Road to the BCS Championship, Week 11

A soft landing for the Tide.

Alabama fell to fourth in the Coaches and Harris Polls, but it remained third in the BCS standings. I have no problem with this; if you lose to the No. 1 team by a field goal in overtime, there's even a case to be made for not falling in the polls at all.

So how did the Crimson Tide edge out the Stanford for third? Glad you asked.

The Pac-12 is absolutely toxic to the BCS computer rankings.

The computers have often disagreed with the human polls about contenders, but almost never to this degree. Stanford is either No. 2 or No. 3 in the human polls, but it's seventh in the aggregate computer rankings behind No. 4 Boise State, No. 5 Oklahoma, and No. 6 Arkansas. The Cardinal is as close to the computers' No. 13 South Carolina as it is to LSU and Oklahoma State, who are tied for first. Only two computer rankings have Stanford even as a top five team, and one of those is the notoriously messed up Billingsley rating.

Here's the thing though: Stanford's reward for beating Oregon State on the road this past weekend was seeing its computer poll score fall from .800 to .770. Yes Stanford will get a bump from playing Oregon, but if it wins, that will limit how much the playing the Ducks helps their cause. Plus the Oklahoma schools will get a bigger bump from playing each other, and Bama's bump for playing Auburn actually won't be all that far off from Stanford's Oregon effect (especially if the Tigers beat Georgia).

Stanford will likely have a healthy deficit in the computer polls relative to any zero- or one-loss SEC or Big 12 teams when everything is said and done. It probably won't be enough to make a one-loss Oklahoma team or Alabama jump an unbeaten Stanford in the final standings, so don't fret over that particular scenario. However right now the two human polls disagree over who is No. 2, and if they continue to do so, the tiebreaking computer vote will swing firmly in Oklahoma State's favor.

Boise State: Laying in the weeds.

To follow up on Boise State's feature here last week, the Broncos are still hanging around the top of the standings. Ultimately, I don't see their fate being so much as a contender as a spoiler.

Despite playing Colorado State, Air Force and UNLV in the interim, Boise's computer ranking has actually increased from .800 to .850 in the time since the BCS standings first debuted. I expect that to fall back down, especially with the dead weight that is New Mexico still yet to come. I honestly am not sure if an undefeated Stanford would be ahead of an undefeated Boise State in the final computer rankings, which says a lot about the computers this year.

Where Boise really has influence is if we end up with a logjam of one-loss teams. The Broncos won't play for the national title unless they are a unanimous No. 2 in the polls, and voters being voters, that won't happen. They could end up spoilers for someone if there really is disagreement in the human polls.

LSU-Alabama rematch odds: 20%.

I had the odds at 10% last week; I'm moving them up to 20% this week. One of the absolute key prerequisites was met: that the teams played a legitimately close game with a final margin of no more than one score. LSU's three-point overtime win certainly qualifies. I might have put the odds higher at maybe 30% or so if Alabama had won; voters are more forgiving of road losses than home losses.

I still think that the chances are better than not that Oregon beats Stanford and Oklahoma beats Oklahoma State, so I do think a rematch will be on the table. The continued winning of the latter set of teams helps the BCS out as a format, because imagine what would happen if Oregon wins out and those two each had two losses. You're looking at these options: LSU rematches Oregon, LSU rematches Alabama, and LSU faces a Boise State team that will have effectively played 11 nobodies. That's not appetizing.

Let's imagine Stanford and Oklahoma State do fall (which has a 25% chance of happening if each's big game remaining is a toss up). Let's also say LSU, Bama and Boise win out.

My educated guess is that in that situation, Alabama would end up second in the computer polls. It would probably be a narrow second over Oklahoma, but second nonetheless. Keep in mind though that the computers are basically a tiebreaker if the humans disagree.

The serious candidates to play LSU would be one-loss Oklahoma, one-loss Alabama, and undefeated Boise State. Aversion to rematches and a bad computer poll ranking basically would eliminate one-loss Oregon. Oklahoma would seem to be the obvious choice here, with Bama having lost to LSU and Boise having a rich history of being shut out of the national title game. Plus, Boise would be behind both of the other two in the computers. The human voters have shown no inclination to vote overwhelmingly for the Broncos over one-loss AQ conference champions, and that is what it would take for them to face LSU. They're done, essentially.

Some voters might boost Boise to make a point though, especially coaches from non-AQ conferences. How many sympathetic voters BSU has will play a role in sorting out the Oklahoma-Alabama positioning. Who will the Bronco-leaning rebels put third and who will they put fourth? This is where BSU's spoiler potential really comes through.

As for the Oklahoma-Alabama debate, it will be a good one. OU will have won what is arguably the toughest conference in the country this season. It will have also lost to an increasingly sketchy Texas Tech team, which, at present, is not a lock to make a bowl game. Meanwhile, Bama's sole blemish will be a three-point loss to the No. 1 team in overtime. Alabama will win every defensive statistical argument; Oklahoma will win most every offensive statistical argument. They don't even have anything resembling a common opponent.

I can't predict who will win that argument. We may not even get that far. And for as wild as that situation will be, just imagine if Arkansas knocks off LSU on the last weekend of the regular season. We're a long way from this issue being settled either way.