WOW! No Official Ranking Thread. I Will Start One

[iggymcfly]

MyTurn, one thing I don't like about your system is that it punishes teams for winning by solid margins. i.e. If LSU had beaten Virginia Tech by 1 instead of by 41, the corresponding rise in Virginia Tech's rating would bump up LSU's "wins over expectation" significantly.

I don't really think this method is good for ranking a group of teams at all, but where I do think it could be useful is to compare two teams. For instance, say that BC and LSU were in competition for a spot in the BCS title game. Well, then you could compare what a #7 team (LSU's predictor rank) would do with BC's schedule to how BC did and how a #16 team (BC's predictor rank) would do with LSU's schedule to how LSU did.

That way, the margin of victory is still included somewhat so teams aren't punished for winning big and you get a fairer picture of how the the teams' accomplishments were affected by their schedule.

[rwperu34]

[ QUOTE ]
MyTurn, one thing I don't like about your system is that it punishes teams for winning by solid margins. i.e. If LSU had beaten Virginia Tech by 1 instead of by 41, the corresponding rise in Virginia Tech's rating would bump up LSU's "wins over expectation" significantly.

I don't really think this method is good for ranking a group of teams at all, but where I do think it could be useful is to compare two teams. For instance, say that BC and LSU were in competition for a spot in the BCS title game. Well, then you could compare what a #7 team (LSU's predictor rank) would do with BC's schedule to how BC did and how a #16 team (BC's predictor rank) would do with LSU's schedule to how LSU did.

That way, the margin of victory is still included somewhat so teams aren't punished for winning big and you get a fairer picture of how the the teams' accomplishments were affected by their schedule.

[/ QUOTE ]

Margin of victory against any particular opponent is a small factor in the overall scheme. The predictor is used as a gauge for how good the opponent is. If Virgina Tech had only lost to LSU by 1, their whole would look a lot better and they would be considered a tougher opponent. The fact that they lost by 41 is a part of the whole that makes up VT's ranking. We'll never have a big enough sample to make things 100% accurate, but this is as good of a system for determining schedule strenth and conference strength as I've seen yet.

[Austiger]

[ QUOTE ]
alright, my ranking system, OTFRR#5, is up for the top six.

compared to a #5 team, the teams are the following games ahead of pace on their respective schedules
Kansas +1.005W
Ohio State +0.887W
Arizona State +0.844W
Boston College +0.774W
Oregon +0.210W
LSU +0.204W

For those who haven't followed so far. My ranking system is purely about who has the best win-loss on-the-field results. It compares how each team has done verse their schedule compared to the #5 team in the country. It uses sagarin predictor scores and a spread to moneyline conversion based on all NCAAF results from 1993 to 2006.

Ohio State remains second due to playing one extra game. The other 3 undefeated teams would all pass Ohio State by a significant margin if each wins the rest of their games.

Oregon has now moved ahead of LSU, which is interesting.

[/ QUOTE ]

I gather by this that a #5 team would have, on average, 1.005 losses against Kansas's schedule and 1.204 losses against LSU's schedule. Kansas has the 101st rated schedule according to Sagarin and LSU the 16th toughtest. That doesn't add up to me.

Maybe the easier question is, if you are basing these off Sagarin, how do you have Kansas ahead of BC and ASU if Sagarin has their schedules rated much tougher?

[MyTurn2Raise]

because Sagarin and most other rating systems screw up strentgh of schedule

they'll often just do an arithmetic or geometric mean of the rankings of all opponents played

that doesn't work when trying to evaluate how tough it is to win a game

for instance, a home favortie of 21.5 has won 94.7% of their games from 1993 to 2006 while a home favorite of 27.5 has won 97.5%. That isn't much difference for 6 points. However, a home favorite of 3 has won 54.5% of their games while a home favorite of 9 has won 74.1% of the time. That difference is huge. Here, almost all rating systems just look at the opponent being 6 points different. A team that played the 21.5 and 9 point favorite would be getting just as much credit as the team playing as a 27.5 and 3 point favorite. However, it was much, much, much more difficult for the latter team to win both games.

