If Michigan Wins More Games, Does It Really Matter What Kind Offense They Run?

If you're here, you've probably already read MGoBlog's latest missive on Brady Hoke's and Al Borges' installation of a more traditional pro-style offense at Michigan. Being a true believer in the spread offense, MGo is, shall we say, upset about putting Denard Robinson under center and returning to the "boring" "mustachioed" Michigan offenses of old.

When I first read Brian's piece I didn't agree with it at all. But these days I rarely do. However, I consider Brian's opinions to be representative of segment of the Michigan fan base that truly believes the spread offense was and is the best offense we've ever, or will ever, have. And because these people care deeply about Michigan and Michigan's success, it would be wrong to dismiss their oppinions or to belittle them. I think they have a point, whether I agree with it or not.

As you would imagine with any MGo editorial, there are the usual spate of statistics that butress his point. Not surprisingly, last year's team had the highest yards per carry of any team in the last 10 years. Brian is correct in pointing out that a reason for this is because Michigan possesses the most dangerous offensive player in its history. He is also within the bounds of sound reason in asserting that a pro-style set may not be best way to utilize him.

There is, however a rather large leap of faith that Brian takes in making his point. Yards per carry (YPC). In the context of his argument, YPC equals success. While I'm sure this is an over simplification of his (perhaps) larger point, it is the argument's main support. Ignoring for a moment my personal reservations about the efficacy of the spread Michigan ran the last three years, my biggest qualm is ignoring the results of the seasons he cites. While the ground numbers are truly impressive, the year end results that accompany them are no where near as impressive. Each year Michigan ran the Rodriguez spread ranks highly in yardage for the last 10 years. Sadly, the won loss record is 15-22. (Caveats apply, of course)

However, the old, boring pro-style attack still occupies the remaining 7 spots and is 64-24. In terms of statistics, the obvious conclusion is that Rodriguez spread teams were between half a yard and yard and a half more productive on the ground. The other obvious conclusion is that the extra yards didn't turn into wins, and the old boring teams won a lot more games.

But does this really mean anything at all? Not really. Statistics only give you so much information. You would think that if your running game and passing game were unstoppable then you'd win every game, based on numbers alone. I decided to look at Michigan's passing yards over the last ten season to see if there was any correlation between wins and passing. I got the following:

These are the numbers. Based on this you would think that going to a kick ass passing game would ensure success. Of the top five passing teams of the last ten years, the only team without 9+ wins is the 2010 team. By this rationale, we should focus entirely on passing, hire Mike Leach and buy pirate patches. But that's not how it works. Any football coach worth his salt will tell you he wants a balanced attack so that he can keep the other team off balance, guessing. And that was a pillar of Rodriguez' game plan as well as Borges.

Looking at the numbers, the 2003 and 2006 teams appear to be the most accomplished and balanced teams in recent memory. Why? because they have ten wins and rank in the top ten in both YP carry, yards per completion and most importantly, Wins. In terms of total offense, I will concede the point that on paper the 2010 offense was the most prolific offense anyone has ever counted at Michigan. But in terms of efficacy, I and many others would argue that the offense was not as good overall as its total yardage would indicate.

more after the jump.....

But what did the team actually look like when they played, the so-called "eyeball" test. Was the offense effective when it mattered? Could it dictate the play of a game against talented opposition? I would argue it could not and did not. I've voiced these opinoins in the past and they remain relevant to this discussion. Brian's right, though, if we had a kicking game maybe the results might have been different. As he says:

Getting Michigan's offense to go from explosive but inconsistent to world-destroying is a matter of getting a kicker, finding a good running back, working on Denard's reads and accuracy, and leaving everything else the hell alone.

I don't think "leaving everything else the hell alone" will do it. I would argue that he should include cutting down on dumb penalties (Personal fouls, procedure, lining up correctly, etc.), not turning the ball over, finding more tight ends that can block, and coming up with another running play other than Denard left, right, middle. Adding all these things together, they are not an insubstantial set of hurdles in any way. That's a lot of stuff that has to happen for this to be a world destroying offense. For all the four star tailbacks Michigan's recruited.... yeah, you know the results recently. But arguing over hypotheticals gets us no where. That is my opinion contrasted with his. You may make your own judgments.

Here's the thing, for all the hand wringing over what's happened to the supposed death machine of a offense we "would've" had, I think something's getting lost. That something is that the pro-style offense works. The pro-style offense gave the 1997 team a national championship. The pro style offense gave us three 10+ win seasons since 2001. The pro style offense gave us some of our most revered players like Brady, Hart, Manningham, Braylon, Touchdown Tim, Jake, etc.

It's not like we're trading in a finely tuned Lamborghini for a Model T. I think it's more like we're trading in a Lamborghini for a Vette. Sure the Lambo is faster... when it works, isn't in the shop or having other issues. The Vette is still impressive, explosive, and works consistently when you turn the ignition. On paper the Lambo is mind bending, but in reality you'll never have it out of the shop enough to enjoy it. The Vette may not have the numbers of the Lambo, but it'll still blow your hair back and put a smile on your face. More importantly, it'll work.

In supporting the spread, there seems to be this insistence on selectively applying numbers to butress a biased argument. Baseball may work this way, but football doesn't. The physical and scheme variables in football are too great for numbers to tell the tale of a game. How many times have you seen a game where the numbers in the box score were incredibly close, but the game you watched was not. Think of OSU/Miami this past year. Think of the Wisconsin or Ohio State games. Numbers lie as easily as they tell the truth.

