Fee Fi Foe Film: Michigan State Offense Comment Count

Ace October 5th, 2017 at 4:07 PM

Previously: Michigan State Defense

not bad

Michigan State's offense is still looking to put it all together. Against Notre Dame, the Spartans gained 496 yards but acquired nearly half of them in garbage time, making the outcome look closer than it should've (sound familar?). They also committed two disastrous first-half turnovers, a pick-six by Brian Lewerke and a fumble by LJ Scott right before he crossed ND's goal line, that skewed the outcome in the other direction.

Then Iowa simply shut MSU down, holding them to 4.1 yards per play and 2.4 yards per non-sack carry. The Spartans managed to beat the Hawkeyes in large part because they won the turnover battle 2-0, but it wasn't pretty.

Personnel. Seth's diagram [click to embiggen]:

On the Michigan side, Chase Winovich has sustained his remarkable play long enough to earn a shield, and he should be in for a big day against MSU's young, skinny tackles.

On the MSU side, right guard David Beedle is questionable after missing last week's game with an undisclosed injury. True freshman Kevin Jarvis, who started last week against Iowa, would take his place if he can't give it a go. Either a banged-up guy who was a sore spot last year or an 18-year-old kid will have to block Maurice Hurst on occasion. That projects to go rather well for Michigan.

Spread, Pro-Style, or Hybrid? It's a hybrid at this point, for whatever that means. State spends most of their time in the shotgun but will go to more manball-y I-form and heavy Ace formations. They'll bring out an extra OL—usually Chase Gianacokos—on occasion, and not just in short-yardage situations.

Basketball on Grass or MANBALL? State runs a relatively even mix of inside zone and power. Neither has proven effective. Most of MSU's best runs have been Lewerke keepers or scrambles.

Hurry it up or grind it out? While MSU is right about average in adjusted tempo, they feel slower than that—they usually huddle up and take their time at the line.

[Hit THE JUMP for the rest of the breakdown.]

Quarterback Dilithium Level (Scale: 1 [Navarre] to 10 [Denard]): As BiSB has noted in Opponent Watch, one of the most notable statistical oddities of the young season is Brian Lewerke's 8.5 yards per carry, which ranks him second in the conference. He somehow managed to break into the open field for a 52-yard gain on a third-and-one QB sneak against the Irish, and while he's been limited as a runner otherwise in the last two games, he all but sealed the Iowa game with a third-and-long scramble:

Just a thought: maybe don't have everyone turn their back to the mobile quarterback when the other team is trying to run out the clock. You may get a sense during this post that Iowa's defensive approach drove me bonkers both live and upon further review. You wouldn't be wrong.

Anyway, Lewerke isn't a Denard, but he's a pretty adept runner in the read-option game and as a scrambler. He's not just a straight-line guy; he set up a short second-and-goal against the Hawkeyes by juking a defender to the ground after breaking the pocket. He doesn't look to scramble first—I was actually quite impressed with his ability to go through progressions—but he's usually decisive when he does. That said, Lewerke's mistakes tend to be big ones, and that extended to his scrambling against Notre Dame, when he broke the pocket to the left, then fumbled when he tried to reverse field while lifting the ball over his head like a basketball player attempting to set up a crossover:

As it turns out, there's a reason they don't coach you to do that.

Even with that taken into account, Lewerke is a dangerous player because of his ability to make big plays with his legs. He gets an 8.

Dangerman: As this team is currently constructed, Lewerke is the dangerman. He's improved as a passer, forces defenses to account for his legs, and shows a strong command of the offense.

We've given running back LJ Scott a star out of pure respekt. Scott is an enigmatic NFL-level talent who's had a brutal start to the season. Running behind a bad run-blocking line, he's averaged only 3.7 yards per carry, but he can't blame his teammates for all his woes; he's lost three fumbles, two as he was about to score touchdowns, and that's been a problem throughout his career.

