2009 Polls - Regular and Pre-Season

[muffler dragon]

Agreed in our present situation, but I believe our sentiment would be different if . . .
The final poll ranking hypothetically would look like this:
. . . . . . . . Harris . . . Coaches. . . . Comp. . . . Total pts.
1. Florida . . 1 . . . . . . 1 . . . . . . . . 1 . . . . . . 3 (1st)
2. Texas . . . 3 . . . . . . 3 . . . . . . . . 2 . . . . . . 8 (3rd)
3. tOSU . . . 2 . . . . . . 2 . . . . . . . . 3 . . . . . . 7 (2nd)

Precisely why I'm not talking about any other teams. The polls are irrelevant for the Buckeyes this year.

Haven't you heard? The Pac-10 representative is going to destroy tOSU because they want the win more!!!!

For me, personally, regardless of who the opponent will be, I'm nervous. I'm excitedly nervous, but nervous nonetheless. Living out here, I get to see the teams quite a bit. The Pac-10 is a sound conference this year, IMO.

 

[Muck]

Sounds like you are looking at last week's. I don't think this week's is out until tomorrow.

I was....they had the wrong date (Nov 15th) on the poll.

Still...fuck USC.

 

[Redhawk]

Newest BCS Standings

OSU at 10

 

[muffler dragon]

I'm confused as to how Oregon jumped us by two positions, but I don't know how to decipher the information. Ultimately, it's irrelevant.

 

[BuckeyeTillIDie]

I'm confused as to how Oregon jumped us by two positions, but I don't know how to decipher the information. Ultimately, it's irrelevant.

It's simple. Oregon both scores and allows an impressive amount of points.

 

[OregonBuckeye]

For me, personally, regardless of who the opponent will be, I'm nervous. I'm excitedly nervous, but nervous nonetheless. Living out here, I get to see the teams quite a bit. The Pac-10 is a sound conference this year, IMO.

Agreed but I really like how we match up with Oregon. Oregon's DL can be shoved around. The way our OL is playing I think they will oblige. They have aggressive DB's so it probably wouldn't be too smart to make this the game Pryor throws 40 times. On D, I think we need to be more aggressive with our DB's or Masoli will pick us apart. Run the ball down their throats, keep their O off the field, play disciplined D, and we'll win.

I'm confused as to how Oregon jumped us by two positions, but I don't know how to decipher the information. Ultimately, it's irrelevant.

Win over Arizona is more impressive than a win over scUM. Welcome to the Rich Rod era.

 

[muffler dragon]

Agreed but I really like how we match up with Oregon. Oregon's DL can be shoved around. The way our OL is playing I think they will oblige. They have aggressive DB's so it probably wouldn't be too smart to make this the game Pryor throws 40 times. On D, I think we need to be more aggressive with our DB's or Masoli will pick us apart. Run the ball down their throats, keep their O off the field, play disciplined D, and we'll win.

I, too, think that is the key on D. If Masoli's reads are off just a hair; then it throws off the rhythm.

Win over Arizona is more impressive than a win over scUM. Welcome to the Rich Rod era.

I was meaning by the breakdown on the link. We're 8 in the Harris and Coaches Polls versus 10 for Oregon. Then we're an average of 11 by the computers and Oregon is 7. I'm not seeing how that equals out to a 3 point jump by Oregon while we stay the same. Granted, Oregon mopped USC while we lost close, but I'm not seeing how any other determination can be made. I'm just not informed on the whole BCS ranking system.

What's been intriguing for me at work the last week is that both the Duck and Beaver fans that I work with are anxious in their anticipation at the possibility of playing the Buckeyes. The only statement I've made with regard to either possibility is that I believe the Buckeyes front seven on D will be the most impressive group that either team will have faced all year.

 

[gracelhink]

. . .
I was meaning by the breakdown on the link. We're 8 in the Harris and Coaches Polls versus 10 for Oregon. Then we're an average of 11 by the computers and Oregon is 7. I'm not seeing how that equals out to a 3 point jump by Oregon while we stay the same. Granted, Oregon mopped USC while we lost close, but I'm not seeing how any other determination can be made. I'm just not informed on the whole BCS ranking system.

What's been intriguing for me at work the last week is that both the Duck and Beaver fans that I work with are anxious in their anticipation at the possibility of playing the Buckeyes. The only statement I've made with regard to either possibility is that I believe the Buckeyes front seven on D will be the most impressive group that either team will have faced all year.

