Thursday, September 30, 2010

Zone reading down the depth chart

After the Bowling Green game, I feel pretty confident saying that Tate Forcier is Michigan's unquestioned #2 on the quarterback depth chart, surpassing Devin Gardner who, coming out of camp, was reportedly first off the bench--something we frustratingly saw as he burned his redshirt against UConn. But against Bowling Green, Gardner showed a propensity to make bad reads on the zone read and generally have trouble gaining yards with his feet. Here's one of the more egregious examples of Gardner on the zone read:

Michigan is in a three wide set with Vincent Smith lined up next to Gardner in the backfield and Kevin Koger lined up as an H-back. This is the triple option read that Michigan runs a lot: a basic zone read play but if Gardner decides to keep the ball, he also has the option to throw a bubble screen to the slot receiver.

Right off of the snap, you can see that the weakside defensive end is being unblocked. This is the player Gardner needs to read. Remember, if he stays high, Gardner's job is to hand the ball off to Vincent Smith. If the defensive end crashes down on the hand off, Smith needs to pull the ball.

Gardner and Smith are still engaged right now and you can see that the defensive end hasn't committed to the halfback run. This is a pretty clear sign that Gardner should hand the ball off. If he doesn't, he needs to pull the ball out quickly and make a quick snap throw to the slot receiver who is rolling out into space with 10 yards between he and the closest defender. (This play--the triple option bubble screen--may be Tate Forcier's most consistent threat. He makes quick reads and solid throws to the slot receiver, a play that usually nets five to seven yards.)

Despite having great blocking for Smith, Gardner decides to pull the ball out and is staring down two defenders. At this point, he still has the bubble screen open, but not by as much. Instead, Gardner, with no blocking ahead of him, tries to juke through two defenders as well as the free safety coming down from his position.

This is how the play ends. As you can see, Bowling Green has Gardner boxed in. He won't be able to bounce it outside and he has two defenders on the inside as well. The most disappointing part of this play is that, as you can see at the bottom of the screen, Smith has a cannon shot through the offensive line with Koger about to block off the final defender.

Until Gardner can consistently make the correct read here, he won't surpass Forcier who appeared to make the right decision each time he took a snap against Bowling Green. I questioned Gardner's viability as a running threat, which I think was warranted, but part of the problem may also be that he's putting himself in really difficult positions like the one above. It looks like most teams are preparing to force Michigan quarterbacks to hand the ball off because of Denard. As such, making the proper read and handing the ball off when it's this obvious is a must.

Wednesday, September 29, 2010

Why Michigan struggles against rollouts

After the UMass game, I got on Greg Robinson's case because he more or less allowed UMass run a quarterback rollout pass at will on the defense. What I couldn't understand was a) why he didn't ever change his scheme to counter the play and b) why Michigan's defenders couldn't figure it out on their own; a basic zone coverage scheme is not susceptible to a rollout on every play, you'd imagine. Bowling Green obviously watched the UMass film and saw the same thing, going to a rollout a number of times in the game. Below is one example of what usually goes wrong, and something that hopefully they can fix. (For the record, the man coverage nickel package that I'm oh-so happy with will mitigate a lot of these issues if it's called at the right time.)

Michigan is in their typical 3-3-5 stack against a trips look for Bowling Green. Michigan is going to rush their three down linemen as well as Craig Roh from his linebacker spot (the stand up defender on the line of scrimmage). They will drop the other seven players into zone coverage. Bowling Green has called a designed roll out to the strong side of the field.

After the snap, you can see the Michigan rush as well as the zone coverage. Michigan will drop into a cover-2 look here with JT Floyd playing the deep weakside of the field and James Rogers (top of the screen) playing the deep strongside of the field. The linebackers, Cam Gordon, and Jordan Kovacs will all drop into underneath zones.

This is where the trouble starts on the play. Highlighted in red is James Rogers whose job it is not to get beaten deep. With three Bowling Green defenders streaking downfield (the third receiver is partially cut off at the top of the screen), Rogers starts heading deep. Highlighted in blue is Cameron Gordon. As the widest underneath defender, it's Gordon's job to defend the boundary. Kevin Leach is the playside linebacker and is between Bowling Green's two inside receivers. Ezeh is trying to get over to the playside of the field to defend. Since he's not there yet, Michigan essentially has two defenders (Gordon and Leach) against three receivers.

With Rogers deep and Ezeh still getting over to that side of the field, Gordon and Leach are covering the two inside receivers. The third receiver, whose foot is highlighted in red at the top of the screen, has recognized Michigan's coverage and is going to sit down in a vacated zone seven yards from the line of scrimmage. I'm not entirely sure who's at fault here: Ezeh for not getting over soon enough, forcing Leach to carry the inside receiver through his zone, or Gordon for not playing wide enough of a zone to attack the outside receiver. Either way, he's wide open. Bowling Green's quaterback sees it and throws:

The ball is highlighted in red and is headed toward the outside receiver who has about 10 yards of space between he and Gordon. Rogers still has coverage over the top, so that's definitely not his responsibility. Michigan has three players (Gordon, Leach, and Ezeh) within five yards of one another when the ball is released.

When the receiver catches the ball, Gordon is still three to five yards away from him. Leach and Ezeh are in pursuit but it feels like Gordon was out of position on this play. Otherwise, Ezeh and Leach would be further behind him. Gordon eventually makes the tackle but only after an easy seven yards for Bowling Green.

As I said above, I don't really know whose at fault here. It's definitely not Rogers who was playing the deep half of the field. I'm inclined to say it was Gordon's fault for not covering the boundary, but then again, Ezeh's late rotation (not entirely his fault, he got off the snap quickly and just had a lot of ground to cover) seemed to necessitate Leach covering the inside receiver through his zone and Gordon shading over to the middle receiver. UMass torched Michigan on this same play repeatedly and I have no doubt that other teams will as well. And as Michigan faces more talented quarterbacks through the Big Ten schedule, this looks to be the team's Achilles' heel.

The nickel package that Greg Robinson debuted in the Bowling Green game is one way to remedy this problem. Without gaping holes in the zone coverage, Bowling Green doesn't get this easy completion. Having man coverage here probably necessitates three vertical routes with two over-the-top defenders in Floyd and Rogers. I'd prefer this over pitch-and-catch completions, but Robinson is probably worried about the deep ball more. This is bend don't break football.

