Mattison and RPS numbers

Submitted by UMQuadz05 on September 12th, 2011 at 1:07 PM

Thinking about the game, is anyone expecting huge RPS scores (positive and negative) from the defensive UFR?  The schemes seemed very creative and effective on the two INTs, but there were also many, many picked up blitzes and runs right into a backpedalling Martin.

This reminds me of what Mike Lombardi calls "Battleship Defense" in the NFL.  The way he tells it, defenses that can't keep up will try to call the perfect play every time.  When it hits, boom INT.  When it misses...



September 12th, 2011 at 1:20 PM ^

After the first two games of the year it is nice to see that after the first few drives for the opposite team our D is able to adjust to what they are doing. Both games so far have started out with me thinking this looks no different from last year but as the games go on the D makes it adjustments and is able to come up with some stops or turn overs.

Blue in Yarmouth

September 12th, 2011 at 1:53 PM ^

That has ben the biggest difference in the entire team this year as opposed to last IMHE, on both sides of the ball. What has me genuinely excited this season is that we are seeing coaches make adjustments. We are also seeing some improvement.

I expect to continue seeing adjustments within game times and improvement over the course of the season, which my biggest issue with the previous staff. I loved RR (and still do actually), but a team should be constantly improving throughout the season, not just during the offseason and remaining stagnant throughout the season. 


September 12th, 2011 at 2:33 PM ^

High variance, but neutral to slightly negative overall, I think.  Lots of zone blitzes where either Rees checked to a run or rollout away from the blitz, or they just ran it up the middle and we vacated the center of the line.  That will more than negate the 3rd and one stops, at least on the RPS number.  We won, so the RPS number means little to me today.

Franz Schubert

September 12th, 2011 at 5:33 PM ^

How many big plays (25 yards or more) did ND have in the whole game? UM is doing a great job of balancing blitzing without giving up huge plays. This is a bend but dont break defense and im fairly encouraged. Another major difference from last year is the turnovers, this D is creating turnovers.

Something lost in the crazy ending is that after the first two ND possessions, the UM defense only allowed 10 points until the last drive of the game, and only 17 including it. As any UM fan can tell you after last year, yards dont win games, its all about scoring points and this Defense did its job very well for at least 3 quarters last night.


September 12th, 2011 at 1:28 PM ^

the plays where martin dropped into coverage seemed to go horribly. they were basically gambling it would be a pass play, and when it ended up run, they were woefully unprepared to stop it for anything less than 10 yards. i remember at least two instances of that.

Deep Under Cover

September 12th, 2011 at 1:55 PM ^

Not sure how he is taught to do that, but his first move is always a few steps back on the zone blitz where he drops for pass coverage (like there is no read involved, which I think would be tough from the DT position).  His strength is, well, his strength, so we do ourselves no favors by getting him moving BACKWARD at the snap of the ball if it is a run.

I like the idea of disguising our defense, but it seems parts need better execution.


September 12th, 2011 at 2:08 PM ^

should be dropping back on pass play. Maybe what you saw was a stunt where Martin drops as a LB and a LB was supposed to go into the gap. But if the LB thinks "pass" and forgets his stunt and drops too then there is no rush. If they are doubling Martin they need to run stunts to get him and hit talent free.


September 12th, 2011 at 2:12 PM ^

watch the tape. they definitely dropped martin back into a short zone a couple times. both times were passing downs if i recall correctly, but they ran anyway. the vikings did the same thing with jared allen against the chargers yesterday and came up with a pick because the qb didn't expect a DE to be in that spot.


September 12th, 2011 at 3:48 PM ^

of MM dropping - that is the stunt I was talking about in my original post. It would be foolish just to have the DT drop back with no LB stunting in his place.

A DT dropping back isn't going to panic any QB. He is busy checking off downfield receivers. The point of the stunt - if a LB did fill in - is to confuse the OL, not the QB.



September 12th, 2011 at 3:43 PM ^

you're not getting it. i'm not hypothesizing what happened. mike martin actually did drop into a short zone on a couple plays.

as michigan mike pointed out, it's a way to disguise the coverage. he's basically swapping duties with a lb, who takes over his pass rush. the idea is that the qb sees the lb coming in, thinks it's a blitz doesn't see a lb in the flat, and throws a short pass without noticing martin dropping back into coverage. or something thereabouts. it's not a bad playcall, but when nd chose to run on those downs instead of pass, it does not turn out well.


September 12th, 2011 at 2:48 PM ^

I don't think they are being too cute. I think they have a problem. The problem is, their defensive personnel are just not that good overall. They have GOT to try something to anticipate, even to an excessive degree, what the offense is doing, even if it means getting horribly beat on occasion. When they have a front four that can penetrate, or even just hold the line of scrimmage on a run play, then maybe less extreme guesswork is in order. When they have two corners who can cover a division 1 receiver, maybe less guesswork is in order. But for now, they don't have those things.

I'd rather do it this way, force some TOs and TFLs, and get killed occasionally than just sit back in a vanilla defense and get murdered all day (2009/2010).



September 12th, 2011 at 3:03 PM ^

Agree, and seeing the results of the play in total, it's odd to say, but it kind of worked.  I believe (no way to check if I am right of course) that the D play was designed to stop the quick slant for the 5 yard first down throw..  most likely to Floyd.  Tough to make that throw with a large DT in your throwing lane.

AND, in the end it could not have worked out better for UM.  A five yard first down and ND might score with less than 30 seconds left.  Game over.

