stephenrjking

May 24th, 2018 at 8:03 PM ^

I appreciate some of the deep diving this article does. I question some of its applicability, because it doesn't seem to account for talent (ie, Shea's success to the right side, under 10 yards, how much of that is due to whom he's throwing to) or offenses, which are widely different. At least, it seems, he can make throws to physical locations listed, so that's good.

This stat caught my attention:

Michigan's playaction pass usage dramatically dropped last season after Wilton Speight went down against Purdue. Averaging the output of John O'Korn and Brandon Peters, it comes out to somewhere around 26% of the team's passing snaps overall. The numbers for each quarterback were varying but none had much success.

This seems strange to me, because after Speight went down Michigan started to run the ball more effectively as they transitioned back to gap blocking.

Speight had the best grasp of the offense, of course, but PA isn't supposed to be complicated--it's something that can help QBs that are having trouble with more complex reads. It would seem to make sense to include more of it, not less.

So that leads me to consider possibilities why such a reduction in PA attempts would occur:

  1. The OL plain couldn't block PA plays effectively and the QBs risked getting killed (remember John Navarre getting destroyed at MSU on that one PA call?)
  2. The switch to gap blocking, done midseason, did not allow for a corresponding installation of play-action calls that used Michigan's new running action. If you show gap when you run and zone when you do play-action, that's a pretty obvious tip. Thus, the plays aren't as effective.
  3. O'Korn and Peters, or perhaps the receivers, weren't making good reads and the plays were guaranteed disasters.
  4. The offensive staff forgot about play-action and didn't realize they were doing this.

Given that O'Korn and Peters had trouble with reads anyway, and that the OL had blocking problems anyway (some of which could be deal with by rollouts), and given that Michigan's offense became more run-heavy as the season went on, I strongly favor option 2.

It's weird regardless, but option 2 both fits the facts on the ground and helps answer those who question why the offensive plays don't seem to "fit together." It also seems true, or at least I recall learning, that zone blocking does a good job setting up play-action with relatively conventional pass blocking, one of the potential motives to install it in the first place. Moving away from it may have hamstrung the play-action game last year.

You can do really tricky stuff like pull guards on PA using gap blocking schemes, but those take work to install and the shuffling can make pass blocking more complicated. So option 2 makes lots of sense to me. And suggests that next year, with just one scheme all season, things will be better.

Stringer Bell

May 24th, 2018 at 8:17 PM ^

Didn't we run play action on like 4th and 20 one time?  I don't think 4 was the issue.  And like you said, PA should simplify the reads for the QBs.  I think OL blocking was probably the biggest issue in that regard.  Shea and Ed Warriner are, without a doubt, the 2 biggest keys to a successful season.

stephenrjking

May 24th, 2018 at 8:47 PM ^

PA on passing downs gets attention when it doesn't work, but this isn't unusual for Harbaugh. That night game at Minnesota, for example, was won by a Speight TD on a PA call on a clear passing down. The play fit the criteria the coaches wanted (presumably, route combo, read, depth of routes, etc) and the coaches called it. People wouldn't have the same kind of reaction if it was a straight drop-back with exactly the same action by the receivers, but because there was that fake hand-off, fans tend to focus on it.

That said, I do find the reduction in PA calls interesting in part for that reason: the coaches are willing to call PA even in situations where the run fake isn't likely to have much effect.

UMfan21

May 24th, 2018 at 11:18 PM ^

PA generally requires the QB take his eye off the defense to sell the run. given how bad our OL was last year, it doesnt surprise me the coaches were hesitant to have the QB look away from the rush. furthermore, one devastating hit could have taken us to our 3-4 strong QB.

DoubleB

May 25th, 2018 at 9:58 AM ^

1) If under center, the QB will lose, for a substantial amount of time eyes on the defense. Most PAP routes would require the QB to then find the safeties and make their reads based on that, but it's possible O'Korn and Peters weren't very good at this. Recalling the bowl game, Peters had some curls that he hit for 8-10 yards off play action that he seemed most comfortable running, but those aren't prototypical down the field reads. The QBs may also have struggled with pocket presence off play action protection.

2) If the protection is done to actually influence the LBs and safeties to read run, the protection is just not as good as straight dropback. There's a reason all teams kick-slide or vertical set their OL on dropback pass pro--it's just better protection.

Perkis-Size Me

May 24th, 2018 at 8:12 PM ^

How can Shea improve the offense? If he can throw the ball more than 100 yards a game and not play a game of 500 with the opposing secondary, he’s already an upgrade.

MadMatt

May 24th, 2018 at 8:25 PM ^

Excellent insights relying on data.  This would hold up well during the season, and it is pure ambrosia in the dull days of the off season.

Rafiki

May 24th, 2018 at 9:31 PM ^

Shae doesn’t need to be great. The team around the QB will be as good the 2016 team so if Shae is as good as 2016 Speight the floor should be 9 not counting the bowl game. Shae being great would mean this team should finish no lower than 5 like the Athlon poll. The 2016 team played the same home/away schedule against the other top teams in the conference. The only difference is ND.

Eng1980

May 24th, 2018 at 9:38 PM ^

I remember a few of those dropped passes but geez.  Does anyone have some insight as to whether or not our new wide receivers coach(es) are certain to fix this?  I suspect that with more experience the drops become completions.

stephenrjking

May 24th, 2018 at 9:43 PM ^

Well, all of the receivers were really young, and our best receiver was hurt most of the year. Crawford, the worst culprit, will play less of a role this season. So the question to be asked is, will DPJ catch more of the passes thrown his way? 

I think the answer is yes. Coaching? Sure, but just plain experience and reps also.

Hail-Storm

May 25th, 2018 at 11:56 AM ^

Black had an outright drop.  I think he batted it down trying to run before he grabbed it.  I have to imagine the freshem to sophmore leap for recievers will be big.  Crawford doesn't seem to be the answer, and I don't think he will have a late resergence in his career.  Should have kept his name as Dylan (as a fellow Dylan, I think this was his biggest mistake). 

Tyler1495

May 25th, 2018 at 3:28 AM ^

Last year with a young recieving group. They were coached by a grad assistant. This season getting coached by a former head coach. He talked about every practice having them learn how to get better releases something that they struggled with. They def have a couple of homerun hitters with DPJ and Black. this offense can be explosive by the end of the season

Tyler1495

May 25th, 2018 at 2:50 PM ^

I think people writing of mcdoom too soon. Hes got great hands his problem like many of the other recievers was getting seperation from the line. At worst he can be a depth guy used in certain packages. like a percy harvin type of player

noahtahl

May 24th, 2018 at 10:47 PM ^

Okorn had 18 of his 151 pass attempts dropped. He caused all 18 with his bad reads.blah blah blah.so it is decreed by his higness brian of arbordale.

MgoWood

May 25th, 2018 at 10:08 AM ^

Guys..let's remember that Harbaugh himself, even admitted that HE underprepared the quarterbacks. He noted this during the Florida/Shark(now good shark) game. Trust is there that AN play will be on the up and up regardless of line play. Getting the ball out faster helps oline deficiencies.