QB Run Game

Luke Rubenzer was the hot discussion topic after Saturday's game. Let's break down what he did, why it was effective, and where we need to go from here.

Math

Running QB's are awesome. For a long time they were seen as a gimmick, which is absolutely absurd given the facts we're about to go through, but I can't say enough about how much a running QB can help an offense's run game. Let's start with some football math. Both teams have 11 guys on the field. Before the zone read took off, QB's were basically just distributors. They couldn't run the ball themselves, and they couldn't block for anybody. That meant that the offense had 1 ball distributor, 1 ball carrier, and no more than 9 blockers. Meanwhile the defense had 11 potential tacklers. Because the QB is useless once he's let go of the ball, and because the ball carrier can't block for himself, that means that the defense has a built in +2 advantage in classic, pro-style football. No matter how good the offense is at blocking, there are always going to be two guys that they can't account for.

Against pro-style offenses, defenses tended to use their two extra defenders in the following way: one of them would be a deep center-field safety, and the other would rock down into the box. Hence, a typical “pro-style” defense against the I-formation would play a three deep shell with 2 CB's lined up over the two WR's, a deep safety in the middle, and 8 men in the box. This 8-man box was going against seven blockers (5 OL + 1 TE + 1 FB). So, the defense was able to keep a +1 advantage against the run while also having a safety deep in case anything went wrong. We can call that extra box player a “free hitter,” because he'll be unblocked and conceivably able to make the tackle no matter what the offense does:

This, or something like it, is what Saban wants to run. This is what the Seahawks want to run. This is as good as it gets for a pro-style, run-stuffing, 8-men-in-the-box-and-keep-everything-in-front-of-you kind of defense.

Enter the mobile QB. If the QB is more than a distributor and can carry the ball himself, then the defense loses that critical extra player. If the defense wants to keep one deep safety, then a match-up that was 8 box defenders against 7 blockers + 1 ball carrier becomes 8 defenders against 7 blockers + 2 potential ball-carriers. Without a mobile QB, the defense had a free hitter. With a mobile QB, they don't. It's not just that the QB gives you an extra guy, it's that he gives you the critical extra guy. If the defense wants to stop the run, they now have to beat a block at the point of attack. This problem only gets worse for the defense if they want to play with two deep safeties, at which point they're actually minus-1 against a mobile QB (2 CB's and 2 deep safeties leave 7 box defenders against seven blockers and two ball carriers). The only way the defense can regain their numerical advantage against a mobile QB is by playing with no deep safety, which has obvious drawbacks.

The best way to get both safeties involved against the run without playing a true Cover-0 is to use quarters. In my previous post I discussed the reasons that quarters is a favorite coverage against the spread. I mentioned that quarters lets the safeties play more aggressively against the run. Quarters is also a good coverage against mobile quarterbacks. If both safeties can be involved against the run, than the defense can actually get nine defenders in the box (the front 7 + 2 safeties). This regains a free-hitter for the defense. The same goes for Cover-0, which is straight man-coverage with no deep safety.

Luke Rubenzer and the Spread

This is where the spread comes in. Recall that in Cover-4 against the spread, the safeties are responsible for the inside receivers if they run deep routes. This pass assignment obviously means that the safeties have to line up more or less across from the inside receivers:

If the offense is in a spread formation, therefore, this means that the safeties will be aligned pretty wide off the ball. This isn't a problem if the QB isn't mobile. We can see how the defense can use their safeties to fit an outside run:

If the OL blocks this perfectly and the RB takes this ball to the edge, then the defense will still have the SLB and SS in a 2-on-1 against one blocker (Y), leaving one guy free to make the tackle. The defense can also get good numbers against plays up the middle:

Even if the OL blocks this perfectly, the defense can get a free hitter by folding the WLB into the hole, or by doing some stunt involving the RE and the WLB: 

 

In terms of the math, by involving an OLB against the run, they can get 6 run defenders against 5 OL. So, we can't run to the edge without the safeties causing problems, and we're still probably looking at 6 box defenders against 5 blockers on inside runs.

Well, you can guess where this is going. Everything that I've just said shifts +1 in favor of the offense with a running QB. The play we used most often with Rubenzer was QB power:

The OL blocking here is the exact same as it is for our regular1-back power play; most of the OL down blocks (blocks away from the playside) while the backside guard pulls. The only difference is that now the RB leads the QB through the hole. In his press-conference after the game Sonny mentioned that our QB runs help us get better angles in the run game, and you can see why. The RB will meet whichever defender is in the hole head-on. He doesn't have to track him down, because the LB's reads are taking him downhill, straight into his blocker. By lining the RB up directly over the hole, we eliminate one of our major weaknesses: blocking LB's in space. Furthermore, because the safeties are pulled out of the box in order to cover the inside receivers against the spread passing attack, they have a harder time being involved against a run through the C-gap. They could still fill aggressively against outside runs, but they won't be a big factor on a play like QB Power. This would contrast with a pro-style offense with a mobile QB (think of the 49ers), where the offensive formation is more compressed, and so the safeties can be more involved in all run fits.

