Mailbag: Uptempo Effect On D, Next Year On O, Personnel Bits

Submitted by Brian on November 28th, 2012 at 3:46 PM



Reading your 11/26 post about "The Game", I noticed you made mention of the question of whether huddling is necessary anymore.  As a high school coach and former small college player who has used both the huddle and the no-huddle, my first reaction is to disagree with your point that the huddle is archaic and a bit of a dinosaur.  But I am also intrigued.

My overall response would be that there are obvious advantages to both.  Points and tempo are the obvious gains, but tempo is also a negative from the no huddle (as we learned circa 2009-2011) when an offense has too many three and outs and a team's defense bears the brunt of it all.  In short, it's hard to praise a defense, especially an overachieving bunch like our boys, while touting an offensive style that so often seems to lead to a loss in defensive production.

In short, what data is out there to suggest that the no huddle would not obliterate our defensive gains?


There are two issues here being conflated here. Not huddling is not necessarily synonymous with going at a high speed. Ohio State runs the same no-huddle style as Oregon but does not push the pedal down nearly as much:

Team Offensive Plays Defensive Plays Total % Plays on offense
Oregon 989 915 1904 52%
Michigan 738 791 1529 48%
Ohio State 837 849 1686 49%

Oregon games had 24% more plays in them than Michigan games; Ohio State games had just 10% more, and I'm guessing the difference there is more Michigan being exceptionally slow than Ohio State being fast: decidedly MANBALL Wisconsin games featured a little over 1600 plays this year. If OSU is over the national average for plays run it's not by much.

The benefits of getting to the line immediately are the same whether you're going fast or slow: the defense is hampered in its ability to substitute and you can see how they align as you make your playcall. You can sprinkle in tempo plays when you have an advantage without cracking the whip up and down the field.

The other issue is how a high tempo affects your defense. This is the reason people invented tempo-free basketball statistics: how fast you go can distort how your offense and defense look. A high-tempo approach artificially inflates the former and deflates the latter.

Take the most extreme possible example: Oregon. The Ducks are are decent but essentially mediocre in yardage statistics. They're 47th in rushing, 60th in passing, and 46th in total D. But they faced 915 plays. Michigan's defense faced 791, and in there is a large part of the gap between the two defenses. Michigan is still better on a yards-per-play basis, but there's no question that Oregon played more good offenses.

Advanced stats that try to account for tempo look a lot more favorably on Oregon's defense than conventional ones. The Ducks are eighth in FEI*, one spot in front of Alabama. They're 21st in S&P, which I don't like as much because it's play-focused instead of drive-focused.

As this year's Michigan team proved against Nebraska and OSU, a three and out is a three and out and you're in trouble no matter what if that's what your offense is doing. In games you can win, the price you're extracting from your defense is going to be similar to the price you extract from the opposition.

And, like, I don't think it would have mattered if a Greg Robinson defense played opposite the Lombardi Packers. They were cooked.

*[A note on FEI: that ranking looked pretty strange on offense last year; this year it passes any sanity test you want to give it with Florida/ND/Stanford/MSU as the top 4. It seems to be overrating some small schools but that's inevitable.]

Are we running the Air Raid next year?  Taking out Denard, Michigan had a pretty atrocious rushing offense this year.  It's obviously going to look different next season without a primarily rushing QB but even still, with a starting RB coming off a serious leg injury, and replacing 3 or 4 offensive lineman, this looks bleak right?


Look on the bright side: at least Borges has shown to be more willing to adapt to a throw-first mentality than Carr-era coordinators. And the interior line literally cannot be worse! Woo!

It does look bleak. Michigan has to hope that Mike Schofield can hold up at left tackle (I think he can against non-elite pass-rushers and will probably be a Stenavich-level player) and then fill in the rest of the line with n00bs. There are some assets:

  • Extremely fast QB.
  • "College-ready" five star Kyle Kalis at guard.
  • Guy who has played center all the way, Jack Miller, at center, presumably making much better line calls.
  • Enormous guy Chris Bryant
  • Enormous guy Ben Braden

It… yeah, it looks grim-ish. While any of the above could work out Michigan is still working through the disastrous RR OL recruiting and will have a similar problem to the one they had this year: few, if any options to turn to if the starters are not performing. Things are little better at tackle, where Erik Magnuson is available to back up instead of nobody, but on the interior you've got a redshirt freshman on the 3/4 star borderline (Bars) and Joey Burzynski, who is still 6'1".

