BCS Top 10

[milohimself]

Riveting banter, chaps.

Thread cleaned.

Save all of that for your HS debate squad (or the pub).

 

[wheaton4prez]

I've already admitted Florida is a poor team. Oregon is better than any of us believed they would be. I don't quite understand what you're gloating about?

You spent several days and 20+ posts arguing against people suggesting that it wasn't absolutely determined that Alabama "WOULD" beat Oregon no matter what.

And I stand by that statement. Alabama can beat anybody, anywhere. Including and especially Oregon.

Not the argument you made. In fact, you're now trying to change your position to the one that the Oregon fans disagreeing with you held:

You = "Alabama would beat Oregon no matter what."
Several Others = "I don't know for certain. But, Oregon could win that game."

Losing a game doesn't automatically render you a second rate team.

Thank you for agreeing with one of my supporting arguments for Sagarin's ratings. But, then, if you accept this, why do you find it so outrageous that a one loss Stanford is rated above an undefeated team?

They played three ranked teams in a row, and in their final week of that stretch, they were pitted against South Carolina(who had an extra week to prepare), and they dropped a game. It happens.

So, you're now admitting that Oregon could beat Alabama? Which is it?

You would spend a bunch of time arguing with people that Oregon couldn't win that game. But, then if they did, you would just make up an excuse and write "it happens"?

Oh, ok.

Hey Wheaton, your rankings suck, and your rationalizations for those rankings are a joke.

You obviously have not read and learned as I suggested. This is still the same logical fallacy.

Alabama lost Wheaton. They'd still beat Oregon, probably in a similar fashion to how Ohio State did.

Right. Right. But, if someone would have told you when this came up that South Carolina could beat Alabama, you would have agreed. Even though you spent days arguing that Oregon couldn't. Right Bumi? That's what you want us to believe?

How would you know? The discussion didn't even occur here. :ermm:

Wow. Are you really denying that you wrote that?

You are mistaken. Saying USC is #1 is calling them national champ, just as the AP, BCS, Berryman, Billingsley, Colley, DeVold, Eck, FACT, FB News, FW, Massey, NFF, NY Times, Seattle Times, Sporting News, USA/ESPN, and Wolfe called Ohio State national champs. The NCAA recognizes them as national champions.

No it isn't. I could make my own rankings and put whoever I want at #1. That wouldn't make that team the NC. There are defined rules that determine who gets to claim NC status. The rule is not "whoever Sagarin rates as #1 at the end of the season."

The goal of Sagarin's ratings is not to come up with a formula that always matches the BCS formula at the end of the year. It rates football team performance based on specific criteria.

As I pointed out before, the best team doesn't beat an inferior opponent every time.

If you do not think the system is good enough to define the best team in the country after all of the games have been played, why do you believe it is good enough to define the teams that should be playing for the national championship?

Good question.

The best way to determine outcomes in football is for teams to play each other. Since it's not feasible to have a giant play-off with every team involved, the next best thing is to have an objective, merit-based rating system to determine who should play in the limited post-season match-ups that we have.

Because he's a homer, and those terrible Sagarin ratings support his opinions on the superiority of the Pac-10. :yes:

I've supported the Sagarin ratings for years, regardless of which team or conference they gave credit to.

That's one of the significant things about a rating system. It's objective rather than subjective like human polls.

 

[FREE EARL]

You spent several days and 20+ posts arguing against people suggesting that it wasn't absolutely determined that Alabama "WOULD" beat Oregon no matter what.

And I still think Bama would(can, will, etc. Whatever verb you wish to use) beat Oregon; Today, tomorrow, and in two weeks. Losing to South Carolina doesn't change that. And if and when Oregon loses, which is given their history, is highly probable, it'd completely asinine for me to scramble in here with faulty "See, I told you so" logic.

Not the argument you made. In fact, you're now trying to change your position to the one that the Oregon fans disagreeing with you held:

You = "Alabama would beat Oregon no matter what."
Several Others = "I don't know for certain. But, Oregon could win that game."

I honestly don't recall which verb I used, and given the greater picture, it's irrelevant. I still think Alabama can and would beat Oregon. Does that help? I've eliminated any fault you found with semantics.

Thank you for agreeing with one of my supporting arguments for Sagarin's ratings. But, then, if you accept this, why do you find it so outrageous that a one loss Stanford is rated above an undefeated team

Whoa... now that was bit of a leap now wasn't it? If anyone were ever in need of an example of a logical fallacy, you're opening sentence would be perfectly suitable.

That aside, a few people have already explained why having Stanford at #2 is laughably ridiculous. They haven't beaten anyone first and foremost, and secondly, a number of teams have performed better(that includes 1 loss teams) under tougher schedules.

