The NCAA is making changes with its main metrics model for team performance again. Just two years ago, it introduced the NET Rankings to replace the Ratings Percentage Index. Both aren’t used to qualify or disqualify a team from the postseason, but are used as one of the main tools used to evaluate and stack rank teams.
The first year of the NET Rankings resulted in some confusion and criticism — especially in the early releases — and now it appears that three of the original five components are being discontinued. The two that will be used going forward are Team Value Index (TVI) and adjusted efficiency. The three components being discontinued include winning percentage, adjusted winning percentage, and scoring margin.
According to Joe Lunardi of ESPN, the best way to think of these two components going forward are “who” did you play, and “how” did you play.
From Lunardi:
“RPI was primarily a measure of the “‘who,'” whereas the first iteration of NET was a reasonable attempt at measuring the “how.” This newly revised and still undisclosed NET formula appears to be an attempt to combine the two, while retaining an increased emphasis on the “‘how'” and perhaps eliminating some of the less-than-credible early-season NET releases.”
By the way, both the men’s game and women’s game will be affected by this change. The NCAA women had been still using the RPI even with the move to the men using the NET Rankings. However, the women announced earlier this month that they would also be switching over.
Breaking down the tweaks to the NET with Dan Gavitt for @marchmadness @NCAA: https://t.co/Om9gg4kZwe
— Andy Katz (@TheAndyKatz) May 12, 2020
So there you have it. The ever-evolving way of quantifying and qualifying teams for the NCAA Tournament continued to get tweaked. One school of thought is that correcting things is the right way to go so that the secret sauce is more reliable. Another is that if you keep changing metrics, you simply don’t know what you are doing.
Which side of the fence do you land on?