But there's a bunch of hype for the freshman class and Bryant was reported looking good before the leg injury. Michigan may have to go with more Gardner running than they might want. I certainly hope they don't ditch the veer, for one.

Hi Brian,

First of all, I'm very surprised by Hoke's prediction that Devin's redshirt will go through, given his history of never saying anything specific, ever.

With that in mind, do you think Devin is a good fit for Borges' style of West Coast offense?  Or will we still be "making due" with some Frankenoffense for the next two years, give or take, until Shane is ready? 

Like most fans, I've been pleasantly surprised by Devin over the past month, but I don't know enough about Borges' dream offense to know how well his QB skills translate.


I was surprised, too, and assume he was told by the people who had talked to the Big Ten that they would get it. Which hurray, one less argument about how dumb Rich Rodriguez is and extra year for starting QB.

Gardner is clearly a better fit for what Borges wants to do than Denard was. He's tall, he can stand in the pocket, and he has an excellent deep ball when he's not being asked to throw it on the run for some strange reason. He also tends to run when he should run, thus rescuing various plays that aren't going so well.

I am actually hoping for something of a Frankenoffense, though. Running big epic "play action" from an I-form that is really just a max-protect setup doesn't use Gardner's legs particularly effectively, and we've seen that when you get a guy who can run and throw trying to stop both is super hard. Hell, run-and-kinda-throw is pretty hard. I'd like Michigan to still run most of its offense from the shotgun and use Gardner's legs to mitigate some of the problems that will arise on the offensive line. With Morris backing Gardner up you won't have to be as overcautious as it seemed Michigan was this year.

Maybe the OL will surprise and DeVeon Smith or Derrick Green will show up as a grinder and it will work. If a traditional manball running game isn't in the cards, though, the Frankenoffense may be the best one available.


Do you think Hopkins gets a long look in the spring back at tailback with Fitz's status up in the air?

Seems like they have some depth now at fullback with Kerridge  and Houma.  I know he has ball security issues so maybe he has to walk around campus all spring, summer and fall carrying a football. Could he have the potential to be a Leroy Hoard-type in 1988, '89 as a featured, ball-carrying back?


While that's a possibility I doubt it will amount to much if they do try it. Hopkins may be marginally better than Rawls if he doesn't fumble, he has. Rawls hasn't done much but he also has not fumbled.

A Hoard-back requires Hoard-blocking, and more speed than Hopkins brings to the table. If Derrick Green ends up committing he's the early favorite to get a plurality of carries.

Any chance that Michigan starts Pipkins and Washington in a similar way that they used Washington and Campbell this year?


Probably not. Nose tackle is a draining position staffed by enormous men and requires that two people play it. Michigan probably wanted to use Pipkins more than they did a year ago; they couldn't because he wasn't very good. Michigan will probably stick with him as the backup nose in preparation for a two year starting run. Big guys take some time.


M Fanfare

November 28th, 2012 at 3:58 PM ^

I've always thought the argument that fast-tempo offenses kill their defense when they go 3-and-out is a strange one. If the defense gets thrown back on the field quickly but forces a 3-and-out of the other team, they're right back on the sideline. I guess it doesn't matter for a good defense, but it would compund the problems of an already bad defense?


November 28th, 2012 at 4:21 PM ^

The better argument, I think, is that Oregon's defense often has to return to the field quickly after an offensive position on which Oregon scored.   Three-and-outs are basically the same, but there's a big difference (obviously) between a nine-minute TD drive and a two-minute TD drive. 


November 28th, 2012 at 6:58 PM ^

has been successful for years despite not having elite talent though they have excellent players like Devin Harris, Jordan Taylor in the past.  They slow it down and is very efficient which is a good recipe for regular season success, but not for postseason where you're more prone to get upset as a higher seed because it gives the underdogs a better chance of beating Wisconsin in the NCAA tourney.


November 28th, 2012 at 7:07 PM ^

Actually, if talent is the issue, I'd make it part of general strategy.  If you have a deep defense, you can play up tempo -- you wouldn't necessarily substitute every play (especially if the other team does the same), but on a 3-and-out you can just rotate in some or all of the second unit.  Fatigue issue, we don't need no stinkin' fatigue issue.  If your defense is thin, then huddle or no-huddle, it's in your best interest to pace your offense.  If you play MANBALL, even a 3-and-out is still several minutes of breather time.