To further illustrate this point; Stanford has played a single ranked team thus far this season. They lost by 21.

And clearly, based on pretty much every single poll with the exception of Sagarin's ratings, my opinion is spot on. As no other credible poll has Stanford in the top 10.

So, you're now admitting that Oregon could beat Alabama? Which is it?

The world could end tomorrow. You could stumble upon a billion dollars. I could be struck by lightning. Milo could one day end up the GM of the Blazers. You could realize that employing the "could" argument, is elementary and faulty, as I've so clearly demonstrated.

Yes, Oregon could beat Alabama. Would they? I don't think so. In fact, it "could" get brutal in favor of Alabama given both teams strengths and weaknesses.

You would spend a bunch of time arguing with people that Oregon couldn't win that game. But, then if they did, you would just make up an excuse and write "it happens"?

I'd admit I was wrong, and that Oregon surprised me. :ermm:

You obviously have not read and learned as I suggested. This is still the same logical fallacy.

I'm not even going to get into what is and isn't a logical fallacy again. I do find your opening sentence cute though.

Anyhow, no amount of "learning" is going to convince me that those rankings are any less of a joke. And I'm not the only one that pointed out how terrible they are. Even your little buddy from Tennessee considers having Stanford at #2 laughter inducing.

Right. Right. But, if someone would have told you when this came up that South Carolina could beat Alabama, you would have agreed. Even though you spent days arguing that Oregon couldn't. Right Bumi? That's what you want us to believe?

If someone would've told me before it happened(south Carolina beating Alabama), I would have argued in favor of Alabama.

By the way, I don't want "us" to believe anything. I'm having a discussion with you. I understand groupthink is a mechanism of security, but It's unneeded here.

I've supported the Sagarin ratings for years, regardless of which team or conference they gave credit to.

That's one of the significant things about a rating system. It's objective rather than subjective like human polls.

Really? Because I still think it's because you're a homer.

 

[wheaton4prez]

And I still think Bama would(can, will, etc. Whatever verb you wish to use) beat Oregon; Today, tomorrow, and in two weeks. Losing to South Carolina doesn't change that. And if and when Oregon loses, which is given their history, is highly probable, it'd completely asinine for me to scramble in here with faulty "See, I told you so" logic.

Bumi. There is a significant difference between saying a team "can" beat a team and that it "will" beat a team. You need to pick the word and meaning you want. That's not my job.

Of course it would be asinine for you to come in here with that logic. But, that's only because nobody has claimed that Oregon "will" win every game.

You, on the other hand, claimed that Alabama would beat Oregon no matter what. Had someone brought up South Carolina, you would have written the same thing.

I honestly don't recall which verb I used, and given the greater picture, it's irrelevant. I still think Alabama can and would beat Oregon. Does that help? I've eliminated any fault you found with semantics.

It doesn't really help. It leads to many more questions:

Is your position that you think Alabama would beat every team except for the ones that they lose to?

Do you think that Oregon should be below Alabama in the BCS ranking?

Do you think that South Carolina should be ranked above Oregon?

Should South Carolina be ranked above Alabama?

Whoa... now that was bit of a leap now wasn't it? If anyone were ever in need of an example of a logical fallacy, you're opening sentence would be perfectly suitable.

Which logical fallacy was it, specifically?

That aside, a few people have already explained why having Stanford at #2 is laughably ridiculous. They haven't beaten anyone first and foremost, and secondly, a number of teams have performed better(that includes 1 loss teams) under tougher schedules.

They have performed better relative to their schedule. At least based on the criteria of margin of victory. That's why Sagarin's rating puts them higher.

To further illustrate this point; Stanford has played a single ranked team thus far this season. They lost by 21.

Yes. They lost to the #1 rated team in the country that has the widest average margin of victory of any other team. The #2 team can lose to the #1 team and still be the #2 team. That's why one of them is #1 and the other #2, we expect #1 to beat them. That doesn't mean that they would most likely lose to #3 or anyone below.

And clearly, based on pretty much every single poll with the exception of Sagarin's ratings, my opinion is spot on. As no other credible poll has Stanford in the top 10.

You probably should have researched that one a little first. Of the BCS ratings:

Colley has them at 10.
Massey has them at 4.
Wolfe has them at 9.
Sagarin has them at 5 (in his BCS-friendly column).

Four of the six used in the BCS have Stanford in the top 10.

Yes, Oregon could beat Alabama. Would they? I don't think so. In fact, it "could" get brutal in favor of Alabama given both teams strengths and weaknesses.

Thank you for coming around. Hopefully, it won't take weeks to happen next time.

I'm not even going to get into what is and isn't a logical fallacy again. I do find your opening sentence cute though.

