I don’t disagree with a lot of what you are saying but one thing I wonder about is if the computer models will have enough data in the future.
If the SEC goes to 9 conference games and then they schedule this non-conference game of B1G-SEC like they are talking about, then I assume the last 2 games on the schedule are going to be cupcake home games against Sunbelt level teams.
I’m not sure how many remaining games there will be against other conferences for the computers to even know how the conferences compare in terms of strength.
But to be clear, I don’t disagree with the broader point that we need a non-biased way to make the decisions and maybe the computers will still be the best way to do it.
If the Big Ten/SEC "disconnect" from the rest of college football to that degree (which I think they will at some point) save maybe one de facto exhibition game per year against a woefully overmatched FCS or G5 school (not a Big 12 or current ACC school, do note) then the computer being unable to make valid comparisons between the "P4" conferences at that point isn't going to matter in the absolute slightest.
The P4 would be in de facto divisions at that point and will probably end up having separate championships. Which at some level... is fine by me. It might be inevitable. Let the Big Ten and SEC and ND go and let the Big 12 and ACC remnants, G5, and the best few FCS conferences be at different levels.