I don't respect one-trick ponies no matter how much experience they have.  If you have the defense that works well with an up-tempo offense, use an up-tempo offense if you can use it to your advantage.  If you rely on keeping a few playmakers fresh on defense, then slow the hell down.  While the process of training the players into smoothly running one or the other isn't something you do in the middle of a season, this isn't rocket science either.  Look at your roster in spring camp, make a decision and implement.  I mean, of all the various things to teach in an incredibly complex game, this should be a relative no-brainer.


November 28th, 2012 at 4:37 PM ^

Defense tends to be mentally harder in that you're being reactive and going "uphill" at times.  Also on the offensive side of the ball, players can take plays slower when they need a breather and they know they're not the target.  Sometimes you just have to run your route hard enough to suck a pair of DBs along in your wake, but you don't actually need seperation in that you're not really a target.  Whereas defensive players have to assume their guy is always the target.

Personally I think the real issue is denying the defense a chance to sub and exploting favorable personal groups.  There is a difference between no huddle, snap the ball on the ready for play signal vs no huddle, but still bleed 15 seconds off the play clock as you hard count.   I think the former sucks, but the latter is awesome.  

The huddle's demise is also overrated for times when you're say Michigan and playing some team you physically outmatch.  Take the time to set the play up perfectly, because honestly who cares if UMass gets a chance to sub.  Lewan will destroy whoever puts his hand on the ground across from him.  Also when the defense has the proper package in against you, I'd argue you might as well take the five seconds to huddle and reduce the odds of route running problems, etc.

I've seen good success at the high school level out of playing to deny defensive subs for a number of plays when you catch the defense with poor personnel out there, but also not being to slow it down and let both sides sub when your offensive guys need to get their breath or you want a special package in.  

Kilgore Trout

November 28th, 2012 at 4:02 PM ^

One thing I was thinking is that UM's super slow tempo may be hurting the defense when they face up tempo teams. Air Force and Northwestern both seemed to really throw UM for a loop when they hurried. If Michigan is going this slow in practice, it has to be kind of hard for the defense to know how to react to that when it comes to game times. I think Hoke said something about playing a spread hurt your defense when they had to play against "manball" type offenses. It seems that he didn't consider that it appears to work the other way too.


November 28th, 2012 at 7:06 PM ^

I'm guessing most teams go slower in practice then they do in games.  I never played football at a high level but I played basketball through college.  When you're running certain plays the coach would often stop it after each play to go over what went right and what went wrong.  In basketball we definitely did some full court scrimmaging where play isn't stopped for 10 minutes or so, but I'm not sure how much 'scrimmaging' goes on at a football practice.  I know they practice two minute drills, end of half, end of game situations.

Teams that go no huddle and especially those that go uptempo no huddle probably do a bit more 'scrimmaging' or at least run a few plays in a row to try and replicate game conditions.  I'm guessing when the coaches know the defense is going to face an offense that likes to go uptempo, the scout team at least goes uptempo.


November 28th, 2012 at 4:05 PM ^

Per FEI, we played the #6 (Ohio), #8 (ND), #9 (Nebraska), #20 (Alabama), and #27 (NW) offenses, so I don't think there's "no question Oregon played better offenses."


November 29th, 2012 at 2:00 AM ^

and use tempo free football metrics just like KenPom.  Just because they don't pass the eye test, doesn't mean that they're false.  Oregon uses a lot of possessions and the defense are allowing less than 21 points in 8 of the 11 games.  Throw in Oregon's offense that loves to run uptempo no huddle offense, that's pretty damn impressive.

Bama defense is excellent but they struggled against a bad LSU offense who put up yards and points against them, Ole Miss who moved the ball on them, Texas A&M shredded Bama defense.  Look at Bama's schedule. They played no one good on nonconference and some of the SEC teams they faced aren't good offensively.

It's not surprising that Oregon defense is better than Bama.  Oregon lost because their offense couldn't score and the defense did a great job of limiting Stanford's running game.


November 29th, 2012 at 3:02 PM ^

Doesn't make it right.

I don't see LSU as a bad offense. They beat the Huskies 41-3, who Oregon beat 52-21. Put up 23 on South Carolina, 37 on 8-4 Mississippi State. And Bama still held them to 17 points in Death Valley with two weeks to prepare for Bama.  Bama laid an egg in a let down game after their biggest game of the season, and A&M was better than anyone thought they'd be. 