I probably wouldn't want to get into that with me either, if I were you.

Anyhow, no amount of "learning" is going to convince me that those rankings are any less of a joke. And I'm not the only one that pointed out how terrible they are. Even your little buddy from Tennessee considers having Stanford at #2 laughter inducing.

I'm not suggesting that learning will lead to you agreeing with me about what makes the best rating system. But, it will help you avoid making logical fallacies and make our conversation more constructive.

I believe that the reason adhering to an objective rating to rank teams is shocking to some people is that we are so used to sports outcomes being determined by play-offs. It doesn't seem "right" to most people that a team can lose and still be ranked higher than a team that hasn't lost because that never happens in the traditional play-off bracket we are all familiarized with since 1st grade soccer.

However, if you take the time to study the way the different types of ratings work and are familiar with how people have voted in the human BCS polls, you will find that the margin of victory rating that Sagarin uses is a more accurate and unbiased depiction of what a team has accomplished relative to their schedule than any of the other sources of ranking that the BCS uses. Short of head-to-head match-ups, it's the next best thing.

If someone would've told me before it happened(south Carolina beating Alabama), I would have argued in favor of Alabama.

Exactly. Proving that you really don't know who "would" win and are just wasting peoples time insisting that they are wrong to think that any other team has a chance.

South Carolina beating Alabama proves that your arguments on this subject were a waste of time and did not demonstrate the sports knowledge that you repeatedly claim as greater than others.

 

[spudvol]

Begun, this Quote War has.

 

[FREE EARL]

Wheat, I chose to start a new post entirely in which I'll address your affinity for Sagarins ratings, and why using those ratings as a base of reference is fallacious, fallible, and nonsensical.

To begin, your poll is based off a flawed system. Which, as Sagarin himself points out, is the “politically correct” version, which he claims is “less accurate” than another version he calculates. You've argued in favor of the credibility of his ratings due to the basis behind them being sound. I think you fail to understand that the only thing that makes a ratings system meaningful is what comes out of it. What's it matter if the logic behind the formula is sound if the results have no credibility? For example, not even two weeks ago Sagarins ratings were as follows.

1. Boise St
2. TCU
3. Oklahoma
4. Oregon
5. Michigan St
6. Oregon St
7. Stanford
8. LSU
9. Auburn
10. Nebraska
11. Arizona
12. Nevada
13. Air Force
14. Missouri
15. South Carolina
16. California
17. Alabama
18. Delaware(AA)
19. Florida St
20. Michigan
21. Kansas St
22. Wisconsin
23. Ohio St
24. Virginia Tech
25. James Madison(AA)

And his predictor ratings were equally as baffling.

1. Oregon
2. Alabama
3. Stanford
4. Florida St
5. TCU
6. California
7. Boise St
8. Ohio St
9. Nebraska
10. Arizona
11. LSU
12. South Carolina
13. Arizona St
14. Missouri
15. Arkansas
16. Miami
17. Iowa
18. Oregon St
19. Nevada
20. Virginia Tech
21. USC
22. Florida
23. NC State
24. Texas A&M
25. Auburn

Of course, as you see, there are a number of issues.

1) Boise and TCU were ranked as the top two teams in the nation.

2) Why is a two loss Oregon St. team ranked #6, above both Alabama and Ohio State?

3) Boise, TCU, Nevada, Air Force are all in the top 13. Why are Nevada and Air Force ranked above Alabama, Ohio State, Arkansas(who isn't ranked), Wisconsin, South Carolina, etc?

4) Why are James Madison and Delaware ranked at all, let alone as high as 18? Delaware being ranked a single slot below Alabama.

And then of course, there's the clear bias his "formula" possesses for Pac-10 teams. Their SOS etches them the #1 spot in conference ratings, and they'll likely maintain that position until bowl season. Very much like they did last season prior to going 2-5 in bowl games. He gives the PAC-10 far too much credit for scheduling "tough" OOC games, neglecting the fact that their handicapped into doing so because the PAC-10 is/was limited to 10 teams. Furthermore, he doesn't put enough emphasis on the importance of losing those said games. Common sense tells you that scheduling "tough" OOC games is irrelevant if you're losing those games. Sagarins ratings of course defy logic and tell us otherwise. With 2 losses, Oregon State is at #6. Obviously, because the computers reflect it, they are far better than Ohio State and Alabama. I mean, it says so right there in Sagarins ratings. Also, Oklahoma, despite 5 losses in 2009, was ranked #9 (ahead of BCS bowl winners Boise State, Ohio State, and Iowa) in MPT's final regular season ranking. However, neither the AP nor the BCS had Oklahoma in the top 25.

