Upon Further Review 2016: Offense vs UCF

SPONSOR NOTES: Oh man Sauce Castillo, you're in for it. You already turned El Assico(!) into a blowout. I'm supposed to talk about mortgages. Right: low rates right now, and Matt will take these rates and turn them into a home if you qualify for things such as loans.

In addition to being a gentleman replete with Michigan tickets, he is also a good man to know if you need a mortgage. It's striking that we actually get non-astroturfed comments about positive experiences with Matt not infrequently.

If you're buying a home or refinancing, he's the right guy to call.

FORMATION NOTES: UCF was a 3-4 front with a couple of adjustments. This is their base front; Michigan is in "ace diamond TE," with Asiasi at one of the FB spots.

On passing downs UCF would go to a nickel with two DL on the field and standup ends:

And they'd frequently line up their three DL right next to each other and shifted to the run strength of the formation:

Called this "pinched 3-4."

PERSONNEL NOTES: Michigan cut down on the rotation severely despite having a huge lead. With the exception of left guard, the starting OL got almost every snap. Non-LG starters (Newsome, Cole, Kalis, Magnuson) got all 81 snaps. Braden and Bredeson platooned at LG with Bredeson(49 snaps) getting the plurality of time. Bushell-Beatty and Onwenu came in very late in a 7 OL package.

At WR, Chesson and Darboh got most of the run in a game featuring a lot of heavy packages. Grant Perry got just 15 snaps. Butt was near omnipresent; Bunting was the next-most utilized blocky/catchy guy. Poggi and Hill are still splitting things down the middle.

Smith got about half the work at RB(37 snaps), with Evans, Isaac, and Higdon splitting the rest about down the middle.

[After THE JUMP: pass great, run not so much]

PANIC

YES THAT

Maybe panic. On the other hand, this is what happens when you're that gung ho about the run:

Maybe not so panic.

EVEN SO I AM NOT TAKING KYLE KALIS'S ADVICE.

To relax?

YES. I DECLINE.

I don't think you're totally crazy. Let's look at the—

CHRT

OL Chart.

*[PFF has Magnuson in a pass route on one snap. ]

There were four TEAM pass minuses as I couldn't assign blame on some blitzes.

In the aftermath PFF came away calling the line's performance "dominant":

After some initial struggles in the first quarter, Michigan’s offensive line put together a really dominant performance. Grant Newsome and Erik Magnuson, the two starting tackles, did not allow a single pressure on 89 combined pass-blocking snaps. In addition, with the exception of left guard Ben Braden, all members of the unit were solid in run-blocking, too. Center Mason Cole was the best player on the Michigan offense, as he had only one negatively graded run-blocking snap and had multiple plays where he drove UCF linemen several yards downfield.

I didn't come out that positive but once you separate out run and pass we're close to the same wavelength. Let's discount the protection minuses from Hill and "team" since the former is not an OL and the latter were blitz issues applied to RPS. That leaves you with an 85% protection rate against a team blitzing a ton—which I may not have taken into account sufficiently. That's top-notch, and PFF put a lot of weight on that.

The situation versus the run is a lot murkier. Michigan did not hit or approach our Mendoza line; I thought the right side of the line was mostly good and the left side, especially left guard, a major issue. PFF is mostly on board, less high on the right side and far less skeptical about Bredeson.

You're both in agreement that the left guard spot is a problem.

Yes, the rotating left guard, whoever he was, was a clear issue. Michigan started out with Braden and he struggled from the get-go. This is the first offensive snap:

NT does a good job there to turn his body so he's not too exposed to the chip but nonetheless that chip fails and he's in the backfield. There was a lot more of this kind of stuff than I'd like to see against UCF.

On third and eleven Braden would way overplay a DL slanting away from him and allow a blitz through untouched:

That guy's blitzing from some depth and leaning forward enough to tip an aware player off; that's pretty grim. On the next offensive snap he'd straight up tackle a dude and somehow avoid a holding call; on the snap after that his man came around him so quickly on a double iso that he made a tackle at the LOS. (To be clear, that last play was doomed for several different reasons.)

Braden was so far off where he was by midseason last year that you have to wonder how healthy he is, and how much practice time he missed with whatever injury he sustained. This was very rusty version of Braden nowhere near an established level of performance.