Scott is the most talented of State's offensive players yet he's stuck in an underwhelming timeshare with Gerald Holmes and Madre London. Going by what he's done this season, he shouldn't be in this section. I remember the first drive of last year's game, however, when Scott almost single-handedly took State down the field for a touchdown. He still has that in him.

Up front, senior center Brian Allen is an impressive player, though it's often hard to tell given the chaos that surrounds him. He's strong in both the pass and run game, and while he's not quite Mason Cole he's a good blocker in space.

Zook Factor: Mark Dantonio still vacillates between extremely conservative and extremely aggressive. After a third-and-three run came up a two yards short, he called for a punt from the Iowa 44-yard line—a bizarre sequence if you're going to run on third down. Later in the game, he got aggressive when it may not have been the right choice. More on that later, as it has more to do with the offensive line.

This is the Michigan game, so Dantonio will make impeccably aggressive game theory decisions. He should get called out for his defeated with dignity approach, though; he kept the starters in long after the ND game was functionally over and Lewerke came close to getting his knee blown out.

HenneChart: I charted Lewerke against both Iowa and ND, as MSU's approach was different based on the opposing defenses. The Irish were more aggressive, so MSU dialed up a lot of quick passes and screens while mostly keeping Lewerke in the pocket. Iowa sat back, as they do, so they called for more rolling pockets to allow longer routes to develop without exposing Lewerke to too much pressure.

Notre Dame 3 9++ (3) 1 4 (1) 2x 2x 1 3 1 59%
Iowa 1 15+ (1) 2 7 2 3 1 -- 2 58%

While the downfield success rates came out nearly identical, the mistakes Lewerke made against Iowa weren't nearly as dangerous—having half-field reads on many of his throws helped there. His biggest mistake against ND, on the other hand, was biiiiiiiiig:

For the most part, though, I think the DSRs here undersell Lewerke's talent. He's working with inexperienced receivers behind a leaky O-line, and his coaches tend to tip plays by formation. His talent is obvious; much of the time he looks really dang good:

It's the few moments when Lewerke still looks like a first-year starter (remember, he sat for most of last year) that keep him from consistently being one of the best quarterbacks in the conference, but he's on his way.


So, yeah, about formations tipping plays—this is the Iowa game:

Formations Run Pass PA
Gun 14 18 11
I-Form 13 1 3
Ace 6 1 1
Pistol -- -- --
Heavy 1 -- --

That's not just a down-and-distance thing; MSU will break out the I-form or Ace in situations other than short yardage, but they're still really likely to run out of them. Ace formations tended to feature jet motion frippery.

There's also a very noticeable split in how plays are called by down:

Down Run Pass PA
1st 17 -- 11
2nd 12 9 4
3rd 5 11 --

MSU has started incorporating RPOs, mostly an inside zone read with a slant on the backside, so several of those play-action passes have a run element involved—which only serves to skew the playcalling even more in the run-run-pass direction.

If not for the fact they were winning almost wire-to-wire, the Spartans almost certainly would've had to abandon the run instead of stubbornly sticking to it. The maybe-you-shouldn't-have-done-that fourth down call by Dantonio was fine by game theory standards; leading by a touchdown, MSU needed a half-yard with the ball at the Iowa 30. This offensive line simply cannot run block, however. Here's second-and-one:

Here's third-and-one:

Guess what happened on fourth-and-one?

Hindsight is 20/20 but this stuff wasn't hard to predict with the power of foresight. No matter the running back, there weren't any yards available to get without considerable effort or a cheeky quick toss to the edge (and that worked precisely once before MSU tried to go back to the well and got stuffed). The tackles are both sub-300-pounders who have a hard time getting any push; left guard Tyler Higby is average at best and prone to getting knocked back; right guard has been a mess whether it's manned by David Beedle or Kevin Jarvis. Even Allen has had a rough go at times; he picked up a couple penalties against Iowa. It's hard to see how MSU will move the ball on the ground against M's defensive front.