Bowling Green: What to hope for recap

The Bowling Green game was Michigan's last real cupcake opponent on the schedule and as such, they had the ability to try out a few new things and get younger players some game experience. I laid out what I was hoping to see last week and I figured it would be good to follow up to see how it all went down.
  • Most importantly for me (and probably most Michigan fans) was seeing playing time for the freshmen cornerbacks (Cullen Christian, Courtney Avery, and Terrence Talbott). Michigan got them all into the game and in the case of the latter two, they saw a significant amount of snaps. Avery and Talbott were both part of a nickel package that Michigan implemented frequently on passing downs. So not only was Michigan able to get playing experience for some of their young need-to-be-contributors, but they also debuted a man coverage nickel package that I had been begging for. Christian's performance was a little disappointing. He got beat a few times in coverage, though Bowling Green's quarterbacks weren't able to exploit it. One time in particular, Christian tried playing bump and run coverage when standing five yards off the line of scrimmage. It ended in a blown coverage, but it was a good learned experience. On the next play, Christian was in far better position and was able to jam his receiver at the line.
  • Snaps for Devin Gardner was probably the next most critical goal of the Bowling Green game. After Denard went down early, Gardner came into the game and closed out the first quarter, as well as played a little time in the second before being replaced by Tate Forcier. Gardner was, in a word, inconsistent. He had a few very nice throws downfield (about more which later) but struggled mightily on the zone read and looked overmatched occasionally. Once, he even reverted to that hideous sidearm throwing motion he uses when he's pressured. I was hoping that Forcier would play the third quarter and Gardner would be afforded the entire fourth quarter, but instead Forcier played at least one whole possession in the fourth, taking away a good number of snaps from Gardner. Then again, Forcier is clearly the team's #2 quarterback right now, so getting him more snaps is not altogether bad.
  • The defense looked improved albeit against significantly worse competition. I'm still far more pessimistic about this team's chances because of the defense than most around the Michigan blogosphere. While a lot of people are recalibrating preseason expectations up to 8-4 or 9-3 (the former, how I predicted the team would finish before the season), I'm actually being a little more hesitant and looking at 7-5. I hope I'm wrong. Regardless, the defense looked improved, got a number of sacks, and had a few nice pass break ups, one of which came on a quick slant that was defended expertly by Courtney Avery.
  • As mentioned above, Michigan took a few shots downfield. I knew that Bowling Green was a team that Michigan wouldn't struggle to move the ball against, so taking a few gambles down the field and potentially throwing away plays wasn't something Michigan really had to worry about. Gardner proved he's probably the best downfield passer on the team by throwing two perfectly thrown deep shots. Denard didn't really have a chance to go downfield, and Forcier's only long ball was an underthrown lob because of a scramble (that ended amicably in a defensive pass interference call). In any case, you can see Rodriguez sending outside receivers on fly routes down the sideline in an effort to keep defenses honest. This will be a big advantage in the future.
Those are really the main things I identified before the game. I was happy to see Will Campbell get some snaps in the game because he's someone who I've been advocating gets more playing time this season. If Michigan goes to more four-man defensive lines, I'd like to see him on the field more. And seeing Drew Dileo take back a few punts was a nice introduction to his skillset. In the end, the Bowling Green game is a tough one to pull many negatives out of.

Tuesday, September 28, 2010

Denard touchdown run

Ed. note: Sorry about the delay. I officially have internet again and can function properly. But on with the show.

You probably all remember the 47-yard Denard touchdown run early in the Bowling Green game, but there were a few interesting things that happened that you may have missed.

It's third and one early in the first quarter. Michigan is lined up in a max protect set with two tight ends in the game. This is a quarterback draw all the way. Bowling Green is in a 4-3 defensive front.

After the snap, Denard starts running left on the QB draw. (More after the jump)

Programming note: Evil internets

Due to a lack of internet for whatever reason the last three days--AT&T U-Verse is a very strange system--I don't have anything prepared for today and haven't been able to download the Bowling Green game. The internet fixer person is coming between 12 and 4 today and I should be able to get something up later. Apologies for the delay.

Monday, September 27, 2010

Bowling Green: Where a QB is a QB

Despite all of the horror of the 2009 Big Ten schedule and the relative uncertainty of the 2010 team, the last 16 games have done at least one thing: They have effectively demolished the memory of 2008's offensive ineptitude. No longer is the team having trouble finding a quarterback who can complete a pass, run the zone read, or having trouble getting the offensive line to know their blocking assignments. If the first four games of this season have proved anything it's that the offense has arrived. And it's more than just having the right players. The schemes are exact, the players know what they're supposed to do, and it is unquestionably lethal.

Caveats about Bowling Green's talent level aside, as a team, Michigan completed 23 of 26 passes for 255 yards (9.8 YPA), with two touchdowns and no interceptions. Almost more impressively, nine different players carried the ball for a total of 466 yards on 56 carries (8.3 YPC) and combined for 7 rushing touchdowns. The point is, Michigan has weapons. This will be the most effective Michigan offense of recent memory--there's really no debate about that anymore.

The real question heading into the Bowling Green game was the defense and how they would bounce back from a 2009-like performance against UMass. They responded... adequately. Despite the television commentators harping on Michigan's defensive troubles in the game, Bowling Green ended the game with 283 total yards, 71 of which came on a screen play that ended in a touchdown because of some bad tackling and poorer angles. They completely shut down Bowling Green's ground attack, which was promising. However, BG's backup and third-string quarterbacks were able to rack up a fair amount of yards through the air, despite not being particularly efficient (19-31 with 2 interceptions).

What all of this means for the rest of 2010 is still up in the air. The defense still looks really shaky and may cost Michigan a game or two that they should realistically win (I'm looking at you, Indiana). Otherwise, the offense looks good enough that winning shootouts shouldn't be a problem.

  • Obviously, a tip of the hat goes to Tate Forcier for his performance. After all of the drama involving Forcier this year, he stepped up and proved that he's still a legitimate threat in this offense. In fact, his performance may have solidified him as #2 on the depth chart. But we'll worry about that if Denard ever gets seriously injured (knocks on wood).
  • Speaking of QB depth chartin: Does anyone else think that Devin Gardner is severely lacking in speed? Obviously it's hard to compare him to Denard, but Gardner looks slow and lumbering in the backfield. He appears to have very little in the way of acceleration, and I'm starting to get concerned about his viability as a runner. He moves well, but this offense works best with a home run threat runner. Gardner does not look to be it.
  • Speaking of not a home run threat: Fitzgerald Toussaint. I really wanted to see Toussaint get a few carries in the game and while he proved himself to be productive (66 yards on two carries), getting caught from behind, and convincingly so, by a Bowling Green defender on his 61 yard carry was not encouraging. For someone who was being heralded as the next great Michigan running back, that was not a great introduction.
  • Will Campbell looked nigh unblockable on the goal line. I continue to push for more playing time for him.
  • It was good to see Terrence Talbott, Courtney Avery, and Cullen Christian get a lot of snaps during the game. The other encouraging sign was that Greg Robinson called a lot of man coverage. It returned mixed results, but the fact that he recognized the issues with the defensive zones is encouraging.

