Statistical Analysis: Egg Bowl

[ED: For a brief explanation of what's going on here, check out the introductory piece in this series. Also, it's early in the morning, so who doesn't want to write and read about numbers right now, right guys?!]

So it's down to this. With just one game left in the regular season, the Ole Miss Rebels need just one more win to reach bowl eligibility, something which would itself be an impressive feat in Hugh Freeze's first year in Oxford. The opponent in this final game is, of course, the arch-rival Mississippi State Bulldogs, a team which both loves to beat Ole Miss and has done so every season since 2009.

So how do those two stack up in the statistics we like to measure?

So looking at that, with State's statistical advantages on defense and demonstrated ability to protect quarterback Tyler Russell, why is it that they, at 8-3, are the one point underdogs to our Rebels, at 5-6? Why is it that largely neutral oddsmakers and sports betters feel better about our Rebels than perhaps many of us do?

To attempt to answer this, I wanted to look at our team's statistics in a what is perhaps a more relevant manner. I looked at these statistical measures (Well, most of them at least - I didn't want to pore through the numbers to dig up passer efficiency ratings, third down conversion rates, etc. Hey, you get what you pay for.) for our teams against our common opponents, or everyone in the SEC West outside of the state of Mississippi.

What you'll see then may help to answer the question as to how on earth we could be favored to win this game.

The difference in this (admittedly incomplete) set of statistics and the former set is pretty pronounced. Ole Miss has more advantages in the second set of measures and, where the Rebels do have them, they're stronger. Against Alabama, LSU, Auburn, Arkansas, and Texas A&M, the Rebels have had more offensive yards, given up fewer defensive years, scored more points, gotten more sacks, et cetera, while State, as per the rest of the season, has done a better job of protecting the ball, the quarterback, and the endzone.

What accounts for these differences? To put it simply, State had Tennessee, Kentucky, MTSU and a trio of Sun Belt Schools on their schedule, where we had Georgia, Vanderbilt, Texas, an FCS team and a pair of CUSA teams. Playing Georgia and Vanderbilt is, itself, more daunting than playing Tennessee and Kentucky, and the Texas game certainly skews defensive statistics against the Rebs.

Of course, State's paltry defensive performance against Texas A&M could skew those statistics against the Bulldogs, but they do have "normalizing" performances against Arkansas and Auburn, opponents against which they largely outperformed the Rebels.

When looking at these two sets of data, there are three areas which switch from a Mississippi State advantage to an Ole Miss advantage when changing the range of data collected: scoring offense, total defense, and pass defense. No statistical categories which were Ole Miss advantages in the first set become State advantages in the second, although State did improve ever so slightly slightly against Ole Miss in the second set in their rush defense and passing offense, while improving fairly significantly in sacks.

Still, this is all an imperfect way to really compare these two teams, a statement which itself is largely vapid because there is no "perfect" way to do so. I simply wanted to avoid simple looks at records or particular statistical categories or position groups in an attempt to paint a broader picture, something which I have kinda halfway done so far.

FootballOutsiders.com, a website run in part by SB Nation's Bill Connelly, looks at these statistics and more - oh so much more - in a deep and robust way which attempts to analyze, and not elect, the who the best college football teams in America are. To do so, they look at a team's "efficiency" (basically, how well did they move the ball against their opponents), "success rate" (a measure looked at in Me and Paul's statistics posts we all love so much), "points per play", and so on and so forth. These statistics are then adjusted against a team's strength of schedule in a way that, as they put it, "is designed to reward tougher schedules and punish weaker ones."

Sound complex? Well that's because it is, and that's by design. The complexities of their analysis allows for a depth and robustness not present in the run-of-the-mill looks at football stats, something which they could certainly do a much better job of explaining than I could.

I bring them up because I think their adjustment for strength of schedules is important in understanding just how good a football team is. It's that sort of adjustment that could perhaps answer the question posed at the beginning of this article.

So where does FootballOutsiders see the Bulldogs as stacked up against the Rebels? Well, they have State as the 49th best team in the country and Ole Miss as the 62nd. Thirteen spots, in a vacuum, seems like a lot to separate our two rivals, but when broken into offense, defense, and special teams ratings the differences between our two teams become more pronounced. On offense, State is ranked 61st, whereas Ole Miss is ranked 35th; on defense, the teams are 44th and 46th respectively; and on special teams - wait for it - State is 46th while Ole Miss is a stunning 115th out of 124 teams in the FBS. So, if those ratings are to be believed, Ole Miss has the better offense, our defenses are more-or-less the same, and the Rebel special teams are literally some of the worst in the country.

Okay, so what does this all mean? And why did I bombard you with it on a Saturday (I don't know)? It means that the folks who are picking the Rebels to win, are doing so because they're saying to themselves something like "the Rebs played LSU, Alabama, and A&M much better than State did," an argument which itself can be demonstrated statistically. What can this lead us to conclude? That there's a good chance that this game is terribly close and nerve-wracking from start to finish, and that we are perhaps more evenly matched than a lot of us would like to admit.

It also means that if the Rebels can minimize turnovers and special teams mistakes, they are undeniably more than just capable of beating Mississippi State at home tonight.