BCS Top 10

[wheaton4prez]

It's a simple copy and paste. Doesn't take much effort, no need for self flattery.

You're missing the point. Why do you care enough to make any effort toward copying and pasting my comments between forums?

It's stalkerish.

And Alabama would likely kill them. Those rankings are crap.

Based on human opinion. The statistics get it right more often than humans do.

Same reason why index funds tend to outperform managed funds over-all, in the long-run.

The "who would beat who" game doesn't work. The only method that works properly for ranking is comparing the accomplishments a team has had in that particular season. Who has beat who and who would beat who follow respectively in importance.
Posted via VolNation Mobile

That's true and that is what Sagarin's points ranking is based on. Which teams scored the most points subtracting the points that were scored against them, on average.

The "who would beat who" aspect is only useful in later looking at how accurate the rankings were. It's the proof of the equation and Sagarin's points ranking pencils out as more accurate than the others.

Why are you so offended? :ermm:

It was a sincere question.

and stanford has beaten whom exactly?

That ranking is based purely on margin of victory. So, it's not based on "who they beat". "Who they beat" is a very limited view in determining football performance because it depends on scheduling. Using margin of victory, you can look at how a team beat their opponents. Performance can then be measured against any schedule.

Agreed. If I was going to take one team against any other team, it would be Bama. They shouldn't be number one though.

So, which should the rankings be based on? How good a team is on the field or where they deserve to be based on wins?

This where the objective aspect of Sagarin's ranking makes sense to me. It goes by the most basic measure in football: points. 6 points is 6 points of credit. Nobody gets to decide later how much each touchdown should be worth.

His poll sucks. Stanford above a no loss Auburn team with a superior schedule? Hell, Stanford at #2? Those are "smh" rankings.

It's not a competition of schedules. It's a competition of football. And, according to their performances against their schedules, Stanford has performed better than Auburn.

 

[FREE EARL]

If we're going off of most talented team...

1. Alabama
2. Ohio State
3. Oklahoma
4. LSU
5. Florida
6. Miami(FL)


After that, couldn't really decide

Ohio State isn't that talented.

Oklahoma should be behind Miami and Florida.

 

[DuckinTN]

Sagarin's rankings are usually very accurate, but having Stanford at #2 is admittedly laughable.

 

[wheaton4prez]

Sagarin's rankings are usually very accurate, but having Stanford at #2 is admittedly laughable.

It's a very limited set of data, getting more accurate as the season moves along and new games are factored in.

However, to try and "correct" it with human opinion about what will happen in the future is to disregard what teams have already accomplished.

In other words, if going by performance on the field produces "laughable" results, they are still fair results based on the same types of rules we have in football. Touchdowns don't get called back because the team that scored them doesn't appear powerful enough to most people.

 

[tvolsfan]

If we're going off of most talented team...

1. Alabama
2. Ohio State
3. Oklahoma
4. LSU
5. Florida
6. Miami(FL)


After that, couldn't really decide

OSU doesn't have nearly the talent of a team like Florida. OSU's talented, but the best thing they have going for them is an experienced defense.

 

[cotton]

Based on human opinion. The statistics get it right more often than humans do.

Sagarin's rankings are usually very accurate, but having Stanford at #2 is admittedly laughable.

You Ducks are going to have to explain this one a bit. The computers, and Sagarin in particular, often come to conclusions very different than the consensus.

A sampling:

In 2002, 14-0 Ohio State was a near unanimous NC; Sagarin named 2 loss USC.

In 1998, Tennessee swept every single poll except one; Sagarin named Ohio St.

In '92, he alone named Florida St over Bama.

In '90, he pegged Miami over Colorado or Ga Tech.

In '89, he picked ND over Miami; '87 FSU over Miami; '85 Michigan over Oklahoma.

If you really want to see some weirdosity, check out the national championships he retroactively awarded before his abacus was invented in the early '80s.

 

[KingNick865]

There are Ohio State fans who still ***** and moan that the Buckeyes would have beaten Tennessee by a comfortable margin.

That just reminded me of that.

 

[FREE EARL]

You're missing the point. Why do you care enough to make any effort toward copying and pasting my comments between forums?

It's stalkerish.



Based on human opinion. The statistics get it right more often than humans do.

Same reason why index funds tend to outperform managed funds over-all, in the long-run.



That's true and that is what Sagarin's points ranking is based on. Which teams scored the most points subtracting the points that were scored against them, on average.

