Monday, November 26, 2018

Coaching Points: Michigan vs OSU, 2018

I used to do more of these back in the day. Admittedly, this is more about Michigan and a response to some of the criticisms of Michigan from this game. So sorry it's light on the OSU POV.

Kirthmon F. Dozier, Detroit Free Press


Michigan Offense - Wrinkles
It was clear from the onset that Michigan believed they could be successful on offense with mostly their core playbook. Underdogs, or teams that believe they will struggle to generate yards, typically are the ones that come out with drastically different looks. Teams don't like to hide major portions of the playbook only for a single game, because for the most part, it's counter-intuitive. Those plays that are hidden generally a repped less and executed worse, and so if you believe your execution and talent is good enough to win, you are more likely to stick with your core playbook. Michigan believed that going into the game.

They utilized some of their better plays. Some WR screens, splitting Evans out wide for the slant, some nice route concepts from the WRs and TEs that resulted in a very efficient pass game. They weren't wrong that the core could be quite successful.


That isn't to say Michigan didn't add some wrinkles. They had two interesting counter plays that had mixed results, but you could understand the logic behind each given OSU's film. OSU's LBs had a tendency to plunge into the LOS and erase themselves. One counter play was set up much like Michigan's standard Belly Read play, but had the OG pull across the formation for a long trap. Depending on the movement of the WILL (LB attack the LOS, RB follow the pull; LB follow the pull, RB dive), you can get some yards. Another counter play was based off Michigan's staple Belly Read Arc block play, in which the H-back would initially cross the formation before returning in the direction he came. It is effectively Counter H.



There were a few other wrinkles (a nice play action wheel route that resulted in a dropped TD), but they were more in line with the standard wrinkles Michigan adds from week-to-week, they were not the "new offense" that many fans seem to clamor for come big games. Given OSU's defensive film and Michigan's relative efficiency on O, I don't blame them for going with this approach.



Michigan Offense - Missing Pieces of the Playbook
What was lost out of the core playbook that I didn't understand was the addition of RPOs that Michigan had shown, along with some of the split flow looks they showed off Down G early in the year, and at least some sweeps.

In terms of RPOs, Michigan had started adding RPOs that attack the backside LB. Much of OSU's success against the run came from the WILL attacking at the snap. This is and was OSU's MO on defense. Despite the relative issues on that side of the ball, they have been relatively successful at getting stuffs and TFL on standard downs. Success against them came when you could get behind the LBs. OSU was shooting gaps, and for some reason Michigan never pulled the trigger on some of their RPO looks that I think could have been very successful and started opening up the run game. Whether this was due to the backend coverage or something else, I can't completely confirm, but Michigan needed to find other and more ways to put those guys in conflict other than simple counter plays (which I still think were a good add, just not enough).



The jet sweep action is in line with this. Michigan has not been a great sweep team this year, but they have run it several times every game. They threatened it a few times, but to my recollection never pulled the trigger. This sort of motion was key to pulling LBs out of assignments all year, and they didn't do enough to threaten it. The split flow and other sorts of things are the same idea. Get the LBs reading flow in multiple directions - besides from just to OL - and watch them get out of run lanes. The double Down G is another option in Michigan's playbook that we never saw (albeit, Michigan ran that a few times this year and never really successfully).

I don't think Michigan needed to drastically move away from the playbook, despite what many fans said. The pieces are there, and given Michigan's success on the ground and OSU's issues on D, the idea of taking a huge risk by trying to execute a bunch of unknowns doesn't seem like the best risk balance. But there were parts of the playbook that didn't get touched that could have been, and could have really helped the offense in general.

Michigan Offense - Management
The first down run/pass balance was off. Michigan wants to be run heavy. They want to wear down defenses. This can be great, especially when your defense is getting off the field. Michigan's D was not, and Michigan was forced to play catch up. The theory needed to change earlier than it did, because the offensive philosophy needed to change to "keep up" instead of "wear down".

The gameplan I don't think was bad to be majority run on first down. It was clear once the game started that Michigan had issues with protection, and I think Michigan knew that would be an issue in this one. Michigan is a very bad "behind the chains" offense. Negative plays or incompletions really hurt them from an efficiency standpoint, and really limit what the offense is capable of. Getting into 2nd and 7 isn't great, but it puts the offense in a much better position than 2nd and 10, because of what they are. That's been their bread and butter all year. So the gameplan going into the game was fine, but it needed to adapt quicker.