In 2006 for example, this was Sagarins SOS ratings...

http://www.usatoday.com/sports/sagarin/fbt06.htm
1. Stanford
2. USC
3. Arizona
4. Tenn
5. Wash
6. Oregon
7. Ucla
8. UF
9. Cal
10. WSU
11. ASU
15. Oregon State

All 10 Pac teams in the top 15 of SOS's? Really?

Furthermore, Oregon, TCU, and Boise State all have SOS's ranked over 70 by non Sagarin polls. His ratings actually reward teams with weak SOS's.

In addition to making a mockery out of college football, Sagarin and his BCS formulating colleagues have offended the world of mathematics. Mathematicians have actually called for the boycott of the BCS and the formulas that comprise it due to the potential harm they cause to the credibility of mathematics. See, Hal stern.

But wait, the comical results spat out by Sagarin's rating don't stop there, his inane influences extend to College Basketball.

Last year, Sagarin released his list of "All Time Basketball Schools". It came out a little like this.

1. Kentucky
2. UCLA
3. Kansas
4. North Carolina
5. Indiana
6. Illinois
7. Duke
8. Purdue
9. Ohio State
10. Iowa
11. Louisville
12. Notre Dame
13. Michigan
14. Minnesota
15. Michigan State
16. St. John’s
17. Cincinnati
18. Oklahoma State
19. Utah
20. Oklahoma
21. Villanova
22. NC State
23. Syracuse
24. Marquette
25. Southern California

Now, if you aren't a college basketball fan, this may not seem as ludicrous to you as it has to some.

The first and most blatant issue is that he has Illinois above Duke despite Duke being significantly more accomplished as a program.

Duke
Players drafted: 74
Conf. Championships: 23
Total Wins: 1852
Win pct.: .696
Tournament Wins: 88
Tourney Win pct.: .746
Final Fours: 14
Championship games: 9
Championships: 3

Illinois
Players drafted: 68
Conf. Championships: 17
Total Wins: 1605
Win pct.: .654
Tournament Wins: 38
Tourney Win pct.: .567
Final Fours: 5
Championship games: 1
Championships: 0

By every tangible measure, Duke is a superior program to Illinois. Despite that, Sagarins ratings defy all we know about Basketball and declares Illinois as Dukes superior.

Just to point out a few more head scratchers...

1. There's some screwy BIG X thing going on in that system.

2. Why is MSU ranked below 8 other big 10 schools? They're the second most accomplished program in the conference.

3. Why is MU at 29# and UF at #64?

etc, etc, etc

I could go on and on about the fallaciousness and incompetence of Sagarins ratings, but I'm positive that anyone who reads the above post will come to a similar conclusion that myself and others have already come to. That Sagarin's ratings are horrid and often times defy logic and reality. To use ratings that draw ludicrous conclusions while ignoring common sense through formulas that aren't even fully accesible to the public and are denounced by mathematicians is imbecilic.

And again, your poll sucked.

 

[kptvol]

I have to ask, what was your response to him?

Should have told him we'd put on our cheatin' cleats and whip the hell out of them.

 

[tvolsfan]

I am referring to his unaltered formula, which resulted in wackiness before he had to change it to meet BCS standards. It was the formula used to award a recognized NC before all of them were combined into the NCG system in place now.



They do.

Which is insanity, at best. Any school that feels the need to claim a title awarded to you by a single computer rating system 30 years after the season ended should pretty much drop football.

 

[tvolsfan]

Florida didn't stomp a team at Florida St.'s caliber, unlike Oklahoma. I never even said that Oklahoma was some juggernaut. They're a strong team, but I don't think they're one of the best 5 teams in the country.

You call beating teams 56-10, 26-3, and 30-9 struggling in the early going? Florida was a force in 2008 from the start of the season, sans a poor effort against Ole Miss. I personally witnessed them dominate Tennessee in every facet of the game in 2008.

Florida looked much, much better after the Ole Miss game. And even in the the Tennessee game, they had to make 2 goal line stands to keep UT from adding two touchdowns. They weren't playing nearly as well as they ended up playing post Ole Miss/tearful speech.

 

[FREE EARL]

Florida looked much, much better after the Ole Miss game. And even in the the Tennessee game, they had to make 2 goal line stands to keep UT from adding two touchdowns. They weren't playing nearly as they ended up playing post Ole Miss/tearful speech.

He should know this. He supposedly watched the Gators "dominate" in person.

 

[FREE EARL]

Which is insanity, at best. Any school that feels the need to claim a title awarded to you by a single computer rating system 30 years after the season ended should pretty much drop football.

I can't get over how much of a joke his ratings are. Below are last weeks.

