duke football (seriously)

The other side of the ball from the Duke game that it's all like LOL you watched and cared about.

Surprise! Virginia Tech is aggressive. Let's go back to that Smart Football diagram of the VT defense:

This isn't always what VT runs—a lot of the time they have their front seven in an undershifted front just like Michigan—but the primary thing is that guy labeled "R," who is a "rover"; the "W" is the "whip," and they're both little lightning bastards equally capable of dropping into coverage or blitzing into your face.

As noted in the VEQ last week, corner/rover/nickelback Justin Fuller is a TFL machine with 14.5. This is the equivalent of Jordan Kovacs or Courtney Avery having 14.5 TFLs. Kovacs had eight. Mattison is pretty aggressive, but he's not Bud Foster. Here's Fuller shutting down the zone read:

That is consistent. The end will crash down the line and VT will fill with the rover or whip depending on which side of the formation the play is going to. This means a lot of man, or zones that devolve into man coverage based on keys a la a Saban defense. This is usually very effective. Sometimes it's not…

The secondary. Just Duke caveats apply, but dang they're good. They can get burned by double moves and the like…

…but it is rare to see that much separation. Former Michigan recruit Jayron Hosley has developed into a premiere cover corner and there's not much separating him from the rest of the secondary, including the safeties.

Even when Duke hit stuff VT made it difficult to execute:

While I'm sure there's an athleticism gap between Michigan's WRs and Duke's, Michigan does not have anyone you'd describe as a burner. Precision execution like that Michigan got against Ohio State will be required. The windows are small.

The key matchup: Denard versus fill safety. VT uses that safety fill even on conventional running plays, which allows them to crash hard to the playside without giving up the cutback. Usually, anyway. When the safety screws up there isn't help and runs can go long:

This is an improvisation from the tailback as he sees nothing on the front side of the play, but a conventional zone read pull puts Denard in that space by himself with that safety. If Denard manages to make him miss like the Duke tailback he's not getting tackled.

We saw Denard WOOP an Ohio State linebacker on his 42-yard touchdown against Ohio State; we've also seen him chopped down as he tries to make it outside too often. Getting those decisive upfield cuts that churn out 4/5/6 yards when the big play isn't there will keep the offense in a position to exploit that aggression. Losing two yards is a recipe for punting.

As for the inverted veer: VT runs it plenty, using Thomas as a plowhorse. They are likely to have a better plan for it than OSU did. I expect Michigan to run it a few times just in case it keeps working, find little success, and move on to other stuff.

Other key matchup: Borges pass routes versus Foster. Borges didn't exactly light up Foster in their earlier meeting. The high-powered Campbell-Williams-Brown 2004 Auburn offense eked out 16 points in a narrow win.

It's a different year, though, and a different version of the Hokie D. The 2011 edition is quality (13th nationally); it does not live up to the terrifying 2004 unit, which gave up 268 yards a game en route to finishing 4th nationally. That Hokies outfit had to go up against USC, then at the apex of their power, and Auburn. This one hasn't played anyone of note outside of a weak ACC.

So there will be more opportunities for Borges to wiggle his guys open with the sort of passing routes that Rodriguez never bothered to envision. The heavy emphasis on man and man-like principles leaves VT open to some pick plays:

That is like the circle route above. Both are instances where VT incorrectly diagnoses the route either pre- or post-snap and Duke gets a guy open with the play design. The frequency of Borges RPS+ routes will have a major impact on the efficacy of the offense. If you like bunches and that triple stack, I have good news: expect all of the bunches and triple stack.

Other other key matchup: Denard reading the defense. Here are a couple slants. The Duke guy picks the right one:

The one further inside is a pick waiting to happen thanks to a front seven member dropping out into a robber zone—a pick Denard has thrown this year. VT pairs its man-like principles with random guys in zones in the hopes of getting a big play. Denard managed to cut down his interceptions in the last two games (just one). Maintaining that improvement will likely lead to victory.