The 'puters are in love with the Pac 10. Check out the Massey, Wolfe and Sagarin rankings. If I was a Penn State or Oklahoma State fan and on the BCS bubble, I would be looking for some 'splaining.
Sagarin especially has caught my attention, he ranks tOSU at #17,
behind 2 loss Oregon #6, 3 loss Oregon State #15, 3 loss USC #11, 3 loss California #13, 3 loss Va Tech #9, 3 loss Miami #12, and 3 loss LSU #8??!!!!
As noted earlier, thankfully we are not needing computers to qualify for a BCS bowl and mercifully the Sagarin rankings are dropped when factoring the average for tOSU. The cumulative effect of Sagarin, Massey and Wolfe is still enough to drop us 2 places below the Harris poll and the USA Today poll.
Also of note the team that jumped tOSU last week in the poll, lost the following week (LSU).
Could be a bad omen for Oregon.

 

[Steve19]

The computers don't really rank Ohio State that badly. It's the BCS requirement that margin of victory is removed. Sagarin Predictor, his most accurate predictor of wins, has Ohio State at #9 before the Georgia Tech and Pittsburgh losses this week. Boise State is at #10 and Pittsburgh was ranked at #17.

Just to make sure it's highlighted: Cincy is ranked well below Ohio State at #13 when Sagarin is allowed to make his most accurate prediction. Their SOS will drop into the mid-70s this week, probably #74 or so.

On the BCS formula, Ohio State are penalized for losing to USC, who then lost 3 games that include a stunner almost as bad as Ohio State's other loss to a weak Purdue. It's Ohio State's losses and to whom they lost that cause Ohio State's relative strength against other teams to be masked in that silly BCS formula.

This week, Ohio State probably will raise to #7 in the Coaches and Harris polls and #8 in the BCS. Winning the Rose Bowl will result in at top 5 finish. Not bad in a rebuilding year.

 

[gracelhink]

The computers don't really rank Ohio State that badly. It's the BCS requirement that margin of victory is removed. Sagarin Predictor, his most accurate predictor of wins, has Ohio State at #9 before the Georgia Tech and Pittsburgh losses this week. Boise State is at #10 and Pittsburgh was ranked at #17.

Just to make sure it's highlighted: Cincy is ranked well below Ohio State at #13 when Sagarin is allowed to make his most accurate prediction. Their SOS will drop into the mid-70s this week, probably #74 or so.

On the BCS formula, Ohio State are penalized for losing to USC, who then lost 3 games that include a stunner almost as bad as Ohio State's other loss to a weak Purdue. It's Ohio State's losses and to whom they lost that cause Ohio State's relative strength against other teams to be masked in that silly BCS formula.

This week, Ohio State probably will raise to #7 in the Coaches and Harris polls and #8 in the BCS. Winning the Rose Bowl will result in at top 5 finish. Not bad in a rebuilding year.

The reference to the Jeff Sagarin ranking in the BCS computer rankings is found here.
BCS formula
Maybe I am reading it wrong. :!
The "Sagarin Predictor" you are referencing must not be the same as the "JS" column in the BCS computer rankings. The JS computer gives tOSU a value of "8" or the equivalent of #17 ranking.
If this is not the Sagarin predictor, then someone named Jeff Sagarin has ranked tOSU at #17, and they are using his ranking in the BCS formula.:(
BTW, "JS" (Jeff Sagarin) also ranks Cincy at #2.

 

[Buckeyeskickbuttocks]

The reference to the Jeff Sagarin ranking in the BCS computer rankings is found here.
BCS formula
Maybe I am reading it wrong. :!
The "Sagarin Predictor" you are referencing must not be the same as the "JS" column in the BCS computer rankings. The JS computer gives tOSU a value of "8" or the equivalent of #17 ranking.
If this is not the Sagarin predictor, then someone named Jeff Sagarin has ranked tOSU at #17, and they are using his ranking in the BCS formula.:(
BTW, "JS" (Jeff Sagarin) also ranks Cincy at #2.

It is not the same. Sagarin's ratings can be found here and Steve's description of the difference is accurate. It's the same Jeff Sagarin, but it's a different version of his computer poll that the BCS uses.

In ELO-CHESS, only winning and losing matters; the score margin is of no consequence,
which makes it very "politically correct". However it is less accurate in its predictions for
upcoming games than is the PURE POINTS, in which the score margin is the only thing that matters.
PURE POINTS is also known as PREDICTOR, BALLANTINE, RHEINGOLD, WHITE OWL and is the best single PREDICTOR
of future games. The ELO-CHESS will be utilized by the Bowl Championship Series(BCS).

