...is it wrong that every time i've refreshed this week, i have secretly hoped for the Sugar Bowl UFR?
things go poorly
The number one question about last year's offense was how much it would play to Denard's strengths and how much it would settle into Borges's comfort zone. The answer was mostly the former. While the first real test against Notre Dame was a rocky one and Michigan's under-center experiment against Iowa—against a Hawkeye defense that had just been plowed for a game-winning touchdown by Minnesota—was an outright disaster, those were outliers in a season that saw Michigan hardly budge from its shotgun-oriented ways under Rodriguez. The Sugar Bowl was a big fat raspberry at the end of things, granted.
What they ran from the shotgun was a lot different, but when it came down to the most important game in Brady Hoke's career to date—Ohio State—Michigan's primary gambit was the single most prominent spread play in the game today: the inverted veer, which marries power blocking to spread principles and gets you a lot of carries where Denard is charging hard upfield. The result was 170 rushing yards, a 167 yard, 14/17, 3 TD, 0 INT day passing, and 40 points against Oho State.
That seemed to work pretty well, right?
This blog tracked Michigan's success in various formations all year, and it wasn't even a debate except when the opposing defense was entirely theoretical (think EMU). Against mediocre defenses, the shotgun was far superior. Against good defenses, the shotgun was far superior. Various examples:
You get the idea. For the season Michigan averaged 3.9 YPC from the I and 6.7 from the gun. While ace (not that Ace) actually bested the gun's performance at 7.4 YPC, less than ten percent of Michigan's snaps were from that formation and they were heavily biased against good Ds—no ace snaps against ND or MSU, big chunks against Purdue and Iowa. One 59-yard Fitz run against Purdue explains most of that number, and that was some pretty inexcusable D combined with Fitz being awesome.
When the I worked it was usually due to opponents screwing up…
Three defenders to the left of center vs four blockers plus a FB = 8 yards
…or the tailback making chicken salad out of chicken despair, as in the clips from the Illinois game above.
SHOTGUN SHOTGUN SHOTGUN SHOTGUN SHOTGUNNNNNNNNN. Consider the line: Lewan, Mealer/Kalis, Barnum, Omameh, Schofield—all Rodriguez recruits who can move save the LG. Consider the QB: Denard. Consider the RB: Fitz Toussaint, space jitterbug. Consider the TEs: 404 file not found. Consider the FB: Stephen Hopkins, a guy who can reprise some of the MINOR RAGE if attention is drawn away from him and he's free to run straight at one guy. You've even got leftover RR slots in the WR corps. Just let it ride, man.
Next year is the year you flip over to your multiple pro-style whipsaw offense, next year when Denard is gone and maybe Toussaint heads for the draft and Kalis/Miller/Bryant is your road-grading interior OL and you've got TE depth and a panoply of different rushers for different situations. This year, stick with it and refine what works.
The spring game, which was almost all RR-at-WVU déjà vu 3WR 2RB shotgun set, indicates that's what the coaching staff thinks, too, as does the buzz I've gotten from The Fort. Now about using it a little more smoothly.
[after the JUMP: Borges fusion cuisine, yet more on DG at WR, stupid predictions.]
I promised I'd never talk about this again, but then you had to go and ask me. I know you're me. But let's not dwell on technicalities like that. Let's CEO this mother.
So my deal with bubble screens is not an obsession with that particular play per se. It's just that I got used to thinking about the Rodriguez offense, which was a collection of plays designed to work together that would make the defense wrong every time—at least in theory. Blue Seoul summarized this very well:
…the plays that [Borges] uses are more like "one offs". Whereas with RR, every play was really more like 3 plays that you could choose from depending on what the defense was doing. You can see it in the way that RR calls the plays after lining up. It's all about hitting the right note at the right time and making adjustments during the play. Borges sends in a play, sees what happened, and then sends in another play that makes sense in the sequence. …
Both styles work, and you can't ask one man to magically turn into another. BUT
GOD DAMN, Sometimes you just look at things like that and go "COME ON, AL!" And it doesn't have to even be a bubble screen. A zero yard hook probably gets a first down. Any kind of short combo route is wide open. The bubble probably goes for 20 yards, a Hook n' Ladder is a TD. But instead it's a slam into the LOS for no gain.
Michigan doesn't have the pieces to run Borges's preferred style yet, and they don't have the experience at coordinator to run Rodriguez's preferred style. This was a problem last year. Borges didn't take those edge opportunities, and against top-level defenses Michigan wasn't able to overcome the numerical disadvantage that often put them at. The 2010 offense had similar, if less severe, issues, but with an older quarterback and better talent surrounding him it should have gone the other way.
