Interesting little article-NCAA Selection Tools

[hawkmatt]

An article about how to improve selections:

RealGM Blog: Using The RPI To Group Teams Can Lead To Bias

1. Stop using the RPI as an organizational tool
...
2. Stop giving teams so much credit for close wins and penalizing them for close losses.
...
and more..
Remove the RPI as an Organizational Tool
Of course, as stated at the top, the committee can make these adjustments on the fly, and notice that beating Baylor, Virginia, and Iowa is an accomplishment this year. But given the huge amount of information they have to process, I would not count on it. The time to replace the RPI as an organizational tool is now."

 

[BryceC]

I think there are biases in any rating service. Kenpom and bpi attempt to rate performance. Rpi really only looks at results and ignores performance. There is a place for both, because I think there is something to be said for the ability to win the close ones.

 

[ameshawkeye]

An article about how to improve selections:

RealGM Blog: Using The RPI To Group Teams Can Lead To Bias

1. Stop using the RPI as an organizational tool
...
2. Stop giving teams so much credit for close wins and penalizing them for close losses.
...
and more..
Remove the RPI as an Organizational Tool
Of course, as stated at the top, the committee can make these adjustments on the fly, and notice that beating Baylor, Virginia, and Iowa is an accomplishment this year. But given the huge amount of information they have to process, I would not count on it. The time to replace the RPI as an organizational tool is now."

As far as the RPI goes, I don't think this actually happens. Any win is counted the same in RPI, just like losses.

 

[tm3308]

As far as the RPI goes, I don't think this actually happens. Any win is counted the same in RPI, just like losses.

I think that's the point. A buzzer beater counts the same as a 30 point blowout. Same goes for a loss.

 

[ddiction]

As far as the RPI goes, I don't think this actually happens. Any win is counted the same in RPI, just like losses.

That has the effect of overemphasizing a close win/loss. Pomeroy's rating of a team wouldn't change much if a game-winning shot were to rattle through the hoop or bounce out. In RPI if the shot happens to bounce in, it has the same effect as blowing out a team by 50.

 

[H8IAST8]

Its proven to be a sub par predictor of Tournament performance