1- Oregon
2- TCU
3- Boise ST.
4- Stanford
5- LSU
6- Nebraska
7- Arizona
8- Alabama
9- Callifornia
10- Florida st.
11- Oregon st.
12- Oklahoma
13- South Carolina
14- Nevada
15- Ohio st.
16- Missouri
17- Auburn
18- Michigan st.
19- Va-Tech
20- Arizona st.
21- Iowa
22- NC st.
23- Utah
24- Arkansas
25- Wisconsin

1. 3 Pac-10 teams with losses, two of which had multiple losses, are ranked above no loss Oklahoma, Ohio State, and Michigan State squads.

2. 3-3 ASU is ranked 20th, above both Arkansas Wisconsin(who beat ASU), and Iowa.

3. Why are Arkansas and Wisconsin ranked so low?

4. Why is a two loss Va Tech squad ranked at 19, above Arkansas and Wisconsin?

It's flat out mind boggling. Sagarins formula is a flat out joke.

 

[cotton]

The rule is not "whoever Sagarin rates as #1 at the end of the season."

Actually, it was, at least until the NCAA started pretending that there was no 1-A national champion prior to the BCS. However, that is beside the point. The #1 team selected by Sagarin at the end of any season in which that system gave a silly result simply serves to show why that formula doesn't work.

The best way to determine outcomes in football is for teams to play each other. Since it's not feasible to have a giant play-off with every team involved, the next best thing is to have an objective, merit-based rating system to determine who should play in the limited post-season match-ups that we have.

The objective, merit-based rating system you are promoting sucks. See, again, the objective, merit based results that wrongly identified the most deserving #1 team at the end of the season from 1978-present, or retroactively from 44 BC-1977.

You want objectivity? I'll objectively take the merit-based, objective A&H results, which list LSU at #1. Or maybe I'll take Massey or Colley, who have Oklahoma. Then again, Billingsley has LSU, but Wolfe has Oklahoma. Objectively, scientifically, and mathematically, without any human bias, Oregon is either 11th, or 2nd, or 9th, or 7th.

Last season, those same objective machines would have put Bama and Cincy in the championship game, but that wasn't unanimous--two had Bama/Texas, two had Bama/Cincy, and two, including Sagarin, had a Bama/Florida rematch. If the season ended today, Oklahoma, LSU, Auburn, Boise, Missouri, and Oregon would make for a very crowded championship game.

You are putting too much faith in an algebraic equation to choose participants in the NCG. That method, Sagarin included, gives bad results too often.

 

[cotton]

Which is insanity, at best. Any school that feels the need to claim a title awarded to you by a single computer rating system 30 years after the season ended should pretty much drop football.

I have never accused any UK fan in any sport of being completely sane.

 

[milohimself]

Odd that many statisticians say that an objective poll at the end of the season (untethered by the ridiculous constraints of the BCS, and any other entity, that is) comprised only of computers is the best way to determine who ought to play in a national championship game.

The fact that people complain at all about computer rankings in the middle of the season baffles me. They are running off a very limited set of data, and are explicitly meant only to be correct at the end of the season.

For the purpose of trying to select the two best teams at the end of the season, the set up of the BCS is doing a pretty good job. The actual premise of trying to choose the best two at the end of the season is flawed, but the idea of mixing statistical objectivity with human polls to pick two teams is not a bad one.

Of course, the execution sucks, as well. The coaches don't fill out the polls themselves and we all know it. Look at who submits ballots for the Harris Poll, it's a total joke. And again, the BCS places ridiculous constraints on the computers.

 

[FREE EARL]

Odd that many statisticians say that an objective poll at the end of the season (untethered by the ridiculous constraints of the BCS, and any other entity, that is) comprised only of computers is the best way to determine who ought to play in a national championship game.

The fact that people complain at all about computer rankings in the middle of the season baffles me. They are running off a very limited set of data, and are explicitly meant only to be correct at the end of the season.

For the purpose of trying to select the two best teams at the end of the season, the set up of the BCS is doing a pretty good job. The actual premise of trying to choose the best two at the end of the season is flawed, but the idea of mixing statistical objectivity with human polls to pick two teams is not a bad one.

Of course, the execution sucks, as well. The coaches don't fill out the polls themselves and we all know it. Look at who submits ballots for the Harris Poll, it's a total joke. And again, the BCS places ridiculous constraints on the computers.

'Death to the BCS': Nonsense rules - College Football - Rivals.com

In my opinion, there doesn't need to be any computers involved at all. At this point, I'd prefer the scrap this system and start over from scratch.

 

[droski]

a plus one solves any rational argument.

 

[tvolsfan]

a plus one solves any rational argument.

I'd love a plus 1. I'd prefer what we have now to an 8 team playoff.