Some of this has nothing to do with philosophical differences between Borges and Rodriguez and is the natural result of putting a quarterback in a new system. Freshmen are not known for checking into plays at the LOS, and that's essentially what Denard was in the context of the offense. Checks were rare, checks to get the ball to the perimeter quickly nonexistent. Hopefully Michigan will realize they allowed too many defenses to align in ways they could beat without beating them:
This ended up being a Gallon tunnel screen for zero yards when just throwing it to the dude is a first down…. they gave you a first down by alignment. Take the first down! Don't throw screens into the box when there are extra guys in the box! Death to the tunnel screen!
Taking the free yards is important when you've got Denard. It's super easy to execute, gives you more shots at ripping off the big one, and when it's not there you've just made it easier to run your other stuff.
That said, Al Borges is not a return to the DeBord days. Michigan's last two drives against OSU were SHOULD HAVE SENT A POET moments for anyone who remembers punting from the damn 34 in 2005:
I want to focus on what happened in the fourth quarter. After the punt disaster Michigan gets the ball back on their own 20 up three points. Their drive goes like so:
- QB sweep for 3
- Waggle action rollout to deep, uncovered Hemingway for 20
- Zone stretch for 11
- QB draw for 10
- I-Form power for 5 on second and inches (+13 penalty yards)
- Zone read keeper for 3
- Inverted veer keeper for 6
- Waggle TE corner for 4 yards and a TD
the Boise State "just plays" philosophy? The TD was that incarnate.
So you've got this pro-set sweep thing with counter something something and what the hell is going on? Michigan hasn't aligned in that formation all year. It hasn't run anything like that all year. There is nothing for the defense to key on. They have no idea what's happening in front of them and end up so mesmerized Koger can declare his corner of the endzone Kevin Koger's Kogerland and hold elections without anyone noticing. President for life of Kevin Koger's Kogerland: Kevin Koger. First order of business: a motion to put six points on the board. Vote: unanimously in favor. Ratify that baby, Vice Exchequer Gibbons.
On the next drive Borges dialed up the Dileo corner route for 28 yards on second and eight. This is not a weak-tight calling station of a team anymore. A lot of Michigan's problems were failed execution on easy plays set up by Borges's philosophy.
The philosophy itself has proven itself on every level. I was checking out this great FO article on the rise of double tight end sets, which naturally made me think about the various offensive Swiss army tools the recent recruiting profiles have chronicled, and this also made me think about Borges:
Stephen Hopkins spent some time at the top of a many-wideout formation last year, and Junior Hemingway moved all over in an effort to get a better coverage matchup.
It'll take another couple years before the machine is a assembled. Once it is, Michigan can expect downright weird stuff…
…and a chameleon thing that's good to great. Michigan ran all sorts of business from Hemingway delay routes to that thing above to one of those "packaged plays" that Smart Football is making all the rage in the college football blogosphere and the Oklahoma State branch of the Air Raid tree is making all the rage in plain old college football. Long-term, we're good.
As for this year… you can't ask coaches to be different people. Borges will move a bit towards Denard's strengths but when the split second decisions come they will always be biased towards a comfort level established over 30 years previous.
First, a position paper on whether it's worth the risk to move Gardner to WR given his status as the backup QB: hell yes. Michigan has an excellent shot at the conference championship and should do everything they can to maximize that. Other QBs have managed it without too much damage—Texas A&M's Ryan Tannehill caught over a hundred balls during his first couple years, switched to quarterback mid-year(!), and became the eighth pick in the draft. While MarQuies Gray hasn't been as successful after a stint at WR, it's hard to judge his performance in the desert of talent that is the Gopher offense.
If Gardner is anything like the hype makes him out, ride or die. It's not like Michigan's likely to reach lofty goals if Denard goes out anyway.
As to Gardner's potential impact… aw, man, I don't know. His hands will be inconsistent. He'll make some leaping stabs at balls no one else on the roster can touch. He'll have defensive backs too close to him because his routes aren't polished, and he'll make that not matter so much. If I had to bet he's the leading receiver this year if only because the default mode of operation for some of the offense is "oh shiiii throw it up."
Despite the lack of TEs I don't think it will be a huge issue. Gardner provides a fade target, Hopkins and Rawls is a quality power battery, AJ Williams should be a good inline blocker from day one, and you can always give yourself an extra blocker by running Denard. Rawls + Hopkins + Williams means you just need one extra blocking type guy in your goal line stuff, and Brandon Moore should suffice there.
The OL will be able to pull better as well, and while Barnum to Molk is a downgrade the undersized Molk was not particularly effective on short yardage. They'll be fine here.