With that in mind, let's pile on Fitz, who said that he knew we had a running QB but had no idea what we were going to run. I knew what we were going to run, and I haven't seen a minute of practice.  We only had one game last year where we featured a mobile QB as part of the game plan. That was our use of Boehm against tOSU, and we ran exactly the version of QB power that I've diagrammed above, and did it exactly in short yardage situations like we did on Saturday. We even had Goff run this play on the goal line against UCLA. If Fitz had prepared for QB Power and the Zone read and gotten burned by something else, OK, but with an entire offseason to prepare, if I were a Northwestern fan I'd hope that my coach would figure this out. Seriously, would it have taken that long to have a GA pull all of our QB rushing attempts from last season? You could probably get everything you needed in an hour.

In fairness to Fitz, while we ran some version of QB Power 6 times against Northwestern, it wasn't all that we had. Three times we ran a QB Lead Draw, once again using the RB as a lead blocker. At one point we actually caught Northwestern in this awesome look by running out of a trips formation:

Yes, that's five box defenders against 6 blockers and 1 ball carrier. We, the OFFENSE, actually had a +2 advantage on this play. We were able to double-team the NT and still had a blocker free to take on the WLB in the hole. This is the kind of thing that should never happen if the defense has time to gameplan for it, but even with the right adjustment we're probably still in good shape numerically (although there's a decent chance we would've screwed it up on execution; we didn't look that clean even when we had an advantage).

 

If the defense does pinch inside to stop these plays (which are by far the best additions to our offense), we also showed a QB Sweep, once again using the RB as a load blocker:

We bring in X tight to seal the end, the OT seals the DT, and the C, RG, and RB head around the edge to take on LB's and DB's. Once again, the math on this play was fantastic (as you can see) but our execution only allowed us to get 1 yard.

Basically, you can take any play that we already run, and you can run it with the QB to gain an extra blocker. There's almost no new installation necessary for your OL, but by swinging the numbers +1 in your favor you force the defense to gameplan differently against these runs.

The point of all of this is that even if the defense knows that Rubenzer primarily runs, these plays are sound schematically and give us an advantage. Even better, we don't always run the same play, and each of our plays attacks a different gap. Looking through the diagrams above, you can see that QB power attacks to the side of the RB's alignment, and aims for the C-gap. Our lead draw play attacks away from the RB's alignment and aims for the A-gap. The sweep is an edge run to the D-Gap, and we could easily run a lead draw to an open B-gap as well. So, even if the defense knows that we'll run with Rubenzer, they can't know which gap or which side of the formation we'll be attacking, making it harder (though by no means impossible) to use stunts and run blitzes to take away these plays.

The other comment is that Rubenzer wasn't reading anybody post-snap on any of these runs. So far, at least, we aren't running the kinds of plays you'll see from Oregon, Arizona, or Nevada. This is much more like what you'd see from an Urban Meyer offense, so even within the world of QB runs there's a fair amount of diversity, and what we're doing isn't very common in the PAC-12. Our run game has always been more sophisticated and interesting than it gets credit for, and that sophistication has multiplied considerably with the addition of Rubenzer to the offense. Judging by Sonny's comments over the last two years, this is the kind of thing we've wanted to do all along, but only now have the personnel to pull off.

Rubenzer as a Passer

No surprise here, but we didn't get into our downfield passing game with Rubenzer very much on Saturday. He had two completions. One of them came on one of our common quick-game plays (H-out), and the other came on a packaged play where we ran our normal power play to the right and had X running an out on the backside. These plays are relatively easy completions because, as I've discussed elsewhere, they require the QB to read at most a few simple things. We didn't see these options much from him on Saturday, but it does seem like before too long Rubenzer could potentially (1) do one of the runs I've described above, (2) run a packaged play just like Goff does, or (3) throw any of our basic quick-game concepts. Until we see it we probably shouldn't expect him to be going bombs away with 4-verticals or reading floating LB's and safeties on Y-cross. As a result, if he's in on 3^rd and medium or long we probably know that it's going to be a draw or a roll-out with some kind of 3-level flood route, but on more neutral downs the three kinds of plays that he can run right now make up a decent chunk of our offense and give us some diversity for when he's in the game, in addition to preparing him well to be the starter some day.

The big problem that we'll run into immediately is that teams might start going to Cover-0 on us whenever Rubenzer's in the game (especially at the goal line). Thinking about our football math, 4 DB's to cover 4 WR's leave seven defenders against 5 OL + 1 RB + 1 QB. Man coverage is also effective against our packaged plays and quick game concepts. If we want to beat this, the next thing we'll probaby want to see him develop is a good slant throw. With our WR's, slants are good, safe options against the off-man coverage you usually see with Cover-0 (teams usually won't press without a deep safety), but the timing and the throw still have to be tight for it to work. It's a tougher throw than the zone stretches that he'll see on our quick game concepts, where one defender is being forced to cover two guys who are running in opposite directions.  Speed outs by inside receivers could also be a good option here. When you see Luke start to drill some quick throws in to covered receivers, you'll know that we've really got something good going on. If we can consistently throw for 5-6 yards against Cover-0, that's enough to work with until he can develop deeper throws and become a real passing threat.

Here's the BI thread for questions, comments, and discussion: http://bearinsider.com/forums/showthread.php?85920-Bear-Raid-Breakdown-Luke-Rubenzer&p=842352435#post842352435