I get where you are coming from and in some cases I certainly agree, but in others I do not and it would be nearly impossible to devise an objective rule that could differentiate between the two cases.
A hypothetical example in which I would agree with you:
Suppose that Wisconsin ends up being just a steamroller and destroying everyone in their path this year. In the B1G-E that would be bad for PSU and M who play Wisconsin and good for MSU and tOSU who do not. (Note, I'm assuming here that those four are the only contenders for the B1G-E and that IU, UMD, and RU will be also-rans).
In that case I would agree with you because it is "unfair" for the Nittany Lions and Wolverines to be penalized for playing a tougher schedule than the Buckeyes and Spartans.
A hypothetical example in which I would disagree with you:
Suppose that the Wolverines, Buckeyes, and Nittany Lions go 1-1 against each other and 6-1 against their other conference foes with each of the three losing in a shocking upset, Michigan to Rutgers, Penn State to Indiana, and Ohio State to Minnesota.
In that case I would disagree with you because I don't think that the Buckeyes should be rewarded for losing to a crappy team from the other division rather than losing to a crappy team from the same division. Note, however, that the Buckeyes would win this tie in the current structure anyway because it would go to tiebreaker #2, divisional record in which Ohio State would be 5-1 while Penn State and Michigan were each 4-2.
Where I would STRONGLY disagree with you:
I think we are moving toward and will eventually arrive at a playoff system in which the P5 Champions get auto-bids. One thing that I would NOT like about that is that it would render OOC games nothing more than exhibitions. If we further made cross-divisional games into non-factors in the divisional race then those games would also, effectively, be exhibitions. In that situation Ohio State could literally lose to Oregon State, TCU, Tulane, Minnesota, Purdue, and Nebraska (six games) and still control their own destiny for the National Championship because winning the other six would get them to the B1GCG and winning that would get them to the (enlarged) CFP.
I'm not sure I agree (in the "might disagree" option, OSU lost to Minnesota and went ***undefeated*** within the division. Yes, IMO, they deserve the division crown even if they lost EVERY other interdivisional game).
Then, even if I fully did agree, a lot of the trouble spots you cite can be eliminated by only addressing division record first if there are two or fewer teams at the top.
Your counter examples include three-team ties.
What is an example of a scenario where a team that is (1) alone at the top of the division record or (2) tied with precisely one team at the top of the division record isn't the team most deserving of that division's CCG bid?
So this proposal would have two tiers:
A. Address division record. If two or fewer teams are at the top, proceed to name the champ (if only one) or execute the H2H tiebreaker. If three or more teams are tied at the top, proceed to option B.
B. Move to Conference record. If ties remain: Execute the circa 2017 tiebreaker list.
