For better or worse the precedent set by the committee is that the #1 criteria is "number of losses". The only exceptions have been cases in which SoS was an EXTREME outlier.
Bama's SoS simply wouldn't be an extreme outlier if it would even be #1.
Take the example implied by
@FearlessF :
- Georgia wins out, finishes 13-0 and SEC Champs.
- Bama wins their other games, finishes 11-2 SECCG loser.
- Ohio State wins out, finishes 12-1 and B1G Champs.
- Oklahoma wins out, finishes 13-0 and B12 Champs.
- Oregon wins out, finishes 12-1 and P12 Champs.
- For the sake of this discussion assume that Wake does NOT win out such that the ACC produces a Champion with at least two losses and no serious argument.
I think the first two are obvious. Georgia is the #1 and Oklahoma is the #2. They would both be undefeated P5 Champions. Oklahoma is down in the rankings right now but that is because they've struggled a bit against weaker opponents and because their schedule is seriously back-loaded with their last three regular season games being against the other three B12 contenders. If they get through that AND the B12CG without a loss, they'll easily be in at 13-0.
That leaves:
- 12-1 B1G Champion Ohio State: Best wins are #6 Michigan, #7 Michigan State, #19 Purdue, and (probably) #18 Wisconsin. The loss is a respectable one-score loss to #3 Oregon.
- 12-1 Pac Champion Oregon: Best win is #4 Ohio State followed by #24 Utah (probably twice). Everybody's victims would be expected to drop for losing to them but Oregon is the only one playing theirs twice and thus degrading their own second-best win dramatically. After Utah there isn't much and the loss is to a terrible Stanford team (currently 3-6).
- 11-2 non-Champion Bama: Best wins are #15 Ole Miss, #25 Arkansas, and #17 Auburn. Losses are a close loss to #11 aTm and (per Fearless' example), a close loss to #1 Georgia.
Both Ohio State and Oregon have a better win than Bama's best (#6 Michigan and #7 MSU for tOSU, #4 tOSU for Oregon). Ohio State's loss is also better than Bama's loss to aTm (#3>#11). Additionally, Ohio State has four wins over ranked teams to Bama's three and Ohio State is a P5 Champ where Bama isn't. This really isn't a close call.
Oregon's H2H win over Ohio State is the big monkey-wrench. If you ignored the names (and consequently the H2H) Ohio State would be the obvious choice for #3 and then it would be between Bama and Oregon for #4. That simply comes down to the rather typical dispute between Championships vs losses vs SoS. Oregon has less losses and a Championship but they played a weaker schedule.
It would make no sense to put Bama behind Oregon and ahead of Ohio State so if you feel that Oregon HAS to go ahead of Ohio State due to H2H you are left with two choices:
Either:
- #3 Oregon, plays Oklahoma in a CFP semi-final
- #4 Ohio State, plays Georgia in a CFP semi-final
- #5 Bama, out.
Or:
- #3 Bama, plays Oklahoma in a CFP semi-final
- #4 Oregon, plays Georgia in a CFP semi-final
- #5 Ohio State, out.
The easiest thing to do is to exclude Bama on the basis that they have two losses and aren't a P5 Champion. I do think that in this case the committee would put Ohio State ahead of Oregon simply because they'd want UGA and tOSU on opposite sides of the bracket for a host of reasons.