KenPom / metrics update

[wu_shizzle]

KenPom is helpful for predicting stuff correctly in the long run more often than the committee or the eye test or whatever might. Because upsets are bound to happen, I find it is most useful as a tool for evaluating upset picks.

For example, the seeds said Minnesota over MTSU should've been a slam dunk, but KP said it was much more of a toss-up, even though it still thought MN would win. Because of that, it made it a relatively good upset prediction. Your idea of a bracket with all KP predictions would be interesting for a variety of reasons, but I think it's also important to recognize that KP calling a game a toss-up is probably a "win" for advanced metrics over the typical seeding.

I've done well the last few years using this logic.

I had Middle Tennessee and Xavier and didn't think twice.

[WSUwatcher]

What's also striking in the KenPom overall margins is that only two teams really stand clear of the field: #1 Gonzaga at 32.76, and #2 Villanova at 30.65. (Numbers as of this writing.) The next team down, #3 Carolina, is almost three points behind at 27.71. And #4 Kentucky, at 26.94, is less than half a point ahead of 10th seeded but #5 WSU at 26.49, pending how the numbers change today.

Of course, you still have to win the games -- for example, Gonzaga and Nova have both lost, on the proverbial "given day," to teams way behind them -- and that means WSU has to capitalize on its big edge over #36 Dayton (15.54) and translate it into a victory before they even get the much-discussed shot at Kentucky (or the Norse!). I believe they will, but I'm admittedly apprehensive, and there's no doubt that Dayton feels disrespected and would be thrilled to shut everyone up.