It would have been possible to come up with a new tool which completely ignored margin of victory and other non-victory related statistics that was much better than the RPI. People who didn't understand statistics thought the problem with the RPI was that it didn't take these advanced metrics into account. Statisticians knew that RPI was broken because it was non-predictive - you could easily create situations where a stronger team had a lower RPI in a theoretical situation with infinitely many trials (or just using statistical methods predicting finitely many trials). It's known as a biased statistic. It may not be more of a travesty in a statistical sense as OPS, but overall was probably the weaker metric due to its exploitability, as well as the fact that it was used for such an important task.
To see the NCAA replace the RPI with a tool that takes all those margin of victory metrics into account, as opposed to just coming up with a non-biased statistic based on wins and location only (Sagarin's ELO does this) tells me that this is mostly a P.R. move, meant to make sportswriters say good things and fans say good things about the NCAA, as opposed to actually making positive changes.
That being said - there's a reason to consider metrics lilke Pomeroy's and Sagarin's which account for points and efficiency. It's to prevent a team from getting an unfair matchup, and you may toggle seeding using these tools in order to make the matchups fair for higher seeds. But for purely looking at inclusion into the tournament, I believe those metrics have no place and in fact encourage poor sportsmanship by encouraging teams to run up the score. Therefore, I disapprove of this new metric, although it's entirely predictable by an institution which cares more about its reputation than actual integrity.
But it still appears to be better than RPI.