 

[droski]

I'd love a plus 1. I'd prefer what we have now to an 8 team playoff.

agreed. anything more than a plus 1 and then you ruin the regular season since a team will no doubt be guaranteed a berth if it wins it's conference and the conference could easily be won before the end of the regular season.

 

[droski]

I can't get over how much of a joke his ratings are. Below are last weeks.

1- Oregon
2- TCU
3- Boise ST.
4- Stanford
5- LSU
6- Nebraska
7- Arizona
8- Alabama
9- Callifornia
10- Florida st.
11- Oregon st.
12- Oklahoma
13- South Carolina
14- Nevada
15- Ohio st.
16- Missouri
17- Auburn
18- Michigan st.
19- Va-Tech
20- Arizona st.
21- Iowa
22- NC st.
23- Utah
24- Arkansas
25- Wisconsin

1. 3 Pac-10 teams with losses, two of which had multiple losses, are ranked above no loss Oklahoma, Ohio State, and Michigan State squads.

2. 3-3 ASU is ranked 20th, above both Arkansas Wisconsin(who beat ASU), and Iowa.

3. Why are Arkansas and Wisconsin ranked so low?

4. Why is a two loss Va Tech squad ranked at 19, above Arkansas and Wisconsin?

It's flat out mind boggling. Sagarins formula is a flat out joke.

yup

computers can't factor in teams who regurally crap their pants in big games. see #9

 

[wheaton4prez]

Wheat, I chose to start a new post entirely because I have no valid answer for your other points.

fyp

To begin, your poll is based off a flawed system. Which, as Sagarin himself points out, is the “politically correct” version, which he claims is “less accurate” than another version he calculates. You've argued in favor of the credibility of his ratings due to the basis behind them being sound. I think you fail to understand that the only thing that makes a ratings system meaningful is what comes out of it. What's it matter if the logic behind the formula is sound if the results have no credibility? For example, not even two weeks ago Sagarins ratings were as follows.

To begin, you have completely mangled this one.

My rankings were based on Sagarin's Predictor rating. That is not the flawed, "politically correct", less accurate one. The less accurate one that he is referring to is the wins-only, Elo Chess one that the BCS uses.

Get your facts straight if you're going to show up with guns blazing.

And then of course, there's the clear bias his "formula" possesses for Pac-10 teams. Their SOS etches them the #1 spot in conference ratings, and they'll likely maintain that position until bowl season. Very much like they did last season prior to going 2-5 in bowl games. He gives the PAC-10 far too much credit for scheduling "tough" OOC games, neglecting the fact that their handicapped into doing so because the PAC-10 is/was limited to 10 teams. Furthermore, he doesn't put enough emphasis on the importance of losing those said games. Common sense tells you that scheduling "tough" OOC games is irrelevant if you're losing those games. Sagarins ratings of course defy logic and tell us otherwise. With 2 losses, Oregon State is at #6. Obviously, because the computers reflect it, they are far better than Ohio State and Alabama. I mean, it says so right there in Sagarins ratings. Also, Oklahoma, despite 5 losses in 2009, was ranked #9 (ahead of BCS bowl winners Boise State, Ohio State, and Iowa) in MPT's final regular season ranking. However, neither the AP nor the BCS had Oklahoma in the top 25.

The only thing this quote accomplishes is to expose how little you understand about how the ratings work.

The formula was developed years ago and applies to every team exactly the same. You being upset that the Pac-10 is doing well measured by margin of victory is your bias, not the formulas. There is no Pac-10 adjuster and it has shown other conferences doing well in years past.

He doesn't put any emphasis on losing or winning in the Predictor column. It's all based on margin of victory. The logic for that is that losing to a tough team by a few points can be a greater accomplishment than beating a weak team by a few points. The wins only approach doesn't recognize the difference.

Not even sure where you're getting that it puts Oregon State at #6. Are you looking at baseball statistics or something?

In 2006 for example, this was Sagarins SOS ratings...

All 10 Pac teams in the top 15 of SOS's? Really?

Yes. Really. A big reason for that is that USC dominated Arkansas, Nebraska and Notre Dame that year. In addition to the well documented fact that the Pac-10 schedules relatively more difficult OOC games than most other conferences.

Furthermore, Oregon, TCU, and Boise State all have SOS's ranked over 70 by non Sagarin polls. His ratings actually reward teams with weak SOS's.

Rankings in the 50s are not high SOS relative to other D-1 schools. What makes you think that is a reward?

In addition to making a mockery out of college football, Sagarin and his BCS formulating colleagues have offended the world of mathematics. Mathematicians have actually called for the boycott of the BCS and the formulas that comprise it due to the potential harm they cause to the credibility of mathematics. See, Hal stern.

Bumi, I can't understand you with that foot in your mouth.