It wasn't too long ago that Beauford broke down Michigan's overall offensive effectiveness month by month, and demonstrated that Michigan's overall numbers for 2010 were skewed by the first four games of the season. And that's where the break down occurs between Rodriguez offense's supporters and people who are more open to the pro-style, Points. The argument that Michigan was a hair away from being awesome is based on Michigan's yards per game. The argument that Michigan was good and not great is based on points per game. And there are numbers are there to support both.

If you look at Michigan's overall scoring compared to yardage, the difference is pretty stark. Michigan's overall scoring offense went from 10th in Aug/Sep, to 48th in Oct, to 38th in Nov, to 67th in Dec/Jan. While you can argue that Michigan shouldn't be charged for not playing a game in December, I think it is arguable that Michigan shouldn't get credit for it's extra overtime points in its basketball-on-grass match up against Illinois. Both of these factors skew the numbers.

So to be fair, let's drop the Dec/January comparison for lack of games. We'll also drop the extra 22 points Michigan scored in OT against Illinois because when you're starting from the 25 everytime your chances of scoring... yeah. What that means is that Michigan scored 45 (Ill), 27 (Purdue - and that's counting the Cameron Gordon fumble return for a touchdown!), 28 (Wisc), and 7 (OSU). That averages out to 26.8ish points per game in November, down from the 32.3 with the OT that Michigan averaged. FWIW, 26.8 ppg drops Michigan to 63rd in the country for the month of November. So, adjusted, Michigan's offense went from 10th, to 48th, to 63rd as the three months of the season went forward. In terms of the offense's ability to produce points, the thing that is required to win ball games, Michigan's output collapsed against better competition.

For comparisons sake, I took a look at San Diego State's offensive numbers over the course of the year. I took a look at total offense, passing offense, rushing offense, and scoring offense, overall and month by month. The numbers are will impress you.

I'll let you look at Beauford's piece for all the other numbers, but here are the overall tallies for the Wovlerines in 2010:

I really, really don't understand the handwringing here. I think this goes back to the Vette versus Lambo argument. Maybe we're not going to  be a top ten offense again, but I'm not going to sneeze at scoring 35 ppg and racking up 457 total yards a game. Michigan's rushing game will obviously take a bit of a hit, but not one that should make you run in front of a bus. On the flip side, Michigan's passing offense should improve quite a bit and take advantage of Denard Robinson's arm strength and Michigan's incredibly talented slate of receivers. FWIW, SDSU's offense put these numbers up in only the second year of Hoke's tenure at SDSU. The result was a better scoring offense and passing offense, and an overall SDSU offense that ranked 16th overall compared 8th for Michigan.

If your response to this is that SDSU played no one, you have a point. To an extent. SDSU put 300 yards and 28 points on the board against TCU (the extra 7 came via fumble recovery). For comparison's sake, Wisconsin put up 385 and 20 points against the Horned Frogs. Transitive properties being what they are, SDSU's offense was pretty good. More importantly, SDSU's offense started out pretty good, and stayed pretty good throughout the season. They were consistent.

Every coach at Michigan since Bo patrolled the sidelines has preached one thing: Execution, Execution, Execution. The reason for this is that football is first, and foremost, a game of brute athletic strength and coordination. All things being equal, the bigger stronger guy will usually win. To combat that, there is coaching. For the faster smaller guy to win he must out execute the larger man. As you extrapolate that out, a well coached smaller team can beat a bigger team. However, if the bigger, stronger team is coached equally well as the weaker team, the bigger, stronger team should win. There is an apt analogy in hockey, "big fast beats small fast." You have to consistently execute every aspect of the game to be successful, and the only way to do that is to practice the mundane stuff until your brain turns to jelly. But even if it does, you'll remember it. And you'll execute it consistently.

If Hoke's insistence on repetition and "toughness" and the so-called "manball" means Michigan will return to out-executing it's opponents, I'm good with that. Because it's not something Michigan's done a lot of in the last three years. As much as some people would love to see the Oregon style spread, there are others who like watching a well executed pro set. Oregon's flashy and fun, but it requires a level of execution that hasn't been present at Michigan since the mid part of Lloyd Carr's tenure.

As boring as people like to make Michigan's old offense sound, I think we forget how much fun it really was. What about Brady to Terrell or those long runs from Chris Perry? What about A Train leveling the boom on some fool as he plowed his way into the endzone? I remember bouncing up and down as these things happened. I remember being excited. This dirty, old man, mustachioed offense was pretty damn entertaining to watch. And with a rocket armed passer, who also happens to be rocket legged, I'm just as excited to see Michigan play this season as I was last season.

What it comes down to is winning. I am neither old nor mustachioed, and I don't have a clue who this Beckmann guy is. If this coaching staff can generate and offense anything like the ones we saw from 2001-2007, I'll be happy. I think they've got a pretty good chance of being good based on the return of the majority of an offensive line that averages nearly 300 pounds and all but one starter on offense.

Frankly, it doesn't matter what kind of offense that Michigan runs. Michigan will score points. Why the switch to a proven offense is such a concern is beyond me. What matters is whether the defense will improve. So, to me at least, going into this season not knowing how the offense will operate is far less terrifying than going into another season with Greg Robinson and knowing just how bad the defense will be. And that, more so than the numbers on offense and whether Denard wins the Heisman, will determine the success of this team.