Thursday, September 23, 2010

Bowling Green: What to watch/hope for

With Michigan coming up against what should be a blow out against Bowling Green on Saturday, the team should have an agenda that goes beyond just winning. This is the last game to get experience for some of the younger members of the team before entering Big Ten season when the need for depth will be even more essential. Without further ado:
  • Devin Gardner needs to take a significant amount of snaps. It would've been nice if he had the chance to do so against UMass but we all know that wasn't possible. If Gardner really is #2 on the depth chart, he needs to have some relevant game experience if he's going to be effective. If Denard goes down for any significant period of time, Michigan doesn't have to defense to sit and wait for him to get healthy. They need Gardner to be able to pick up where Denard left off--I know this is unrealistic, but there can't be as significant a falloff as it seems there would be now--without committing countless turnovers and negative yardage plays. Ideally, Gardner starts the second half of the game, but chances are Denard plays at least a few series in the third quarter.
  • Speaking of playing time, Fitzgerald Toussaint, Cullen Christian, Taylor Lewan, Terrence Talbott, and Courtney Avery all need to see the field in this game if we expect them to contribute this year. If Michigan has a decent lead early, I'd like to see the young corner backs get on the field when they're not just doing mop up duty. Lewan will very likely start the game given his performance against UMass, and Toussaint (and the other tailbacks for that matter) needs to get some reps before the Big Ten season starts.
  • Some sort of resurgence on defense. Everyone in the program said that they weren't mentally prepared for the game against UMass, and let's hope that's the case rather than the defense being as poor as it appeared. Bowling Green will be playing without their starting quarterback, and Michigan should be able to stop them fairly easily. If Bowling Green starts putting up points in this game that don't come off of fluke plays (e.g., Hagerup's dropped punt), we need to go into serious panic mode: 2009 looms large.
  • Continue taking deep shots down the field. Bowling Green is the kind of team that Michigan should be able to pick up yards with relatively easily. As such, I'd like to see them take a few more shots down the field like they did against UMass. Denard needs to develop as a deep ball passer and this is going to be one of the last chances he has to do so without too many repercussions (knock on wood).
  • Maybe most importantly, seeing a bounceback performance from Obi Ezeh and James Rogers and the rest of the defense that played poorly against UMass. More than trying to draw conclusions about how good they are in the long run--probably not very good--having confidence and a solid performance under your belt before heading into Big Ten play is for the best.
I know everyone is saying that this year feels different than last because of Denard and so on and so forth, but I can't shake the feeling of last year's 4-0 start quite yet, especially not after that UMass game. Michigan has a lot of questions and holes still and Saturday's game is the last time that they'll have a chance to test out new answers. Until Michigan has a "6" in the wins column, I won't feel comfortable with this team. Maybe a solid performance on Saturday will temper that feeling, but for now, I'm still in panic mode.

Tuesday, September 21, 2010

Corner blitzin'

I mentioned in yesterday's game column that Michigan was content playing the same defense against UMass the entire day. They didn't really show any new blitzing schemes, defensive fronts, or adjustments to try and stop what was a pretty potent UMass attack--at least potent against the shoddy defense Michigan was throwing out there. One of the specific problems I found was this:

Never blitzing a corner on Totally Surprising Quarterback Rollouts that were gashing the defense. It took Michigan until UConn's last offensive series to blitz a cornerback on the one play that UMass ran repeatedly to astounding effect. Over and over again, UMass quarterback Kyle Havens rolled out of the pocket and hit any one of various free receivers downfield that were scampering through the Wolverines' porous zones. He completed 22 of 29 passes and produced a better performance on Saturday than he did the previous two weeks against William and Mary and Holy Cross.

It took a while but I finally found the play when Michigan blitzed Floyd: it was on the Cam Gordon interception/fumble. And while Floyd's blitz wasn't the only reason Havens threw the pick, I'd venture to say it went a long way to causing it. Let's take a look at the film.

With 10 minutes to go in a  42-24 game, if Michigan gets a stop here, they very likely put the game to bed. The final result of this play is really disappointing because if Cam Gordon had held onto the interception this game is probably not quite so terrifying, but alas. Michigan is in their regular 3-3-5 set here and is bluffing a blitz on the strongside of the field. UMass' slot receiver is about to motion across the formation and will end up on the strongside of the field. UMass, as they did much of the game, has one tight end in and a single half back.

Right before the snap, you can see JT Floyd starting in on his corner blitz. The strongside blitz that Michigan was showing pre-snap will pull off the line and drop into coverage. Now, you're probably asking: What about UMass' uncovered defender at the bottom of the screen. Well UMass had been using weakside receivers as the final option on most passing downs and having the strongside receivers typically run drag routes across the field, a few yards behind the Michigan's linebacker zones. So the receiver that Floyd was cover pre-snap would not be a hot read if UMass QB Havens gets into trouble.

At the snap, Michigan rushes three--with Floyd coming off the corner--and drops the linebackers into coverage. UMass' linemen are blocking one-on-one with the defensive ends and double teaming Mike Martin. The running back will be charged with blocking Floyd.

At this point, the left guard realizes that Michigan was bluffing the blitz and doubles Ryan Van Bergen. The rest of Michigan's linemen are tied up on their blocks. The running back sees Floyd's blitz and steps up to block him. This is a critical point for Floyd: As you can see, theres a bit of a gap between the right tackle and the running back. If Floyd takes this route, the half back can simply push him through the pocket and Havens will have a free release out of the pocket. Instead, Floyd stays disciplined and continues to rush to the outside of the pocket.

Michigan's secondary has covered well enough that Greg Banks is getting off of his block and heading into the backfield. Martin is about to do the same. One reason Michigan was able to cover so well on this play was that they blanketed the strongside of the field with defenders. Floyd is continuing around the outside of the pocket and Havens has no idea he's there.

Now Havens sees the pressure coming up the middle and tries to roll out of the pocket like he had been doing all day. But instead of just having a free wheel release, when he turns his head, Floyd is rushing around the edge. Though there is a hole, Havens is forced to cut back upfield. On film, you can actually see this bother him.

Havens has now cut through the hole and is being pursued by Michigan's linemen. You can't see in this picture, but five yards beyond the line of scrimmage, Jonas Mouton is waiting for Havens and covering one of his receivers; he either has to pass it or rush for a minimal gain. Havens decides to pass the ball very late in the play and overshoots his receiver who is dragging across the play and being nominally covered by Obi Ezeh downfield.

