clock menu more-arrow no yes

Filed under:

Quick and dirty BCS update

New, 92 comments

USA Today poll apparently out per CNNSI.

http://sportsillustrated.cnn.com/football/ncaa/polls/coaches/

My projection on the 40 2nd place Alabama votes had them at points of 1397; actual appears to be 1399. My projection for 19 OSU 2nd place votes, 32 3rd place votes, and 8 4th place votes (trailing Stanford) had OSU at 1368 points. Actual appears to be 1367.

If this were identical in the Harris poll, and if OSU were to get its best case computer scenario, the final BCS tally would be #2 Bama 0.9390, #3 OSU 0.9378. Excruciatingly close. That tells you, if the Harris can go just a little more pro-OSU, OSU could pull it out. But OSU needs the computers to go there way. Either Massey or Wolfe goes Bama, and this is over.

Unfortunately, Bama is looking like they may just eek this out. But don't give up hope Poke fans. Maybe the Harris voters are feeling more of an rematch backlash affect.

Earlier in the morning....
Spent a good amount of time analyzing the human polls. The way this goes is all about what you believe the voters will do. And none of us know with any degree of precision how big of a delta there will be based on the OSU/OU result.

Going into the weekend, Bama was getting about 55 out of 59 second place votes. OSU was relegated to getting 3rd, 4th, and 5th place votes (with many below that level as well). Bear in mind while VT did lose (and they will be putting about 20-ish of 3rd place votes for the taking), Stanford still sits there as a wedge this week. I estimate they had about 16 3rd place votes as of last weekend.

So what do you believe?

Here's an easy scenario to think about if you want to have hope for the Pokes: when the polls come out today, Bama takes 75% of 2nd place votes (and gets 25% of 3rd place votes), and OSU gets 25% of 2nd place votes (with 75% of 3rd place votes…. and NO VOTES below 3rd). OSU also wins Wolfe and Massey, giving them their best case 0.04 computer lead. In this scenario, OSU edges out Bama.

Unfortunately, the easy scenario is not realistic in any shape or form. Ballots are messy, and several voters will put OSU lower than 3rd (unfortunately). And several voters will continue to keep Stanford at 3rd. To me, the key number is 40. If Bama gets 40 (out of 59) or more coaches to vote them #2, they will beat OSU even in the bullish OSU computer scenario. If Bama gets less, OSU has a prayer. When I say this, I’m assuming about 8 voters (out of 59) keep Stanford ahead of OSU, meaning that Stanford loses about half of their 3rd place votes last week by virtue of not playing and viewers watching the OU asskicking.

If you think OSU can get 20 or more 2nd place votes, you believe OSU will be ahead of Bama on 1/3 or more of the human ballots. When I put it in those terms, and stop getting scientific about it, that sounds possible.

This is all of course hinging on whether OSU gets Wolfe. Brad Edwards posted that he’s heard Wolfe is pro-Bama. The numbers are a lot harder for OSU if Wolfe goes Bama’s way. Frankly, it starts to get really hard.

For now, given the fact the voters tend to have a strong SEC bias, given the uncertainty of both computers going OSU’s way, and given my feel Stanford can stay ahead of OSU on at least half a dozen ballots, I’m thinking Bama pulls it out. But if there is any “rematch backlash” it could be really tight (my heart is pulling for OSU by the way, and I still have hope). Will update more as other polls/computers get released.

original message from earlier in the morning...

Gents,

I've been taking most of this season off but couldn't help myself reading the news this morning. Bama/OKSt could be interesting. I am working through the computers now, and am posting some thoughts below. Will update this post later this morning when I've dug into the human polls a little more. Clearly, the humans control this decision. And it's safe to say OKSt has a tailwind from the computers, just a question of how much (0.02 advantage or 0.04?).

1. Sagarin is already out, and has Bama #2 and OSU #3.
2. Colley Matrix is safely #2 for OSU, keeping Bama at #3.
3. Billinsley’s ratings keep a score static if a team doesn’t play (the case for Bama), and adds to their score if they win against a strong team. Seems highly likely OSU will jump to #2 here.
4. OSU already had a solid lead in Anderson Hester, and with a win over #6 OU, that poll is not in question.

The two tougher calls are Massey and Wolfe. Massey has a heavy schedule strength influence, and also gives more value to games played late in the season rather than those played early in the season. The first factor is probably slightly bullish for Bama, the second factor is definitely a positive for OkSt.

Wolfe is similar to Massey, connects all CFB teams and ranks based on probability of what losses they have based on their schedule versus actual losses. The Wolfe advantage for Bama was 8.887 to 8.582 coming in, but OSU going to 11-1 (obviously same record as Bama) by beating Wolfe's #10 team will obviously have an impact.

Bottom line: Bama came into this week with a 0.95 to 0.93 advantage over OSU in the computer scores of the BCS. I think OSU will outrank Bama in AH, RB, CM, while Bama has been confirmed to outrank in JS. I think it is unlikely Bama keeps #2 spot in both KM and PW. If Bama can hang on to one, then OSU will have a 0.95 to 0.93 advantage in the computers. If OSU jumps Bama in both KM and PW (possible), OSU will have a 0.96 to 0.92 advantage (twice the computer lead).