Talk about the line play through three games. Some room for improvement, do you think?
“Yeah, absolutely. There’s always room for improvement. They’ve done a good job and we’ve got to keep on getting better just in pass protection, which is—we’ll do that, and communication. But they’ve done a nice job, and we expect a big game this week.”
Do you see guys still playing too high at times?
“Yeah. You always can improve on the offensive line with pad level. You’ve got to get lower and get your eyes in the right spot and be physical with your hands and move your feet and know where you’re going and communication and targets and that sort of thing still needs improvement. Every day you’ve got to work on that and that’s important.”
Do you still see a rotation with the two Bens at left guard?
“Rotation’s always open. Rotation’s always open.”
Talk about Bredeson’s progression.
“Done a nice job. Really has. He plays, like I mentioned before, quick twitch and smart and can process quickly on his feet. Like all those other freshmen, has done a nice job transitioning there and really improved.”
What did Wilton show you in that game? You take the first game interception and then the big hit and coming back in.
“Wilton’s a true competitor and he’s a tough guy. I think anybody who plays the quarterback position’s got to be a tough guy. You’re gonna take hits you don’t want to happen, we don’t want to happen, but it’s how they respond. I mean, that was a tough hit on him and he bounced up and he led us the rest of the game and took us back from being behind. Just really pleased with what he did. It shows his true character and how important it is, the team to him and him being a competitor and winning every down.”
[After THE JUMP: what’s up with the Morris/O’Korn thing, complimenting large gentlemen plying their trade, and a big compliment for Speight]
|Ben Bredeson||Fr.||Ben Braden||Sr.*||Mason Cole||Jr.||Kyle Kalis||Sr.*||Erik Magnuson||Sr.*|
|Grant Newsome||So.||David Dawson||Jr.*||Patrick Kugler||Jr.*||Michael Onwenu||Fr.||Nolan Ulizio||Fr.*|
Michigan's line took a quantum leap in 2014, going from a flaccid crew of confused gibbons to pleasingly mediocre. Last year's edition of this post positively marveled at the fact that these gentlemen got in the way of the opposition frequently enough to be average-ish:
It got better. It really did. The OL nadir is in the past. We can come out of the bunker and rebuild society now.
That assertion was based both on my charting and a bunch of stats, many of them of the advanced line variety. Advanced line stats make total guesses about assigning credit and blame between tailback and line but they're worth peeking at in case they tell a story. Michigan's 2015 stats are mostly about treading water:
|Year||Adj Line Yards||Opportunity Rate||Power Success||Stuff Rate||Adj Sack Rate|
Michigan was less likely to get tackled for loss and less likely to get the 5+ yard carries that opportunity rate tracks. Those were a wash as Michigan's line yards stayed static. Contrary to your memories of the OSU game, pass protection took a big leap forward.
A certain level of treading water is expected when a new coach with a new, complicated system arrives. With four starters back and Mason Cole moving to his natural position, a step forward is likely. It's just that fifth guy who gives pause…
An Editor's Note About Charts
With four returning starters you're going to see a bunch of charts derived from last year's UFRs. Here's how to read them:
|Game||Opponent||+||-||TOT||Pass -||Error Rate||Comment|
|1||Utah||5||8||-3||5||8%||Guy did X|
Game and opponent are self-explanatory. The +, –, and TOT columns are my evaluations of how the player did when run blocking. Keep in mind that zero is not good, or even average. It is the nature of the beast that any successful run has many successful blocks; many unsuccessful ones are submarined by a single error. We're looking for a 2:1 positive-negative ratio to be decently successful. A future pro might be more like 3:1 or 4:1.
"Pass –" is derived from the protection minuses in UFR. Two protection minuses are approximately equivalent to one sack or severe hurry. "Error rate" is the number of protection minuses divided by the number of available protection points. The above line is Ben Braden's from the Utah game, in which he was almost 1:2 in run plus/minus and had protection errors on 8% of snaps. That's terrible; the good news is that Braden got better.
TACKLE: JUST A GUY WOULD BE FINE THANKS
present, he said [Brian Fuller]
Senior ERIK MAGNUSON was thrust into the lineup too early as one the umpteen guys tossed into the maelstrom of the 2013 offensive line. He was a guard then; the next year he played some there and, after an injury cost him his job, as a blocking tight end. Last year he got flipped out to tackle.