I think the best offense to attack OSU was more of a spread look, but that isn't Michigan's base playbook. They utilize more 12 personnel. They utilizes their TEs in ways to attack the defense similar to many spread teams did, but it's less effective from a speed standpoint, and Michigan failed numerous times to execute it (short arming passes, drops, etc.) despite it being schemed well. They also utilized a bunch of 11 personnel. I don't really have a huge issue with the personnel packages that were utilized in this game. People complaining about "walk-ons getting snaps" should be ignored. Wilson got snaps because he's shown this year to be value added for a handful or two of snaps a game. I don't know why McCurry got snaps, but I'm guessing it's because the coaching staff felt there was something he did well at the position he was asked to play, had gotten reps at that position, and they felt that gave them the best chance to be successful.

My biggest gripe is the management of the backup QB situation. Peters has looked solid when playing. Him going out there first indicated to me that the coaches believed he was the true backup heading into the game with McCaffrey out. He came in, threw a great pass, and then they put in Milton. Milton is a super-raw true Freshman. They threw him into a game where Michigan was behind, unlikely to catchup, but was still going to be pass heavy. The protection was struggling. He promptly threw into double coverage for an INT. Even his long completion was a bad decision where the WR made a great play. At least it didn't burn his redshirt, but there was no reason to throw him in that situation. Even if Peters is back next year and you are using him as the primary backup short term, that is putting Milton in a position where upside is "snaps in hostile environment" and downside is hurting your team significantly and doing nothing for his growth. I hated that move in that situation.

Michigan Defense - Gameplan
The initial gameplan was bad. Michigan got beat on crossing routes against IU. They got beat on crossing routes by Haskins last year. To come out and play their City (Cover 1) and let it go was a bad decision.

OSU did a think Michigan hasn't seen much this year that helped drastically. They would play their outside WR off the line, short motion him into a stack, and run both receiver inside as part of their mesh scheme. With the way Michigan runs their banjo coverage against stack, this meant the inside coverage had to follow the up-WR, and Watson had to follow the underneath WR across the field. This means Watson started out further outside, wasn't able to get his hands on the WR off the LOS, and then had to chase him through the wash or underneath the routes. Watson isn't that level of athlete (which is why the previous staff moved him back to safety), and OSU picked on that coverage repeatedly. It was great scheme by OSU and a bad initial gameplan from Michigan.


They also utilized motions and formations to get Gil matched up on crossers, specifically with McCall, a talented receiver for a RB.


Michigan started out with a lot of 5 man pressures because the film says to get Haskins off his spot. For whatever reason, these were not effective. OSU's pass pro held up all day really, really well. The crossing routes help because the QB can buy time in the pocket and only needs to complete short routes, but the pressure never disrupted his timing. Without getting OSU behind the chains, Michigan struggle to get to exotics. That really hurt that part of the game plan.

Michigan Defense - Adjustments
Contrary to popular belief, Michigan adjusted fairly quickly. They went more to their zone, specifically, their trap 2. After a few successful plays, OSU successfully adjusted the offense to bait the trap coverage and get open behind the CBs or suck the safeties up with crossers and then work behind them. OSU scouted this defense well and understood the likely adjustment. Michigan needs to self-scout a bit before the bowl season and especially over the off season to better understand what tendencies are on film and how better to break them so that offenses can't adjust quite as easily.

Coming out of half time, Michigan made further adjustments. They cut some crossers, they ran some other zone variety, and OSU switched up their attack to take advantage of those things. The idea that Michigan didn't make adjustments in this game is false. They
did. They did quickly, and then fairly extensively at half. OSU has a great offensive staff and was very well prepared and Michigan was caught trying to catch up the entire time. And then it snowballed once they were behind.



Michigan Defense - Personnel
Obviously Watson got picked on. A lot are asking why he was out there playing man-to-man. Why? Because he's Michigan's 3rd best CB and was the best option for them at that spot. Ambry Thomas has more athletic ability, but has struggled with technique and executing assignments consistently. Watson was the best option, and you play your best options. Especially after Long got dinged, they had to play him. But there is also a reason Watson was moved to safety by the previous staff, he has athletic limitations. OSU schemed and exposed those by not allowing him to utilize his hands.

I wasn't a fan of Michigan splitting snaps so much between Ross and Gil. Ross is the better player, and OSU attacked Gil when they had an opportunity. They are splitting snaps to keep both fresh, to allow Ross a different vantage point and more coaching on assignments and techniques, and because both are likely starters next year. But in my opinion, the split should have been closer to 3-to-1 in favor of Ross instead of the typical 50-50 split.

Michigan Defense - Going Forward
I don't have the answers to this one. Brown hates quarters coverage. There is a reason Big 12 teams run it so often. Brown has some things in the playbook from a matchup perspective that are similar, but they aren't always great playcalls on standard downs because it leaves the box under-manned from depth. He needs to look at adding some new wrinkles. It probably won't be quarters, but some means of getting inside leverage or switching up auto-checks or rotating coverage on the backend to prevent some of these inside breaking routes from working, and then not allowing the offense to so easily adjust.