With Havens scrambling out of the pocket and under durress, he throws up what will eventually be the Cam Gordon pick.

The point is this: When Havens was given a free release out of the pocket, which was often, he was usually able to find one of his receivers streaking through Michigan's zone. But because they sent a different look at him--Floyd coming off the edge--he had to readjust where he was able to throw the ball. Much of UMass' offense on Saturday was predicated on getting Havens to the edge and allowing him to work Michigan's zones. But with the corner blitz, he was unable to do so.

Again, if Gordon hangs onto this interception, the game is probably over in unceremonious fashion. But more importantly, this was a play where Greg Robinson dialed up something unexpected and it should've ended well for Michigan.

Monday, September 20, 2010

UMass: Where owls

At some point during Michigan's win against UMass on Saturday, I stopped being terrified. I knew how the game was going to end. I watched it for four years while I attended Michigan, the last four of the Lloyd Carr era. Eventually, UMass was going to make it close, Michigan would curl up in the fetal position (read: prevent shell), and Michigan would pull it out in a terrifyingly pedestrian fashion. When Denard Robinson lined up under center in the I-formation on third and one with five minutes left in a two possession game, everyone in the stadium knew what was going to happen, including a UMass linebacker who blew into the backfield untouched and tackled Vincent Smith for a three yard loss. Somewhere, Lloyd Carr was sitting in a plush leather chair and smiling.

But Michigan's offense wasn't the problem. They scored 42 points, racked up 525 yards, only turned the ball over once, and generally looked like the behemouth we've all come to expect despite not debuting any new formations or plays. Michigan--or maybe just Denard Robinson again--did what they were supposed to do despite Rich Rodriguez's recent urge to run half back dives at the end of games. Obviously, the bigger concern of the game was Greg Robinson's refusal to do, well, anything. Among the biggest, obvious complaints:
  • Never blitzing a corner on Totally Surprising Quarterback Rollouts that were gashing the defense
  • Never playing man coverage
  • Never playing a 4-3 defense
  • Think. Adapt. Think
But let's handle these one by one. Never blitzing a corner on Totally Surprising Quarterback Rollouts that were gashing the defense. It took Michigan until UConn's last offensive series to blitz a cornerback on the one play that UMass ran repeatedly to astounding effect. Over and over again, UMass quarterback Kyle Havens rolled out of the pocket and hit any one of various free receivers downfield that were scampering through the Wolverines' porous zones. He completed 22 of 29 passes and produced a better performance on Saturday than he did the previous two weeks against William and Mary and Holy Cross. This defense is just as awful as we thought it was. The only hope (the only hope) is that Robinson was told this week not to show any new formations or fronts. That seems unlikely. Eventually, they blitzed JT Floyd off the edge and it ended in a broken play that went UMass' way, which, fine, that'll happen. But Robinson's complete and utter inability to make adjustments is terrifying. This was the worst called defensive game I've ever watched.

Then again, Michigan's defense might not be so anemic if they could play man coverage. Again, I think it took until UMass' last offensive series for the Michigan defense to show man coverage. UMass receivers were devouring Michigan's zones. I understand that the team has little to no depth in the secondary and putting true freshmen like Cullen Christian out there is a harrowing thought, but was anyone else watching this game?

Of course, the passing game wasn't the only thing that Michigan couldn't defend. They were being shredded on the ground too to the tune of 217 yards and 4.4 YPC. Part of this was because of Robinson's insistence on playing a three-man defensive line that was repeatedly getting pushed back. I understand that the coaches are upset with Will Campbell's conditioning. I get that. But to never show a 4-3 front when it's obvious to everyone watching that the defensive line is a) being put on skates or b) slanting away from the play because they need to get into the backfield somehow, well that's just willful ignorance and stubbornness on behalf of the coaches. Campbell needed to be on the field. He has the size to contend with UMass' offensive line, something that the defensive front desperately needed.

Finally, THINK DUDE. Do something. Please. Robinson seemed much too content sitting back and watching as UMass trounced our piecemeal defense. He didn't show anything new. He didn't seem overly concerned with the beating, either. We all assumed that the defense was going to be bad, and when they seemed to show improvement in the first two games, Michigan fans had every right to be excited. The problem is, we all assumed the problems on defense would stem from personnel, not necessarily awful play calling, an utter lack of adjustments, and frankly, the scheme. It's too early to bury this defense yet, but the grave has been dug.

But a win is a win, and after the last two years, we've all learned to appreciate the good, even when it comes with the most ominous rain clouds. Michigan should have a much easier opponent next week against Bowling Green, but you never know anymore. The team is going to put this game behind them. I suggest you do the same. It'll help you sleep easier. But not much.

Friday, September 17, 2010

Notre Dame and snap counts

Ed. note: In lieu of a UMass preview, something that, unless something awful happens, should be useless, I wanted to do a little breakdown of Notre Dame's defensive line shift that was effective because of Michigan's predictable snap count. For a comprehensive UMass preview, visit MGoBlog.

I mentioned in Wednesday's Good Idea, Bad Idea post that Michigan is continuing to use a predictable snap count. When I was watching the game live, I saw Notre Dame shifting their defensive front right before Michigan snapped the ball, and finally found some video of comparable plays that shows that it may actually be more detrimental to the team's blocking schemes and production--aside from the obvious: allowing defensive teams to time blitzes--than originally thought. First up is a play in which Notre Dame uses the defensive front they originally showed:

Here, Michigan is set up with Denard in the shotgun and Michael Shaw to his right. They're using an H-back (Martell Webb) in this formation. Notre Dame is in the 4-3 base set that they had shown for most of the game. This will eventually be a passing play, but the important thing to watch is the blocking of the offensive line.

After the snap, Denard rolls to his right. David Molk and Patrick Omameh double team the strongside defensive tackle. Perry Dorrestein has one-on-one blocking against the strongside defensive end. Mark Huyge is charged with blocking the weakside defensive tackle and Steve Schilling is being given a free release into the second level. Martell Webb is supposed to get to the second level and block one of Notre Dame's corner backs as this will eventually be a counter screen pass to the weakside of the field. The weakside defensive end is unblocked.

Regardless of how this play ends, you can see that the offensive line is tied up with their appropriate defensive players. Huyge has passed the weakside defensive tackle to Schilling and is releasing to help block on the screen pass. Omameh and Molk are still driving back one of Notre Dame's defensive tackles. How the play ends is beside the point (it ends poorly, for the record). But you can see that there's order. All of the offensive linemen know where they're supposed to be and who they're supposed to block.

