• Follow us on Twitter @buckeyeplanet and @bp_recruiting, like us on Facebook! Enjoy a post or article, recommend it to others! BP is only as strong as its community, and we only promote by word of mouth, so share away!
  • Consider registering! Fewer and higher quality ads, no emails you don't want, access to all the forums, download game torrents, private messages, polls, Sportsbook, etc. Even if you just want to lurk, there are a lot of good reasons to register!

BCS Guru (incorrectly) projected LSU-Oklahoma State in Title Game

Jump to latest Follow Reply
Sort by date Sort by votes
BB73

BB73

Loves Buckeye History
Staff member
Bookie
'16 & '17 Upset Contest Winner
I posted this in another thread, but I'm adding it here so it be be referenced when the polls come out on Sunday, and people start wondering how the final BCS standings might shake out. This is the scenario for Okie State trying to jump Bama.

Oklahoma State would only pass Bama if they were #2 in at least 5 of the 6 computers (which is probably likely), and were voted ahead of Alabama in at least 1/4 of the human voting in both the Coaches and Harris polls. It would be a dead heat if Oklahoma State was #2 and Alabama #3 in 5 (or 6) of the 6 computers (high and low are thrown out); and Alabama was #2 and Okie St #3 in exactly 75% of the human votes, with the remaining 25% having it flipped. That's with LSU being #1 in every computer and for every voter (which they are now), and with no other team getting into the top 3 in anything.

Team....Computers..Coaches..Harris..Total (average of the 3 columns)
Okla St...960.......930......930.....940
Alabama...920.......950......950.....940

If the 'middle 4' of the computer rankings (after tossing out the high and the low) is exactly 2, a team gets .960 in that column, if it's exactly 3, they get .920 (all 1's equals 100%, and each spot drops 4 percentage points).

In the human polls, the number is the vote total divided by how many votes a team would get if everybody voted them #1. So LSU is at 1.000 (100%). A team that got all #2 votes would be at .960 and a team with all #3 votes would be at .920. If you get half #2s and half #3s you'd be at .940; the 75/25 split among #2/#3 votes, with no other team getting any votes in the top-3, would mean one team would be at .950 and the other at .930.

That's how the math has to work to get Oklahoma State up to #2. They were already #2 in two of the computer polls before yesterday. I think they will pass Alabama in all except the Colley Matrix, which will give them the .960 in the computer portion, and Alabama the .920. Then at least 25% of the voters need to make Oklahoma State #2, with nobody having them below #3.

BCS.Guru now thinks this will happen. Of course, this will generate hits like crazy for his site today.

 
Last edited:
Bump. Watching the computers and the polls will be interesting today.

Remember it's the Coaches and the Harris for the human polls - the AP doesn't count in the BCS.
 
Last edited:
Upvote 0
Diego-Bucks;2056625; said:
Sagarin's model has Bama at #2 and OK State at #3. Have to look at the ELO_Chess model which is the one utilized in the BCS rankings.

http://www.usatoday.com/sports/sagarin/fbt11.htm
Click to expand...

Once again, what a need for more transparency. I don't have much of an emotional stake in this, but just looking at the stats Sagarin lists, OK State has the same overall record as Alabama, a tougher strength of schedule, and better records against the top 10 and top 30 by Sagarin's measure. Since margin of victory isn't counted, how in the world can they be behind Alabama? There's clearly some unstated subtlety that is pushing them ahead in Sagarin's model, but we'll never know.
 
Upvote 0
bluechillj;2056630; said:
Once again, what a need for more transparency. I don't have much of an emotional stake in this, but just looking at the stats Sagarin lists, OK State has the same overall record as Alabama, a tougher strength of schedule, and better records against the top 10 and top 30 by Sagarin's measure. Since margin of victory isn't counted, how in the world can they be behind Alabama? There's clearly some unstated subtlety that is pushing them ahead in Sagarin's model, but we'll never know.
Click to expand...

I guess I spoke without thinking it through completely; the answer is probably how much the model 'dings' a team for a bad loss. However, my point about more transparency stands.
 
