Hokepoints Has a Running Quarterback

Submitted by Seth on May 6th, 2014 at 11:08 AM

DSC_3744

read option [Fuller]

I am determined this spring to mine every possible stat for every possible insight. This week I delved into quarterback rushes. Not sacks. I wanted to know which offenses tended to have their quarterbacks take off, or planned runs for them into their game plans.

Baseline: here's Michigan and their opponents last year. Sacks and yardage lost to them are not counted, but I couldn't tell from scrambles and QB sneaks, or stuff like if he took off for 10 yards on 3rd and 15 that defenses are happy to give up:

Season Avg vs Mich
Opponent QB Rush Yards QB Rush Yards
Central Michigan 1.8 8 5 11
Notre Dame 1.5 3 0 0
Akron 4.6 19 3 11
Connecticut 2.5 3 3 -4
Minnesota 14.4 77 18 69
Penn State 2.4 4 2 0
Indiana 8.9 42 10 60
Michigan State 4.5 17 4 1
Nebraska 8.8 35 12 31
Northwestern 10.8 58 24 92
Iowa 5.2 25 5 26
Ohio State 14.7 119 14 165
Kansas State 19.2 97 14 76
NCAA Avg 7.5 40.0 8.8 41.4

Indiana, Nebraska, Northwestern, Ohio State, and Kansas State ran option games. Minnesota's offense was QB power running (thing it is like: Michigan's 2010 offense when Rodriguez gave up on trying to make Denard into a zone reader). According to the UFR database Minnesota quarterback running plays vs Michigan were as follows: 7 QB powers; 2 draws; 2 zone read keepers; a false zone arc sweep thing, a QB sneak, and 7 scrambles.

The stats can't tell the difference between this kind of offense and a dedicated Richrodigan spread 'n shred. There aren't many teams who run this as their base offense, as simple as it may be, but a lot of teams have a mobile change-of-pace quarterback and a small package built around him. Notable teams who deployed a second guy:

Player (2014 Elig) Team % of Snaps % Will Pass Rush Pass
Austin Boucher (graduated) Miami(NTM) 51% 73% 80 211
Austin Gearing (So.) 35% 35% 129 70
Drew Kummer (Jr.) 14% 71% 22 55
Nate Sudfeld (Jr.) Indiana 61% 94% 22 338
Tre Roberson (Jr.) 38% 62% 84 139
C.J. Brown (11th year Sr.) Maryland 73% 72% 119 303
Caleb Rowe (Jr.) 26% 91% 14 136
Philip Nelson (transferred) Minnesota 59% 72% 79 200
Mitch Leidner (So.) 38% 51% 89 91
Gary Nova (Sr.) Rutgers 68% 93% 25 328
Chas Dodd (graduated) 32% 87% 21 143
Tommy Armstrong (So.) Nebraska 39% 68% 63 135
Ron Kellogg III (graduated) 31% 90% 16 141
Taylor Martinez (graduated) 30% 77% 34 116
Trevor Siemian (Sr.) Northwestern 63% 92% 29 315
Kain Colter (graduated) 36% 50% 98 99
Braxton Miller (Sr.) Ohio State 72% 65% 150 276
Kenny Guiton (graduated) 25% 74% 39 110

I included Rutgers to show Chas Dodd wasn't a Drew Henson-ian run threat except in comparison to Gary Nova.

[Jump: Okay spread zealots, do teams with running QBs have an advantage?]

Is an offense where the quarterback's a run threat inherently better?

The operating principle behind QB power and option offenses is if the defense has to account for your quarterback as a run threat, that opens up room for other players to operate. It should follow that teams who use their quarterbacks as regular run threats will have an appreciable advantage in offense over those who don't. Apologies to the spread zealots: very much no. Here are last year's FBS teams by % of plays by their QBs that not passes or sacks:

Sack-Adjusted Stats

Warm colors are teams whose quarterbacks ran enough to be considered a likely threat, with the scattered red dots to the right representing the committed option teams. You'll note there's almost no difference. An r-squared of 0.0008 is a big strong "there is no correlation" between how often you run with your quarterback, and how good you are at accruing yardage.

Again, hard to tell if that's "my quarterback will scramble" or "I put some option in my offense." Michigan State and Florida State have different offenses but had almost exactly the same QB-will-run ratings. The difference was FSU had the Heisman winner at quarterback and MSU floundered until they settled on a freshman and playing safe-ball with him.

What I can't account for here is the talent on hand. Ohio State had all the right parts (senior offensive line, all-conference skill players) to make their offense very dangerous. They were joined in the ranks of 7+ YPA by "my QB will run sometimes" offenses in Baylor, Oregon and Texas A&M, but that blue dot all the way on the left is pro-style Alabama. What those teams had in common was great offensive players. Miami (No not THAT Miami) ran their QB more often than the Buckeyes, but had terrible offensive players and thus the second-worse offense in the country.

But maybe it cuts down on turnovers?