In the next play, Michigan is in the same formation and, at least initially, Notre Dame is in the same defensive formation. But because of Michigan's predictable snap count, Notre Dame was able to change their defensive front just before the snap and confuse the offensive line.

Here you can see that Michigan is in the same alignment they are above. Denard is in the shotgun with Michael Shaw to his right. Martell Webb is in as an H-back. Notre Dame is showing the same 4-3 defensive front. But because they're able to time Michigan's snap, the defensive line shifts from a 4-3 to a 3-4, changing all of the blocking schemes for the offensive line:

As you can see now, Notre Dame has shifted into a 3-4 front with the weakside defensive end becoming a stand-up linebacker, and the rest of the defensive line shifting over so that now they have a nose tackle and two defensive ends. Worse still is that all of this happens with Molk's head down, waiting to snap the ball; when he picks his head up, he'll be greeted with a completely different defensive front.

Because this is a different play than the above example (this is a designed QB run), there will be some differences in how it plays out. But the important thing to watch is the offensive line. If Notre Dame was in a 4-3 front still, Molk would've reach blocked the strongside defensive tackle with Omameh, Dorrestein would've seal blocked the defensive end, and Michael Shaw would block the playside linebacker. Instead, Molk goes for a reach block on the defensive end who has taken place of the defensive tackle. Omameh looks to double team the defensive end and Dorrestein, who was able to see the changing blocking schemes, also hits the defensive end and then releases to the second level.  Long story short, there are three linemen blocking a single defensive end. The nose tackle is being left unblocked by the front line and is now the responsibility of Webb, who is supposed to release to the second level to block one of the middle linebackers (about more which later).

The next thing you see is utter chaos. Molk and Omameh and still tied up double teaming the defensive end. Dorrestein and Schilling have released to the second level to block linebackers, and Huyge is tied up with the weakside defensive end. Shaw is locked up with the playside linebacker. The nose tackle is going completely unblocked and is able to, along with the playside middle linebacker, keep contain on Denard and force him back into the defensive help.

More chaos. Denard has no where to go and is being forced back into unblocked defenders. You can see the nose tackle in the middle of the scrum without a single Michigan player looking at him.

Denard has to cut the play back, and runs into the defensive end that Omameh and Molk were supposed to double team. The nose tackle still hasn't been touched.

This is admittedly a bit of a convoluted example, but the evidence is clear: when Notre Dame was switching defensive alignments just prior to the snap, the offensive line was confused, missed blocks, and didn't know its assignments. We've already seen Michigan State time Michigan's snap count to devastating effect, and if Michigan doesn't change their count, problems like this will continue to arise. Notre Dame has question marks on defense and was able to confuse Michigan's offensive line. Similar moves against Wisconsin, Ohio State, or Penn State could crush Michigan this season.

Thursday, September 16, 2010

Running backs and reads

After the Notre Dame game, Rodriguez was asked a lot about the amount of touches Denard had and whether or not he wanted to see the running backs get involved more often and be more productive. His response was that Notre Dame was doing a lot of things that forced the ball into Denard's hands, which, huh? After watching the game over and over again, I found a few plays that the running backs were actually given the ball (it only happened 13 times all game) and saw that this may not have been the case.

Example 1: Notre Dame neutralizes zone read
Michigan is in the shotgun with Michael Shaw to Denard's left and Martell Webb set up as an H-back. This is a simple zone-read. On this play, Michigan will leave the weakside defensive end unblocked for Denard to read.

After the snap, you can see that Mark Huyge is headed straight for the weakside defensive tackle and leaving the defensive end unblocked. Martell Webb is charged with blocking the weakside linebacker who was shaded over the slot receiver. Denard has to read the weakside defensive end. If he crashes down on the handoff to Shaw, Denard is supposed to pull the ball himself. Notre Dame is also blitzing the middle and strongside linebackers in an effort to plug the middle of the offensive line.

As you can see here, the defensive end is staying high, forcing Denard to hand the ball off. But the Notre Dame linebackers have engaged with the offensive line before they were able to get upfield. If Michigan's linemen were given a free release to the second level, they're fast and smart enough to make the block. But instead, Notre Dame's linebackers were told to plug the holes at the point of attack and make Shaw slow down at the line and pick a hole. In doing so, it gives the weakside defensive end enough time to crash down for the tackle.

Shaw is now headed upfield but there aren't any clear openings because of the blitzing linebackers. Notre Dame's defensive end has now read the play and is crashing down on Shaw. If there were a clear hole at the line of scrimmage, Shaw blows by the defensive end and on to the second level. Unfortunately, there's not, and this happens:

Shaw gets wrapped up by the defensive end that's never supposed to tackle him if Denard makes the right read, which, for the record, he did. In this instance, Notre Dame effectively neutralized the play by blitzing their middle and strongside linebackers. But if they did this too often, a play action pass could very easily burn the Notre Dame defense as Michigan did earlier in the game. It's a chess game and Notre Dame won this time. Shaw doesn't get off without any blame in this game, though.

Wednesday, September 15, 2010

Good Idea, Bad Idea: Burning redshirts

Rodriguez has come in for a lot of praise on this blog already this season. I've been an ardent supporter of Rodriguez since he arrived in Ann Arbor and seem to always look for the best. That's probably no more apparent than in the last two weeks where I've basically hailed his playcalling as unheralded brilliance. But that's not to say I agree with everything the team's doing, so without further ado, the first installment of Good Idea, Bad Idea:

Good idea
  • Hiring Greg Robinson. So maybe this is premature, and given the majority of our sample size (2009), this should probably go under the Bad Idea header, but the improvement that the defense has seen this year feels like it's due in large part to Robinson. Aside from the schematic changes that appear to be for the better, the work that Robinson has done with Stevie Brown, and more relevantly, what he appears to have done with Michigan's current linebackers has helped quickly turn around a talent-barren defense. By the season's end, this may very well flip to the negative side of things, but right now Michigan is coming up all spades.
  • Playcalling. I've mentioned this most often on the blog, but Calvin Magee and Rich Rodriguez have been calling brilliant offensive games. Look no further than the Roy Roundtree touchdown and its setup as proof. Other plays that stick out are the Vincent Smith screen pass touchdown against a fierce UConn blitz and Denard's various QB draws (like the intentional--make no mistake, this was intentional--QB draw on 3rd and 15 deep in Michigan's territory, that earned a first down).
  • Burning Devin Gardner's redshirt. Despite the obvious gains that the program would see from having Gardner around for a fifth year, it's difficult to look at the way the offense functions under Denard and not understand the need for a similar athlete as a backup. Getting him time on the field is going to be essential if Michigan wants to have a real insurance policy should Denard go down.
  • The Tate hug.  Before the Notre Dame game, the cameras panned along the Michigan sidelines and focused on Tate and Rodriguez for a moment as the two shared a pre-game hug. Despite this feeling staged--you couldn't have done that in the locker room or before the game or, ya know, not on national television?--with all of the rumors about the relationship between Tate and Rodriguez, this put the chatter to rest.
Bad idea
  • Burning Devin Gardner's redshirt (SeewhatIdidthere?). The most groan-worthy moment of the season thus far has been watching Devin Gardner come onto the field after Denard took a hip stinger, only to be a nonfactor and mindlessly waste his redshirt. It's understandable that Rodriguez feels the need to win this season, but Gardner is not the kind of athlete that comes along every year. Having him for two seasons after Denard and Tate are gone would've been highly preferable to the one we're going to get.
  • The predictable snap count. MGoBlog has been lamenting this for two years now, but Michigan's snap count continues to be problematic. UConn didn't really exploit it, but Notre Dame did. After Denard drops into the shotgun, he lifts his leg and systematically snaps the ball after it. Notre Dame was actually changing their defensive front before the snap, which I presume changes the reads for Denard. Fortunately, Michigan didn't really get beaten on this, but better defensive teams will be able to key in on this in the future.
  • Whatever Rodriguez is doing that's making players transfer. There has to be something going on here that's making these players transfer at such a high rate. Maybe not. Maybe this doesn't fall on Rodriguez. But it feels like it does.
For posterity

Tuesday, September 14, 2010

QB stretch and the Roundtree touchdown

With Michigan down 7-0 early in the first quarter against Notre Dame and showing little in the way of offensive production, Rich Rodriguez used a little trickeration to get the Wolverines on the scoreboard. At this point, you probably know how Roy Roundtree ran completely unabated to the endzone after a Denard Robinson play action pass, but the way in which it developed is a little more intricate and shows a lot of foresight by Rodriguez.

The setup
Michigan debuted a play that I've referred to as the QB stretch, where, rather than following his lead blocker and running between the tackles, Denard would follow his lead blocker outside the tackles. Michigan is in a 3-wide, 1-back set with tight end Kevin Koger lined up on the weak side of the play. Notre Dame is in a basic 4-3 coverage. One of their linebackers is shaded over Michigan's receivers and the playside cornerback is playing 8 yards off the line of scrimmage.

On the snap, Michael Shaw heads outside of the tackle to block the middle linebacker; Michigan is purposefully leaving the playside, outside linebacker unblocked. David Molk and Steve Schilling quickly double team the playside defensive tackle and Schilling releases to the second level to block the weakside linebacker. Notre Dame's safety and cornerback are dropping back into coverage because the playside linebackers are spying Denard and sitting on the run.

This is the brilliant part. After the Roundtree touchdown, one of my friends questioned whether or not the play was intentional or just a broken play. As you can see in the image below, none of the receivers are looking to block anyone downfield, something that they do excellently on almost all other QB draws. Here, they're just running three vertical routes against Notre Dame's two defensive backs. At this point in the play, Michigan isn't concerned with gaining yards. This was a fact-finding mission. If Denard breaks this play for good yardage, even better. But right now, with the outside linebacker (standing on the 35 yard line between Michigan's receivers) already selling out on the run, Rodriguez knows that the next time they show this play, they can score with a simple play action pass.

Despite Robinson basically being wrapped up here, this play can be considered a net win for Michigan. They have three receivers running free and Rodriguez knows how to attack the defense. (The touchdown after the jump.)

Monday, September 13, 2010

Fortune cookies

2010 has been undoubtedly the most difficult year of my adult life. In the span of just a few months, my best friend was diagnosed with cancer, my long-term girlfriend left me, Michigan sports were dissolving, I was increasingly unhappy with my job and friends, and I encountered various issues with my car, taxes, and other pitfalls that a poorly paid editor is unfit to deal with financially. At one point, I had a running count of “wins”; 2010 is holding a commanding lead over me. Since then, I quit my job, moved to Los Angeles, have attempted to get over said ex-girlfriend, and have scraped together enough money to temporarily deal with said tax and car problems.

I lost my first grandparent when I was in seventh grade. My parents woke me up before school, pulled me into the living room, and told me that my grandfather had passed away. He had been in the hospital because he fell off a ladder trying to clean the gutters of his house, and his health worsened after surgery. Of my four grandparents, I was probably closest with him, or at least enjoyed his company the most. He showed all of the grandchildren how to play solitaire and would carve intricate wood sculptures for all of us. His funeral was a solemn affair; it was the first funeral I had ever been to. Before the service, one of my parents, unbeknownst to us, went to the church organ player and requested that as the casket was leaving the church, she play “You Are My Sunshine”. The organist was confused and not entirely sure it was appropriate. My grandpa used to play “You Are My Sunshine” for the family on something that’s like an accordion but not quite an accordion.

Last year was supposed to be the year that everything came together for Rich Rodriguez after the disastrous 2008 campaign. There was Tate Forcier, Weapon of Choice. Next came Western Michigan and Notre Dame. By mid-October, a month later, what had looked so promising was a hollow shell. The team was in ruins and needed to start again next year.

The weekend that my ex-girlfriend and I broke up was difficult for me. We had been making a long distance relationship work (she in Los Angeles and I in Cleveland) for nearly 15 months and things had started to splinter. I flew to California knowing full well that we had to have a Talk and that it might very well end with us not being together anymore. We broke up shortly after my plane landed. On the last day that I was in California, she and I were getting dinner together at P.F. Chang's--yes, in Los Angeles, where there are certainly better Chinese restaurants, but I digress. After dinner, I cracked open my fortune cookie and it told me, “Remember this date. Three months from now something good will happen to you” (or something to that effect). I’m not superstitious, but this was obviously a good sign as I wasn’t the one that initiated the breakup. A month and a half later, we were having a discussion about her new boyfriend. We stopped talking shortly after that.

In two games, more than leading the Wolverines to impressive, imperative victories, Denard Robinson has inspired hope in the Michigan faithful. After two years of endless piling on, there seems to be light at the end of the tunnel. People said that last year about Tate Forcier, but this feels different. You can see it in the childishly playful way Brian Cook, leader of the Michigan faithful, Tweets or the two-part comic strip that Denard’s UConn performance pulled from the depths of his crazies. Or in Denard's brilliant smile that everyone who’s seen in him a press conference raves about. Denard is the promise of Pat White, 11-1 conference-winning seasons, contending for a national championship; everything that people have come to know Rich Rodriguez for seems to be possible with Denard. Whether or not that comes to fruition is beside the point. Right now, there’s hope.