Brown is still a great DC. He still has a very stout scheme. OSU picked on matchups (mostly Watson, also Gil and to a lesser extent Kinnel). They knew those were the matchups they wanted and attacked them. I'd like to see Brown do somethings to change the checks and who follows due to motions so that offenses can't as easily dictate those matchups. And against teams with more speed, start with the zone looks primarily, and utilize the Cover 1 as the changeup.

Michigan Defense - Defensive Line
We saw Michigan's interior DL really struggle against ND. OSU I don't think has an ND caliber OL, but OSU had similar success in terms of pass pro as ND. It really starts on the interior, where Michigan is fairly limited. But the outside rush wasn't getting home either. Michigan was utilizing Winovich early by spiking him inside a bit, this in an effort to mitigate OSU's rush threat. That was mostly very successful (against the run), but limited the pass rush, and they were unsuccessful with other methods of getting home (and the ball was out quick, so blitzers usually didn't have much time). I don't think there is an obvious or huge thing that led to this, OSU just played their asses off and executed.

So this is likely a combination of scheme, initial game plan (which has some logic but was adjusted later as things started getting out of hand), and just not playing well enough.

Overall
OSU had a great game plan in this one. Michigan had I think an understandable offensive gameplan that didn't adjust quick enough and left out some questionable aspects, and a very flawed defensive gameplan that was left playing catch up with OSU's adjustments.

In the end, Michigan got flat beat. OSU had playmakers and a game plan that Michigan couldn't match, and they knew how to take advantage of it. I get annoyed that some Michigan fans apparently had the expectation that this would be easy. OSU has some of the best offensive coaches in the country. They have the 2nd best HC in modern football. They have an offensive coach that was recently a great offensive HC and a previously great OC. They have an up-and-coming OC that will soon be a HC. They upgraded at WR coach in the off season. The offensive staff is as good as any in the country. Having a good defense isn't enough. And when OSU found a great gameplan, it exposed the defense for the deficiencies it has. Every team has them, some defenses get blitzed, it happens, it sucks, but you have to adjust, clean up some techniques, self-scout, and come back at it. I have little doubt Brown will do that.

There are no excuses in this one. The refs were fair to kind for Michigan. Michigan was as healthy as they've been in a long time, and a few players getting hurt during the game didn't leave Michigan in a worse injury situation as OSU (who was without Bosa). Michigan got flat whooped, no other way to say it. Whooped schematically, whooped from adjustments, whooped from talent. It sucks for Michigan fans because it's a regular thing. But willing it to change isn't enough to change it. Michigan is in a better position now than they were when Harbaugh arrived. They are in a better position now than they were last year. They'll try again next year, hopefully with a better plan, and one of these days it'll be enough. Until then, the monkey will stay on their back, and they'll get the best of OSU, and it'll be tough. Just don't expect the tides to turn or some other BS. OSU isn't going away, but Michigan can find ways to win going forward, and frankly needs to. Getting pissy won't help do that though.

19 comments:

  1. This was a *fantastic* article. Thank you for taking the time to break everything down like that. Understanding where M failed helps me feel a little less hopeless about the outcome. Heading into the game, I felt that M was about to get exploited through the air and lose a shoot out. I enjoyed getting a deeper understanding of the X's and O's and the specific match-ups that OSU targeted. Thanks again.

    ReplyDelete
  2. Great post! Quick question, What is your opinion on the match-up zone as a defense (similar to what Wade Phillips has done in Denver, Dallas and LA)? Does Michigan have the personnel to run this scheme? Has Don Brown ever run match-up zones in the past?

    ReplyDelete
    Replies
    1. They ran their version of Saban's Rip/Liz, a single-high pattern match scheme. It wasn't really any more successful. Think they would benefit from looking at a two-high version. Brown isn't a fan of quarters, so not sure how that will go. They have some bracket coverages that help, but those don't work as well on standard downs.

      Delete
  3. Thanks again Space Coyote- I suspected quite a bit of what you wrote, but it is always nice to have confirmation from a such a knowledgeable source. It sucks to say it every year, but if we continue to move forward and improve, the gap will narrow enough that we will eventually get a win.

    ReplyDelete
  4. Great article. You explained in detail what my gut was telling me during the game: that every adjustment we made was already planned for, and OSU countered quickly and efficiently. In the end, no matter the schemes, if your D-line can't get pressure on a young QB, none of the other Don Brown stuff works very well. Great game prep by OSU, I hate to say.

    ReplyDelete
  5. Can you explain what you mean when you say Michigan "cut some crossers"?

