I present the following plan. I know it looks complicated at first. Its not at all. It favors conference championships, playing a P5 supermajority schedule, and head to head results. It penalizes playing cupcakes (especially non-D1 teams), losing in any form, and using money-making schemes like conference championship games. It also deemphasizes "the better loss" argument. Thats a stupid argument that should never matter. It also establishes the idea that playing a supermajority of P5 teams is whats important, rather than "strenghth of schedule". strength of schedule depends greatly on what method you use to determine what that means. Its too complicated. using the supermajority method, its binary; either you did or you didn't. end of discussion.
8 team playoff
1. Power 5 conference champions;
1.1. Its up to each conference how it determines its champion. If you continue to have a conference championship game, YOU ACCEPT THE RISK THAT A PREVIOUSLY UNDEFEATED TEAM MAY BE UPSET BY A TEAM WITH SEVERAL LOSSES IN SAID CHAMPIONSHIP GAME. If you don’t like that possibility, then don’t have a conference championship game, and let the regular season results be the sole determinant of the conference championship status. Your choice, don't bitch.
2. 3 wildcards
2.1. Total wins team playing a supermajority power 5 schedule is the top criteria for wildcard consideration
2.1.1. Teams playing a supermajority of its schedule vs. power 5 teams will receive first consideration for wildcard bids
2.1.2. Teams playing the type of schedule defined in section 2.1.1. will get bids in inverse order of the number of losses REGARDLES OF WHO THOSE LOSSES WERE TO (i.e., undefeated teams vs. P5 supermajority schedule will receive a bid before any one loss team, regardless of whether that loss was to a sun belt team or a number 1 ranked sec team). Undefeated vs. P5 supermajority schedule matters most.
2.1.3. If there are a 4 or more teams meeting these first two criteria, the first tiebreaker is number of NON-power 5 teams were included in those teams schedules. A team playing one NON-power 5 team will get in over a team playing 2 NON-power 5 teams, regardless of who they are. If any teams in consideration played a non-D1 team, that game counts as 2 NON-power 5 games played for the purpose of this tiebreaker (you don’t get rewarded for playing the Citadel. Sorry)
2.1.4. Once ALL wildcards are filled using the above criteria, one loss (then 2 loss, 3 loss, etc) teams will be evaluated. The first tiebreaker is head-to-head results. If two of the teams in consideration played head to head, the TEAM THAT WON THAT MATCHUP GETS IN OVER THE ONE WHICH LOST, REGARDLESS of who the other team lost to. Absolutely no exceptions to this rule. The two teams played, and we know which one was better.
2.1.5. If the teams did not play head-to-head, then the next tiebreaker is number of NON-power 5 teams were included in those teams schedules. A team playing one NON-power 5 team will get in over a team playing 2 NON-power 5 teams, regardless of who they are. If any of the teams in consideration played a non-D1 team, that game counts as 2 NON-power 5 games played for the purpose of this tiebreaker (you don’t get rewarded for playing the Citadel. Sorry)
2.1.6. If the above tiebreakers are insufficient, then the committee can make ANY DECISION THEY LIKE USING ANY DATA POINTS THEY LIKE to break the tie. The most important factors have been considered (conference championships, being undefeated vs. a P5 supermajority schedule, number of wins, playing a supermajority p5 schedule, NOT playing cupcakes, and head to head) so other decisions are more nuanced than a formula can determine. The committees judgement is acceptable beyond this point.
I believe this method gives you the best OBJECTIVE representatives in the playoff. Not “most deserving” or “the best team” or “most 5 star recruits” or whatever. Using the above rules, what happened on the field is what mattered most, and will give you the RIGHT teams in the playoff based objectively on what happened on gamedays during the season.