There is some validity to what you say. FSU is #1 on the Worst Lost list, can’t get any higher; but there’s only 1 spot difference between them and the best team with a single loss. I should devise a way to weight the list, so that being undefeated gives you more breathing room over the 1-loss teams than a single ranking point.
From your keyboard to the Committee’s ears!!! 
I guess you could say that if you could keep a straight face while referring to a team with only four wins over power conference teams and only one win against a team which has a winning record against power conference teams as “the #8 team in the country”. I guess they get a whole lot of credit for that win over Alabama, which itself has only one win against a currently ranked team.
I get that it is very difficult to compare teams across conferences, but it isn’t difficult to have some basic consistency. For example, Alabama has a 1-1 record against Top 25 teams (both of which were also in the Top 10), and has two wins over teams which were over .500 in their conference as of last week, when the last CFP poll came out. Oregon is 3-1 against Top 25 and 1-0 against Top 10, with 3 wins against winning teams.
It seems to me that if the committee wants to say that Alabama is better than Oregon, they need to either put some more of those teams that Alabama beat into its ratings or demote some of the teams that Oregon beat. Although there aren’t too many of these sort of glaring inconsistencies in the ratings, it does seem that when they do happen, the SEC seems to benefit more often than not.
Miss St.'s best win is over #15 Auburn
Rats. I miss something every week.
This puts State into the top 4 of the jsc1953 committee’s rankings, so it’s a significant oversight.
1t. Alabama 4.5
1t Oregon 4.5
1t TCU 4.5
4. Miss St 5
5. Baylor 5.5
6. FSU 5.5
7. Ohio State 8.5
But if we take Zakalwe’s point into consideration and apply more weight to FSU’s unbeaten record – say, by giving them a zero in the Worst Loss ranking, instead of a 1, the list is:
1t Alabama 4.5
1t Oregon 4.5
1t TCU 4.5
4t Miss St 5
4t FSU 5
6. Baylor 5.5
7. Ohio State 8.5
Incredibly tight, just like in the real world.
Pac-12 says: oops, kinda blew that one. Twice.
Another fun little toy: Here are the current AP top 25, with records adjusted according to the following rules: wins over top 10 teams count triple, wins over teams ranked 11-25 count double, and wins over non-power conference teams don’t count at all. Conversely, losses to unranked teams count triple and losses to teams ranked 11-25 count double. Losses to non-PC teams count quadruple. This is designed to reward tough wins while punishing bad losses, and also discourage filling up on cupcakes.
- FSU 12-0 1.000
- Bama 9-2 .818
- Oregon 12-2 .857
- MS St 7-1 .875 an underrated SEC team! OMG!
- Baylor 8-3 .727
- TCU 9-1 .900
- OhSU 10-3 .769
- UGA 10-6 .625 ah, there’s that SEC bias we know and love!
- UCLA 10-4 .714
10, MichSt 6-2 .750 - KSU 6-3 .667
- Zona 8-4 .667
- ASU 7-4 .636
- WI 6-6 .500 Cheesehead bias?!
- Auburn 7-6 .538
- GATech 7-6 .538
- Mizzou 6-4 .600
- Ole MS 6-8 .428 how’d they get in here? SECURITY!
- Marshall 0-0 .000 this system isn’t kind to minor league teams; bug or feature?
- OK 6-4 .600
- CSU 0-4 .000
- Minny 5-5 .500
- Clem 7-5 .583
- L’ville 6-6 .500
- Boise St. 0-4 .000
This seems to support my point that Ole Miss have no business still being ranked at all, and Alabama, while highly ranked, isn’t close to #1. TCU looks very good by this metric; its wins aren’t great, but it has a lot of them, and their only loss is to Baylor (the system doesn’t take into account that their loss was by 3 points on the road, either); they also get rewarded for putting Minnesota on the schedule. In general, I think that I like these rankings slightly better than the “official” ones, although I find it kind of hard to wrap my head around #2 TCU (and they would bump even a bit higher if we counted KSU as a top 10 team, as this system suggests we should).
Also, I note that Nate Silver has now turned his talents to predicting the playoff participants, although predicting the behavior of a small group of people with minimal past data to go on isn’t really what he usually does.
I like the way you think, TF. But I do believe there’s a flaw in your methodology, although I haven’t yet come up with the solution.
If I understand your algorithm correctly – if Marshall had beaten a single Top 25 team they’d be tied with FSU for #1. As you say … not sure if that’s a bug or a feature.
Actually, if Marshall had beaten ANY PC team, even, say, Colorado, they would be 1-0 and tied with FSU, so I think it’s safe to call that a bug. It won’t be relevant to this season, since none of the ranked minor league teams have any PC teams on their schedule. I guess you could set some minimum number of games required to have the rating “count”, but that would effectively prevent any minor league team from ever being ranked at all.
In general, I think it is even harder to compare non-PC teams to PC teams than it is to compare across power conferences, especially when the non-PC teams don’t play any PC teams at all (and of course this is often not for lack of trying to get them on the schedule!). The least bad solution I can come up with, as I said upthread, is to put them in what we used to call a BCS bowl if they go undefeated, but not to consider them for the 4-team playoff.
Just read that, in fact, the new regime guarantees one spot in the six major bowls for a non-PC team, specifically the highest ranked conference champion among them; and they have now ranked Boise State ahead of Marshall.
I hate it when I provide evidence of something and the other poster never comes back to acknowledge, so here I am:
1 point jsc1953
Wait, no, looks like we’re going to be reviewing this one
There is a reasonable chance that the ACC might sweep the four in-state rivalry games this weekend.
UGA is the only favorite (against Ga Tech).
FSU, Clemson and Louisville are all favorites against Florida, USCe, and Kentucky.
I hope the SEC fanboys get a rude awakening.
If anyone else watched that LSU - Tx A&M game… That was a terrible no call. Terrible.
Yeah, I have no dog in that hunt, but that was bad. Frankly, I don’t think it would have changed the outcome of the game, but sucks for the Aggies to have that chance taken away from them.
I’m a Cal fan. I expect the worst.
Come on, Wolverines!
Come on, Yellowjackets! You can dooo eeet!
So, if three ranked teams (including a team that hasn’t lost since the 2012-2013 season) beat three unranked teams, that’s a rude awakening?
No, getting swept would be. That was the point of the post; note the first sentence. When was the last time the SEC got swept by OOC rivals, let alone by one other conference?
WOOT! Way to go Georgia Tech!