    ReplyDelete
    Replies
    1. In this case it's a robber scheme. For a crossing route, the outside DB will replace the deep middle safety. The deep middle safety, when he identifies the crosser, will break forward in an effort to "cut off" the crossing route. It keeps you in a single high, Cover 1, but depending on the route structure, a different player will be that middle defender.

      That's what I'm showing in the "Mesh Out" diagram above, two players seeing the inward breaking route, reacting, and then the play breaking back outside.

      Delete
  6. Who were teams that had better success against the OSU O (MSU?) and I guess they ran quarters and played better? Why does Don B hate Quarters and what talent upgrades do you think M need to make on the Defensive side of the ball?

    ReplyDelete
    Replies
    1. Not sure what Brown has against quarters, but trap-2 is effectively a Cover 2/4 hybrid, so it's not like he's abandoned it all together. MSU did have more success against OSU, but OSU also made strides during the season to continue to improve. They were the #2 offense in the country, so it's not like they were bad in the rest of the games.

      As far as talent, they just need to continue to build depth. Better athletes at CB will help. I don't think it's a talent issue on the DL as much as they just need to get better, but talent always helps.

      Delete
  7. Space Coyote, a couple of questions:
    1. I am still puzzled by the lack of a pass rush. Chase didn't look right to me. Gary seemed non-existent. I would have thought Michigan would have picked on Isaiah Prince. Solomon didn't play either did he?

    2. Zone read -- I think Shea ran twice all day (the second was when he was injured?). Was it me or did he not pull the ball a few times when it appeared he could have? Perhaps I misread it, but it seemed like at least twice he had situations where if he pulls the ball he has 10 - 15 yards of green space, Thoughts?

    THANKKS!

    ReplyDelete
    Replies
    1. 1. No Solomon. Chase didn't look 100%, and I'm guessing Gary still isn't. But think OSU executed really well as well.

      2. Didn't really look for more opportunities for Q run. Wasn't really a huge part of the game plan, as it typically isn't. Think the score prevented them getting into it more (Michigan typically ran it on 3rd and short and in the 4th quarter, so lack of runs isn't completely surprising).

      Delete
  8. Great writeup SC, follow your comments regularly and always feel well informed, always helping instill a good sober attitude about things.

    One thing I've wondered about for a while is the whole "check with me" thing OSU has done for years (as have other teams like PSU recently, etc). Basically seeing our D alignment/personnel and then adjusting the play as necessary. I generally find this to be a somewhat lame and annoying way to play football, but in certain situations I do think it can potentially help gain an advantage, especially when you have 20 year olds out there who can only remember so many things. So my question is, do you think this is a useful approach that Michigan has stubbornly refused to experiment much with (outside of the Rich Rod era), and that it inherently disadvantages the defense? I get that you don't want to overthink things and you can get overly into a cat and mouse game (as Al Borges would say), and I do love when they make some long check only to be stuffed once again, but it is pretty frustrating when its working so well against your supposedly top defense in a game like last Saturday. Does Michigan need better disguising of the D? Moving around/stemming more? Should their offense be doing more of this, and they are wasting too much time in the huddle? Any thoughts you have on the topic would be appreciated.

    ReplyDelete
    Replies
    1. I know in that past Harbaugh would often send in multiple plays and have the QB kill plays where they didn't like the alignment. I didn't see much of that this year (though I believe it happened in certain situations).

      Brown is known for checking with the "check with me", so from that standpoint, Michigan's D is usually pretty sound. Movement is great when it works, but it also tends to leave you in a disadvantageous spot at the snap. So it's a balance. Don't think that would have made a huge difference.

      I would like to see more pace added to the offense. I don't hate the huddle like a lot of people, think it has its usefulness. But it would be nice to alter the pace a little from down-to-down to dictate terms a little more.

      Delete
  9. Great article! In addition to the issues Watson had, I was alarmed to see OSU RBs routinely match up against UM linebackers. While watching the game, I was horrified to see OSU's P. Campbell or M. Webber motion out of the backfield and a UM lineback (Gil or Ross) follow him out in man coverage. Did Michigan move to a Nickel or Dime look during the game? I just could not understand all the linebackers playing man in pass coverage.

    ReplyDelete
    Replies
    1. Michigan ran quite a bit of Nickel in the game, but when Long got hurt it limited their ability there.

      It's kind of a damned if you do, damned if you don't. You put a DB in instead of a LB and OSU has an advantage in the run game. But a DB out on a RB then a LB has to cover someone else.

      Delete
  10. Great article well explains what happen thank you now I see

    ReplyDelete
  11. Great article. Loved the depth of it which is well beyond the drivel you read on ESpin. Thank you.

    ReplyDelete
  12. I know this is late, but I appreciated the write up. It helped me understand the game more and also think about things with a level-head. These take A LOT of time to write, and I thank you for it.

    ReplyDelete