I think the people ranking and seeding both use results in addition to the predictive metrics, which pushes us down. Probably recency too even though they claim they don't. That's why is UConn and Purdue are both 2 seeds and Illinois is a 3 seed. Which is probably the right way to do it to be honest.
Yes, the seeds are more correlated with the resume rankings, especially SOR and WAB, than the efficiency ranks, but it isn't consistent. There were claims that efficiency was going to be the primary seeding criteria, so part of my surprise at some seeds is based on that expectation. And I wasn't complaining because of our seed. Here are some outliers:
Vandy (5 seed): #9 resume, #11 efficiency. Non-blueblood prejudice?
St John's (5 seed): #13 resume, #15 efficiency. Favored their efficiency and are only off by one seed, but this contradicts the recency theory
Kansas (4 seed): #16 resume, #21 efficiency. Favored their resume and/or blueblood bias?
Utah St (9 seed): #28 resume, #30 efficiency. Non-blueblood prejudice?
Miami FL (7 seed): #26 resume, #34 efficiency. Favored their resume
VCU (11 seed): #34 resume, #41 efficiency. Favored their efficiency
Maybe there are circumstances for each of these teams (or the teams around them) that I'm not considering, like injuries affecting a stretch of the season or their opponents during key victories, but I'd at least prefer if the committee would commit to something like an average of the ranking systems (ideally weighted towards efficiency) as a hard starting point, and releasing brief explanations of any deviations from that.