NCAA 2003 RPI

[Rich Kern]

I have the NCAA 2003 RPI up on my site if you are interested along with a comparison of the RKPI.

www.richkern.com/d1wvb03/RPI2003/index.asp

There is one BIG bust in the data associate with Murray State. As you may recall, they had inadvertently played with an illegal player for part of the season and had to forfeit several conference matches. I counted those forfeits as losses but the NCAA counted them as wins (they had actually won the matches).

Another difference is with Washington State. I wasn't sure how to handle matches against provisional teams and decided to not count them. It appears the NCAA counts teams such as Utah Valley State as Division I opponents (wins and losses against them do count) but does not count them as Division I teams (they are not listed in the RPI). Since Washington State had lost to Utah Valley State, I had ignored this in my RKPI computation but the NCAA counted it as another loss.

This gave Murray State 74 more points than I had computed and subtracted 29 points from Washington State. This filters down to all teams they had forfeited to as well as all their opponents and their opponent's opponents.

Even with that major difference, 301 out of 311 are within 10 places in the NCAA RPI and RKPI.

[vb1midwest]

explain to me how to understand these rankings???
alot of teams are ranked higher then others when the other teams beat them. what is this all tied too???

doesnt strength of schedule mean anything?

[Rich Kern]

vb1midwest said:

explain to me how to understand these rankings???
alot of teams are ranked higher then others when the other teams beat them. what is this all tied too???

doesnt strength of schedule mean anything?

I'm not exactly sure what your questions are but I'll try to answer some of them anyway.

The RPI is one of the things the NCAA looks at when trying to select and rank teams for the NCAA tournament. It is based on a team's record, their opponent's record and their opponents' opponents records so YES, strength of schedule does mean something. The RKPI is my attempt at trying to simulate the RPI. I can see that I missed a couple of things where the NCAA counted things one way and I counted them another.

However, (isn't there always a "however"?), once you start looking at why teams that lost to lower ranked teams but are still ranked higher, you are lierally in for a "losing" battle. You will get into a lot of circular arguments where A beats B who beats C who beats A. Which team is the best? The RPI ignores the head-to-head battles and concentrates on total season performance and weighs every match even from the beginning of the season to the end.

Another "however" is that this is just one of the things they look at for the selections and rankings. When they considered all things this season, I don't think they did too bad compared to some previous years questionable calls.