B1G vs. SEC

[tschu]

computer rankings take into account margin of victory and opponent strength.

I'm not going to argue with someone who doesn't even understand the basics of how the computer rankings are built.

 

[SandhillshuskerW]

computer rankings take into account margin of victory and opponent strength.

I'm not going to argue with someone who doesn't even understand the basics of how the computer rankings are built.

Where do the computers get the opponent strength? If it's from the human polls and factors in those rankings to get schedule strength, then it's skewed slightly because of human bias in how they seem to rank some teams higher because of which conference they are in.

 

[Creed]

Wow. The point being made is the wording ESecPN uses to describe teams in the SEC and teams that are not. Similar to the Tennessee v Oklahoma/Georgia from earlier in this thread.

I watched the Arkansas - Miss St. game Saturday night. The ESPN announcers kept referring to how you just can't take a week off in the SEC. Keep in mind #1 Miss St was being handled by a pathetic Arkansas team. Instead of saying maybe Miss St isn't ready for prime time yet after struggling with Arkansas and Kentucky, ESPN spun it as a meat-grinder SEC schedule.

Don't get me wrong there are several strong teams in the SEC and until they lose more OOC games, we can all just b!^@h and moan. Also we need to stop saying this is cyclical. I think the SEC is experience the "flywheel concept" (google it; from Jim Collins Good to Great).

• They invested in their programs (facilities, coaches, etc)
• Benefited by geography most African americans in the southeast, favorable weather to attract Midwest kids
• Lax signing rules - oversigning and lax edumacation requirements - we all know education standards are the worst in the south east.

This has led to an ultimate fan interest and getting the best athletes to wear SEC uniforms.

I think SEC teams do a great job at scheduling too. Very rare true away games in hostile environments. They have a knack for their top teams scheduling an historically middle of the road P5 team (Auburn vs K-state, Alabama vs WVU, VT, Mich) which gives them a pretty safe win on paper over a good name team. I will give Tenn and to some extent LSU props for some decent scheduling although most of LSU big OOC have been neutral site games.

 

[tschu]

computer rankings take into account margin of victory and opponent strength.

I'm not going to argue with someone who doesn't even understand the basics of how the computer rankings are built.

You're 50% correct....

Sagarin doesn't use Margin of Victory

Massey only started using Margin of Victory this year...it could be that he doesn't have it nailed down yet...from his FAQ:

The BCS compliant version does not use MOV at all. There is no distinction between a 21-20 nailbiter, and a 63-0 blowout.The main version does consider scoring margin, but its effect is diminished as the game becomes a blowout. The score of each game is translated into a number between 0 and 1. For example 30-29 might give 0.5270, while 45-21 gives 0.9433 and 56-3 gives around 0.9998

The maximum is topped at 1, so the curve flattens out for blowout scores. In addition, I do a Bayesian correction to reward each winner, regardless of the game's score.

The net effect is that there is no incentive to run up the score. However, a "comfortable" margin (say 10 points) is preferred to a narrow margin (say 3 points).

In summary, winning games against quality competition overshadows blowout scores against inferior opponents. Each week, the results from the entire season are re-evaluated based on the latest results. Consistent winners are rewarded, and a blowout score has only marginal effect on a team's rating.

Lol. You're mostly wrong again.

Sagarin had to do that for BCS purposes.

The ranking that Sagarin posts to his site, the good "Predictor" one, uses margin of victory - in fact, it's the main component. He had to make a separate rating in order to appease the BCS. And that ranking without margin of victory was complete and utter garbage.

That's why I look at the Predictor rating. It uses MOV and is the most accurate. The ELO one is garbage.

 

[StPaulHusker]

Isn't margin of victory kind of a bullsh#t factor? We often hear that the score isn't indicative of the game that was played. Matter of fact, we hear it a lot right now with MSU vs NU and MSU vs Oregon.

 

[VA Husker Fan]

computer rankings take into account margin of victory and opponent strength.

I'm not going to argue with someone who doesn't even understand the basics of how the computer rankings are built.

Where do the computers get the opponent strength? If it's from the human polls and factors in those rankings to get schedule strength, then it's skewed slightly because of human bias in how they seem to rank some teams higher because of which conference they are in.

It's not from rankings. It can't be. There are only about 30-35 or so teams that get votes. How would they do anything with the other 90 teams.

 

[VA Husker Fan]

Isn't margin of victory kind of a bullsh#t factor? We often hear that the score isn't indicative of the game that was played. Matter of fact, we hear it a lot right now with MSU vs NU and MSU vs Oregon.

