The Most Balanced Team in College Football Is... Eastern Michigan?

We all know by now, or at least you should because I've told you countless times, that Baylor had the best offense in college football in 2012 in terms of total yards. Who had the most balanced attack is, of course, a different question entirely and one I hope to answer today. The only problem is that the concept of "balance" is in and of itself at least somewhat subjective. Does a "balanced" offense mean one that averages close to the same amount of yards passing and rushing? If so, as you'll see, you may not like the results you get defining things that way. Is balance instead the team that ranks highest in the two pertinent categories? I think that gets us closer, but that approach also has it issues.

The honest answer is that I don't really know who is the most balanced offense in the country for the very reason I said above: how you define the term basically dictates your answer. What I've done, however, is try to put all the information I can in one place so that others (that's you) may help find the answer as closely as we can.

It's worth noting at the outset that I didn't intend for this to be another way of stroking Baylor's offensive ego, even if the results in some instances look favorable to the school I love. I didn't compile the information I compiled as a form of propaganda, and in fact, one of the reasons I didn't finalize my initial process was that it ended with Baylor on top. I'm sensitive to accusations of systemic bias like that. But all that said, here's where I started, with the top 10 offenses in the country ranked by total yards per game. This is the baseline.*

*Note: I'm just including the top ten at this point to show you were I went from here. My spreadsheet has all 120 FBS schools in it with compete information should you want to look.

Like I said, I didn't intend for this to be a "OMG, Baylor is so awesome" post. That's just how it turned out. Of the top ten offenses in the country, the worst individual units-- meaning the teams that are most one-dimensional from the looks of things -- are the Oklahoma and Oklahoma St. rushing offenses and the Oregon passing offense. I suspect the latter is going to get a lot better as Marcus Mariotta gains experience and becomes not-a-freshman. The document I linked has the entire list sorted by total offense with whatever ranks you want to see.

Just for the heck of it, here's the worst ten offenses, too.

And here's the Big 12 as a whole:

Now that the fun part is over:

My first thought upon getting the data entered was to simply add the two ranks together (rushing and passing offense) and use that as a crude proxy for overall balance. Here's how it turned out--

This is where the definition of "balance" becomes a problem. Though the rankings seem legitimate in that the best offenses are predictably near the top, I'm not comfortable with the idea that Baylor (or LA Tech) is ranked above Texas A&M in this system. The reason that happens is that simply adding the two ranks together favors lower numbers significantly. I'm not sure if that's a proper view of what "balance" really is. Why should Texas A&M be seemingly punished for ranking 13th and 14th in running and passing offense, respectively, when that appears to be more balanced than a Baylor offense that is 19th and 3rd? At the same time, however, A&M's rankings in the midst of the entire country don't really show how balanced they are, either. To steal a phrase from Inception, we need to go deeper.

This time I added another metric taking the quadratic mean of the actual yardage going into each ranking. National ranks themselves don't matter so much as the yards they represent. Again, however, the system slants toward higher individual numbers. That may not be a problem-- I'm not an expert in statistics-- but it made me uneasy.

This time I went back and took the quadratic mean of the rankings themselves instead of the yards going into the rankings. This brought me back to a ranking system I felt much better about, and one that eliminated some of the bias from simply adding the ranks together. By this process, Louisiana Tech wins and the Aggies come in second. They'll love that. Let's keep going.

It was at this point that I got a tip from a friend to to try to normalize the yardage itself, so I did. But things got really weird, really fast.

Fans of the SEC probably just wet themselves at this table before they even knew what I did, so I'll explain. This time, I took the quadratic mean of total yardage and then subtracted the actual mean for each team in an attempt to normalize great offenses versus the not-so-great. Lower numbers in the second column from the right means a lower spread between the two numbers. The problem is that we ended up with a ranking of the teams whose yardage totals were the most similar in the two columns. Some might think that's actual balance and that I'd done what I set out to do. Any system that ranks the 103rd-best offense as the "most balanced", though, is going to be rejected. Sorry.

To prove that's what it did, though, here's the resulting table from taking each team's run/pass ratio and finding the standard deviations away from the mean of each spread. It yields a ranking of each team in terms of who was the most equal in their yardage.

So there you have it: Eastern Michigan might well be the most balanced team in the country.

I'm going to keep working on this tonight, but in the meantime, if you have any suggestions for things I might do, please let me know. Until then, you can check out the Google Docs spreadsheet I'm working from (I actually built it in Excel).