The line. In this game it was exceedingly young with three sophomores and a freshman. The two ends do get some pressure. Sophomores JR Collins and James Gayle combined for 13 sacks this year. In this game Collins—who seemed like the better player—smoked the Duke LT to force an interception:

I don't think he'll be able to manage that against Lewan. IIRC, Lewan has no pass protection minuses all year save for stunts on which there was poor communication here and there.

While Huyge is more vulnerable, the bet here is Michigan does not give up much pass rush against four men. If they do it will be a long day. I don't think Michigan's receivers are well suited to a quick passing game against a secondary with this many cover guys.

What about a conventional running game? Duke hardly ran and is very much Just Duke in this department, so I'll punt on that.

Takeaways. Virginia Tech's D in bullet points:

• Young, undersized line
• Extremely aggressive safeties
• Four solid cover guys
• Good open field tacklers
• Can get burned when opponents confuse their coverages

This will be a tough game for Borges and the offense.

So I downloaded and scouted VT's game against Duke. Literally everybody I've told this has laughed about the depth of my obsession, but there are good reasons for this. They are:

1. Duke runs a spread offense and even brings in an underclass athletic QB type from time to time, so this was a rare opportunity to see VT operate against a zone read.
2. The game finished 14-10. How does a team only score 14 on Duke? \
3. I could be reasonably assured this content would not overlap with that of BWS or Ace.
4. Hey, man, ACC Network action. Can't pass that up.

I have completed this process and will now go through a series of bullet points organized into offense and defense; this is not a FFFF so a more formal treatment of what Virginia Tech does will have to wait for that.

Apologies in advance for the video quality; ACC cappers are not up to the high standards of MGoVideo. Also, this game was originally broadcast in something called "standard definition" because the ACC Network is actually run out of Estonia.

Offense

QB Logan Thomas: legs. Thomas is a tank engine, more in the Tebow (huge and slow) mold than Denard (small and fast) or Cam Newton (huge and fast). His advantage in the run game comes when he can thunder straight ahead, which came on an assortment of inverted veers. This is one half of the Logan Thomas run game:

That sort of bulldozer running has the potential to neutralize Michigan's excellent third and short defense. If you don't have to move off the line of scrimmage to get a 6'6", 250 pound guy moving forward, a yard seems assured.

On the other hand, he's no Denard in space. This scramble picks up a decent chunk of yards but just lacks… oomph. This is against Duke, mind you, and Thomas seems plain slow:

While that tank thing still sees him pick up nine yards it's hard to imagine him getting any more than nine. Later in the game he'd break outside the pocket and lumber for the same nine yards John Navarre would have gotten in that situation. He's not a Miller-like threat.

Excise sacks and Thomas has 500 or so yards at 4.3 YPC. He's someone you have to account for on the ground, but he's not going to blow up for 100 yards in the Sugar.

QB Logan Thomas: arm. After watching Posey run circles around the Michigan secondary, Thomas's primary asset—an excellent deep ball—is a worrying one.

But wait, there's more, in the form of a 60-yarder to Danny Coale that hits his WR in stride.

We'll see if his uncanny accuracy over the top was a one-game phenomenon or a consistent thing as more games are added to the dossier. Extremely rough survey says: somewhere in between. Thomas averaged 7.7 YPA this year, which is above average but not spectacular. And Virginia Tech's schedule was not a high mountain—if you think the Big Ten was bad this year (and it was), the ACC was probably worse and VT's toughest nonconference opponent was East Carolina. East Carolina is trying to get fans to buy "virtual bowl" tickets, which are like real bowl tickets except there's no football game and the school selling them actually ends up making money. So… yeah. Not a tough schedule.

Anyway, if Thomas's distressing tendency to drop inch-perfect balls over the top of the secondary is stressing you out, this should help things:

He also forces things into places they cannot go, and does so without any semblance of a rush in his face. This shouldn't be overstated. Thomas had just 9 interceptions on the season. Sometimes, though, he unleashes the dragon.