The "who would beat who" aspect is only useful in later looking at how accurate the rankings were. It's the proof of the equation and Sagarin's points ranking pencils out as more accurate than the others.



It was a sincere question.



That ranking is based purely on margin of victory. So, it's not based on "who they beat". "Who they beat" is a very limited view in determining football performance because it depends on scheduling. Using margin of victory, you can look at how a team beat their opponents. Performance can then be measured against any schedule.



So, which should the rankings be based on? How good a team is on the field or where they deserve to be based on wins?

This where the objective aspect of Sagarin's ranking makes sense to me. It goes by the most basic measure in football: points. 6 points is 6 points of credit. Nobody gets to decide later how much each touchdown should be worth.



It's not a competition of schedules. It's a competition of football. And, according to their performances against their schedules, Stanford has performed better than Auburn.

This entire post is a joke. :yes:

 

[milohimself]

Are you using the ranking sagarin submits to the bcs? All of the computer rankings designers (except billingsley) will tell you that their bcs submission is far from the best rankings they can generate.
Posted via VolNation Mobile

 

[KingNick865]

Are you using the ranking sagarin submits to the bcs? All of the computer rankings designers (except billingsley) will tell you that their bcs submission is far from the best rankings they can generate.
Posted via VolNation Mobile

This is pretty much correct.

A few of the guys who compute these ratings also have another rating system that includes margin of victory.

 

[tvolsfan]

If you really want to see some weirdosity, check out the national championships he retroactively awarded before his abacus was invented in the early '80s.

I believe UK actually claims one in 1950.

 

[wheaton4prez]

You Ducks are going to have to explain this one a bit. The computers, and Sagarin in particular, often come to conclusions very different than the consensus.

A sampling:

In 2002, 14-0 Ohio State was a near unanimous NC; Sagarin named 2 loss USC.

In 1998, Tennessee swept every single poll except one; Sagarin named Ohio St.

In '92, he alone named Florida St over Bama.

In '90, he pegged Miami over Colorado or Ga Tech.

In '89, he picked ND over Miami; '87 FSU over Miami; '85 Michigan over Oklahoma.

If you really want to see some weirdosity, check out the national championships he retroactively awarded before his abacus was invented in the early '80s.

Which of his rankings are you referring to? He has more than one because the BCS prohibits using margin of victory.

Are you using the ranking sagarin submits to the bcs? All of the computer rankings designers (except billingsley) will tell you that their bcs submission is far from the best rankings they can generate.
Posted via VolNation Mobile

I'm using the one he does not submit to the BCS for that very reason.

Sagarin stands by the same ranking that I posted. He put together the wins-only ranking to submit to the BCS after they made margin of victory off limits.

He also has a hybrid version that averages the wins-only and points-only results together.

 

[DuckinTN]

Are you using the ranking sagarin submits to the bcs? All of the computer rankings designers (except billingsley) will tell you that their bcs submission is far from the best rankings they can generate.
Posted via VolNation Mobile

Sagarin's main complaint is that he isn't permitted to use MOV in his computer rankings. Those rankings that have Stanford at #2 aren't the one used in the BCS.

 

[wheaton4prez]

This entire post is a joke. :yes:

Ad hominem - Wikipedia, the free encyclopedia

Read. Learn. Make valid statements.

 

[wheaton4prez]

You Ducks are going to have to explain this one a bit. The computers, and Sagarin in particular, often come to conclusions very different than the consensus.
...

I tried to find some of his older results that you refer to but could only find back to 98.

The answer is that a rating is not a consensus. His ratings are measurements of what teams accomplished on the field in terms of points scored vs scored against them, averaged and then adjusted based on SOS.

The difference in ratings and consensus about who should be #1 is explained by one team being perceived to be the best by people (for most likely a million different reasons ranging from talent on hand, being undefeated to sheer opinion and bias) and the other simply having a higher margin of victory relative to their SOS than the consensus selection.

One is a rating based on statistics. The other is a selection based on nothing specific.

I think that Sagarin himself would agree that whoever wins the NC game should be considered the NC champ, regardless of their final rating after the game. However, that doesn't change the validity of his rating in terms of accomplishment over the season or that his rating is an objective, merit-based way to seed the NC game.

 

[DuckinTN]

Ad hominem - Wikipedia, the free encyclopedia

Read. Learn. Make valid statements.

I might dare to consider Bumi the King of Logical Fallacies.

