(03-13-2018 01:27 PM)The Cutter of Bish Wrote:  

(03-13-2018 01:09 PM)ken d Wrote:  So, when someone is included, or excluded, I'm going to want to see a reason other than RPI for why the committee got it wrong.

Yeah...but it's the one in which the committee wraps itself. I agree with you...I don't hate RPI like others do, but I am over its misapplications. And it's easy to knock it down, because it's a composite that takes way too much into account without proper portioning. And I'm over the conspiracy theories coming from major conference schools against non-majors for manipulating things.

This team at 26 isn't worthy, look at their SOS...but RPI already takes SOS into consideration to get to that 26 number. Well, look at their NC-SOS...ditto there, too. Yeah, but, but...2-3 versus Group 1, and a bunch of those aren't going to the tournament...again, somehow the algorithm spits out a result at the 26th place.

If there wasn't such resistance to "show me your work," I think many people could drop it. The problem is, and I think Jay Bilas nailed this one...you have some who don't watch games, look at paper and numbers only, and then don't apply a consistent method across the board. That's where I soften up to RPI...it's not a terrible metric...it's the endorsed metric in a bad committee member's hands when other metrics might be better.

I used to think RPI wasn't such a terrible metric, but crunching the numbers has me questioning that. Some examples involving bubble teams:

Team..................RPI....Consensus

St Bonaventure.....24......62
Providence...........31......59
Middle Tennessee..33......54
Oklahoma St........88......52

Then I see South Dakota State with an RPI of 35, when the other ranking services average around #80. A "good" metric shouldn't be that much of an outlier.