Kind of. There certainly are times where a 1 point game wasn't really close and a 17 point game was in big doubt until late, but over time it's at least some indication. The BCS first required it's computer rating systems to limit the effect to 21 points or less to keep the powers from pouring it on the weak teams, and then eventually told them to take it out completely.

 

[tschu]

Isn't margin of victory kind of a bullsh#t factor? We often hear that the score isn't indicative of the game that was played. Matter of fact, we hear it a lot right now with MSU vs NU and MSU vs Oregon.

A single game by itself, yeah. There's a lot of chaos. Variance is a b!^@h. But it still means basically the most when dealing with all of them at once

That's why models get better and more accurate as the year goes on. More information to work with. Regression to the mean means that over time, overperformances and underperformances will even out.

 

[tschu]

computer rankings take into account margin of victory and opponent strength.

I'm not going to argue with someone who doesn't even understand the basics of how the computer rankings are built.

Where do the computers get the opponent strength? If it's from the human polls and factors in those rankings to get schedule strength, then it's skewed slightly because of human bias in how they seem to rank some teams higher because of which conference they are in.

It's not from rankings. It can't be. There are only about 30-35 or so teams that get votes. How would they do anything with the other 90 teams.

It's not from the human rankings. Long post coming later on how this is done

 

[SandhillshuskerW]

computer rankings take into account margin of victory and opponent strength.

I'm not going to argue with someone who doesn't even understand the basics of how the computer rankings are built.

Where do the computers get the opponent strength? If it's from the human polls and factors in those rankings to get schedule strength, then it's skewed slightly because of human bias in how they seem to rank some teams higher because of which conference they are in.

It's not from rankings. It can't be. There are only about 30-35 or so teams that get votes. How would they do anything with the other 90 teams.

It's not from the human rankings. Long post coming later on how this is done

I would like to see that, I have never really known how they figure strength of schedule.

 

[VA Husker Fan]

It's not from the human rankings. Long post coming later on how this is done

That would probably be worth it's own thread, since so many people have this misconception, even though it makes no sense.

 

[Stumpy1]

The SEC plays against the least number of teams from other Power 5 conferences plus ND and BYU. Here are the numbers for every conference.

SEC: 11

B1G: 13 + ND (3x) = 16

Big 12: 10 + BYU = 11

PAC: 8 + ND (3x) + BYU = 12

ACC: 13 + ND (4x) = 17

The SEC has 14 teams just like the ACC, B1G and PAC, the Big12 has 10.

 

[VA Husker Fan]

The SEC plays against the least number of teams from other Power 5 conferences plus ND and BYU. Here are the numbers for every conference.

SEC: 11

B1G: 13 + ND (3x) = 16

Big 12: 10 + BYU = 11

PAC: 8 + ND (3x) + BYU = 12

ACC: 13 + ND (4x) = 17

The SEC has 14 teams just like the ACC, B1G and PAC, the Big12 has 10.

And what are the records of those games?

 

[BigRedBuster]

"They haven't proved it against anyone. They only play each other"

*immediately lists 3 wins by SEC teams over very good power-5 conference teams*

lol

there is a bias. and it is that the SEC is held to a much higher standard. who has proven anything then? certainly not the big 10. not the pac 12. not the big 12.

if the argument is that the SEC is not that good, i want to know, compared to what?

You are right. What has anyone proved?

So....if we take this train of thought, the entire basis for what conference gets the possibility of multiple teams in the playoffs boils down to three games played in the first month of the season. We have something like 125 teams playing 12 games. That's 1,500 games and it all boils down to three games played in the first month.

Wow....that's a great system.

 

[sd'sker]

You are right. What has anyone proved?

So....if we take this train of thought, the entire basis for what conference gets the possibility of multiple teams in the playoffs boils down to three games played in the first month of the season. We have something like 125 teams playing 12 games. That's 1,500 games and it all boils down to three games played in the first month.


Wow....that's a great system.

you pick three games because that fits the idea that this SEC bias is based on next to nothing. but a lot of games have been played. a lot of good teams lost to not so good teams and no one really talks about that (osu losing to va. tech, usc losing to bc). there is a whole body of work at play and it becomes clearer every week. but some would rather act like some how the aTm/scar game framed the whole playoff race or it comes down to just 3 games. it does not. i expect tschu to demonstrate this, because i believe he gets this better than anyone else.