a major indicator of a quarterback who is not comfortable with the offense and isn't seeing guys get open. I was frustrated with him for both being too conservative (he had a one on one matchup on the outside on second and one and instead scrambled for a few yards) and too aggressive (he had a wide open conversion on third and three and instead went for a relatively well-covered deep ball.)

I'm all for taking shots. Third and three when the side of the field you're surveying has a blitheringly wide open conversion with YAC potential is not the time.

The throws he did hit were mostly super easy, from the smoke/flash screens (four for 43 yards) to a slant on which a confused UNLV wasn't even in the area code. I'm glad the smoke stuff is still exploiting open edges, but that is going to go away against teams with better corners who think they can match up with Michigan's WRs—and that will be a lot of teams. The pressure to improve the actual downfield part is going to ramp up. Michigan needs Rudock to be the guy who always hits the stuff he should hit; if he does not do that then he does not make sense as a QB.

I will kind of defend him on the interception.

He has a flare route underneath his target that he does induce the LB to take two steps to; once he comes off the flare he probably is not expecting the UNLV player to recover and get to the ball. That does not happen if he is Ryan Mallett or if he holds the guy a beat longer, yes. It is still an excellent play from UNLV.

Also in faint praise, this little pop pass is spooky:

That corner gives zero indication he's coming. Michigan must have seen something on film that gave away the blitz there.

I am dousing myself in acid.

I won't stop you.

Hey.

I mean that Michigan needs much better play out of Rudock if they're going to have a decent season.

This weeks Stanfordization of the week?

Multiple options. Michigan debuted an iso-ish variant where the offset fullback starts pulling to the back of the play:

DTs flipped out about this and created some big gaps up the gut. I think I know why: this kind of fullback action is most commonly split zone. DL sees that, figures they are defending zone to their side of the ball, and get upfield. If you look closely at the playside DT in the above clip, he redirects hard once the FB crosses his face. Then the FB redirects and there are some big ol' gaps on the interior line.

And this is… something?

Michigan ran this twice. The first time was not particularly successful since Houma was taking on a 330 pound NT at the point of attack; the second saw that Isaac spin move above make up for the fact that Michigan was trying to block that dude with Jake Butt.

I have no idea what this is. I have not seen it before. It looks like three separate trap-type blocks across the front with immediate releases from the guys DL normally expect to block them. I called it "triple trap" but that is obviously a dumb and bad name. Taking advice on whatever this is.

We also saw more of the traps (rather successful) and FB dives (not so successful) that the Oregon State game brought us. I thought this trap was particularly interesting:

The interesting people are the fullback and Kalis. The fullback heads to the back side of the line and the UNLV LB reacts in Pavlovian fashion. You see that fullback, you go hit him. So he removes himself from the play; Kalis chases him instead of hitting the other guy. Otherwise Isaac is again one on one with a safety.

Does this feel qualitatively different than other manball stuff you've seen?

It does. Michigan keeps deploying these little tweaks. The split iso thing is very cool to me; it's not something I've seen before and you can see how it moves the defensive tackles. Michigan's previous manball offenses did very little of this kind of thing. Rich Rodriguez was also a tinkerer—remember that period where Carlos Brown was suddenly zooming off the backside for 80 yard runs?—and it's no coincidence that Michigan was better at running under him than they had been since Tim Biakabutuka.

Harbaugh's style is diametrically opposed to Rodriguez but his approach is similar. He is always looking for ways to screw with people. He has a base; he never wants people to know what's coming. What Harbaugh has with his style that Rodriguez did not are options. Each guy is a gap, and each gap can be moved. I still think in most situations it's better to remove blockers instead of add them since one missed block can end a play, but Harbaugh has a track record of making the extra guys an asset.

I have an inkling of why now.

Have any more Layman's Rules Of Blocking?

No, but I'd like to bring forth another example of #2, Stay Attached:

In this life you're occasionally going to get hurled to and fro by large, angry men with flowing manes of curly hair. If you have the balance and footwork to continue annoying them after they have done this you can push them past where they want to go and create a cutback lane. It's the tailback's job to run around the guys. Just stay attached.

Receivers?

Nothing of note, which is unsurprising given the passing game.

[0 = uncatchable, 1 = circus catch, 2 = moderate difficulty, 3 = routine]

Kerridge is a nice receiver for a fullback.

I didn't put that last Darboh deep ball on him particularly. Thought it was more Rudock.

Heroes?

All blockers.

Maybe not so heroic?

Rudock, Smith.

What does it mean for BYU and beyond?

We get to test out the line's improvement. At this instant it looks like BYU NT Travis Tuiloma will not play. That leaves BYU considerably short of Utah's fierce line but well above the other two teams Michigan has faced. They look a lot better. I'm not saying they're going to be great, but I have hopes they can get a clear win.

Rudock needs to be Iowa Rudock please.  Let us all look back on this as a hilarious aberration.

RB battle is on. Smith gave up much of the commanding lead he had after Oregon State, and Isaac looked quality.

Tight ends are coming on. Poggi, Butt, and Williams look very credible as blockers.

Braden is probably going to hold on to his job. Big turnaround, and it was Kugler who got the weird 7 OL snaps in this game.

Tom Strobel is looking to set a Michigan record for the most numbers worn in a single season. Do it.

Wide receivers… eh. Hard to judge with limited opportunities.