In summary, all the strength of schedules I've seen publicized fail because they do not account for the non-linearity of win/loss pcts in relation to the ranking of the opponent.

I use the same inputs as sagarin, but feel it's a much better methodology for getting the output.

The other methods fail Kansas due to Kansas playing some really bad squads...so bad that I gave Kansas an outright 100% chance to win some games instead of capping at 99%. However, there shouldn't be a huge difference due to Kansas playing a 100% gimme like SE Louisiana and Arizona St playing SDST at home (a game they over 97% of the time). Yet, ASU is getting credit for playing the #100 overall (#73 predictor) team while Kansas gets credit for #203 overall (#186 predictor). When the schedules are weighted, that will plummet Kansas even though the onfield win/loss difference is minimal.

Kansas has had the toughest schedule of the undefeated so far because they have played @ K St, @ Colorado, and @ Texas A&M...all of which are very losable games even for a #5 team.


This idea had buzzed around my head for a long time, but I finally broke through conceptualizing it when I read the coltonindex criticism of the rpi in college basketball. The rpi is a sham that has teams gaming the system to secure automatic wins from teams rated 130 instead of 270. The same phenomena is showing up in college football only the football teams haven't really thought of 'gaming' the system yet.

[MyTurn2Raise]

[ QUOTE ]
MyTurn, one thing I don't like about your system is that it punishes teams for winning by solid margins. i.e. If LSU had beaten Virginia Tech by 1 instead of by 41, the corresponding rise in Virginia Tech's rating would bump up LSU's "wins over expectation" significantly.

I don't really think this method is good for ranking a group of teams at all, but where I do think it could be useful is to compare two teams. For instance, say that BC and LSU were in competition for a spot in the BCS title game. Well, then you could compare what a #7 team (LSU's predictor rank) would do with BC's schedule to how BC did and how a #16 team (BC's predictor rank) would do with LSU's schedule to how LSU did.

That way, the margin of victory is still included somewhat so teams aren't punished for winning big and you get a fairer picture of how the the teams' accomplishments were affected by their schedule.

[/ QUOTE ]

yep...no system is perfect
the system I've put forth is only a tool and looks at win/loss proven on the field results only... it does not look at what the appropriate margin of victory is for those teams. As rwperu states, the difference would be small.
I thought about coming up with a MOV based result as well, but all had more flaws than they are worth. Teams play to win games. They don't play to beat the #90 team by more than 32 because that's the key number.
It's tough to imagine a MOV result still wouldn't have Kansas on top. They blew out the crap. They dominated @ Colorado and @ A&M (it took a controversial late TD and 3 missed FGs for Kansas to keep that 19-11) and they won @ K St in the most impressive win for any of the title contenders.

most important:
champion teams are determined in nearly every sport by their win/loss results...not by margin of victory

I'm not trying to get the 'best' team. I'm trying to get at the most 'deserving' team.

[iggymcfly]

I guess the thing I didn't like about your rankings is how big the gap was between the undefeateds and the one-loss teams. Kansas on top didn't bother me as much as the gap between BC and LSU when the Tigers are obviously better, and at least in my eyes have accomplished more this season as well. That's why I started thinking about comparing how the teams with MOV would do against the others schedules since I think that LSU would have a better chance of being undefeated with BC's schedule than BC would of having one loss or better.

Also, I think a lot of the reason the system likes Kansas so much is simply because the Sagarin predictor likes Kansas State so much. Sagarin has the Wildcats ranked #9 this week which is much higher than anyone in the media would put KSU. I'm not even arguing the Wildcats as being overrated by Sagarin, just saying that's where a lot of the differences come from.

Also, while they're not as efficient since they don't use MOV for opponents either, I do think that the Sagarin ELO-Chess ratings weight opponents in the way you described. They have Kansas #2 ahead of Ohio State despite Ohio State having the better raw SOS, and the only reason Kansas isn't #1 is that the ELO-Chess ratings give BC credit for beating a #6 team on the road in Virginia Tech.

