(11-03-2019 09:18 AM)DavidSt Wrote:  

(11-03-2019 08:55 AM)quo vadis Wrote:  

(11-03-2019 06:42 AM)DavidSt Wrote:  1.Boise State 95% of winning all out.
2.Memphis 65% of winning all out.
3.SMU 55% of winning all out.
4.Appalachian State 75% of winning all out. Be ranked outside the top 25.
5.San Diego State 65% of winning out.

IMO, the AAC champ has an overwhelming chance of being the G5 rep.

The AAC is way ahead of the other G5 this year, and the committee will recognize that.

But we shall see.

AAC still games:
Cincinnati still have Temple and Memphis.
Memphis still have Houston and Cincinnati.
SMU still have Navy and Tulane.
NAvy still have Notre Dame, SMU and Houston.
Boise State still have Wyoming and Utah State.
The chances for all 4 AAC schools to lose again is a lot higher since they could take each other out. Lets say Temple wins the AAC championship game? That will take the AAC out of the picture for good.

Oh, my, where to begin...

For probabilities, both FPI and Massey (with his own rating system, not the composite rankings) offer an accessible look at statistical probabilities of winning each future game. The two are generally within 5% of each other so we can use either. FPI seems more favorable to Boise, so let's use that.

FPI gives Boise's chances in four remaining games as: 80.4%, 97.1%, 70.2%, and 85%. Cumulative probability is 46%. FPI's main page has a win-out probability of 38.2%, telling us that that column has also factored in a probability of winning against the most likely CCG opponent.
So while Boise is favored to win out, only one game is at your WAG of 95%, much less the cumulative.

You know who else is favored in remaining games? Memphis.
And Cincinnati is favored to win out until the finale against Memphis.
Compared to Boise's 46% chance of getting to the CCG unblemished, it is likely that Memphis (48.6%) will meet Cincinnati (67.5%) on Black Friday, both ahead of Boise. They could have a CCG rematch, or depending on other results, the CCG would likely be the Black Friday victor vs SMU, Navy, or UCF.

The reasonably likely worst case for the AAC is a two-loss champ, with a CCG win and possibly another November win over a team or teams currently ranked ahead of Boise.

Your hypothetical of a three-loss Temple as the AAC champ, including their bad loss to Buffalo? You did correctly identify the AAC worst case.
Temple winning out is a 2.5% probability, AND winning the East requires two more UCF losses plus a Cincinnati loss in addition to Temple beating them. The cumulative probability of that is about 6/100th of 1%, BEFORE their win probability in the CCG. Then they would still have November-December wins over Cincinnati plus a ranked West champ among Memphis, Navy, SMU...all four of which are better wins than any and all of Boise's regular season or CCG opponents.