Page 1 of 1

QRF vs. RPI vs. Computer Ranking vs. Polls

PostPosted: Wed Jan 10, 2018 11:24 am
by Bison-Vikes #1
Over the years, there has been a lot of talk on various threads concerning media polls, QRF, RPI, etc. Thought it would be informative to get a thread going to get an explanation of each and which seems to be the most accurate. If someone has insight on the basic concepts of each, what are the driving factors of each, please put on here. Also, which one seems to be the best "predictor" from the start of the season to how the season ends up and which seems to be the weakest. I for one have noticed that many times they are very skewed at beginning of year and start "predicting" closer at the end of the year, which is real no help because all the games have been played so we already know the best teams. Looking for the one that "predicts" the most accurately in the early-mid portion of the year.

RPI does use: Winning % of opponent's opponents.
QRF does use: Margin of victory and SOS of opponents.

Re: QRF vs. RPI vs. Computer Ranking vs. Polls

PostPosted: Wed Jan 10, 2018 11:34 am
by Flip
if the QRF is the same formula MN uses margin of victory is not used. Run will know more.

Re: QRF vs. RPI vs. Computer Ranking vs. Polls

PostPosted: Wed Jan 10, 2018 11:58 am
by Bison-Vikes #1
Here is an example, looked at a RPI listing of ND teams a week ago. A team was listed as #2 in RPI ranking. Their record was 4-2. They were 6 & 8 spots ahead of 2 undefeated teams that had beaten them. In my opinion, something is wrong with the RPI formula at that point. Therefore, should RPI ever be used for ranking purposes at the end of the year, when there are certainly some flaws in the formula? The thing is, love following the media polls and these sites which put this information out there. With having hindsight of past year's early "predictions" and how the years finished, just wondering if some of these formulas can be fine tuned to be more accurate early on.

Re: QRF vs. RPI vs. Computer Ranking vs. Polls

PostPosted: Wed Jan 10, 2018 12:28 pm
by Run4Fun2009
Flip wrote:if the QRF is the same formula MN uses margin of victory is not used. Run will know more.


margin of victory is not used in QRF...confirmed

Re: QRF vs. RPI vs. Computer Ranking vs. Polls

PostPosted: Wed Jan 10, 2018 1:50 pm
by The Schwab
I use the computer rankings/QRF/RPI/Polls as a conversation piece, that's about all of the stock I put in them. There's no way to make a perfect ranking formula because of the parity across the state. There are so many conflicting styles across the state that lead to upsets and closer scores than you would expect in a lot of cases. There's also no real way to determine the quality of a win. In South Dakota they use power points which gives a point value to a win or loss based on the winning percentage of the opponent. They also reward you for playing teams in larger classes with extra power points. I like this formula a lot, but to full implement their system would require us to look at a 3 class system...but alas, I digress....

Re: QRF vs. RPI vs. Computer Ranking vs. Polls

PostPosted: Wed Jan 10, 2018 2:36 pm
by Bison-Vikes #1
Run4Fun2009 wrote:
Flip wrote:if the QRF is the same formula MN uses margin of victory is not used. Run will know more.


margin of victory is not used in QRF...confirmed


I went back and dug a bit deeper. They quit using margin of victory in 2011-12. Thanks for input.