Pass protection is also going to be an adventure. MSU is 86th in passing down sack rate despite attempting to gameplan around their offensive line. I didn't see enough bad moments from right tackle Luke Campbell to make him a sore spot; I also can't imagine matchups against Rashan Gary and Chase Winovich will go well for him. The other tackle, Cole Chewins, is perhaps generously listed at 284 pounds. He's 6'8". Mere physics dictates he be a sore spot and the film confirms—he can get pushed around. I also don't think these types of passes, which kept the chains moving against Iowa, are going to work when Don Brown is dialing up pressure from all angles:

Those who saw the Iowa game may wonder why wideout Felton Davis, who had nine catches for 114 yards and two touchdowns, isn't a dangerman. While he was certainly considered, he hasn't produced at that level in the other three games, and this felt like MSU exploiting a matchup—Iowa's Michael Ojemudia had a really tough game primarily lining up against Davis. While I need to see more before declaring Davis a star, he's impressive physically; he's definitely a jump-ball threat with strong hands, but I want to see if he can consistently get separation against good corners.

Lewerke's favorite target prior to last week was slot guy Darrell Stewart, who's bigger than your average slot bug and otherwise unremarkable. The other starting outside receiver, Trishton Jackson, was invisible against Iowa. Tight end Matt Sokol is a downgrade from Josiah Price as a receiver and, while it's tough to tell given the O-line mess, doesn't seem to be an improvement upon him as a blocker either. They do have some younger receivers, namely Hunter Rison and Cody White, who show some excellent athletic potential, but it seems like they're still getting the offense down—both have had limited roles.

I don't see MSU faring well against Michigan's defense. The combination of bad offensive line and lack of creative playcalling is bad news against Don Brown, and the traditional run game is flat-out moribund. Lewerke is going to make some plays, but unless he plays a mistake-free game—unlikely given his style—it's going to be tough to the Spartans to keep up.



October 5th, 2017 at 4:18 PM ^

We all know this - Dantonio treats this game as Tressel did.  Expect their offense to do a complete 180 based on what they've shown thus far in the season.  They are a team that will actually hold back a ton of stuff for Michigan.  A lot of fans speculate that teams do that, but Dantonio actually does it.


Brian made a great point last week about how Michigan probably wished they'd saved the Evans touchdown run for another game - Dantonio would have never ran that against Purdue.

Big Boutros

October 5th, 2017 at 4:30 PM ^

Agree -- my guess is the RPOs that have been sprinkled into the offense become the primary scheme against Michigan.

Lewerke's usage right now is at 56.1%, by far the highest in the Big Ten. The only other player in the conference over 50% is Chayce Crouch.

I don't think Dantonio is so callous that he will purposefully jeopardize his quarterback's health, but he won't protect him, either. There is no game that matters more to him. I would expect 20+ designed carries for Lewerke.


October 5th, 2017 at 4:39 PM ^

The entire season essentially rests on his legs. Weaver is not ready and Terry is not good.

I think MSU will selectively run him and use th threat of his left to throw deep passes which UM has shown some vulnerability to.

MSU has 7 more games after this. That cannot or at least shouldn’t overuse Lewerke in the run game Saturday.

Space Coyote

October 5th, 2017 at 4:41 PM ^

So I don't think that's the thing they'll turn to. Those are relatively stanard practice for them on first down now. What's more likely is true play action or actual drop back passing. What is likely is probably a lot more motions and shifts. What is likely is breaking tendencies based on formation, and things of that nature.


October 5th, 2017 at 5:28 PM ^

"Dantonio treats this game like Tressel"

"They are holding stuff back"

I imagine Don Brown is salty AF about Sparty. Probably, Harbaugh as well.

I suspect they are going to do what Michigan and the best defense in the country feel like doing. Cole Chewins gets abused. If they run Lewerke 20x he doesn't walk off the field.

Sparty can plan whatever they want. I'm sure they have no idea what Brown is doing or, for that matter Harbaugh.


