Michigan Football, Team 139, 2018 Season

[Jeff Buchanan]

We've seen the 2018 version of Harbaughffense. It wasn't good v. ND although my take and confirming it by looking around the web, the framework is there and so are key parts.

Here's a couple of takes from SpaceCyote who posta regularly on mgoboard. He is at least a HS coach at some level but he's never revealed where he coaches. He could be a lower division college coach because his level of knowledge about the game is high. He posted two takes on how to fix what ails Michigan football:

One:

I know, I know, "execution" is a 4 letter word around these parts, but "creativity" is so far down the list of what the problem is. Michigan's scheme, which isn't majorly changed, worked fine previously. Harbaugh's offense worked great in San Fran. He's plenty creative. They need to get guys coached up on some basics first though.

Two:

My personal take for what I would do differently, or at least try (without insight into what has been repped and practiced and done well/poorly during camp, etc.)

I would get under center more. Not like 2017, but nearer to 50-50 under center and shotgun. I think Michigan is more comfortable with a more diverse run game under center and like the more down hill attack. I think that helps the RBs and helps the OL because the RB's hit the LOS harder. This forces the defensive front to play the run more, which hopefully helps pass pro a little.

I would really limit empty formations. Michigan was very successful with empty with Speight in 2016, but it looked very bad against ND. Part of that is because the defensive front knows they are going to get 1-on-1 blocking, they can press on the outside, they can push LBs up to the LOS, and then they can drop underneath. And while Patterson is good at getting the ball out fast, he does struggle a little to read defenses and he tends to fade against pressure. I don't think the offense's strength is in empty. At least keep in a TE or RB.

Strategically, you have to try to get the ball downfield more. This is a risk/reward thing, but is where getting under center and getting a little more comfortable running the ball can help your pass pro and give you the time to do this. You're going to take sacks. You're going to run 7/8 man protections and eat incompletions when your guy gets doubled. But you have to do it at least a handful of times because the offense isn't going to function without getting the safeties to play honest. If UM can hit over the top 3 more times against ND, the rest of the offense opens up. The offense as it was run wasn't unsound, it was that they couldn't get the safeties to ever move backward. Get those guys not playing flat footed and the inside zone gets better because you don't have safeties meeting RBs at 3-5 yards and the run game starts getting chunks instead of cut down for efficient but short gains. You don't have safeties cutting down the WR screen game at the same distance as part of the RPO package. Get them back, even if it means taking a few sacks in order to do so. This is going to sound a lot like 2013 Borges, who continued to try to take shots despite poor pass pro. To a degree, yes. But I think Michigan can run block better than that team, and I trust Harbaugh and Co. to better dial in a mix between shots and efficiency. But they have to take a step or two that direction in order to open up their base offense.

Once you get that, then you can start attacking the short/intermediate voids in the defense. Because then you have LBs creeping up on the run but the safeties staying back and you aren't getting killed in pass pro. And that's what allows you to get YAC. Now you've spread out the defense horizontally and vertically. But to get to that point, you have to condense some formations sometimes and get under center sometimes and then maintain that personnel in the spread looks when you go there (which Michigan did against ND), then that opens up what you really want to do (which I think is close to what Michigan ran against ND, but they have to add some things to get back to that and be successful with it).

So, in summary, get under center a little more, mix in more gap schemes again, help the pass pro with a more confident run game so you can take a handful more shots over the top to get the safeties honest, and then the offense opens up.

[Hannibal]

I would never snap a ball from under center with such shitty tackle play unless it's short yardage. I want the QB square to the LOS on his first step so that he can see the guy who blows around the LT for a quick sack. Under center it's hard enough to do just a straight drop. Play action is basically suicide.

[AlabamAlum]

Under center more for an offense that is struggling with protection seems curious to me. The gun (or pistol) buys the QB a second or maybe two to start the progressions. While the RB may hit the hole a bit faster with the QB under center (and I have never studied that), it seems that benefit would be negated when you pass from that set with the pass pro deficits you guys are already experiencing.

[Jeff Buchanan]

I was a little surprised about his thought that M should take more snaps from under center formations. However, the why the offense should do that makes sense.

In doing this, you're not mis-using Patterson's skill set. What you're doing, according to the recommendations (go deep more often to get the Ss to step back and combine this with more 1-Gap runs to get the LBs to step forward on the PA), is to create zones over the heads of the LBs and in front of the Ss to run TEs and slot receivers into. This is perfect for Patterson's quick release off 3 and 5 step drops and his accuracy.