Through a collection of mutual friends, I now watch football games with my ex-girlfriend. I spend most of the day trying not to look in her direction--which isn’t too terribly difficult as I’m glued to the TV, yelling at random linebackers to tackle people--due in large part to my not being totally comfortable with her there for various reasons. After Denard's performance in the Notre Dame game, we shared a two-handed, over-the-head high five.

Notre Dame: Where it feels like 2009... only better

My recent relocation to California has meant a number of things: an absolute drain of my bank account, the general tardiness of this blog as I search for jobs (e.g., this post), and the meeting of my friends' friends. After the UConn game last week, one of said strangers commented to our mutual friend, "That Chris kid is really intense about football." "Intense" is a nice way of saying, "I don't want to watch football with him anymore," I think.

In the first quarter of the game against Notre Dame, Michigan started running a new QB draw--which, for the sake of continuity, I will call the QB stretch--wherein Denard would follow his lead blocker and run outside of the tackles instead of between them. Notre Dame didn't have too much trouble stopping it because they were blitzing their nickel back with abandon. After a third-down stop, I started jumping around saying, "They can't keep blitzing their nickel back. It's going to burn them. We're going to burn them." My friends looked at me and I presume the only word that came to their mind was Intense. Then on the first play of Michigan's next drive, this happened:
Notre Dame selling out their secondary like that was bound to cost them sooner rather than later. It was blatantly obvious to me and I'd bet it was obvious to Calvin Magee and Rich Rodriguez too, hence, above. The incredible thing was that it actually worked. For the last two years, you could see the lifeless form of something that Rodriguez has been trying to craft, a mess of blocking schemes and counters and You Didn't See That Coming daggers that either just missed or failed miserably. Denard Robinson has not only made these plays possible, he's turned this Michigan squad into NCAA Football 2010. If you see someone blitzing from the corners, you casually throw quick slants or screen passes (as they did against UConn). Want to run around in the pocket for 10 seconds and then scramble for 15 yards? Sure, we can do that. Watching Rodriguez and Magee call these games has been the most joyous experience of the last two years. It just works now.

Which is not to say everything's fixed. This team has a lot of flaws and eventually, people will start daring someone, anyone other than the fastest football player on earth to beat them. But for now, we get to enjoy what was promised to us when Bill Martin was sailing on a yacht and Michigan grabbed a southern boy from West Virginia to be their head coach.

  • Remember when I said this?
Aside from not having much information on the opponent yet, making a prediction based on last week's performance is near sighted and assumes that Denard Robinson is going to once again be the most prolific quarterback in Michigan history. Like Forcier last year, teams will adjust to his style and learn how to shut him down. Notre Dame wasn't able to do that last year and chances are they won't be able to do it against Robinson tomorrow either. Regardless, seeing a more human side of Robinson seems realistic.

          Yeah, nevermind.
  • Michigan's defense isn't good but it's going to confuse a lot of quarterbacks. Michigan's three interceptions yesterday were a 60/40 split of general quaterback incompetence and confusing zone drops. The linebackers/spur/bandit are blanketing under zones really well and making it difficult for quarterbacks to read where they're going to be. This is a huge positive. Unfortunately...
  • Cam Gordon needs to not allow 6'6", 265 lbs tight ends run right by him for touchdowns. That's a bad thing.
  • For all of the unlucky stuff that's happened to this program over the last two years, the temporary blindness of Dayne Crist is at least a few Karma Points back in our favor. If he's healthy the whole game, Michigan goes down in flames. Games against MSU and Wisconsin now look a little more harrowing.
  • As much as I want to declare Michigan "back" or "real" or any of the other terms that would indicate we're not the worst team in school history again, I'm reminded of last year. I'm being very cautious here. You should too. One or two more injuries (and not just to Denard) and they basically can't even field a defense.

Friday, September 10, 2010

Notre Dame preview

Tomorrow, Michigan heads into South Bend for the first time since 2008's disastrous fumble fest, a game that saw Michigan outgain Notre Dame by 128 yards yet still lose by a commanding 18 points thanks in large part to 6 Michigan turnovers. Last year's shootout was a far different affair and one that has become legendary in Michigan football lore--a game defined by the heroics and unflappable style of now-third stringer Tate Forcier. And tomorrow's game looks to be similarly unrecognizable: Denard Robinson is now at the helm of a diabolical offense, Michigan's defense looked functional in its season opener, and Notre Dame is now being run by somebody that has half a clue what he's doing.

Like last year, this game is supposed to define the paths of the two teams' respective seasons. The winner, so people are suggesting, may be on their road to redemption in the face of football armageddon. But I don't think this game will be quite so prophetic for a number of reasons. But on to the numbers...

[Numbers go here]

In lieu of my usual charting of the team's national standings and per-game averages, I'm going to take a bit of a different approach, due primarily to the fact that national averages mean exactly bupkis right now. Last week, Notre Dame faced off against in-state "rival" Purdue, beating them 23-12 on the strength of Armando Allen and Dayne Crist. Allen rushed 18 times for 96 yards (5.2 YPC) and one touchdown, but looked particularly bulldozery doing it. Allen wreaked havoc on Michigan last year--as did most competent running backs--gaining 139 yards on 21 carries (6.6 YPC). And were it not for his heel landing gingerly on the sideline, he would've added a very long touchdown reception to his 24 yards receiving.

Slowing Allen will be vital if Michigan wants to have any chance at winning this game. Fortunately, Michigan's defensive line is the most experienced part of the defense and should be able to attack Notre Dame's offensive front. If Allen is held to the 5.2 YPC that Purdue did (certainly an attainable goal), Michigan will have a good chance to win the game. If he starts rumbling through to the second level endlessly, the Wolverines will be in trouble because...

Quarterback Dayne Crist, first-time starter but former 5-star recruit, managed to complete 19 of 26 passes for 205 yards and a touchdown. And at his disposal is all-purpose pterodactyl Michael Floyd, who you may remember from The Shaming of Boubacar Cissoko starring Boubacar Cissoko, last year in the Big House. Floyd had seven receptions for 131 yards against Michigan last year and looks to repeat that performance against an even weaker secondary. Michigan will be playing a lot of zone coverage in this game and hoping they can create enough pressure that Crist doesn't have time to hurl the ball downfield to Floyd who, make no mistake, will blow right by everyone in the secondary on more than one occasion. Stopping the Crist-to-Floyd connection (a duo that produced 5 receptions for 82 yards against Purdue last week) will be the other key to slowing what should be a potent Notre Dame offense that will look to set up big plays through play action (be terrified of junior tight end Kyle Rudolph who could very easily run seam routes all over Michigan's defense) and the vertical passing game.