I also looked to see if running with your quarterback will lead to more scoring (because the safeties have to run support?) or less scoring (because more defenders are between the ball and the goal line?). Also I tried the Woody-Bo belief that taking away throws to have your quarterback tuck and run will cut down on turnovers. To account for fumble randomness I used all fumbles by the offense and halved them, then added interceptions. Result:

Sack-Adjusted Stats2

No. Friggin. Difference. Maybe a tiny bit on the turnovers but an r-squared of 0.0903 says there's very little correlation.

Comments

massblue

May 6th, 2014 at 11:24 AM ^

I believe the plays run by an option QB look exciting and that is why without hard evidence most of us think they are more effective.  I would suggest to look at injuries with option QB as well.

JeepinBen

May 6th, 2014 at 11:46 AM ^

I like what your data says "Running your QB doesn't add TDs, take away Turnovers or make your offense great".   For a long time the spread has been thought of as somewhat of an equalizer - a way for little schools to even the playing field against the big guys, kind of like the 3 in basketball. Is that true? I think that if we cross your data here with a few other databases we could really get some answers. For example: The Mathlete's talent-by-recruiting-star database - do teams that run their QBs more overacheive more frequently than teams that dont? And I'm not sure if it was Mathlete or Brian, but it'd be interesting to see how this data combines with tempo - does hurrying up change the calculus on if a running QB helps? The theory there is that defenses play more basic schemes since they can't sub and call plays as frequently.

Like I said, I think this is good stuff, and I think it could provide even more info if cross referenced.

dragonchild

May 6th, 2014 at 12:18 PM ^

Not that I have any data to back this up, just watching games, but the real benefit to the spread n' shred was NOT that it was inherently superior to other offenses, but rather the novelty of the concept, and the level at which that novelty operated.

Consider an adjustment to a pro-style scheme.  Pro-style is a mature concept, thoroughly scouted, so any "adjustment" is typically incremental.  It could be as simple as how the receivers run their routes.  I'd call this a "high level" change as it's still built off a number of existing layers in the core concept.  As a surprise it's not guaranteed to work for more than a half against a decent DC, and the fix can be as simple as moving one guy a few yards and/or changing his read.  At some point you just have to out-coach or out-talent your opponent, because you can't consistently rely on deception.

The spread n' shred combines of a bunch of ideas that were already in practice for a long time; the stroke of genius was combining these elements into something completely different as a whole.  It preyed on some pretty low-level concepts defenses took for granted.  This is not the sort of thing a DC can fix just by moving a guy or two around; it makes a mockery of entire seasons of coaching and preparation.  The reason why it was so effective in David-vs.-Goliath matchups is that you didn't need all-world talent with this sort of advantage.  If the defense has never seen QB Oh Noes, what does it matter if it's a five-star safety going up against a 2-star RS freshman if the safety's run himself twenty yards out of the play because the DC coached him the wrong read?

It's the exact same reason Michigan struggled against Air Force's triple option.  Not only is that a very old concept, it's one that was abandoned because it's basically out of tricks.  But Air Force gets a lot of mileage out of it because it's rare.  It takes a certain kind of defense to stop the triple option and it's very difficult to re-mold today's units into a triple option killer in the span of a week.  Except the spread was an all-new look so it took years for DCs just to figure out what it was doing.  It's not like RichRod was calling DCs up and telling them he's optioning off the DE; they had to figure it out.

I guess to concede a bit, the spread WAS an inherently superior offense, but not for reasons it's credited for.  Its inherent superiority was the way it forced defenses to learn it; all the while it was scoring TDs and upsetting favorites by the truckload.  Even after they figured it out, it took an entirely different set of personnel (skilled edge defenders, bigger nickelbacks et. al.) to defend it effectively.  But we've reached a point where there's enough knowledge floating out there about its strengths & weaknesses that no DC has an excuse to not know how to stop it.  Efforts to recruit and develop athletes suited to stopping it are mature operations.  The first guys tasked with defending this offense have long since departed the college level.  So now it's just another concept, and as a concept it's shown to be perfectly viable but not inherently superior.  Life goes on.

ST3

May 6th, 2014 at 11:58 AM ^

no foolies. (How could you leave out the "no foolies" part?)

If you take away Miami OH and that other horrible running team at 3.5 YPA, you have zero QB-run heavy teams (>25% designed runs) under 5 YPA, and 20 non QB-run heavy teams (<25% designed runs) under 5 YPA. After sitting through 27 for 27, I'd gladly take a guaranteed 5 YPA. You just have to have a competent backup QB for when your running QB inevitably gets dinged up. 

Indiana Blue

May 6th, 2014 at 1:25 PM ^

I included Rutgers to show Chas Dodd wasn't a Drew Henson-ian run threat except in comparison to Gary Nova.

Who ???? Wow - gonna take some time before Rutgers and Maryland players are recognized ...

Go Blue!

hfhmilkman

May 6th, 2014 at 2:26 PM ^

I think it makes sense from a thousand miles away that the numbers come up a wash.  If you have players that fit a particular scheme and adjust to them they will do well.  Conversely players who are forced to play a scheme they are not good at will do poorly.  My guess is that an analysis of perimeter oriented teams verses power for basketball would generate the same results.  