October 5th, 2017 at 7:13 PM ^

"formations tipping plays" and "lack of creative playcalling" almost made me cringe because those assessments won't apply to this game.  They know their own tendencies, and have established them to counter them against us.  They'll have some very creative play calls and get some chunk plays.


October 5th, 2017 at 8:17 PM ^

Exactly, he spends time every week, all year, working on Michigan. The end result of all that was last year's opening drive that scored a touchdown and then .... they had nothing else until they started running out the clock on themselves and salting the game away whiel they slowly score meaningless points. If he spends that time on opponents he could have actually beaten last year maybe they would have one a couple more games. But Dantonio doesn't care and will gladly sacrifice wins in other games to do better against Michigan.


October 6th, 2017 at 6:20 AM ^

One reason why we can't play to our fullest against MSU is trash officiating.  Whether it's our linebacker getting ejected for being thrown to the ground, or our D-linemen practically getting their jerseys ripped off, or our D-backs eating PI penalties because they're getting picked, it's a good idea to mentally spot MSU about 2-3 TDs because relatively speaking, this will NOT be a good day for our defense.  Dantonio will brandish the bouquet of plays he spent all offseason working on and our defenders will be ticky-tacked until they're not sure if they can tie their shoelaces without a zebra throwing a flag.

The Maizer

October 5th, 2017 at 4:39 PM ^

On that 3rd and 12 scramble by Lewerke, Iowa's safeties are 12ish and 16ish yards deep at the snap.

Ace, are we sure that's a scramble and not a designed draw? Didn't really seem like he was reading coverage and he took off without any pressure (I mean there was a monstrous sea of defenderless grass ahead of him so running seems like a good plan).

Space Coyote

October 5th, 2017 at 4:39 PM ^

While he wasn't productive against Iowa, teams are looking to stop him now because I think he's a pretty damn good WR. Really good in and out of breaks, pretty good hands, attacks the ball well in the air, and picks up decent YAC as well. Honestly, I think if MSU has success on offense, it's because they can find ways to get the ball to Stewart. The run game will get some yards, probably mostly on gap schemes, but will likely be unspectacular. I don't think the other guys are experienced/explosive enough to be a great threat to Michigan's CBs (FDIII is the definition of JAG, in my opinion). Stewart can change the outlook, he's the wildcard.

I also think MSU's pass pro has held up better than this implies, though certainly with guys like Gary and Winovich on the outside, I suspect pass pro issues will become more apparent. The outside guys mirror well enough. In fact, the whole OL mirrors well enough. They just aren't very physical. So they can generally move the pocket or give some time to Lewerke to work, so maybe the pressure doesn't get home immediately, but it's going to get there still I'd think.

Space Coyote

October 5th, 2017 at 4:50 PM ^

As for Stewart, he had two catches on MSU's first drive. Was wide, wide open on his slant routes most of the night. Made an very contested jump ball TD on MSU's 2nd drive. Had a mundane catch that he was able to make enough people miss and force a face mask to get a first down. I'm not calling him an All-American, but I think he can get matched up on our safeties and be dangerous because he does a lot of things well to really well. He's a good player who I think would be playing on basically every team in the B1G at this point (not that the WR talent in the league is great).


October 5th, 2017 at 4:44 PM ^

Not to put you on the spot, but I both respect and enjoy your opinions about matchups, games, etc.

What do you think: 1) Michigan needs to do to win; and, 2) Do you think Michigan will actually do it?!

The more I read about this game, the more anxious I am becoming (probably irrational...but - hey - Michigan football, right?). So any thoughts you have about scheming, strategy, and outcome would be greatly appreciated!

Space Coyote

October 5th, 2017 at 5:06 PM ^

But I do believe MSU is a threat, mostly because I don't trust the offense's consistency and any major swing could be devastating. 