Bredeson meanwhile seemed just as unready as he did in the opener. He again just stopped on a pull, this one even less understandable than the one last week. This is well blocked on the interior and there's a clear gap for him to go into:

He changes his mind in the middle of the play. That's Layman's Blocking Axiom #3: Never Change Your Mind Halfway Through A Play. If you're wrong, you're wrong. You might block someone relevant all the same. Football's weird. Turning around means you're not blocking anybody.

This zone play was reminiscent of the Funk lines where guys would just run downfield versus guys they absolutely have to block:

That guy is almost head-up on you. You have to combo through. (If that looks familiar, Kalis did the same thing in the bowl game.) I get why that happened: UCF is playing a 3-4 and usually that means the end is further outside and you can go get whoever. Not adjusting to a change in front of you is freshman stuff. Or I-never-had-Tim-Drevno-until-now stuff, if you're Kalis.

In addition to the mental stuff Bredeson got beat up a couple times:

I'm confused why he was in this game over Kugler, who looked functional against Hawaii. I have no doubt he's got bigger upside; I'm skeptical he's going to reach it this year. There's a reason you don't want to play freshman OL.

I'm going to go buy a shotgun now.

This is an excessive level of concern.

Ice cream cone?

Okay. Yeah, that's okay. There was some good news on the blocking front.

Probably related to Wilton Speight mostly standing unmolested for hours at a time?

Yes, and thanks in no small part to Grant Newsome. The most important thing so far: Newsome has been excellent in pass protection. He had one –2 in the opener and was perfect here. All caveats about level of competition apply, but all he can do is block the guys in front of him. So far so good.

In that context a few run issues are nitpicks. They do exist. A couple of times he pulled oddly wide and deep on Michigan's perimeter runs. On this one I'm about 90% sure he's not even supposed to pull since the DE is lined up directly over him:

Michigan gets lucky that doesn't bite them, and then Newsome doesn't read the play in front of him, going outside of a force guy and blocking nobody. He had a similar pull to nowhere on a later jet sweep.

Also Cole and Kalis were thumping people. Cole blew a guy to the sideline. Those two were repeatedly used on short yardage. Hill dive + Cole + Kalis == easy. I didn't clip much of Kalis because unsuccessful runs aren't good platforms on which to convince people that a guy is playing well; he is. Again I don't think I caught much in the way of mental errors.

Left guard is an issue but Newsome and Kalis are hitting the upper end of reasonable projections so far. I'd take that trade if you offered it to me preseason.

So if we can pass protect, and Wilton Speight is good…

Should we just be the world's most hippopotamus-shaped air raid?

I mean…

I want to see Newsome go up against some tougher customers but it's a consideration, because Speight's performance didn't look any less confident on review. His incompletions, categorized:

• Hit on throw: 2
• Drops: 3
• Intentional throwaways: 1
• Receiver disrupted by D: 2
• Missed deep balls: 2
• Bad read leads to PBU: 2

"Receiver disrupted by D" contains two events, one a deep crossing route to Chesson that was probably Speight's best option and almost complete, the other an attempted dumpoff to Evans in the face of heavy pressure where Evans gets hung up on an OL trying to get into his route and can't get his head around in time. Everything else is self-explanatory.

One thing does not make an appearance here: a straight-up miss on a throw shorter than 30 yards. Speight's accuracy was excellent, with vanishingly few throws where the receiver had to do anything even mildly difficult. The receiver chart has a ton of throws marked routine and almost nothing else. I actually clipped a catch of mild difficulty for Darboh because it reinforced how things like this were not happening often:

That's still routine in my book but I had to think about it. Everything else was on point. There haven't been any passes to dig out or guys taken off their feet or throws on which the receiver has to reach behind his body. Everything short has been almost 100% on point. The only exception in this game was a screen to Evans that should have been a TD; Speight lofted the ball far too high and to the outside, allowing the opposition to recover.

Speight's deep accuracy, meanwhile, was still good. The misses stand out because there were so few otherwise. On the other hand he nailed Darboh on a 45-yard ball you can't throw better…

…and hit Chesson on a post that is maybe a yard too far in front:

He was perfect on the blizzard of routes Michigan ran at the numbers about 25 yards downfield.

Thus this chart:

[Hennechart orientation: mouse over column headers for explanations of the categories. + is handed out for a good throw under duress. * is handed out for a very bad version of a bad thing. Numbers in parens are screens. DSR is an attempt to compress the numbers into one overall number. PFF is PFF's grade.]