There he... well, he was there. He was neither forceful nor overrun. He didn't shut down elite pass rushers or get blown through by mediocre ones. His UFR chart from last year is decidedly sparse when compared to Cole's:
|2||Oregon State||5||0.5||4.5||0||0%||Not as involved as others but got his job done.|
|3||UNLV||6||6||0||0%||Clean positive sweep from the OL.|
|4||BYU||2.5||3||-0.5||0||0%||M clearly left-handed when it wants to rely on tackles.|
|5||Maryland||4||4||0||2||4%||Clear left handed bias again.|
|6||Northwestern||6.5||1||5.5||4||10%||End of game was pretty.|
|7||MSU||3||2||1||3||7%||A little frustrated with his second level blocking.|
|8||Minnesota||5||1.5||3.5||1||2%||Good day for him although M is clearly left-handed.|
|9||Rutgers||3||4.5||-1.5||0||0%||Not real good on perimeter.|
|11||PSU||3||3||4||6%||Also took advantage of weak edge.|
|TOTALS||55||29||26||24||4%||65% run blocking|
It's not so much that Magnuson didn't execute, it's that he wasn't called on to do much. He's right around our run-blocking Mendoza line thanks to some good days against the overmatched bit of the nonconference schedule. 24 pass protection minuses over the course of a season isn't anything to write home about, but Cole's maturation and Magnuson's move to tackle are the top two reasons Michigan's pass blocking got a lot better a year ago. When I started to talk about the OL individually in the middle of last year this was the conclusion:
Magnuson is [just a guy] right now. He's okay at blocking. They don't run to him very much. There are not many plays on which he has a big role and that seems to be about half Cole and half Magnuson. He is the Jarrod Wilson of the offensive line.
He's boring. We appreciate this immensely, because we are well aware of the alternatives to boring after the past half-decade.
It's maybe a little disappointing that Magnuson seems to be topping out at boring. I usually pick out the particularly good or bad plays to embed in these previews; Magnuson doesn't have anything to embed either way. On the ground I had him for zero +2 blocks a year ago and one –2 block. Part of the reason he doesn't have a lot of magnitude in that chart above is that he usually does something completely adequate and not that notable. When he does score a plus it's frequently for excellent awareness. Here he reads a blitz and manages to redirect enough to hit the linebacker who would otherwise be burying Smith in the backfield:
When Magnuson does move a guy it's usually because the guy is already moving. He was good at reading and staying attached on slants in Michigan's zone game; a bunch of cutbacks opened up last year because he was able to shove a guy past his intended destination.
This is a power play but it's the same principle and from a camera angle that makes it very clear:
The other times Magnuson moves a guy is because he's already engaged with Kalis:
Magnuson was effective at doubling a guy and popping out to the second level.
These are all real assets. They are offset by what I described as a "lack of oomph" after the Indiana game. Magnuson is not likely to get drive in a one-on-one block, and occasionally he ends up looking a bit… finesse.
That play was an outlier but I don't have anything in the way of a one-on-one drive block in an entire season of clips. This is an area he should get incrementally better in since he's got another year of weight training behind him; the time for big leaps forward is likely past.
Not everyone is as indifferent as this space was. CBS NFL draft analyst Dane Brugler called him a "legitimate NFL prospect" and "one of the top ten senior offensive tackles in the country."
...moves with a smooth shuffle and wide base, transferring his weight well in his kickslide to mirror edge rushers. He stays low off the snap and prefers to use his hands to control the point of attack to out-leverage and out-power defenders. Magnuson is able to secure downblocks and anchor at shallow depth, driving his legs to finish in the Wolverines' power offense.
I disagree with this take, but it's out there. NFL.com's Chad Reuter told Mike Spath that Magnuson could work his way into the first or second round with a good 2016; I disagree with that take as well… but it's out there.
Magnuson was relatively advanced mentally a year ago and will benefit less than some of his compatriots from increasing familiarity with the offense. Improvement should be clear but not transformative; a good goal is for Magnuson to move beyond Just A Guy status, get on the All Big Ten team in a very down year for the tackle spot conference-wide, and get drafted late.