A little convoluted?  sure.


September 12th, 2011 at 1:49 PM ^

...and only the first half, Michigan was often unlucky on the RPS front.  The most noticeable problem to me ( I've posted this w/in the last day or so) is that Michigan sometimes lines up with all of its LBs on the LOS.  No one is behind the DTs at that point.  ND and WMU both used those opportunities to gash Michigan with draws up the middle.  This is the only major systemic flaw that I saw in the first half.*

The problem that Mattison has is that he doesn't have a DL that can get a rush with four men and he doesn't have corners whom he can trust to put on an island.  The only thing he can really rely on is that Martin, Van Bergen, Heininger, and Demens aren't going to get pushed around too much. 

*I say that as if I know what I'm talking about, which I do not.

Mr. Yost

September 12th, 2011 at 2:01 PM ^

#1 Craig Roh is either sick or loves bearhugs...he simply can't get off blocks.
#2 Mike Martin should be an all American
#3 Ryan Van Bergan inside is Pat Massey 2.0, that's not a good thing. Unfortunately Campbell hasn't been consistant enough
#4 Campbell is getting better, but can get be "starting" by the MSU came to allow RVB to return outside?
#5 We have no WLB --- Jones hasn't gotten a shot, Hawthorne is better than expected, Herron is boom or bust, Morgan isn't ready. No consistancy.
#6 Demens, yes.
#7 Ryan, yes.
#8 Woolfolk :( ...please get healthy!
#9 J.T. Floyd really can NOT tackle
#10 Avery is still short, but still better than Floyd
#11 Marvin Robinson can't cover...use him in the box stopping the run or rushing the QB only!
#12 Kovacs may be an All-American...and not just academic.

#1 When does Campbell get better?
#2 Where are Q and Ash?
#3 Can Cam play WLB?
#4 Is it time to try Blake Countess?
#5 Is it time to try Tony Anderson?
#6 Why do we take a LB off and bring in M-Rob as a safety in passing situations?


September 12th, 2011 at 2:53 PM ^

On your questions- For 3 years I've been asking these kinds of questions, and had no faith in the ability of the defensive coaches to answer them- I've snorted loudly when people would say stuff like "The coaches know better than you or I".

Now I believe it- If Countess and Richard Ash aren't getting in, I have a strong suspicion it's cause they either aren't good enough period, or aren't ready yet.

I think RVB has been better than Pat Massey, even inside. I agree, he's an end, not a tackle, but he still was in the backfield a decent amount on Saturday, including some darn big plays.


Indiana Blue

September 12th, 2011 at 9:05 PM ^

would totally disagree with your assessment of JT Floyd, and I certainly do.  GM played nd wide outs man to man almost exclusively.  Michael Floyd will be the best receiver Michigan will face all season ... and GM put Floyd on him when we were man to man, all game long.  Floyd's longest catch was that 30 yarder, and that was a perfect throw.  Overall I was very please with JT's performance, including tackling.  Perhaps Woolfolk is our best cover DB, but not with a hand in a cast ... so Mattison gave it to JT.

JT had 5 solo tackles & 1 asst, 1 PBU (U of M had 2 total), and 1 INT ... covering Michael Floyd.  I say this is pretty damn good for a guy who hasn't seen that much game action witht  the injury last year.  Troy's in the same position in term of game action.  I truly believe they will get better every week.

Go Blue! 

turd ferguson

September 12th, 2011 at 2:34 PM ^

Did anyone else think that the blitzes seemed to be coming from farther behind the line than they did against WMU? Slow-developing blitzes are a pet peeve of mine, since they rarely seem to work, and it seemed like we ran a lot of them on Saturday.


September 12th, 2011 at 3:46 PM ^

I believe during Martin's post game interview he mentions that ND was changing up their snap counts to mess with Michigan's timing.  This would definitely affect the ability to generate 'quick' blitzes.

More than the slow blitzing issue is the fact that ND consistently checked plays away from the blitz.  They did a good job or running away from the pressure or throwing over the blitzers.  I think the RPS should be negative for the day, but I believe Mattison is running a high rish / high reward defense on purpose. 

Also it was mentioned by someone in the post game presser that they wanted to keep Michael Floyd bottled up.  In essense that did work to some extent.  Floyd didn't get to break off any deep routes on us.  So you might give up the catch in hopes to keep it in front of you.  I thought Michigan could have jumped more of those out routes, as Kovacs did for his interception and Hawthorne did for a pass breakup on one of the late drives for ND.


September 12th, 2011 at 4:25 PM ^

I guess what I don't understand is don't most teams change their snap counts?  One play they might snap it on the first 'go', next play snap it on the 3rd 'go', next play snap it on ready. 

Do teams really say - we're going to go 'Ready, Set, Hut' and snap the ball every time on Hut???


September 12th, 2011 at 2:54 PM ^

When I saw the replay breakdown of the Notre Dame's go-ahead touchdown, I thought "What's the worst RPS number Brian is willing to give?"



September 12th, 2011 at 4:30 PM ^

Some of those ND wins need to go to Rees (unless he was getting signals from the sidelines).  It seemed there were a handful of times where Rees successfully changed the play at the line of scrimmage after seeing UM's alignment.


September 12th, 2011 at 6:06 PM ^

but watching the replay of kovacs pick was terrifying. The play may have been designed to do this, but kovacs left his man to make the pick which left the WR open by about 15 yards downfield. if Rees hadn't been locked on Floyd, he might've scored a dagger TD to make it 21-0