I can see where M tried to do this v. ND but didn't work on it hard enough because Patterson threw downfield so infrequently. As SpaceCyote points out, there's a risk/reward calculus for going long but I don't think there is any doubt that M's receivers are good enough to do that and the Ozone comments by Gerdman about DPJ fits right in with the wrongness of Harbaugh not going deep to him, Nico and Oliver more often.

I'll repeat this for emphasis but the long ball and the threat of it off PA opens up the run game becasue you're not getting S.s in your face at 3-4y downfield. Getting Higdon or Evans in space with maneuvering room before they get clobbered is likely to produce some big gains ...... gains currently not happening becasue M isn't alternating the run/pass combos as contraints to each other enough. See also SpaceCyote's comments on playing under center more as this helps the OT problem by bringing backs and TEs into the backfield for pass-pro.

Something I didn't think of that is also pretty clear to me now is that using all spread formations, while it sounds sexy, makes it much harder to protect the QB when you have the reality of M's OTs. It's a five on five propostion in this formation and the OL has to perform perfectly. With Patterson in teh gun and 3-5 receivers, I noticed ND's Ss either stepping up right away or using a delayed blitz through an open outside rush lane. This puts into perspective what appeared to us was poor technique by JBB abd Runyan. It was just as much scheme to fault (i.e., the spread) as it was OTs getting beat and I saw that on my video review but didn't make the connection until SpaceCyote pointed it out.

On another note, the idea of Jim handing the keys over to an OC is a big one and as Hanni points out it is going to be the silver bullet criticism for 2018 if M continues to plod along like it seems there's a pretty big risk that it will. So, hire an OC? Maybe not and this quote makes that case:

Michigan's scheme, which isn't majorly changed, worked fine previously. Harbaugh's offense worked great in San Fran. He's plenty creative. They need to get guys coached up on some basics first though.

[Jeff Buchanan]

I stole this from mgoboard. Interesting:

To follow-up on my hit post below, I wanted to present a little more info that shows how rare a major win over a ranked opponent truly is for most teams. I looked at our main rivals in chasing Ohio State (as they were going to be the obvious outlier) to see how they had fared since 2007 against teams that finished the season ranked.

Michigan State: 2-11

Michigan State's two wins both came in 2015. One, as we all know, came after one of the luckiest plays in college football history went their way against Michigan. The other was against Ohio State when Urban Meyer decided that Ezekiel Elliot was good at football for a night.

Penn State: 3-15

Had one truly impressive road win: 2008 against a good Ohio State team 13-6. They beat Northwestern in 2017 and surprised Wisconsin in 2013.

Wisconsin: 0-12

Wisconsin has not beaten a single team that finished the season ranked on the road since 2007.

This is something that just does not happen very often for most teams. These are three very successful programs and between them they have two elite wins on the road (2008 Penn State vs. Ohio State and 2015 Michigan State vs. Ohio State) as the others were either the result of absolute craziness or were against good, not elite teams.

IF I HAD KNOWN THIS BEFORE I MADE MY 14-0 AND AN NC FOR M PREDICTION, I WOULD HAVE WISELY GONE WITH 9-3 AND A BOWL WIN TO GO 10-3. LOSSES TO ND AND 2 OF MSU, PSU AND osu

[Hannibal]

Under center is a terrible idea unless your offensive line can block long enough to protect the QB on plays where he has his back to the line of scrimmage for a couple of seconds. In our case, that doesn't happen.

And there are a lot more ways to keep the defense honest than to throw low percentage passes fifty yards down field. If I'm a defensive coordinator I'll gladly concede a couple of arm punts a game if it means that you never run the football effectively.

[hack]

Agreed that under center more seems counterintuitive.

[Jeff Buchanan]

Under center is a terrible idea unless your offensive line can block long enough to protect the QB on plays where he has his back to the line of scrimmage for a couple of seconds. In our case, that doesn't happen.

Running a pass play from under center out of an 11 formation (1 TE and 1 RB), using 1-1 gap blocking, creates this:

(1) Initially, there are no open rushing lanes.The OL is lined up 5-5.

(2) You have two or threee extra blockers (the RB and a TE, two TEs if you want in a 21). You're starting out with 7 or 8 on 4, maybe 5 depending on what the defense actually does with it's front 7 (rush 4 or 5 or some variation of a LB or S blitz to make 5 or even 6 rushers).

In my mind this kind of approach seems to protect weak OT play allowing a rusher to get around outside and pressure the QB. The RB and TE(s) have to read the defense's intent if they rush more than 4 and/or pick up leakers to give the QB at least 5 counts to read the DBs and dropping LBs and run his check downs to the open man. The PA fake in a passing play, regardless that it puts the QB's back to where the play is going, is absolutely key to freezing the LBs.