RB David Wilson. Scary dude. There is no obvious analogue that pops to mind, but if you're thinking Anthony Thomas or Chris Perry you're as close as Michigan backs can come. Perry's probably the best comparison since Wilson has better balance than Thomas. He's also got a wicked stiffarm.

Those quick cuts that juke charging defenders are a regular occurrence, and you can see the size/speed combo that Thomas lacks. He'll be a load on a few plays where he bursts into the secondary. Concede his 120 yards and hope it doesn't hit 150 and it takes 30 carries to get there.

Offensive line. I just don't know about these guys. On the one hand, Duke rarely got a pass rusher within ten feet of Thomas. On the other, Duke's defensive line seemed to hold up pretty well. After running up a bunch of yards but not many points in the first half, VT got stoned more often than not in the second half.

On this play Duke's DE dives inside and gets pancaked by the tackle because he is a Duke DE and not very good; also I think the VT tackles may be solid. A big play threatens after Thomas cuts past a contain guy, but watch the backside guard give a ton of room, allowing the backside DT to run down the line and ankle tackle:

That would have me throwing a –2 at that G, and this was not an isolated incident. They couldn't convert a third and short to save their lives in the second half.

As for the pass rush, yes, "just Duke" caveats apply. VT is heavily run slanted (59%) as well. Still, they gave up only 15 sacks on the year. I did not look at this game in sufficient detail to say exactly why that's the case because the just Duke effect is at its strongest here—they're 93rd in sacks and 110th in TFLs. But to hazard a guess I'd say the tackles are good pass protectors and that's enough when you're rarely giving anyone else reason to believe you'll pass.

Receiving corps. I did not get much of a vibe in this game. A half dozen or more of Thomas's attempts were screens (frequently bubbles) and underneath stuff was usually swarmed. Danny Coale* got open deep for the 60-yard bomb linked above and then disappeared; Jarrett Boykin the 32-yard reception and then averaged 6 yards a catch on five others because they were mostly screens and dink hitches. I plead not enough data.

One guy who did make an impression was TE/H-back Chris Drager, a James-Rogers-like vagabond who bounced to defense and back over the course of his career. He was no backup, though. Drager started all but one game last year at DE. By all rights this should have been a signal of disaster but Drager has amazingly good hands for a guy who spent the last two years playing D. He made a few tough catches; he'll be a priority when VT is trying to move the chains on third and medium.

VT's backup tight end is a mess.

This may be useful in third and not-sneakable and goal line situations.

*[Who VT folk say will be punting(!) in the Sugar Bowl due to a dearth of other options.]

Keys for Michigan

Get to Thomas. Durrr QB pressure good. Yes. This is analysis just one step above the "score more points" school of Keys to Victory.

Three things make it better here. Thomas is not that mobile and if flushed out of the pocket is going to get some number of yards under ten unless something seriously wrong has gone down. He has a long, long delivery, which makes that window when he's made up his mind and has the ball in a dangerous position invitingly open. The Bank of Logan Thomas's Chest has extended hours for helmet deposits. And if Thomas is left alone to survey deep, the results will be not so good.

There are two schools of thought when it comes to mobile QBs: contain (think Iowa vs M) and attack (think MSU). Thomas is a guy to attack. The consequences of providing a running lane are less scary than those of providing time to survey and the chances of success are relatively higher.

Exploit the interior OL. I am of the opinion they kind of suck. When VT sprung a big play it was usually on the linebackers and secondary being slow or befuddled; several times Duke players made block-beating plays to hold down Wilson runs.

Have a centerfielder. You have to see if you can defend the VT run game straight up because you probably can, and then you have to take away that deep middle stuff Thomas can nail. I don't know if this is Woolfolk or Gordon after what we saw against Ohio State—but it's probably Gordon.

Gratuitous Appreciation of Duke Safety

The boom!

That is all. I just wanted more than six people to see that.