What is more important than scheme for me is can a staff adjust to what they have?  R^2 has not had the electric running QB in Arizona and now he is making due with the yards coming from a RB.  Too bad his offensive flexibility did not include defense.

Conversely I am bothered by Hoke's lack of flexibiliy and insisting that Michigan only runs pro power and QB's should only sit in the pocket.  Bo seemed to do pretty well with running QB's.  Only Thomas, Grbac, and Wangler could be described as pure pocket passers.  So when Hoke states this is "Michigan", I get riled up as that is a Carr tradition not a Bo tradition.

 

 

Reader71

May 6th, 2014 at 2:50 PM ^

Have you watched a Michigan game in the last few years?

Between Denard and Devin, I don't think a person who has seen a game who can honestly say that Brady Hoke is against running his QBs.

This world is the one we are concerned with. Not the one in your head.

Space Coyote

May 6th, 2014 at 2:53 PM ^

Hoke (more Borges for that matter) may have implemented some of their pass game preferences rather than the typical spread alternatives, but to claim that they insisted to only run pro power and QBs should only sit in the pocket isn't true at all.

Now, they certainly wanted to install a mostly Power O blocking scheme instead of zone, but they ran a bit of zone the first year on some early zone read stuff. They ran possibly too much of a combination of zone and man blocking last year. And Michigan's QBs maybe haven't run as much as they did under Rich Rod, but Denard went from 19.7 attempts/game in 2010, to 17 att/game and 16 att/game in 2011 and 2012 respectively (not exactly under-utilizing a QB's legs). Likewise, Gardner almost ran the ball 14 times per game last year, which is probably closer to how Rich Rod would have used him as well (he's a different type of QB rushing the football).

Seth

May 6th, 2014 at 11:19 PM ^

Michigan wasn't extreme but we ran with the quarterback a lot more than most teams. Hoke gives out hokum; his teams have yet to run anything remotely like a Carr offense, and with Shane Morris (who's pretty athletic) I don't think they will unless Speight becomes the starter too soon.

Flexibility wasn't really the problem. It was more that Borges/Hoke had no idea how to run the offense they inherited, and they badly bungled the transition to something they would know how to run.

bronxblue

May 6th, 2014 at 2:45 PM ^

Nice breakdown.  It pretty much proves what you'd expect - if you are good at running your offense it doesn't really matter the style, and if you don't have much skill/experience you could trot out anything and it would be a rough go.

It will be interesting to see if the offense highlights Devin running more, as Nuss never had an athlete like him at Alabama and Locker wasn't really that type of runner while Price was still figuring it all out.

The FannMan

May 6th, 2014 at 2:49 PM ^

This may be semantics, but not every spread team runs their QB. Some like to spread the D out and pass the ball all over the place. See also the Big XII.

Seth

May 6th, 2014 at 11:22 PM ^

Running QB - / -> Spread, as our fine servicemen can attest. There were a lot of extreme QB-rum teams that weren't spread offenses. Minnesota was an extremely run-with-the-QB oriented team, and they weren't very spread.

I guess I could calculate based on 3rd WR versus heavies (FB/TE etc.) targets which teams are spreading the field.

MCalibur

May 6th, 2014 at 3:49 PM ^

I think its reasonable to say that offensive efficacy is scheme independent but I think there's a problem with how this data is presented. There's an underlying assumption that the variance between styles is the same. Turn each color in your first chart into a histogram and I bet you'll see that the warmer colors have a tighter distribution which would indicates that they might be high performers more consistently than the cool colors. That is a good and worthy thing.

If you share your data, I'll show you what I mean. 

*I do not consider myself a spread zealot

gbdub

May 6th, 2014 at 8:14 PM ^

Are spread option teams evenly distributed among the conferences? If spread option offenses are much more common among mid-majors, that would skew the results. Might be interesting to confine the analysis to the top 25, or to the BCS AQ conferences.

Others have commented on the novelty of the spread as key to its success, and that's true. But the spread has staying power at the college level because of its ability to exploit one-on-one matchups. It gives electric playmakers the opportunity to get the ball in space. This is a boon to smaller schools that can't hope to outmatch the big boys top to bottom. It also forces defenders to make tougher decisions, something that college players will struggle at.

Of course, in the NFL, with greater physical parity, more experience, and more practice time, the benefits of the spread option are more muted.

Mary Anne

May 7th, 2014 at 2:27 AM ^

Newly-minted offensive coordinator and quarterbacks coach Doug Nussmeier has already made an impact in Ann Arbor

Aside from bringing a renewed enthusiasm to the Michigan offense, he's also upped the Wolverines chances of landing 4-star running back Jeff Jones. But there's something Nussmeier hasn't done since assuming the job that has slowly become a story.

http://img.bleacherreport.net/img/images/photos/002/709/131/hi-res-181479651-brady-hoke-coach-of-the-michigan-wolverines-talks-with_crop_north.jpg?w=630&h=420&q=75