[After the JUMP: the biggest question mark on the team. And Jabrill Peppers! (Not really. But maybe!)]
How does a mediocre running team do this:
…to one of the best run defenses in the country? Let's discuss.
Entering this game Florida was fifth in rush D to S&P+, giving up just 4.4 YPC with sacks removed, which means they turned their opponents' running games into basically Florida's freshman-infested running game. In our diagram for Ace's FFFF we gave 7/11 defensive starters a "dangerman" star, and three(!) got the shield that's reserved for the top players at their position in the country. Among them was end Jonathan Bullard, perhaps the best run-defending player in the nation, and strong safety Marcus Maye.
And yet Michigan, whose running game was barely better than Florida's this year, ran on that: 225 yards on 46 carries. Brian mentioned in the game column that a lot of this came from a reinvigorated De'Veon Smith, and from my rewatch I bet you the UFR will bear that out. There was also a little RPS, some great plays by individual offensive linemen (Mason Cole and Graham Glasgow had very good games), and some Harbaugh games.
Part I: Florida throws paper
The run above, which came mid-way through the 2nd quarter, must be taken in context. Smith by this point was already well up in +'s running Michigan's base stuff, and Rudock was well on his way to an excellent day.
Florida's pass defense is just as good as their run D, and in passing downs they're lethal. So it's worth it to them to try some surprising/a little unsound things on 1st and 2nd down to keep Michigan's offense behind schedule, and bring up those long situations. Against Michigan in particular it appeared Florida's gameplan was to take away inside runs, trusting the Wolverine backs not to find any holes that left elsewhere.
Let's first go back to an earlier thing Florida did that should have earned a TFL:
This is the one where the broadcast crew infamously put up the ND and OSU scores mid-play. What Florida called (I think) is below:
It's an under but the 5-tech is actually in a 7-tech.
Michigan tried to run "Power" into this, pulling Kalis, blocking down on the DL and using the fullback as lead blocker. Florida is ready, as the NT shoots up the back of Glasgow before Braden can get an angle on him; Braden blocks air, and Kalis runs into the nose in the middle of his pull:
Smith did a very good job to dance around that NT, then to bounce outside the unblocked MLB whom Kalis was heading toward, turning minus two yards into four.
The lesson: Florida was messing with Michigan's power running attack by sending rushers into the A gaps, leaving the 'B' gaps open, but only as a funnel directly to the linebackers. Yes, you'd like Braden to get a better block. Yes, that was a great play by the NT to get into the path of Kalis. But this was Florida's plan against Michigan's base thing, and they got the base thing. RPS'd.
[After the jump we RPS right back]
back back back back not back (but there's another guy also back) [Bryan Fuller]
|Mason Cole||So.||Ben Braden||Jr.*||Graham Glasgow||Sr.*||Kyle Kalis||Jr.*||Erik Magnuson||Jr.*|
|Logan Tuley-Tillman||So.*||David Dawson||So.*||Patrick Kugler||So.*||Blake Bars||Jr.*||Juwann Bushell-Beatty||Fr.*|
It got better. It really did. The OL nadir is in the past. We can come out of the bunker and rebuild society now.
By any reasonable metric it in fact got a lot better. Michigan's YPC leapt 1.3 yards, going from 11th in a 12-team league to 8th in a 14-team one. If you only look at Big Ten stats Michigan is still 8th, and that's in the division that had the MSU, OSU, and PSU defenses. Sacks allowed had a near-identical improvement, going from barely better than Purdue to a spot in the respectable midsection of the conference. Advanced stats saw something similar. Michigan finished 50th in adjusted line yards*, 32nd in power success rate, and 72nd in adjusted sack rate. And that's with the running backs going the wrong direction constantly.
None of these numbers stand out, but neither do they linger in the seedy parts of the list next to Temple and Penn State and Louisiana-Monroe. They were average-ish. Most of the time they felt average-ish.
That's not a bad place to be with zero seniors and just two upperclassmen. While the unexpected departure of Jack Miller puts a small dent in the front's depth, his likely replacement, Erik Magnuson, is a redshirt junior who came in as a touted recruit and has a season's worth of starts to his name already. If you're willing to fudge a bit, Michigan has five starters back from that okay line. They've replaced Brady Hoke with Jim Harbaugh and Darrell Funk with Tim Drevno.
