One of this week's early goals was an analysis of the AP and USA Today polls and of the OFFICIAL DO NOT DISTURB vanguards' evolution since the preseason. But SMQ, frankly, pays near zero attention to these rankings and continues to find it difficult to convince himself they make any difference whatsoever. Certainly, in the mythical championship, big money sense, they matter on a disproportionate level, but at this stage of the season - after a little over a month, when the rankings remain largely an esoteric goo compared to the tangible effects of the later, inherently better-informed postseason votes - they don't reflect the evolving nature of the ranking beast.
Mainly, this is just because SMQ personally doesn't agree with the end results. Cupcake-devouring West Virginia over meat-eating Michigan in the USA Today? At fifth in the AP, but still getting two first place votes, the only top votes not reserved for Ohio State? Bah. Virginia Tech, last seen garnering the good fortune to lose by only 11 to the only opponent to date approaching either poll, continuing to appear at all, much less six spots ahead of Arkansas, last seen running roughshod over a consensus top five pick in the USA Today? Georgia, gutted by Tennessee, stressed to the limit by winless Colorado, its marquee win over South Carolina, still in the top sixteen of both polls? Bah.
But that's just SMQ's opinion, you know, which is the fundamental problem with the concept of a poll - a poll that will, at some point, matter, anyway, as opposed to the benign versions in other, blessedly playoff-based NCAA sports - that it's an amalgam, a compromise of many opinions reached by messy, idiosyncratic personal methods and resulting in a product with which, in all likelihood, no single voter is very satisfied. Brian, new sidebar addition Dan Shanoff and others made note of the individual voter ballots recently added to the AP's poll site, and this is moderately useful - for example, we can see that Chris Fowler's ballot is unusually sensible in relative terms (that is, it's close to SMQ's), and Auburn's approximate "home" voter, Joe Medley of the Anniston Star, is one of few writers this week to omit Arkansas from his ballot altogether, while Jon Solomon of the Birmingham News, in Crimson Tide country, ranks Arkansas 12th, ahead of the Tigers - but none of it will be as telling in the long term as the "Mr. Bold" and "Mr. Numb Existence" Brian already calculates for the Blog Poll, which is the greatest evidence of the widespread disagreement that makes the poll system such an inconsistent crapshoot and wholly inadequate for determining a legitimate champion. Right now, because opinions vary so widely and there's no existing method for systematic rankings, they don't even do that great of a job determining who's better - or who's been better, or who will be better at the end...what are we measuring again? - anyway. Paul Kislanko says even the voters have pretty much no idea what they really think, depending on how you decide to count, and these slightly deranged but very smart people have completely different ideas that do not count whatsoever.
So SMQ does not get worked up about polls, because it's mostly run of the mill disagreement. The only real issue he has before the ranking system inevitably robs a team of its official, corporately-designated mythical championship opportunity is the mass movement of number one votes towards Ohio State, and the perception - enforced for several weeks now in various outlets - that OSU is the "clear number one." SMQ has had Ohio State on top since the Buckeyes beat Texas, and, because its impressive wins over UT and Iowa and competent handling of the remaining rabble, will probably keep them there this week. But this year was regarded by nearly everyone as a wide open season on the polls, dominated by no one team, and that should still be the case: Michigan, for one, has been particularly impressive, and should be considered basically an equal of OSU; ditto Florida, which has knocked out otherwise very impressive Tennessee and LSU and Alabama, and also handled the rabble with ease. This not including Southern Cal, Louisville, et al, any of whom could put OSU and its manufactured Heisman candidate down for the count without tremendous surprise - unless, of course, you're running right out to buy the massive Buckeye hype, which others take as a mild conspiracy of ratings-boosting narrative and SMQ just finds distastefully, haphazardly narrow. Even if you concede OSU is the top team to date using whatever standard personally suits your sensibilities, do you believe your opinion deserves 97 percent consensus?
Update [2006-10-9 15:26:52 by SMQ]: Kislanko e-mails with another potentially useful grid, the vote distribution by team and rank for this week's AP poll, reproduced here in slightly unwieldly form:
Count of Votes by Rank for the Associated Press Poll
9 Oct 2006 2:49pm (Central)
|North Carolina St||—||—||—||—||—||—||—||—||—||—||—||—||—||—||—||—||—||—||—||—||—||—||—||1||1|
This demonstrates the lack of consensus SMQ mentioned above - after Ohio State at one, slots two through five are basically up for grabs. A few teams are even in positions that a majority of pollsters would disagree with: while each of the top three is ranked in the spot in which it was placed by the highest number of voters (i.e., USC had more votes for third place than any other position, and wound up third), this is true for only nine teams ranked 4-25.
For example, 17 voters ranked Arkansas higher than any single voter ranked Oregon, and Arkansas got more votes at #12 than at any other position. Yet Oregon is only one spot back at 18, and Arkansas is at 17, because seven fewer voters listed the Hogs at all, and almost all of Oregon's votes were concentrated from #16 to #20. But only 3 of 57 voters who included Arkansas on the ballot had the Hogs at 17, where it wound up in the collective average, five spots behind the position in which it actually had the most votes.