On offense, Michigan needs to execute on the outside to open up the rest of the offense I think. Get the ball out of JOK's hands, be consistent. Stay on pace. Get it into the flats on TE crossing routes and PA with bench routes. Utilize the slot to complete short/intermediate outs. When safeties start cheating it or the LBs start creeping outside, then you've opened up the D and you run the rest of your offense. Either you get some consistency or some chunk plays from the run game (I think Michigan can scheme some nice chunk plays out of the run), or you have to be able to attack the middle of the field with post and dig routes (from both the slot but also with outside WRs). Much of the ability to execute the latter will fall onto the OL not allowing free pressure so that a) JOK can actually step into throws and have time to make throws; and b) stay ahead of the chains so they aren't stuck in obvious pass situations.

On defense, press the pocket. I was quite unimpressed with the push MSU was able to generate in the run game. I think they are going to try to run more gap schemes against Michigan's 3 man front and try to gain leverage. Have to win the LOS to form a wall at the LOS. Pretty confident about that. Against the pass, push the pocket. Don't let Lewerke get his feet set. He is accurate and has good velocity when he can step into throws, but when he losses his footwork he loses a lot of his arm strength and accuracy. Don't let him break the pocket vertically, you've won if he's retreating out the sides of the pocket, especially with Michigan's speed. Then take advantage of plays. Michigan will be able to switch up some coverages and confuse Lewerke's reads/eyes. Make plays when they are available.

I think it'll be relatively low scoring for the offenses. I doubt either offense can put up points in the 30s. But I think the defenses can win field position and allow the offenses to score enough points. I think Michigan is better set up to consistently win field position and take some shots to get chunk plays, which is why I think they win.

Interesting to note, MSU's punter has been more consistent than Michigan's this year, and their kicking game as a whole is improved (not hard compared to the last few years), but that should make special teams more even. Michigan has been awful at kick returns this year. MSU terrible in kick coverage. That'll be interesting. 


October 5th, 2017 at 5:29 PM ^

Much appreciated. Personally, I am very, very interested to see how JOK performs in this game. I feel like if he can get into a rhythm with the TEs / Slots and put the MSU LBs on roller skates, it could be a fun night! If he can't, I'm probably going to punish my liver.

Thanks for your perspective!


October 5th, 2017 at 6:04 PM ^

Bush is going to spy him all day and blitz the guy into JAG or worse. Lewerke's worst nightmare is Don Brown unleashing Bush-McCray-Winovich-Hurst-Gary against that putrid as Sparty OL. Going to be a beat-down like Purdue. I fear for Lewerke's health.


October 5th, 2017 at 11:20 PM ^

I have this feeling that we are going to see MSU put a drive together that's about 15 plays and over 10 minutes, and will involve a lot of hair pulling 3rd down conversions on Lewerke's legs. I really do expect us to win, but I also expect a heart pounding game that is uncomfortably close.


October 5th, 2017 at 5:04 PM ^

Agree.  Looking at the chart:

Winovich will feast on Chewins.  They'll need to double with Higby to neutralize.

Hurst will crush whoever starts at RG, whether an injured player or a backup, so Allen will need to double here.

Finally, Gary will beat any freshman tackle, even if they're not marked a trouble spot.  They'll need to double with the TE or RB.

So basically, they don't have a single one-on-one matchup on the OL they can win.  Any time they try to block with 5, Gary or Winovich is going to be one-on-one against an underclassman tackle, which is a BIG WIN.  On the other hand, when they are blocking 6 on 3, our blitzers will be free and clear.

It is going to take a miracle for Sparty to move the ball with any consistency.  They'll get some chunk plays, but sustained drives seem incredibly unlikely.



October 5th, 2017 at 5:33 PM ^

My only concern - which I believe may be correctable by Don Brown as the game goes on - is if we are overly aggressive. Purdue's only success against us on offense was when our defense was essentially too aggressive. Similarly, it looks like on some of the videos Ace posted (above), MSU and Lewerke, specifically, had success with defenses over pursuing him. Given the size of MSU's tackles, I wonder if we see Gary bullrushing more as opposed to getting upfield (???)...Like I said though, I think this is correctable by Don Brown if it does happen.