KU has had a practice under Self of always playing projected conference winners in the noncon. That has helped our RPI/SOS a ton. Putting together the champions classic where we are guaranteed a game against either UK, MSU, or Duke each year is also great for our schedule. You could make an argument that one of the worst games on our schedule each year is the PAC team we always seem to play.
Like I said earlier, if you are a P5 team and you aren't getting into the tournament then you either 1. Suck something truly shitty or 2. Had an administrative failure with scheduling. There are very few teams in the P5 that are solid at large teams with a SOS worse than 45. This year those teams are Tech, Ole Miss, Washington, and ASU. Those teams won 26, 20, 26, and 22 games, respectively. I'd say the only "solid" at large team in that pile is Tech. Those other 3 spots should have gone to mid-major teams that won 26+ games in my opinion.
My thinking has evolved greatly over the last 5 years of really closely looking at team sheets and resumes. I think my opinion has come to of the 36 at large bids 12-14 should go to mid-major/low major teams. Let's stop using eye ball tests and hypothetical scenarios between a small/mid major and a P5 team. The P5 has all year to prove they are good and the smaller teams have about 4 weeks total to prove the same. I say just give 12-14 at large bids to the small teams and let the P5 be compared against each other for the other 22 bids.