We're also not talking going to all under center. We're talking about a 50/50 split. I think M was probably using about a 70 spread/30 under center split v. ND.

[Hannibal]

That describes the Carr/Debord approach and it doesn't work.

[Jeff Buchanan]

Sure ...... but that is all LC and Mike Debord did. It was heavy sets all the time and Mike Hart running between whatever opening developed between the LT and the LG. The PA was the only way the pass was set up. I remember most of us complaining about a lack of "creativity." What we actually meant and the phrasing we should have used was a lack of constraints. The Debord offensive strategy was, and the strategy that we all joked about, is the one that said, "we're going to run this play, go ahead and stop us." By the time defenses started to get really good and really fast and every OC knew exactly what M wanted to do, Hart and every other RB got stoned more often than not.

While Harbaughffense is basically power oriented, it is a ton more diverse than anything LC did. What SpaceCyote is talking about is having a strategy that forces the defense to pay attention to a play-fake from a tight under center formation or RPO from the gun spread out sideline to sideline so as to present to the defense multiple formations and play options as the play unfolds that they have to recognize and properly defend. That's hard. A ton harder than anything Debord did with LC.

I think the biggest issue for M's offense right now is execution. I don't think there is a lot wrong with the offense except the tweaks SpaceCyote suggests. He is suggesting the use of more under center formations and calling a few more down hill run plays using the PA (or not) to (1) shore up the tackle problem and (2) get the defense to have to honor the run or the pass off the PA or RPO by not bringing the Ss up. We're not talking about going all heavy sets here. SpaceCyote still likes the spread out of the gun as far as I can tell using the RPO. The two types of formations could be considered constraints to each other as defending the spread/RPO in a formation that goes sideline to side line is completely different than defending a play in a tight formation from under center using the PA.

[WingsFan]

[iam416]

I mean, sure, if you're able to complete 3 or more bombs every game and execute way better then the offense will look better.

[WingsFan]

said

https://sportsnaut.com/2018/09/michigan-president-jim-harbaugh-not-on-the-hot-seat/

[AlabamAlum]

Harbaugh was 17-20 heading into his final year at Stanford. That last year there was pretty spectacular, they had a great o-line, BUT let's remember, it was a very down year for the Pac-10.

Look at his opponents that year: USC was 8-5 and on probation, Washington had 7 wins, out of conference opponent Notre Dame had 8 wins. In fact, Stanford played only 1 team in the regular season with more than 8 wins (Oregon) and the Cardinal lost by 3 TDs. The Pac-10 finished that year with only two ranked teams, Stanford and Oregon.

Before Stanford, he did kick ass with the Toreos in the Pioneer League for 3 seasons and he had a great NFL run after Stanford, I am just not sure that either of those facts translates into what is needed for a good major college coach. Most great college coaches learn from a great college coach in a mentor-student relationship as an assistant. Harbaugh was never a college assistant or GA. His only time as an assistant in any capacity was with the Raiders as a QB coach for two seasons.

I guess what I am saying is it is not too late to get Hoke back.

[Jeff Buchanan]

Shut up ....... look, the fact is that none of us here in the peanut gallery no shit about football. We are rank amateurs. I'll give credit to those that either played the game or take the time to study it (like me ) but basically we don't know shit about the game or what the coaches know about how the team actually performed in a game.

To your point, Harbaugh knows football. I don't care if he wasn't any of those things you think could be essential. Despite claims to the contrary, I believe the offense he puts on the field is pretty good ..... if one takes the time to understand it. It is far advanced beyond the typical, uninformed spread v. power arguments that predominate in the argument is it a good offense or not.

Here's a link to a guy who does an excellent job of diagramming and explaining plays. In this current piece of many he's done in the past that he calls Neck Sharpies, he highlights some of the errors the tackles make that submarine plays despite those plays being exactly the right call given down and distance, clock, etc. At the outset, he comes right out and says he's not writing this article to nail Jon Runyan. It is to highlight the degree to which the play design for ND was intended to mitigate what the coaches knew, well in advance, were weakness at the tackle positions.

Again, contrary to the typically uninformed hot takes following the game, you could call the initial offensive plan v. ND brilliantly designed or, if you don't like hyperbole, it was pretty damn good. The problem for Harbaughffense v. ND was execution. Fixable? In our postions, we just don't know enough to say with any certainty.

https://mgoblog.com/content/neck-sharpies-mitigating-left-tackle