Some caveats: this is not to say that Denard Robinson is incapable of becoming a pocket-passing quarterback with a deadly ability to run built in. This is saying that the offense, as it's built right now, is not as effective from under center as it is out of the gun. Let's take a look at every play run from under center against Notre Dame, courtesy of mgoblog:
| Qtr | DRV | Ball | Dwn | Dst | Play | Player | Yds | Note |
|---|---|---|---|---|---|---|---|---|
| 1 | 1 | M20 | 1st | 10 | SACK | Robinson, D. | 0 | |
| 1 | 3 | M33 | 1st | 10 | RUSH | Hopkins, S. | 2 | MANBALL 1 |
| 1 | 3 | M31 | 2nd | 10 | PASS | INCOMPLETE | 0 | |
| 2 | 1 | N45 | 1st | 10 | RUSH | Hopkins, S. | 2 | MANBALL 2 |
| 2 | 2 | N43 | 2nd | 8 | PASS | Hemingway, J | 43 | TOUCHDOWN |
| 3 | 2 | M29 | 2nd | 10 | RUSH | Hopkins, S. | 3 | MANBALL 3 |
| 3 | 4 | M43 | 1st | 10 | PASS | INTERCEPTION | 0 | INT |
| 3 | 2 | N6 | 1st | G | RUSH | Shaw, M. | -2 | |
| 3 | 3 | N8 | 2nd | G | RUSH | Robinson, D. | 7 | |
| 4 | 4 | N1 | 3rd | G | RUSH | Hopkins, S. | 0 | TOUCHDOWN |
| 4 | 2 | N45 | 2nd | 15 | PASS | McColgan, J | 15 | 1st Down |
| 4 | 5 | N14 | 2nd | 7 | PASS | Gallon, J. | 14 | TOUCHDOWN |
| 4 | 1 | M13 | 1st | 10 | PENALTY | PENALTY | -4 | |
| 4 | 3 | N26 | 1st | 10 | PASS | INTERCEPTION | 0 | INT |
| 4 | 5 | N21 | 1st | 10 | PASS | Smith, V. | 21 | TOUCHDOWN |
| 4 | 3 | N15 | 1st | 10 | PASS | Roundtree, R | 16 | TOUCHDOWN |
That comes to 16 total plays run from under center, of which 4 were TD's, 1 was a TD that I think we can all agree shouldn't be replicated (Denard's fumble recovery), 1 went for a first down, 4 were so-so gains, 4 went for no yards, and 2 were INTs. My disagreement with Mgo's analysis of this is that I'm not sure you can count Denard's fumble recovery as a "positive" play. The result was positive, the means were a huge negative. I'm going to throw it out of the equation altogether. Even with it in, it doesn't change much.
6 of these plays were unmitigated "bad" results: 2 INTs and 4 zero yard gains. 3 of these plays were good, but not nearly as good as "average" out of the shotgun - about which more later. These include those designated with misopogon's MANBALL reference. 6 of these plays were positive results, including the the FB wheel route, all TD's and a Denard scramble for 7 yards. That means, with fumble-whoops-TD! excised, that you're looking at results like this from under center:
bad: 40%
Below Average: 20%
Positive: 40%
35% of Michigan's plays were run from under center, with the remaining offensive plays being run from shotgun. Let's take a look at the Yards Per Attempt from each:
| Formation | PASS YPA | RUSH YPA | TOTAL YPA |
|---|---|---|---|
| I-Form | 12.00 | 2.33 | 7.64 |
| Shotgun | 14.63 | 7.46 | 10.19 |
| Total | 13.50 | 6.06 | 9.31 |
The big number to pay attention to there is rush yards per attempt. That is a significantly lower percentage from under center than from out of shotgun. What this does is render the "under center playfake" pointless. We're not a rushing threat from under center. We are from the shotgun. When 35% of your plays are run from a formation that you're not very good at running from, you become 1 dimensional, and when you become 1 dimensional you throw interceptions. The majority of plays run from under-center were either turnovers, negative yards, or middling gains that were far out-gained by their shotgun counterparts.
Apparently, you also throw touchdowns. This argument is not to take away from those - those were great plays. Let's take a look at the scoring plays:
1) Denard throws a jump-ball to Hemingway who makes the catch and dives for the pylon
2) Denard drops back, throws a jump-ball to Gallon who goes up and gets it.
3) Denard runs a very well drawn-up screen to Smith who picks his way through defenders for the TD. This play was an awesome call, but not exactly awesomely executed. Smith made a tremendous individual effort to find the endzone as his blockers totally whiffed in front of him.
4) Denard throws jump-ball to Roundtree in the corner, pandemonium ensues.
Now I understand that scoring plays are scoring plays, and bitching about them is the equivalent of bitching about rainbows. But which of these scoring plays feels like something that is sustainable as the primary weapon of the offense? Jump balls to tall receivers? This might actually work, I'm not saying it won't, but it is a concern. The one scoring play I felt really good about - the screen to Smith - wasn't even executed well from a blocking standpoint:
Fast forward to 9:08 to see two linemen whiff on one guy.
Do you really feel particularly good about running a system that relies so heavily on jump balls? You might! I'm not saying that it's 100% bad, I mean, we have the receivers that can do it, and Denard throws a pretty good jump ball. I'm still leery, and for reasons why see "Denard throws pick number 2 in endzone." FWIW, here's Al Borges' thoughts on jumpballs, and he's the only one who really matters:
What are your thoughts on throwing jump balls? "I’ve changed my thinking on this as a coach over the years, particularly on deep balls. I remember way back when I was coaching at Oregon with Chris Peterson, and we were talking about throwing the ball deep, and I always used to have the philosophy that if you throw the ball deep, overthrow them so the ball’s not intercepted. And I remember Pete telling me, he says, ‘We got a couple guys who can go get it. Let us touch it.’ I argued with him. Today, he was totally right. The ball has to go up to an area where [the receiver] can touch it. Now, you have to make it so that only the good guy can get it, and when it is a jump ball, the worst you can get is an incomplete pass.
Is throwing jump balls part of your game plan? "We don’t want to throw it up for grabs. But we want to give our receivers, who are good receivers. Jeremy Gallon, who’s not a big tall guy, but can go get high balls, and Junior Hemingway, who is a big guy. We want to give them a chance. And Roy Roundtree, for that matter."
Not to be lost in this whole thing is the fact that we did indeed run 65% of our plays out of the shotgun. These plays were, on aggregate, more effective than the 35% from under center. When Michigan had to make hay, we were largely operating out of the gun. What I don't understand then is why we're not out of the gun even more. It has better YPA, utilizes our primary weapon more effectively, and isn't relying on the "under center look" to keep defenses honest. The mere fact that our QB is deadly on the ground keeps defenses honest, and from under center we lose that.
The numbers show under center doesn't work as well as shotgun. That's the evidence to this point. Where do we go from here? Well, Al Borges is going to continue to operate out of both formations. This means that execution has to get better in order to have a viable offense from under center. Borges and the coaching staff will work on this, and I have no doubt that they'll get better as the year goes on. Fortunately we have Eastern Michigan coming in this weekend that will hopefully allow for some more gametime reps. I'm not saying that the under-center offense is "bad" and shotgun is "good." I'm saying that the shotgun offense is more productive right now than from under center. I'm also saying that I believe our personnel are better suited to generate yards and points from shotgun than they are from under center. This is not to say that our personnel can't improve from under center, it's just that as of right now - with this extremely limited sample size - I think we should be operating even less from under center.