Thursday, August 28, 2008

2008 A-U Preview: 2007 by the numbers

Aaaaaand fate shifts things the other way.

Ah, luck. It's been one of the best parts of the Rise of the Sports Blog (and/or shrewd MSM writers like, say, John Hollinger or, more appropriately, Phil Steele) and associated stat wonkery.

Not that we've all started finding Sacajawea dollars on every sidewalk and make it through every yellow light that catches the suckers behind us, of course, but that good old-fashioned luck has been spotlighted as such an important predictive tool--take a look at Steele's "Turnovers = Turnaround" or net close games marks over the last couple of years, for instance. Or the Pythagorean Win calculations being done by OTS; last year the Pythagorean marked three SEC teams in particular for regression, and all three regressed. Luck matters.

Which is disheartening if taken the wrong way, but now a close game is also a kind of win-win situation, at least when viewed in a certain sober light that usually doesn't arrive until several days afterward: If My Team wins, Hooray, they've won!, someone buy Wes Byrum a "nonalcoholic" (read: secretly alcoholic) beer; if My Team loses, it's just luck, pure dumb luck, and luck'll make up for it on the rebound at some point.

Getting back to the Pythagorean, one of those three teams it expected a regression from was Auburn. Around this time last year I wrote a post detailing how the 2006 team had overachieved (outgained by 33.4 yards an SEC game, 6-0 in one-possession games!) and said that if the Tigers didn't start dominating people statistically, they might have a rough go of it. I concluded that returns to form by Cox and Borges portended that kind of dominance (why I felt free to overlook the mewling babes on the o-line and the general "Eh, whatever"-ness of the receiving corps remains a mystery even to myself), but when it became obvious that wasn't happening it wasn't entirely a surprise when razor-thin contests against USF, Miss. St., and LSU became Auburn losses. Frankly, after that 6-0 mark, going even 4-3 in one-possession games as Auburn did in 2007 is testament that Tubby may buck the regression to the mean somehow: he's now 14-4 in close games since the start of 2004.

So if 2006 had hints about what lay in store in 2007 (namely, those few close losses), what does 2007 say about 2008? It says some very nice things, fortunately. Looking at some commonly applied metrics ...

Conference yardage margin: A Steele favorite, in which we find that unlike the minus-33.4 mark Auburn posted in 2006, the 2007 Tigers outgained their SEC opponents by around 37 yards a game (offense minus defense). This was one of the better marks in the conference, and if the defense can maintain that same stout level of play--not a given, since Auburn has no choice but to line up a lightly-regarded true freshman across from the slot receiver, but likely nonetheless on paper--what seems like inevitable improvement from the offense should lead to an even wider margin this year. Sweet. (For compare/contrast's sake, I give you Tennessee, whose atrocious defending led them to a hideous minus-43 yards-a-game deficit. Yikes.)

Turnovers: Another of Steele's pet metrics, where random fluctuations that push a team's turnover margin too high or too low (see LSU's plus-20 in this department last season) can portend a swing of the pendulum the other way the following season. Again, no reason for Auburn to worry here: they finished dead even in turnover margin last year, with precisely the same number of giveaways as takeaways. If, again, you assume the defense can continue its ballhawking ways while our quarterbacks don't, ahem, occasionally toss off their back foot into triple coverage, Auburn should finish in the black this year. (It probably won't hurt if OTS's projections as they apply to fumble luck prove prophetic as well.)

Pythagorean wins: Not a whole lot to see here, really. Auburn overachieved by less than half a win in SEC play, neither notably lucky nor unlucky (and not drawing mention at all from OTS in his write-up). Five SEC Pythagorean wins (rounded up) isn't bad: given that they came with a frightfully young team that returns tons of key starters and depth all over the two-deep, a full six--which even LSU didn't manage last year--shouldn't be out of the question.

The bottom line: There's nothing here to suggest that if the 9-4 team of a year ago improves its down-to-down performance, as it should, that Auburn's record--particularly given the cushier SEC schedule--won't improve right along with it.


J.M. said...

Hear! Hear!

My head spins in anticipation.

Rod said...

This football season better start soon.
Hey! It starts today! Woohoo!