While this wasn't a high degree of difficulty game I'm wondering if PFF is still a little pro football focused—I was way more impressed with Speight's outing than a +1 passing grade would indicate. Eddie McDoom got +1.5 on one jet screen.

Speight gives off an aura of confidence. At times it's clear he knows where he's going early in a play. On the first Butt touchdown he takes a couple steps to the right before he throws, improving his angle:

He's done this a few times as he tries to shorten redzone throws, which often have very tight windows. He's comfortable throwing with his feet in motion. That's not MOBILITY, but neither is he a statue who breaks into pieces when you move him off his spot.

His grasp of where he's going led to things like a no-problem touchdown on third and twelve from the 14:

Again he has a great feel for who's going to be open when and why on this play; ball is out exactly on time and Michigan converts in a tough spot to do so.

This feel only let him down twice. On one play he came off a Darboh hitch because he thought the flare to Evans would be open; on the other he threw at a covered Butt instead of an open Chesson. Both of those resulted in PBUs and BRs in the chart, but they're innocuous BRs as they go: neither PBU was anywhere close to an interception because the defenders were being boxed out by Michigan WRs.

His poise under pressure isn't flawless but the eye test is much kinder to him than PFF's (admittedly low sample size) numbers. UCF got in on him in a flash repeatedly in this game and the worst that happened was a throw that could have gone anywhere since he was hit on it. He also took a sack he didn't have to. He did not throw it into coverage and he occasionally made something out of nothing:

I think the play design there is supposed to overload coverage by shooting both TEs to the left side of the field and then both RBs; Smith gets picked up. Poggi does not. That seems far more like a guy Speight knows is open from looking at the TEs than a Mallett event where he's just chucking and praying.

Similarly, the dump to Evans that didn't go well is the right decision in the face of a ton of pressure:

When Speight wasn't totally clean in this game guys were coming in fast, and nothing dumb happened as a result. Let's not go overboard about Speight's start—he passed sophomore John Navarre for most touchdown passes in his first two games, which is a proper dose of perspective—but it almost literally could not be more encouraging against this opposition.

You had some complaints about the running backs again?

After looking at things in detail, not really. Smith did miss that cut I talked about in the game column, about which more in a second, but then he plowed guys flat. When given an opportunity he did very well with it.

RB chart (WR grades are run only):

Michigan mostly got buried by RPS and some iffy blocking. Those plays I mentioned in the game column did stand out as WTF moments. The Smith run where he went stomping through six different guys is one:

Many commenters thought it was unreasonable to expect Smith to change direction here. Nope! Nope nope nope! To feel the location of the hole based on the data in front of your face is a thing that some running backs have. Running backs like Citrus Bowl De'Veon Smith:

Decisive one cut running does exist, and it would have put Smith into the secondary without having to flatten a fishing village on the way. Right here the LB is committed and right in Smith's field of vision:

This is the moment for a hard cut to the backside hole. He does flatten those fishermen so I gave him a plus anyway, but that was a callback to his vision early last year. I also thought his other notable tackle-breaking run could have used a cut behind Jake Butt's iffy block. But, hey, notable tackle-breaking runs.

Higdon got some early playing time. He mostly got buried through little fault of his own, as per usual with him. He did display some blitz pickup chops .

Evans didn't have anything that leapt out either way. I don't know where they got the negative grade for him. He did bang into Mason Cole on a crack sweep:

Downshifting for a step or two is preferable there. Still managed to cut past some chaos and get something out of it. He could be a bit more patient, especially with his burst.

Darboh is fast now?

Just a sec. Chart:

Catch chart?

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

ROUTES: Darboh ++, Chesson +, Evans –, Bunting –, Perry –.

As discussed in the Speight section above, this was a day of routine opportunities for the receiver corps. The most notable events were the three drops, two by Jake Butt. I will continue to ignore these as a thing until they become a thing. We have a ton of data on Butt's hands and it is not describing a guy who regularly drops routine balls.

Other than that, Chesson had that nice diving catch and Darboh blew by a fringe NFL draft pick. On review, the drag routes he took long distances were 1) a defensive biff and 2) good play design instead of Darboh doing something crazy. However, he continued to stay even with or even outdistance Griffin on the second:

It feels like he's got more downfield upside after this game.