Could they be… good?
*[A stat that weighs the first few yards you get heavily and discounts long runs in an attempt to get a feel for how the line is doing.]
Cole coped [Eric Upchurch]
Rating: 3 of 5
Last year I gave this a 1 because Michigan was staring down a starting lineup consisting of a true freshman and a third-year guy who didn't even get a sniff during the chaos of 2013. They seemed to acquire these positions almost by default since the only other tackles on the roster were freshmen (redshirt or true) regarded as huge projects. Or they were starting at guard.
The 2014 edition of this post took a quick gander at what happened when football teams started freshman tackles. It was almost universally ugly. The best case scenario was Ole Miss, which deployed ultra-blue-chip Laremy Tunsil as a true freshman and was middling in both YPC and sacks allowed. All others trundled their way to seasons that were more or less disastrous.
Michigan was not disastrous last year. In what is certainly a first for offensive line projections in recent history, that prediction was pessimistic. MASON COLE, the true freshman, just about hit the top end of reasonable projections, those being:
The occasional freshman tackle can cope. I think Cole is one of those guys. But is he going to blow a guy off the ball and provide a big ol' lane at 292 pounds? Probably not. Our hope here is that Cole is a solid, agile pass protector in year one who is a meh run blocker.
Mason Cole coped. This is what that looks like in the run game according to UFR:
|App St||9.5||2||7.5||A fine debut|
|Notre Dame||4||2||2||Didn't seem overwhelmed at all.|
|Miami||8||5||3||Okay; also had one very bad pass pro.|
|Utah||3||2.5||0.5||Nice seal block ignored.|
|Rutgers||5||5||0||I'll take it.|
|Penn State||1.5||5||-3.5||Hull and Zettel bad matchup.|
|Minnesota||4.5||2.5||2||Overpowered a little but still okay day.|
|Indiana||3||0.5||2.5||Didn't mess anything up.|
It was rougher in pass protection, where I have had him for 16 protection minuses in 8 games. Those are worth about half a QB pressure/sack each. Otherwise things were pretty okay against non-elite defenses. Cole's debut season was… eh.
When it didn't go well, the usual reason was that Cole got blown backwards because he was a true freshman. These were the kind of things that were happening against Zettel and company:
This was a game in which Cole's inexperience and lack of big skrongness really hurt. Here Butt gets an excellent shoulder spear on Zettel, knocking him off balance. This should provide ample opportunity for Cole to step around and wall Zettel off, creating a crease Smith will hit trying to beat a safety for a big gainer. Instead Zettel comprehensively wins:
Cole got whipped on the Norfleet catch that put Michigan in position for the winning field goal; Michigan was fortunate that ball went to the WR instead of getting knocked anywhere.
Ennui prevented me from charting him against Joey Bosa, but I've gone back and rewatched the OSU game. Cole was okay. By the second half of The Game Ohio State decided that they should either blitz outside of Bosa and send him against the interior line (which only kind of worked) or have him go at Ben Braden (which very much worked). Don't get me wrong, Cole did get beat. It was a struggle; it was not one in which he was completely overwhelmed.
All of that is terrific for a true freshman. Cole grabbed the job, held the job all season, and played reasonably well. There has been not a whisper that he would go anywhere except for a blip during spring practice when he was playing center, that because Glasgow was suspended and Michigan was figuring out if he could play there a la Barrett Jones. All practice reports have held that he is a sure thing, greatly improved, etc. 247 heard this from multiple people in a 24-hour window:
Mason Cole has unsurprisingly established himself as a rock at the left tackle spot, and is primed to be the next great four-year starter to play the position.
Cole's added 13 pounds and should make a major leap in year two, what with the rare true-freshman-to-true-sophomore OL transition coupled with the general HARBAUGH.
That will probably still leave him short of dominant. Michigan's most recent really good left tackle, Taylor Lewan, took off in his third year. In year two Cole should slash the protection issues considerably and do okay against the better defensive ends in the league. A year like early Graham Glasgow—reliable, somewhat short on raw power—would set Cole up to be excellent as an upperclassman.