Jeff Sagarin has a math degree from MIT. He is paid over $100k just to determine with math which free agents to sign and line-ups to use for the Dallas Mavericks.

Sagarin objects to the BCS rule that prohibits the use of margin of victory. He is the opposite of being a mockery in the world of mathematics. He's one of the most visible proponents of what I've been arguing in this thread, that margin of victory ratings are the mathematically correct way to measure team performance. The same thing that the mathematicians argue.

I could go on and on about the fallaciousness and incompetence of Sagarins ratings, but I'm positive that anyone who reads the above post will come to a similar conclusion that myself and others have already come to. That Sagarin's ratings are horrid and often times defy logic and reality. To use ratings that draw ludicrous conclusions while ignoring common sense through formulas that aren't even fully accesible to the public and are denounced by mathematicians is imbecilic.

How you have confidence that anyone takes your bumbling around with a subject you clearly don't understand is beyond me.

Actually, it was, at least until the NCAA started pretending that there was no 1-A national champion prior to the BCS. However, that is beside the point. The #1 team selected by Sagarin at the end of any season in which that system gave a silly result simply serves to show why that formula doesn't work.

If they were going to use a rating rather than a champ game, Sagarin's Predictor would be the best choice. These days, though, the BCS uses their own system to determine the championship team. It's not Sagarin's job to come up with a formula that predicts what other peoples ratings and polls are also going to predict.

He has made a mathematically sound, objective, merit-based rating system that uses margin of victory to rate teams. The way to be rated better in his system is to score more points and get scored on less against better teams. It doesn't get any more football based than that.

The objective, merit-based rating system you are promoting sucks. See, again, the objective, merit based results that wrongly identified the most deserving #1 team at the end of the season from 1978-present, or retroactively from 44 BC-1977.

It's a rating. It's not a selection. It doesn't determine who is most "deserving", it determines who had the best performance on average over the season.

As I've pointed out a couple of times already, nobody argues with the result of the super bowl. Head to head play-offs are the ideal. However, most people will agree that the team that had the best over-all performance on average is not always the super-bowl winner. Some teams dominate all season but drop the big game. Some scrape into the play-offs and win the super-bowl. "Deserving" and "highly rated" are two different things.

You want objectivity? I'll objectively take the merit-based, objective A&H results, which list LSU at #1. Or maybe I'll take Massey or Colley, who have Oklahoma. Then again, Billingsley has LSU, but Wolfe has Oklahoma. Objectively, scientifically, and mathematically, without any human bias, Oregon is either 11th, or 2nd, or 9th, or 7th.

Those are wins-only ratings which are inherently flawed. Beating Alabama by 30 points is a greater accomplishment than beating them by 1.

That said, this isn't about allegiance to a school. If Oregon was #9 in Sagarin's Predictor I would still be for using that rating to seed bowl games.

Last season, those same objective machines would have put Bama and Cincy in the championship game, but that wasn't unanimous--two had Bama/Texas, two had Bama/Cincy, and two, including Sagarin, had a Bama/Florida rematch. If the season ended today, Oklahoma, LSU, Auburn, Boise, Missouri, and Oregon would make for a very crowded championship game.

You're not seeing the distinctions between the formulas. I am against the wins-only formulas that the BCS uses. Sagarin's Predictor rating, that I am for, had Alabama playing Florida last year.

You are putting too much faith in an algebraic equation to choose participants in the NCG. That method, Sagarin included, gives bad results too often.

I think that the rules for seeding bowl games should be as unbiased and concrete as the rules of football.

It's not a matter of faith. It's a matter of understanding how the different formulas work and which ones accurately measure accomplishment in football terms.

Odd that many statisticians say that an objective poll at the end of the season (untethered by the ridiculous constraints of the BCS, and any other entity, that is) comprised only of computers is the best way to determine who ought to play in a national championship game.

The fact that people complain at all about computer rankings in the middle of the season baffles me. They are running off a very limited set of data, and are explicitly meant only to be correct at the end of the season.

For the purpose of trying to select the two best teams at the end of the season, the set up of the BCS is doing a pretty good job. The actual premise of trying to choose the best two at the end of the season is flawed, but the idea of mixing statistical objectivity with human polls to pick two teams is not a bad one.

Of course, the execution sucks, as well. The coaches don't fill out the polls themselves and we all know it. Look at who submits ballots for the Harris Poll, it's a total joke. And again, the BCS places ridiculous constraints on the computers.

You make many great points.

I think that including the human voters isn't necessarily a bad idea in theory. However, it's impractical. In addition to being prone to bias and incompetence, it's not feasible that any voter can sufficiently watch and analyze all of the games they are voting on each week.

 

[FREE EARL]

fyp

Yep, that's it Pete. :thumbsup:

To begin, you have completely mangled this one.