Perry's opportunity at a deep ball was underthrown, but like Darboh in the opener I think he can do better with it:

Neither of those guys is looking back for the ball, so if Perry reads the flight of it correctly and slows up to high-point it, he gets run over and legitimately deserves the flag he pleads for afterwards. Instead he fades away. If there's nobody here he's catching this at his waist. It's not good when I'm reminded of Jake Rudock's receivers at Iowa.

Can I go back to something? Jake Butt got a lot of blocking points. Other tight ends… not so much.

Bunting had inline issues with the little quick guys UCF presented him with:

That's a shock to the system for a guy expecting to get an easy kickout block, one that was facilitated by UCF's aggression. They're relying on the safety to get up on any run outside the tackle to that side of the field. Bunting did adapt pretty well as the game went along. On McDoom's jet sweep he stayed annoying a long way downfield and got a second hit in that facilitated some of McDoom's wicked sweet moves:

He didn't get many more opportunities; it was fine.

Jake Butt bounced back after a tough opener. He had one downblock on a crack sweep that he really should have gotten and did not; otherwise he was moving people and staying attached to them. This style of defense fit his blocking skillset. He's good at taking on DB sorts as he can mirror their movements and also move them:

UCF had a lot of DB sorts. Butt got a lot of play long blocks where nobody was overwhelming anyone else but neither was anyone going anywhere.

The Kaiju were up and down, mostly down. Asiasi got more playing time than Wheatley in this one. He airballed on that early double iso and had a couple of seeming mental errors. On the other hand he wiped the edge clear on a late crack sweep:

Wheatley had a poor day, with some Sophomore Braden-style lunges at people:

"Work in progress" is the applicable term.

Any Stanfordization or is it still in the garage?

Mostly still in the garage, or stuff we've already seen. They ran a trap. They brought the double iso back out. These are Harbaugh things but not new Harbaugh things. Most of their RPS+ plays were play action on which chunk passing was easy because UCF sold out against the run. It didn't have to get more complicated than that.

I did notice one little tweak that was effective. Michigan ran McDoom on a jet sweep. To set it up they motioned him across the formation, threatening jet, and then ran him back. The relevant unblocked UCF defender went straight upfield:

The motion there coupled with last week probably has the defense thinking "jet," and then once they don't see it they think "not jet," only to get the jet with little time to react since McDoom is coming from so close. McDoom gets outside and picks up a big gain despite Newsome not blocking anyone.

Also, dang McDoom. There wasn't a ton of that on his high school tape. If he's got that Breaston in him and he's threatening Jehu Chesson for title of "fastest guy on the team," look out.

And if we can go back to that Smith run:

That might not be double iso. It might be something I don't even know what to name. Asiasi is damn near OL sized and he approaches this play like he's going to combo block this DT. And he does. The NT gets sealed away and then he moves to the second level, where there's nothing because the aforementioned LB shot the gap.

Could be an error from a true freshman. Could be a new wrinkle. If it is a new wrinkle than it's even more evidence Smith's got to cut.

While we're taking about tactical stuff, RPS number? Eating blitzes?

I thought many of the blitz issues were OL problems, not Michigan getting RPSed. This was a notable exception:

Nobody can do anything about that except make sure your snap count isn't as predictable. Safety already moving half-speed forward at the snap is not a blockable player most of the time.

Meanwhile the overall RPS number is huge amplitude both ways and ends up just about even. The run blitzing was almost entirely offset by the fact that every play action pass was wide open.

Heroes?

Speight is best case scenario Speight so far. Long may it continue. Cole and Kalis did good work that didn't really come through. Chesson and Darboh were perfect and their athleticism outclassed the UCF secondary. Poggi had a nice day at FB. Both tackles pass-blocked perfectly.

Maybe not so heroic?

The left guard, whoever it was, was real bad.

What does it mean for Colorado and beyond?

Left guard has to get fixed and is now up for grabs. Braden's probably still hurt. Bredeson is either going to get a lot better or it's Kugler time, whether that's Cole moving over or not.

Best case starts for Kalis, Newsome, and Speight. Three of the biggest question marks on the team have started out about as well as can be expected minus some mental hiccups from Newsome.

Butt's blocking returned to decent. Better than that in this game, but an easy matchup.

Darboh might be fast? Ask again later.

Kaiju are a work in progress. As mentioned, the bet here is that they alternate great and awful for big chunks of the year.

Harbaugh will pass on your face if you load up against the run. Not that we didn't know that already, but there's no such thing as slowing down with him, and if you dare him to pass six consecutive times he'll do that.