Upvote 0
bluechillj said:
Once again, what a need for more transparency. I don't have much of an emotional stake in this, but just looking at the stats Sagarin lists, OK State has the same overall record as Alabama, a tougher strength of schedule, and better records against the top 10 and top 30 by Sagarin's measure. Since margin of victory isn't counted, how in the world can they be behind Alabama? There's clearly some unstated subtlety that is pushing them ahead in Sagarin's model, but we'll never know.
Click to expand...
I don't know exactly. Its based on the ELO rating model for chess according to Wikipedia. http://en.wikipedia.org/wiki/Elo_rating_system#History

Reading that, I gauge that there is an underlying assumption on the relative rating of each team, and how that team performs over the course of the season (wins or losses) dictates how that rating changes. Assuming that each team has a skill rating and it is normally distributed, then when two teams of uneven skill play each other, the high-ranking team would have to perform below average to lose, or the low ranking-team would have to perform above average to win. Sometimes that may require playing at standard deviation or more above/below your rating. I would guess that OK State's loss to Iowa State resulted in a performance that was well below their average skill rating. Compared with Alabama's loss to LSU which shows that Alabama played at or near their own skill rating to lose by 3. LSU's rating is 108 to Iowa State's 82.

Alabama's and OK State's overall SOS may be too similar for it to factor into play thus the overall ranking comes down to the performance in their loss.
 
Upvote 0
FWIW, Mandel weighs in.

http://sportsillustrated.cnn.com/20...oma-state-bedlam-bcs/index.html?sct=cf_t11_a2

STILLWATER, Okla. -- Were you watching Saturday night, Harris and Coaches' Poll voters? Did you see BCS No. 3 Oklahoma State (11-1) demolish its archrival Oklahoma (9-3) 44-10, and in doing so -- pay close attention to this part -- win its conference championship? You have a vexing decision to make Sunday morning, and be warned, if you choose wrong a certain multibillionaire may come after you.
"If we don't get in the BCS [title game], if I have the power to do it, I'm going to have an investigation," Oklahoma State mega-booster T. Boone Pickens said after Saturday night's game. "After the way we handled OU tonight, I just can't imagine LSU playing Alabama twice. It looks like an SEC-closed system if that happens."
There are no right answers when it comes to the BCS. Under a more sensible system, Oklahoma State and Alabama could play each other in a football game to determine which would meet No. 1 LSU. (At this point the Tigers deserve a bye, but if we're being fair, they'd meet 11-1 Stanford in a hypothetical semifinal.) Instead, the decision is left in the hands of mostly semi-informed voters.
But this is the system we have. So all we can ask, dear voters, is that you do the right thing: Give us the matchup we haven't already seen. But don't do it just to prevent a rematch. Do it because Mike Gundy's Cowboys earned it.
"I don't think there's any question Oklahoma State should play in the big game," Gundy said after his Cowboys handed the Sooners their most lopsided defeat in seven years. "After what they accomplished tonight and the way they did it, I don't think there's a question they deserve an opportunity to play for it all."
In ending an eight-game losing streak to its archrival and clinching its first outright Big 8 or Big 12 title, Oklahoma State beat its seventh opponent with a winning record and its fifth that was ranked in last week's BCS Top 25. By comparison, Alabama can claim three and two, respectively. And yet, if you watched television Saturday, you would have thought the Tide had some sort of overwhelming claim to the No. 2 spot.
The second half of CBS' broadcast of the LSU-Georgia game turned into a predictable infomercial for an LSU-'Bama rematch. Then, during ABC's broadcast of Bedlam -- with the No. 3 team in the standings still playing its last game -- a "BCS Lookahead" item scrolled on the ticker, providing assorted LSU-'Bama related nuggets.
cont.
Click to expand...
 
Upvote 0
Bucknut24;2056648; said:
Brad Edwards saying this is prob gonna be the closest finish since 04 (when they redid formula)
Click to expand...

Brad clearly has it working out for the SEC rematch right... ES(ec)PN doesn't have enough chutzpah to go against them
 
Upvote 0
Please get Okla St in the title game. Michigan may play Alabama and get killed but I don't give a damn. I don't want to see a damn rematch. It basically says that LSU win over Bama really meant nothing. That is just great for a #1 vs #2 match up.
 
Upvote 0
You must log in or register to reply here.
Back
Top