We played poorly on our OOC schedule against P6 schools with exception of Missouri. We should take a metrics hit for that.
However, we played other mid and low majors. We took care of the games but we did not blow most of them out, nor were we generally efficient. You cannot convince me that our performance against these mid-majors in Nov and Dec are in any way predictive of how we will perform in March against an NCAA tourney field. Yet, that is what it is used for - predictive metrics for seeding.
Read how the Big 12 gamed the NET this year. 5 or 6 Big 12 teams generally played low to mid majors in OOC pretty much exclusively and destroyed those schools in the 30s. They are riding high in the NET and also helping the conference performance. It is broken.
Any metric that tries to equalize all competition to a single number has flaws and needs to be modified. Horse racing has Beyers figures that attempt to factor into the many variances affecting past performance to help handicapping a race from track, speed, times, pace, distance, etc... However, they have to regularly be modified. They also have to be taken into consideration with other variables - like winning and improving. No team is the same at the beginning of the year and the end. To suggest the Jan, Feb March performance should not be looked at a bit diffently is nonsensical.