Just a notch. Here's a helpful scatterplot from The Only Colors charting sack-adjusted YPC and YPA against each other:
Michigan was one of four teams that stood out as effective rushers and one of three that stood out as effective passers. They return their quarterback, unlike the other two passing standouts (and fourth-place Northwestern), and that quarterback is also their ground game. They've got four guys back on the line and the only really worrying hole created by the departures is Guy Who Will Catch Jump Balls Guy. Gardner may fill that. With Wisconsin losing seven offensive starters and the entirety of their braintrust and MSU losing their entire passing offense, your early favorite for best offense in the league resides in Ann Arbor.
The open question is how much of that projected production will come against high-level defenses. You've heard it before, but when Michigan was bad last year they were awful…
Good hopping Lord in a pickle can.
I mean, sweet clod-kicking Jesus knickers.
That is just… something.
It was a—
Holy baboon-faced god of ancient river peoples spinning around on a pogo stick screaming "hey dilly dilly hey-o."
We get it.
…when it was good, though…
ALL OF THE POINTS
All of the points. Michigan had not scored that many points against Ohio State since a 58-6 whipping by Fritz Crisler and company in 1946. If you give the safety to the defense 2006 beats it and 2000 ties it, but then you've got the whole touchdown fiasco.
And what's more, that was a short game. Michigan had only ten drives. None of them were turnover-spawned and many of them were long. Michigan put up 460 yards of offense. Against Ohio State. In ten drives.
…it was really good. Kicking it up a notch is all on Denard and Borges. Denard needs to not turn the ball over. Borges needs to use Denard's legs more effectively when trying to set up play action. You'd think that would be a natural result of progress. The rest of it is about the same.
Let's hit shift and comma!
Yards per carry drop quite a bit but nose above 5.
Shaw claims the starting job to himself in week four, gets injured shortly after, and Toussaint takes over. Both are much better than Smith at making extra yards. At the end of the year they've all got somewhere between 400 and 800 yards.
Not quite. Toussaint and Smith fought for the job early with Shaw quickly relegated to garbage time, but Toussaint did emerge to take the job and was much better than Smith at generating extra yards.
Denard rushes for 1200 yards. His interception rate falls significantly but is still not great.
Almost exact and completely wrong.
Michigan runs more zone blocking than gap blocking. When they do gap block they are a left-handed team thanks to Taylor Lewan.
They did run more zone, almost all of it inside zone, but when gap blocking they preferred to pull Schofield since Omameh was really bad at it until late in the year.
Koger's production is up a bit but total TE catches only go up slightly: 20 last year, 30 this year.
Koger went from 14 to 23 catches; Watson and Moore added two more to get to 25.
Huyge gives way to Schofield mid-year.
Michigan finishes around 15th in FEI and other advanced metrics. By yardage they drop to about the same spot; scoring offense increases from 25th to match.
Michigan dropped from 2nd but only to 9th. Total offense fell a lot further, to 42nd—I underestimated the strength Ds that Michigan would oppose. Scoring stayed just about even at 27th.
...is it wrong that every time i've refreshed this week, i have secretly hoped for the Sugar Bowl UFR?
Hoping Roundtree will be in the "better" category anyways. They'll target him more, naturally, and hopefully he shouldn't come down with as many cases of the dropsies as before. I also think DG will be the leading receiver being such a big, safe target to check to. Hopefully Denard checks to his feet more this year, though.
Did we actually make it through this year's prevew without the dreaded Bo/Canham/Bump photo? Whew...
I think Taylor Lewan will be about the same himself, but the fact that Omameh can now pull means we will see more plays headed behind Lewan, which is obviously an improvement.
Although I recall from the OL preview that Lewan's run blocking was more uneven. He donkeyed sometimes and other times, he did not donkey.
But still, yes. Mealer and Barnum blocking down, Schofield sealing off the back side, and Omameh pulling to the left seems like a good plan.
Do you think Kalis, Magnuson, and Braden play this year? Part of the rationale being: they are going to have to play next year, so might as well get them a little seasoning this year.
I don't agree with Gardner/Robinson/Jackson/Chesson/Darboh == Junior Hemingway, at all. Some guys just have a knack for grabbing jump balls and I'd be shocked if any of our current receivers do it as well as Junior did this year.
I think a better description for the Taylor Lewan push would be "Still hates donkeys == Hates donkeys"...
And I also think the Hemingway replacements will be a downgrade - it's still too optimistic IMO to think that group can be a push considering how many times he bailed us out - Hemingway may well have been worth 2 of the wins (ND and VT) and was an eyelash from giving us an OT shot against Iowa.
1. Probably the most important thing a coach can do is maximize the talent out of the players you have. Kudos to Borges for recognizing how Robinson should be used and ending up with 11 wins and more points scored/game than the previous 3 seasons. In sum, Borges wasn't an idiot.
2. We tend to go under center more in short yardage situations, so there is some bias in the numbers. I like Drob in the gun, but in short yardage, we'll likely see less gun than in non-short yardage situations.