[After THE JUMP: four-ish additional returning starters who happen to be upperclassmen already]
While I was chatting with Brian last week he happened to pull up the top 7 composite recruits from the 2013 season. I followed and…
Woof. Green has obvious vision problems and hasn't emerged from a pile of guys among whom the most statistically effective last year was Drake Johnson. Dymonte Thomas and Shane Morris are already juniors and to date still seem to be at least a year's worth of good coaching away from ready.
That leaves us the offensive line class. Kugler seems to be on track to start when Glasgow surrenders his job—I've heard the same suite of nice things you have. Bosch transferred after performing about how you'd expect a true freshman thrust into a Borges-coached OL would. Fox hasn't been mentioned since a staff ago. Dawson we have only a little more data, much of that getting owned by Maurice Hurst in the spring game (if Hurst does that against Utah's OL I'll happily rescind that as a criticism).
On the other hand we caution all the time about giving up on OL when they're too young.
So when do you know about an offensive lineman?
This is a question I've been interested in a long time, going back to an article one of my Daily colleagues did on OL recruiting to highlight the injuries plaguing the classes Michigan took while I was there. I could never find the article but in January 2013 I tried to recreate some of that information, plus a 12-year update. I did a thing about a year ago on growth tracks to reset expectations for those 2012 and 2013 line classes. Let's check in again, this time with columns.
The towers shrink because players currently on the roster are included in the data, and obviously our information on them is incomplete. "Not available" is a catch-all for transfers, dismissals, guys playing defense, injuries and medicals and whatnot. "Excellent" is basically all-conference-ish, "Solid" is that, "Liability" are guys who were starting but the fan consensus was they shouldn't be or wouldn't but for things like the 2008 depth chart or gross Borges incompetence.
This time I differentiated between "backups" and "two-deep" (an imperfect thing from memory and pouring through old Wolverine annuals). The former are guys buried on the depth chart and unlikely to play; the latter are only the top backups we are relatively certain would have played if they weren't behind an established starter. It's not about being technically on the two-deep, more like the first one or two guys in if an OL goes down—Erik Magnusson last year, or Leo Henige forever.
- Redshirting is overwhelmingly the normal thing to do as a freshman.
- Only a handful of players are capable of starting (shades of yellow) as redshirt freshmen. If you take the yellow chunk from there and size it against the 4th and 5th years you can see only about a third of the eventually useful players are demonstrably so at that age. Sing the praises of any 2014s already playing; don't give up on any who are not.
- By RS Soph there is a big yellow expansion. The mysterious "backups" region has shrunk considerably. You have a fairly good sense of who these guys are by the end of this year.
- There is very little difference—just a slight improvement—between RS Juniors and 5th year seniors. The backups disappear into unrenewed 5ths.
If you're using this imperfect data set of 82 players, many of whom didn't complete their careers for non-ability-related reasons, to get a feel for when to judge an offensive linemen, you could say it's a half-life. Don't judge a (redshirt) freshman unless he's already playing well, but after their third year in the program if he's not on the two-deep the chances of ever doing so decrease exponentially.
What this means for the 2012-'14 OL classes
Be excited for: Mason Cole.
Be extremely content with: Mags, Kalis and Braden if they seem to be playing well this year.
Keep an eye out for: Kugler, Logan Tuley-Tillman, and David Dawson. These are 2013 guys mentioned as probable two-deep contributors, though our current scouting has Kugler pretty much ready to play, LTT half-way there, and Dawson probably not ready yet. Further data received on them this year will speak volumes about their futures.
Be patient with: Juwann Bushell-Beatty. If he pops up this year he's probably going to be awesome; if he's buried there's plenty of time that this doesn't matter.
Getting late: Bars, Fox, Samuelson. With Bars at least we've heard past mentions of him competing, though he was always kind of the last guy in that 2012 class. He may be on the Huyge track; if he's not on the two-deep this (his redshirt junior) year it doesn't seem very likely he'll be a starter in 2016. Fox has been hurt so much in his career (just going off of game reports) if he's not medicaled he probably deserves some extra time to get caught up. Samuelson I've heard nothing about; even when I ask people with insider-y info I get nothing.
