If you gave better teams harder schedules, you'd have records trending toward .500, resulting in worse rankings... unless someone escapes with 1 or fewer losses. Not sure what would be gained by skewing schedules in this manner, especially if you did this "on the fly."

Now, for larger conferences, this might be a neat way to wrap up the end of the season, rather than begin it:

FiveThirtyEight did an article years ago suggesting the Big Ten go divisionless, and flex the 7 games the week before Thanksgiving, have rivalry games the week of Thanksgiving, and 7 more flex games the week of CCGs. I simulated their model (with 13 days heads-up and preset home sites for all 14 flex games) and when multiple teams finished 8-1, there was a head-to-head tie-breaker 98% of the time... better odds than an 11-team AAC.