I think any formula is fraught with potential problems. I don't see a lot of panic about this other than your post.
I think the main reason for that is that most fans are as yet unaware of how disastrous the tiebreaker they selected actually is. Other than the reddit posts linked in one ESPN column, I haven't seen it discussed. Were it more widely-publicized, I think you'd see more objections and outrage.
Lots of posts about scenarios that are of the WTF variety that may or may not happen.
The trouble is the new tiebreaker seems designed to achieve a specific result using what actually
did happen in 2008. That is a major problem, especially when remembering that the party who felt "screwed" in 2008 is A) the conference's unquestioned power broker and B) absolutely wrong about saying they got "screwed" that season.
Lots of scenarios where a CCG in a ten team league screws things up too. Lots of scenarios where expanding by 2 teams is problematic.
You're dead on with this. I never wanted a conference-championship game, and the evidence so far in the leagues that have one shows roughly twice as many examples of teams who were knocked out of national-championship scenarios by the conference-title games as there were teams who played their way into one because of it.
I'm not that informed though. What do you think they should have done?
There are several more equitable tiebreaking scenarios that could be adopted, all of which would be light years more fair than what they're about to adopt. These are my 2 favorites, although they're not the only ones:
1) Use the "points comparison" tiebreaker they're about to adopt the
proper way, i.e. the best point differential among the 3 tied teams is declared the winner--rather than the worst differential being eliminated as they're going to do. When used correctly, this method neutralizes 2 problems in the current one...it is both more simple (one step instead of two), and it also eliminates any possibility of the team who stands to win needing to shave points in order to prevail in the tiebreaker.
2) Decide between the 3 tied teams based on which one played the strongest
non-conference schedule. Now you could argue that using non-conference games to break a tie for purposes of conference play is contradictory, and that claim does have some merit. However, since the overarching concern in all this seems to be that we as a conference want to send the strongest overall team into a potential playoff scenario, then rewarding the tied team which also tested itself more strenuously in out-of-conference play seems geared to achieving that goal.
Certainly there are other tiebreaker methods as well, e.g. if one of the 3 tied teams had to play both of the games against the other 2 tied teams on the road then they would be declared the winner...and so forth.
Sorry I took so long to reply. Your question was a good one, but it also demanded a lengthy response and I did not have the time to do so until today.