If you like manball there's no better guy to have as your offensive coordinator than Tim Drevno. As both a TE coach and OL coach, he was one of the main architects of the thumping Stanford lines that brought the Cardinal to their recently-elevated level. Afterwards Drevno transitioned to the NFL and got a plum job at USC. Now he rejoins Jim Harbaugh at Michigan.
After a small-school tenure as an offensive lineman, Drevno's coaching career started with a few years coaching TEs and RBs at smaller schools. In 1999 he transitioned to OL at San Jose State. Since then he's been exclusively an OL coach save for his first two years at Stanford, when he handled TEs. He also held the offensive coordinator title for Harbaugh's extremely successful San Diego teams. (Harbaugh in fact inherited Drevno from the previous administration.)
At Stanford, Drevno was a key part of the machine that actually got up and running in Harbaugh year two:
[Ranking out of about 120]
|YEAR||TEAM||DREVNO||Rush S&P||Overall S&P||Main back(s)||Results|
|220 rushes, 900 yards,3.8 YPC|
|280 rushes, 1150 yards, 4.1 YPC|
|330 rushes, 1850 yards, 5.6 YPC|
|2009||Stanford||OL||12||6||Gerhart||340 rushes, 1870 yards, 5.5 YPC|
|370 rushes, 1800 yards, 4.9YPC|
|370 rushes,2100 yards, 5.6 YPC|
It's hard to separate Drevno out from the general Harbaugh effect, but again the continued success of Stanford after coach X's departure bodes very well in this case. This wasn't Texas or Alabama when they were up and running. This was a program transformation that stuck; that Stanford continued to excel after Drevno left is pretty good since he was one of the major molders of guys like David DeCastro and Jonathan Martin.
Drevno went with Harbaugh to San Francisco, where he was the OL coach; oddly, NFL veteran Mike Solari was also the OL coach. The two guys had the same title. In any case, the San Francisco OL was up and down.
[rankings out of 32 teams]
|YEAR||TEAM||Rush DVOA||Line Yards||Power Success||Adj Sack Rate|
After a step back in year one, the 49ers had a terrific rushing offense in year two; they then took a major step back. At no point was their sack rate anything other than bad, but he did inherit that and quarterbacks do have a significant, often-unacknowledged hand in that. Kaepernick is a guy who prefers to extend plays even if that results in additional sacks because when it doesn't he frequently lopes downfield for thirty yards.
Despite those numbers, San Francisco sent two OL to the Pro Bowl in 2013 and had their entire line named as either a starter or an alternate in 2012. Margins in the NFL are razor thin.
L to R: true FR, redshirt FR, junior, true FR, redshirt SO
Last year, Drevno returned to college at USC, picking up a run game coordinator title and inheriting a line that thinks last year's Michigan line is impressively experienced. Three true freshman saw extensive time, with Toa Lobendahn moving to left tackle midseason when sophomore Chad Wheeler went down with injury. Redshirt sophomore Zach Banner moved into the starting RT job; Max Tuerk was the only upperclassman, and even he ended up moving to center.
This is like last year's Michigan line if you replaced the starting guards with freshmen instead of a redshirt junior and redshirt sophomore.
Despite that, the numbers were middling:
|Offense||Adj. LY||Rk||Opp. Rate||Rk||Power Success Rate||Rk||Adj Sack Rate||Rk|
USC was about average in line yards and adjusted sacks, a bit below that in "opportunity rate"—the percentage of run plays that go for five yards—and bad at short yardage. Top USC back Javorius Allen almost hit 1500 yards at 5.4 a carry. That's impressive for what must have been one of the youngest lines in the country.
PREDICTION BASED ON FLIMSY EVIDENCE
Drevno has extensive Harbaugh experience and did very well considering the situation in his single year at USC; he was one of the primary guys driving the Stanford rushing renaissance whether it was as a TE coach or an OL coach. A lack of OC experience is not a problem since Harbaugh has a major role in coordinating his own offense, and Drevno worked with Harbaugh in that capacity at San Diego.
At 45, he's probably looking at this job as an opportunity to impress and get a head job. Given the history there that's not exactly a longshot.
UPSHOT FOR REST OF STAFF
Ain't no more upshot.