[Arnold_O]

[ QUOTE ]
Does Arizona State move up to #1 if they beat Oregon? I don't see how you can't put them there.

On a similar note, does Oregon move above LSU with a win?

[/ QUOTE ]

yes and yes

[Austiger]

[ QUOTE ]
because Sagarin and most other rating systems screw up strentgh of schedule

they'll often just do an arithmetic or geometric mean of the rankings of all opponents played

that doesn't work when trying to evaluate how tough it is to win a game

for instance, a home favortie of 21.5 has won 94.7% of their games from 1993 to 2006 while a home favorite of 27.5 has won 97.5%. That isn't much difference for 6 points. However, a home favorite of 3 has won 54.5% of their games while a home favorite of 9 has won 74.1% of the time. That difference is huge. Here, almost all rating systems just look at the opponent being 6 points different. A team that played the 21.5 and 9 point favorite would be getting just as much credit as the team playing as a 27.5 and 3 point favorite. However, it was much, much, much more difficult for the latter team to win both games.

In summary, all the strength of schedules I've seen publicized fail because they do not account for the non-linearity of win/loss pcts in relation to the ranking of the opponent.

I use the same inputs as sagarin, but feel it's a much better methodology for getting the output.

The other methods fail Kansas due to Kansas playing some really bad squads...so bad that I gave Kansas an outright 100% chance to win some games instead of capping at 99%. However, there shouldn't be a huge difference due to Kansas playing a 100% gimme like SE Louisiana and Arizona St playing SDST at home (a game they over 97% of the time). Yet, ASU is getting credit for playing the #100 overall (#73 predictor) team while Kansas gets credit for #203 overall (#186 predictor). When the schedules are weighted, that will plummet Kansas even though the onfield win/loss difference is minimal.

Kansas has had the toughest schedule of the undefeated so far because they have played @ K St, @ Colorado, and @ Texas A&M...all of which are very losable games even for a #5 team.


This idea had buzzed around my head for a long time, but I finally broke through conceptualizing it when I read the coltonindex criticism of the rpi in college basketball. The rpi is a sham that has teams gaming the system to secure automatic wins from teams rated 130 instead of 270. The same phenomena is showing up in college football only the football teams haven't really thought of 'gaming' the system yet.

[/ QUOTE ]

Okay I agree with all of that, and it's remarkable that Sagarin and others don't fix that. That has always been my argument for Auburn>Oklahoma in '04. Auburn played 4 top 15 teams that year, and Oklahoma played 1. But Auburn gets left out because of the difference between Bowling Green and the Citadel.

I slightly disagree on the RPI though. There is gaming of the system, obv. But almost all the bubble teams have losses to teams outside the top 150, so those aren't guaranteed wins. In fact Gonzaga had 2 losses outside the top 200 and still got in one year (the year the lost the classic to Arizona in OT in round 2 I think.) The system is MUCH more fair now that they use home/road. It was laughable before and that's why so many more mid-majors are on the bubble now. Major conference teams used to be able to schedule all home games with no penalty by the RPI.

[BigSoonerFan]

[ QUOTE ]
Okay I agree with all of that, and it's remarkable that Sagarin and others don't fix that. That has always been my argument for Auburn>Oklahoma in '04. Auburn played 4 top 15 teams that year, and Oklahoma played 1. But Auburn gets left out because of the difference between Bowling Green and the Citadel.

[/ QUOTE ]

It was a little more than that. I notice you cut it off at the top 15 instead of top 20, which would've made it 4-3, as well as OU having the top ranked opponent. None of OU's toughest opponents were at home that year, while Auburn's two toughest were. It was a little more than "Bowling Green vs The Citadel".

[MyTurn2Raise]

Gosh...I wish I had sagarin's ratings pre-bowl games for that season

I'll try emailing and seeing what I can get

it's a fun debate