Oregon on the other hand gave up 34 to Arkansas State, 25 to Fresno, and 21 to Washington at home. 24 to Oregon State that an awful Wisconsin team held to 10. And 51 to a USC team that could only score 13 against a really good defense like ND in LA (without their QB...but was he worth 38 points?!)

I don't know that cherry picking games and saying an offense like LSU's is bad really works in your favor. Oregon's defense is better than in some years, but to say high speed, tempo offense doesn't hurt a defense, but then blame all the points they give up on how many possessions they have is just having it both ways. Maybe Bama gets bombed by Georgia this weekend. And the talk of them being an all time great earlier in the season was obviously overhyped. But if Oregon's defense is ranking higher than Alabama's defense at season's end, there's a flaw in the system.


November 29th, 2012 at 4:42 PM ^

I agree with you.

FEI is a single metric. Going by a single metric is rarely a good idea. Especially one that churns out weird outliers as often as FEI does. If you look at YPP, another metric that new age statisticians (including Cook) like, Alabama is much better than Oregon.

I also think that saying high tempo offense doesn't hurt the defense because of Oregon 2012 won't get you very far in any case. Using the best team you can find to prove a general point is not a good recipe for finding the truth.

Regardless, Michigan is as likely to morph into Oregon as they are Alabama. Blaming Borges for that doesn't seem very reasonable to me. 

STW P. Brabbs

November 28th, 2012 at 4:12 PM ^

First, to me the more convincing argument for a slower pace on offense is that the more time it takes up, the fewer plays the opposing offense can run. Is it not true that you're therefore essentially breaking even in terms of number of plays? This doesn't necessarily dispute your arguments re: adjustments at the line, but I wonder if there's an advantage to be had with the huddle in terms of getting the offensive players on the same page (ideally; it didn't seem to work for UM this year unless the no huddle would have been a total. tire fire as an alternative).

Second: some sophistry in your Oregon argument: 'The difference in the rankings is the number of plays faced. Well, the yards per play are different ... but Oregon played better offenses.' So is it the competition or the number of plays?

Finally: This occurred to me belatedly after '27 tickets' but I think you're caught up by the Morris hype train. I don't think Gardner should be all that much further down the list than, say, Qwash, if not up there with Ryan. Morris' high school career does not seem to suggest anything like a precociously polished product, despite all the talent that dude has in spades. Morris may end up challenging Bellomy for backup duties if Russell can shake off whatever malady or evil spirits afflicted him in Lincoln; I'll be absolutely stunned if there's any controversy-type substance re: the starting spot that isn't wholly manufactured by star-gazing recruiting afficionados or by newspaper writers desperate for attention.

Wolverine 73

November 28th, 2012 at 6:04 PM ^

True freshmen QBs who can play well are few and far between.  Better to redshirt Morris if we can get away with it.  Ideally, he then backs up a senior Devin as a redshirt freshman, and has three years to lead the team thereafter--assuming he lives up to the hype, which we all certainly hope he does.


November 28th, 2012 at 4:17 PM ^

I think Borges wants to run a frankenoffense, mainly due to the fact that he has referred to himself as a scientist tinkering and has said that he doesn't like to be known as a WCO guy. He doesn't want his offense to have a name. When they can name it, they can talk about how to defend it.

Now, what plays the frankenoffense uses is another question. I'd bet he keeps some ZR staples (like the veer) but doesn't use them too often. Actually, I'd love if he chatted with Mike Shanahan about the offense the 'Skins have installed for RGIII.


November 28th, 2012 at 4:44 PM ^

When we are ranked 81st in total offense, 99th in first downs, and had three games without scoring a touchdown (for the first time in over forty years), does there really need to be a name to our offense?  Is that really hampering the ability of the teams we are facing?


November 28th, 2012 at 4:17 PM ^

Remember that Borges seems to prefer to use a number of different alignments and personnel combinations - to your terminology, kind of a Boise State "grab bag" offense.  So one cost of going no-huddle is that we may be limited in terms of our ability to change alignment (to the extent different personnel are required).

(I agree with your basic point, just wanted to point out one additional potential cost to going no-huddle.)

Ali G Bomaye

November 28th, 2012 at 4:19 PM ^

All the posts in this series looking ahead to next season seem to assume that Taylor Lewan is gone to the NFL.  Besides him being really good, is there any evidence to support this (i.e., has he said anything that would indicate he's leaning toward going pro)?  He seems to be a guy who enjoys college, and we've had some other guys like Jake Long come back when they didn't need to.