 

[wheaton4prez]

He did try to tell us in no uncertain terms that Oregon would have no chance ever to beat Alabama.

They lost to South Carolina a few days later by 14.

 

[DuckinTN]

He did try to tell us in no uncertain terms that Oregon would have no chance ever to beat Alabama.

They lost to South Carolina a few days later.

Better than that, he was convinced that Florida would embarrass Oregon.

 

[cotton]

Which of his rankings are you referring to? He has more than one because the BCS prohibits using margin of victory.

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.

I believe UK actually claims one in 1950.

They do.

 

[cotton]

I tried to find some of his older results that you refer to but could only find back to 98.

The answer is that a rating is not a consensus. His ratings are measurements of what teams accomplished on the field in terms of points scored vs scored against them, averaged and then adjusted based on SOS.

The difference in ratings and consensus about who should be #1 is explained by one team being perceived to be the best by people (for most likely a million different reasons ranging from talent on hand, being undefeated to sheer opinion and bias) and the other simply having a higher margin of victory relative to their SOS than the consensus selection.

One is a rating based on statistics. The other is a selection based on nothing specific.

I think that Sagarin himself would agree that whoever wins the NC game should be considered the NC champ, regardless of their final rating after the game. However, that doesn't change the validity of his rating in terms of accomplishment over the season or that his rating is an objective, merit-based way to seed the NC game.

His ratings are an inarguable, unbiased, statistically proven, mathematical measure of something. The problem is that, in many cases, "something" isn't the most deserving team of a national title, or a particular ranking of a certain team.

Try explaining his '02 national champion, and I believe you will see what I mean.

 

[wheaton4prez]

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.

The BCS standards formula had Utah at #1 in 2008 (the points only had Florida at #1). Wins only is more whacky than points only.

I'm not sure what his formula was before the change in 2001. It may be that it was the points only formula...

His ratings are an inarguable, unbiased, statistically proven, mathematical measure of something. The problem is that, in many cases, "something" isn't the most deserving team of a national title, or a particular ranking of another team.

It's based on the most basic unit of measurement in football: points.

Other than points adjusted to strength of schedule, what is the "something" that you think is a better measurement of accomplishment?

Try explaining his '02 national champion, and I believe you will see what I mean.

He's not titling champions. He's rating the teams. They are two different things.

Consider this, for example. What if there was a rating system that was based on everything you think is correct for measuring who should be the champion. The output amazingly had 4 teams with a dead even score in one season. If team A defeats team B in the NC game (they break the tie with a coin flip) by 1 point, is it not possible that team C is better than team A if they beat team D by 21 points?

In other words, the more convoluted the actual system for selecting NC teams gets, the more possible it becomes that the winner of the game won't be the most highly rated team after all games are played.

 

[FREE EARL]

Better than that, he was convinced that Florida would embarrass Oregon.

I see you found a relief blanket of sorts. Two fans who don't know much about football. This tandem should be fun. :yes:

Anyhow, there's no need for lies. I believed Florida could beat Oregon. This was prior to Florida's losses.

 

[FREE EARL]

He did try to tell us in no uncertain terms that Oregon would have no chance ever to beat Alabama.

They lost to South Carolina a few days later by 14.

And Alabama losing to South Carolina has no relevance on a possible matchup between Oregon and Alabama. South Carolina has a few things Oregon does not. A good defense and a proven coach. Furthermore, the last time the Ducks were pitted against a team of a similar talent level to Alabama, they were dominated in the trenches and back handed by Ohio State.

 

[cotton]

He's not titling champions. He's rating the teams.

1. He was titling champions until the system changed to award only 1 NC to the winner of the NCG;

2. Rating the teams is the same as picking a champion if the highest rated team becomes champion.

All the rest is verbose garbage used to defend a formula which, in its mathematically perfect and unbiased justice, gives screwy results. I chose the '02 example because, with all of the data available after the season, Sagarin's system identified USC as the most deserving team of the #1 ranking and, as a consequence, a national title. That calculation is, in my estimation and that of pretty much every other person, computer, robot, and cyborg on the planet, clearly wrong.

 

[FREE EARL]

Ad hominem - Wikipedia, the free encyclopedia

Read. Learn. Make valid statements.

My post was a statement, and based on your post and presumptions in this thread, I, and obviously a few others, see the validity in it. Your rankings suck, and the fact that you're attempting to rationalize them with a system of ratings that have a history of being completely and utterly off base, is a joke.