As far as Michigan's machinations, Notre Dame presents a whole host of challenges that UConn didn't. For starters, the Irish run a 3-4 defense which should help them contain Denard Robinson better than UConn did, and I wouldn't be surprised to see more nickel packages against Michigan's spread to try and better match the speed of the Wolverines offense. Other than that, we don't really know much. Having the UConn game mostly wrapped up in the first half, Michigan stopped showing new tricks and hitches in their offensive strategy. But I have no doubt that Rodriguez has more than a few things up his sleeve for this game.

Prediction: This is a difficult game to call. Aside from not having much information on the opponent yet, making a prediction based on last week's performance is near sighted and assumes that Denard Robinson is going to once again be the most prolific quarterback in Michigan history. Like Forcier last year, teams will adjust to his style and learn how to shut him down. Notre Dame wasn't able to do that last year and chances are they won't be able to do it against Robinson tomorrow either. Regardless, seeing a more human side of Robinson seems realistic. Whether or not Michigan can scrape together a win from that is yet to be seen.

Notre Dame, on the other hand, is going to look to exploit a beleaguered, young, and frankly talent-barren secondary with more weapons than Michigan will likely face all year. This could get dicey if the Irish are passing the ball like they did last year against the Wolverines, which is frankly pretty realistic. And if Armando Allen is the revelation he appeared to be last week, the promising week-one Michigan defense may reveal itself the hollow shell we all thought it was.

Per usual, and until it becomes a habit, winning the turnover battle will be key. If Michigan holds onto the ball and limits Notre Dame's offensive chances (long, sustained drives that wear away a thin defense might be Michigan's key to success), this game could look pretty similar to last week's UConn contest.

Notre Dame strikes early and possibly multiple times, but in the end, Michigan's consistency and Notre Dame's inability to stop the offense will provide the Wolverines with a second win. Michigan 35-34

Wednesday, September 8, 2010

Ezeh vs. Kovacs: Who's responsible?

Obi Ezeh has come in for his fair share of criticism of late, no where more prominent than in MGoBlog's extensive Upon Further Review posts. Brian recently posted the UConn defensive UFR and once again, Ezeh grades out poorly. Brian highlights one play in particular in which Ezeh gets blocked off by a pulling guard and the running back jets through the gap for an 8-yard gain. The play in question:

How are the initial returns on the GERG Linebacker Magic theory?
Good, but not great. It's not much of a surprise when you juuuust barely beat out a walk-on for your job but Obi Ezeh's progress has been incremental at best. He did make a nice play on the last defensive play charted up there but there was an awful lot of Ezeh getting shoved around like a rag doll and even one instance of those horrible times last year when he'd go into a pass drop on a run play. I mean:

In this play, UConn is running power, off tackle (running between your right tackle and right guard, with your left guard pulling across the line and acting as a lead blocker). But despite Ezeh not actually making the tackle, I don't see any way that the responsibility on this play doesn't fall squarely on the shoulders of Jordan Kovacs.

Regular readers of this blog will note that I don't like Kovacs on the field. (Jordan Kovacs is to me, what Obi Ezeh is to Brian Cook at MGoBlog.) In the above play, Kovacs takes a 6-yard drop into pass coverage as the ball is snapped and doesn't even recognize the play as a run until UConn running back Jordan Todman is past the line of scrimmage. In fact, he's so far out of position that the weakside safety Cam Gordon has to come across the field to make the play. Let's break it down:

Michigan is lined up with a three-man front and will be bringing a blitz from the weak side of the play. Ezeh is playing middle linebacker and Jonas Mouton is the playside linebacker. On the near hash at the 21-yard line is Jordan Kovacs. Cam Gordon is lined up on the far hash and is playing as deep as Kovacs.

At the snap, Kovacs immediately starts to drop back into pass coverage and won't stop for a while, as we'll see. Martin is being double teamed by UConn's center and right guard. The UConn left guard has started to pull across the formation to act as the lead blocker in charge of hitting Ezeh. The play side slot receiver is going to block Mouton. Greg Banks (the playside defensive end) is being sealed to the outside by UConn's right tackle.

Here, you can see that Kovacs is still backpedaling, now 4 yards deeper than he was pre-snap. Mouton is being hit by the playside slot receiver. UConn's left guard has now pulled across the formation and is in perfect position to block Ezeh. Ezeh's job here is to plug the hole that Todman is supposed to run through. He does this by hitting that pulling guard. It's then Kovacs' job to come into the play and make the tackle. Unfortunately, Kovacs isn't done backpedaling yet.

Kovacs, after taking a 5-yard drop, now realizes that this is a running play and has started to get into the game. But he's so far off the line of scrimmage that there's almost no way this play doesn't go for decent yardage. Ezeh is attacking the play. He's making contact with both the pulling left guard and Todman, but ends up getting blocked to the ground by the guard. Todman is about to slip past him into the middle of the field that has been completely vacated by Kovacs who took far too long to diagnose the play as a run.

The UConn offensive line has now done their job. The linebackers have been taken out and Todman is releasing to the next level. Kovacs is still 8 yards away from Todman without any momentum. He's a sitting duck. (Have you ever tried to stop someone running full speed at you when you are flat footed?) In fact, if it weren't for the referee standing right in the middle of the play (forcing Todman to cut to his left), I'd venture that this play has about a 50/50 chance of going for a touchdown.

But Todman has to cut around the referee who's trying desperately not to be in everyone's way, and Cam Gordon has come from across the field to make a nice wrap up tackle.

I'm usually with Brian that Ezeh makes the wrong read and generally finds himself on the wrong side of should be no-yardage plays, but this one falls squarely on this blog's goat. If Kovacs diagnoses this play quicker or doesn't drop 5 yards into coverage to compensate for his lack of elite-level speed, this is probably a two-yard gain. Instead, Todman is given an 8-yard cushion to play around with and rumbles his way for as many yards.

Brian's argument against Ezeh here is that he needs to funnel Todman to the outside, which I think is patently wrong. If Ezeh pushes Todman outside by taking a different angle on the block, UConn has two blockers against two Michigan defenders. The numbers are squarely in favor of UConn here, and the play eventually becomes a foot race between Kovacs (coming from the middle of the field) and Todman. All of Ezeh's help in this play (his weak-side blitzers, linebackers, and other safeties) are to the middle of the field. Simply because Mouton was able to get off of his block quickly in this instance does not mean that's where Ezeh was supposed to force the play.