November 28th, 2012 at 4:24 PM ^

Hoke said that the proper paperwork has been submitted to the NFL Draft Advisory Board in order to get a draft evaluation for Lewan, and the coach plans to talk to Lewan about a potential jump to the NFL this week. Hoke insisted that he doesn’t try to steer players to come back to school in these situations.

“I’m going to gear the young man towards what’s best for him,” Hoke said.…


November 28th, 2012 at 6:51 PM ^

He asked:

Besides him being really good, is there any evidence to support this (i.e., has he said anything that would indicate he's leaning toward going pro)?

I gave some evidence:

Hoke said that the proper paperwork has been submitted to the NFL Draft Advisory Board in order to get a draft evaluation for Lewan, and the coach plans to talk to Lewan about a potential jump to the NFL this week.

Hoke insisted that he doesn’t try to steer players to come back to school in these situations.

“I’m going to gear the young man towards what’s best for him,” Hoke said.

I tried answering the question. Maybe I shouldn't have responded at all because I didn't give a direct quote from Lewan.

I think the most significant quote is Hoke is going to "gear the young man towards what's best for him." If Brian saw that quote, he is probably among the people who think that declaring for the draft is what is best for Lewan.

Blue since birth

November 29th, 2012 at 3:16 AM ^

Yeah, that's what I was trying to get across in a tongue in cheek way. I guess it didn't come through as intended.

The point is...

I'm sure you can make good arguments either way for what may be "best" for Lewan. If the "most significant" piece of evidence that he's going pro is that Hoke "wants what's best for him"... I find it to be a stretch of an assumption to take that as a sign he's going pro. My guess (if I were to) is that Hoke would take the other perspective.

I think most players in his shoes would go pro and that's the only reason I can see to be pessimistic about the chances of him returning (that and the typical defense mechanism/CYA pessimism so common around here). None of the quotes provided give a clue to his leanings as far as I can tell.


November 29th, 2012 at 2:46 PM ^

And not just draftnicks.  If he's legitimately a top 5 pick the advice is easy. If he's a first rounder with potential to to move up in the top ten, then it's more difficult. If he's a low first rounder who will stay a low first rounder, it's back to easy. And if he's a 3rd round pick you can say he should stay.

I don't know how he "feels", but if he's a Top 5 pick there's almost no reason for him to come back. That would be what's best for him.  And I feel I can say that Hoke is interested in giving him an honest evaluation for what's best for the kid; not the program or Brady's career. That's why people think that would mean he's gone.  

We've had basketball players who left or thought of leaving that weren't ready for the pros that shouldn't go. If Lewan wants to come back, great.  Makes us better and is a great guy. But if he decides it's best for him to go, via con dios, son.  Thanks for the 4 years and tear it up in the NFL.


November 29th, 2012 at 3:45 PM ^

The consensus among draftniks is that the senior class offensive tackles in the 2013 draft are a really weak group -- few if any first-round talents among them. That's one reason Lewan and the two junior tackles from A&M, Joeckel and Matthews, are all assumed to be early entries -- with so little competition from seniors at such a premium position, Lewan and Joeckel in particular both could easily go in the top 15.


November 28th, 2012 at 5:14 PM ^

When we talk about "sanity checking" numbers, I think that using a metric that puts Oregon, who, in tempo free YPP-land is the 34th best defense, and puts them at 8th...I think that needs to be sanity checked.

We say "well, they played better offenses" - my naked eye says the Pac10 also doesn't really play any defense.

It's almost impossibe to reliably compare across region - we have 130+ teams that only play 12 games, and only play 3 of those outside their immediate cohort (assuming each plays an FCS school, making the 4th irrelevent). It's a vanishingly small sample sized accrued across utterly non-analogous levels of competition. You can't adjust for competition, because nobody has a clue how good the competition actually is.


November 28th, 2012 at 5:56 PM ^

Well put.  This is a concept and pretty much fact that I've been acclimating myself to these past few years.  It's so hard to tell how good/bad any team really is in this game of College Football.  In a way it's a reason why CFB is so big and what makes it so fun; endless speculation and conversations on how good one's team is.  But getting stone-cold facts and answers has never been a process that CFB could complete; the championship game winners of the past 13 years haven't exactly been the most convincing, to me at least.  Year after year we're left asking (again, at least I am) many more questions than we'd like to have to ask at the end of a season.  Like you say, "we're clueless" as to how good anybody really is.