Take a look at Miami's non-conference schedule:
http://warrennolan.com/basketball/2018/ ... /Miami-(FL)
They only played two teams with an RPI better than 170 on their non-con schedule, and that was two really good mid-majors in an MTE. They played only two true road games on their schedule, at Minnesota (part of the Big 10/ACC Challenge) and at George Washington. Yet Miami was far above Loyola in the SOS measurement, mostly because Miami only played one team with an RPI worse than 250--- Loyola played SEVEN teams worse than 230. But one of those teams had a 33 RPI last year and a 52 RPI the year before that.
If Middle Tennessee State and New Mexico State had one or two losses more than they did, their games at neutral court against The U wouldn't have been Q1 games for them. If they happened to have RPIs at 51, they'd be Q2 games. But Miami was going to have 4-8 Q1 games in conference, and just three Q1 wins is going to give them a better at large seed than almost any Mid Major auto bid.