The purple rankings are just as familiar – and often controversial – as every subject about college hockey for the most of three decades.
The term, conceived for the first time by three self-described “Stat Geeks” from the University of Minnesota (who started starting with USCHO about 30 years ago), this season goes the way of the VHS tape, since the NCAA Men’s Division I Ice Hockey Committee changes how teams are selected for the NCAA NCAAA-Kam-Kon-Kalings.
The NCAA Power Index, simply known as the NPI, will replace the purple rankings as the selection process for the 2026 championship in March.
It is a movement that has come for a long time, a movement that coaches have called to change for a number of years in the hope of finding a sort of ranking system that teams quantify reasonably and evenly to collect the NCAA touring field.
So what is different and how much difference will this new system make?
If you come down to it, things can look alike.
Ushcho recently spoke with Tim Danehy, better known in the universe of the college hockey as the person who once operated the beloved website Collegehockeystats.net for almost two decades. He is often called to place his very talented mathematical brain to use in an advisory role from the NCAA and ice hockey, especially when it comes to how teams are selected for the NCAA tournament.
Danehy does acknowledge that the NPI may not produce any other field of 16 teams in March than the pair could be. But he says that one of the most important characteristics of the NPI is the ability to eliminate some statistical abnormalities that impedes it pure.
As an example, Danehy points to last season and the comparison of Ohio State and Princeton. The Buckeyes finished the season in 10th place in pair. The rating point index of the team, or RPI (who measure your success against your opponents, taking into account your schemsterkte) .5594, also in 10th place.
Princeton ended the 43rd season in pairs, with an RPI of .4830. But the Tigers won at home twice at the end of November against Ohio State, both due to the score of 3-1.
Now let’s look at the criteria of the purple ranking to better understand how Princeton and Ohio State relate to each other.
There are three criteria that the pair has been used for a number of years:
- RPI
- Common opponents
- Head-to-head
One point is awarded for every criteria won, insight into that every victory of head-to-head deserves a criteria point. If there is a draw in the total criteria won, the team wins the comparison with the best RPI.
Ohio State wins the RPI criteria well, but because Ohio State and Princeton had no common opponents, those two losses made it impossible for the Buckeyes in November to win the comparison against Princeton.
“The purple comparison says that Princeton is better,” said Danehy, and noted that almost any other measurement would have the state of Ohio ahead. “There is a huge RPI gorge, but [in the PairWise Ranked] That doesn’t matter. “
They are statistical anomalies that the NPI wants to tackle, and this does this by eliminating certain games that can negatively the position of a team – and can also positively influence.
In the NPI system, games that influence the ranking of a team negatively are not only counted against the team, but are systematically evaluated and possibly removed. Here is how it works:
1. Initially, all games are counted in the calculation
2. The system then identifies the victories with the lowest value
3. Games under a certain threshold can be removed from the calculation
4. There is a minimal victory requirement (12 wins) to prevent teams from eliminating their entire schedule
5. When a game is removed, it is as if the game has never been played – teams are not punished, only eliminated
6. The removal process is dynamic and iterative, which means:
– The removal of each game influences the calculations of other teams
– The process repeats itself several times, so that the ranking stabilizes
The aim is to prevent teams from being artificially stimulated or punished by extremely weak or strong individual competitions while maintaining a representative seasonal performance statistics.
Rather, teams are rewarded for playing stronger teams.
“We immediately solve that disconnection,” said Danehy. “If you are ranked higher than me, you are worth more to the power of schedule than me. I can no longer help my strength of schedule by playing someone below.”
These changes are not allowed, and in all probability, not create a radical shift in the ranking. Instead, according to Danehy, it will yield a more defensible mathematical method.
And this system has already been tested. Division III has used the NPI in all sports (in fact here is the link to an excellent podcast that explains NPI in -depth: https://www.youtube.com/watch?v=Xrkp8mcryua). And Women’s Division I Hockey replaced it purely with the NPI last season.
Now that the season is ready to start, we are still a long way to worry about which teams the NCAA tournament will and will not make. But if you are looking for Google for “Purple -like rankings” next March, you will not deliver much, change the search term in “NPI”.
#NPI #replace #pair #wise