This was due today, so here's my initial ballot for people to complain about. A couple thoughts after the jump.
In my first year on the blogpoll, one fact became apparent: nobody knows anything. Every week, I struggled to rank teams that I hadn't watched based on completely arbitrary criteria. Week after week, individual blogpoll voters proved that their supposed collective wisdom over the AP and Coaches' polls were a plain fable. True, blogpoll is more representative, and votes are public, but this is ultimately a fool's errand. I would have have been a much better voter several years ago, when I watched every game that I possibly had access to, but I no longer am able to do that.
In that case, why continue? In my understanding, the blogpoll exists as much for promoting its individual voters' sites as it does for some definitive, optimal standard of ordinal rankings. If I didn't vote, the Big East football conference would be further marginalized. This would be a bad thing, as the college football world in general is full of resounding ignorance (myself included), and a heavy, unsubstantiated, highly-selective prejudice against the Big East. Gotta counterbalance at least one yokel's vote.
I don't claim to be better than the average voter, but at least I'll be able to make the sample more representative to a minute degree. I'll try to be straight forward, open-minded, and honest in my arguments.
The Pac-10 is highly underrated. It's been one of the best conferences over the past several years, but gets little respect on a national stage. Conversely, midwestern teams (the Big Ten and Notre Dame) are severely overrated. This owes to the fact that their teams match the SEC in fan support and media coverage, but are not nearly as talented. The Big Ten has been, statistically, the worst BCS conference in each of the past two years. They are not nearly worth their collective hype, and it's little surprise that B10 teams are about as overmatched as ND every year in their equally undeserved bowl games.
Strength of schedule is important. I was one of the few dissenters on Texas Tech last year for that very reason. This is why I did not vote for Rutgers, even though I think they have top 25 caliber talent.
I put a lot of weight on statistical analysis. Computer polls occasionally generate strange results, but their lack of bias makes them far preferable to human voters. It's important to adjust for context, which means weighing quality wins (up to a certain cutoff), SOS, and turnover margin. The latter, specifically, is very important. Football Outsiders researches the pro game, but they have shown that recovering fumbles (NOT forcing fumbles) is largely dependent on chance. Phil Steele's (who, it should be pointed out, does not have a rigorous statistical background) prediction strategy is largely centered on the mantra that turnover margin is not especially consistent from year to year. It's not absolutely definitive, but something to consider when looking for sleepers and surprise disappointments.
A few further general thoughts -
I'm not on the NC State bandwagon for two reasons. When did everyone decide that Tom O'Brien was suddenly an elite coach? He didn't have much to work with at BC, but he always finished at 8-4 there and in the middle of the conference. Russell Wilson is indeed quite good, but he also missed three games last year with injuries. His body looks brittle, and isn't a great bet to hold up on the rigors of a full season.
For reasons described above, most of the blogosphere is high on Illinois this year, and for good reason. They improved on a per play basis for each of the past three years. Last year's losing season was solely predicated on turnovers. Surely, Ron Zook deserves the blame for a lot of that, but he's brought in too much talent over the past few years to not roll over what should be a bad Big Ten once again. Michigan's offense should be potent under RichRod, but mgoblog has convinced me enough of their limited depth on defense that more than 7-5 will be a stretch.
Mississippi is one of the most overrated teams in the country right now. Their QB Snead is good, but the Rebels play a bad OOC schedule, and are getting too much credit for beating a Texas Tech team that similarly feasted on cupcakes (and speaking of the Raiders, how do polls rank them, and yet Cincy still doesn't get any respect?) Houston Nutt was fairly inconsistent at Arkansas. I'm higher on LSU than most because they've fixed their QB and defensive issues, and I like Georgia a lot too. Everyone forgets that they have been one of the most consistent winners of the past decade.
Notre Dame isn't worth any more of my wasted attention, so I'll be brief. I like their receivers, and Tenuta should help their defense a lot. However, voters are overreacting to their schedule, and too heavily weighing the bowl win over a bad Hawaii team. The Irish are still weak in the trenches. I don't think Weis has had enough time to show whether he can develop players yet or not, but the recruiting sites are fairly generous to the Domers, and tend to overrate and overhype their commits. Hence, it's a mistake to think that they're too talented not to at least be somewhat good. Plus, voters are being inconsistent. Pitt has had a bunch of good recruits too recently, and they beat Notre Dame last year. How can you rank ND and not Pitt? For the record, I don't think either team belongs in the top 25.