My rankings were based on Sagarin's Predictor rating. That is not the flawed, "politically correct", less accurate one. The less accurate one that he is referring to is the wins-only, Elo Chess one that the BCS uses.

Get your facts straight if you're going to show up with guns blazing.

I addressed his predictor ratings Petty, they're equally as ridiculous as his weekly ratings. You, of course, sidestepped the examples or how porous his predictor ratings have already proven to be.

The only thing this quote accomplishes is to expose how little you understand about how the ratings work.

The formula was developed years ago and applies to every team exactly the same. You being upset that the Pac-10 is doing well measured by margin of victory is your bias, not the formulas. There is no Pac-10 adjuster and it has shown other conferences doing well in years past.

I know when his system was developed. It was first published in 1985, and they've consistently sucked since then.

I've already explained how his system caters to the Pac-10. You dodged the point.

He gives the PAC-10 far too much credit for scheduling "tough" OOC games, neglecting the fact that their handicapped into doing so because the PAC-10 is/was limited to 10 teams. Furthermore, he doesn't put enough emphasis on the importance of losing those said games. Common sense tells you that scheduling "tough" OOC games is irrelevant if you're losing those games.

He doesn't put any emphasis on losing or winning in the Predictor column. It's all based on margin of victory. The logic for that is that losing to a tough team by a few points can be a greater accomplishment than beating a weak team by a few points. The wins only approach doesn't recognize the difference.

Which is exactly why it's a joke. How can any system that doesn't account for wins and losses be taken serious?

His system doesn't accurately reflect the reality of what's going on in college football.(Cal #6, ASU #13, etc)

Not even sure where you're getting that it puts Oregon State at #6. Are you looking at baseball statistics or something?

Oregon State was #6 two weeks ago. But lets not pretend as if OSU is the only team who's raking makes absolutely no sense.

Yes. Really. A big reason for that is that USC dominated Arkansas, Nebraska and Notre Dame that year. In addition to the well documented fact that the Pac-10 schedules relatively more difficult OOC games than most other conferences.

So the dominance of a single team justifies completely overrating an entire conference?

No amount of rationalizations is going to make ranking an entire conference in the top 15 of SOS's any less of a joke.

Bumi, I can't understand you with that foot in your mouth.

Then it's a blessing we aren't speaking. Your eyes and hands are all that's necessary. :good!:

Jeff Sagarin has a math degree from MIT. He is paid over $100k just to determine with math which free agents to sign and line-ups to use for the Dallas Mavericks.

Sagarin objects to the BCS rule that prohibits the use of margin of victory. He is the opposite of being a mockery in the world of mathematics. He's one of the most visible proponents of what I've been arguing in this thread, that margin of victory ratings are the mathematically correct way to measure team performance. The same thing that the mathematicians argue.

Reading comprehension Pete. I clearly said his ratings making a mockery of college football, not of mathematics.

And I personally couldn't care less where he graduated from. Being a nerd doesn't qualify someone to rank collegiate athletics. Some of his results are simply inexplicable(His All time basketball schools.) and his formula should be banned from being involved in any system that determines rankings due to the ridiculous results it has and does produce.

As I said, you fail to understand that the only thing that makes a ratings system meaningful is what comes out of it. Regardless of the basis behind it, the system itself is meaningless if the results have no credibility.

How you have confidence that anyone takes your bumbling around with a subject you clearly don't understand is beyond me.

No one fully understands his system as it's partially kept from the public; And frankly, comprehension isn't necessary to know the results his mystery formula produce don't accurately reflect what we're seeing on saturdays.

You may be the only fan of college football(are you a fan?) who uses Sagarin ratings as a basis to fall back on.

 

[milohimself]

this thread =

 

[grimvol]

I got a question..we all know TCU and Utah are in the top 10...and TCU and Utah have to play each other.,,my question is..let's say Utah wins and finishes the season undefeated...Bama loses another game as does THE OSU and LSU..and so on..could this be the year we have 3 non BCS schools in BCS games? Boise goes undefeated...Utah does the same and TCU with a one loss season...and what are the odds of the BCS pitting 2 of those teams together in a bowl just to shut up people...because we ALL know that was what happened last year with the Boise/TCU game...the BCS did not want to risk the chance of 2 BCS schools getting beat by non BCS schools so they put them in a game together...
It might get interesting if 3 non BCS schools get in this year.

 

[cotton]

a plus one solves any rational argument.

Not really. All it does is make the margin of error for correctly selecting the teams to play for the NC seem bigger.

this thread =

Word. WTH is up with the mega posts?

 

[FREE EARL]

Word. WTH is up with the mega posts?

I don't know. I usually hate getting into drawn out posting wars. It's exhausting.