 

[gracelhink]

It is not the same. Sagarin's ratings can be found here and Steve's description of the difference is accurate. It's the same Jeff Sagarin, but it's a different version of his computer poll that the BCS uses.

Thanks!
So what you are saying is that "JS" used by the BCS computer rankings is a different version from the "Sagarin Predictor" and the "ELO Chess" formula that is on the USA Today link.
Any idea where this formula can be found?

It still mystifies me that the Sagarin Predictor had Cincy at #13 last week but somehow JS computers made up enough points to give them a ranking of #2 on the JS computer poll used on the official rankings here.
And on the other hand the Predictor had tOSU #8 but some other factor dropped them to # 17 last week.

 

[Steve19]

Thanks!
So what you are saying is that "JS" used by the BCS computer rankings is a different version from the "Sagarin Predictor" and the "ELO Chess" formula that is on the USA Today link.
Any idea where this formula can be found?

It still mystifies me that the Sagarin Predictor had Cincy at #13 last week but somehow JS computers made up enough points to give them a ranking of #2 on the JS computer poll used on the official rankings here.
And on the other hand the Predictor had tOSU #8 but some other factor dropped them to # 17 last week.

None of the rankings would dream of giving away their formula: it's the family jewel.

No, that is not what we are saying.

• Sagarin has three formulae.
  • The Predictor is the most accurate model and predicts outcomes best. It models the margin of victory primarily.
  • ELO Chess is the Sagarin model used in the BCS. It does not consider margin of victory but only whether a team won or loss when playing another team (i.e, as well as whether the game was at home or away and other variables.)
  • The Sagarin power rating uses a combined set of criteria. As a result, the model generally estimates the power rating of a team somewhere between Predictor and ELO Chess.

The models differ last week because of the BCS requirement that margin of victory be left out of the rating.

Why does the BCS model ignore margin of victory? Because the mid-majors lobbied them to drop it, claiming it was unfair and caused teams to run up scores on over-matched teams, IIRC. In reality, models typically apply a decreasing premium as the score difference increases.

Anyway, if you are using power ratings to predict a game outcome, rely more on Predictor than the other Sagarin indices.

 

[BB73]

Thanks!
So what you are saying is that "JS" used by the BCS computer rankings is a different version from the "Sagarin Predictor" and the "ELO Chess" formula that is on the USA Today link.
Any idea where this formula can be found?

It still mystifies me that the Sagarin Predictor had Cincy at #13 last week but somehow JS computers made up enough points to give them a ranking of #2 on the JS computer poll used on the official rankings here.
And on the other hand the Predictor had tOSU #8 but some other factor dropped them to # 17 last week.

The Sagarin Predictor is based on a computer ranking that Sagarin has been doing since long before the BCS. It factors in margin of victory, among other things, in order to rank teams and then predict a scoring margin after factoring in which team, if any, is at home. I'm fairly sure it also uses data from the previous year, at least for part of the season, something that's not allowed under the BCS rules.

His 'ELO-Chess' ranking has margin of victory removed, per BCS rules, so that's the 'Sagarin ranking' used as part of the BCS formula.

 

[gracelhink]

Thanks for all the help with understanding the Sagarin computer numbers.

ELO Chess is the Sagarin model used in the BCS. It does not consider margin of victory but only whether a team won or loss when playing another team (i.e, as well as whether the game was at home or away and other variables.)

Hey I am still battling with Sagarin's formula. In my mind there is a flaw somewhere.
Example, the Nov. 22, JS computer formula ranked Iowa (11 points) ahead of tOSU (8 points).
So either the "home-ness" of our win over Iowa or "other variables" (family jewels?) must be influential enough to lift Iowa and sink tOSU.
Sagarin claims it is unbiased and I have trouble accepting that. I suppose it is likely my own bias that is contaminating the numbers.

The Week 14 polls and rankings were released today.

Week 14 Rankings

tOSU is #7 in the USA Today poll, leading #8 Oregon by 43 points.
In the Harris tOSU sits at a solid #7, ahead of Oregon by 64 points.
In the AP Poll Oregon leads #8 tOSU by a margin of 1104 to 1103.

In a close battle for a possible 2nd Big Ten team selected for a BCS bowl, Penn State leads Iowa by 93 points in the Harris, by 86 in USA Today, but trails Iowa by 23 in the AP.
Anyone know what influence these poll rankings will have when BCS bowls begin the selection process? i.e. Are the Bowl committees committed to a conference "pecking order" based on BCS rankings?