NPI: Breaking down the newest ranking system in college hockey
Pairwise is gone, so what system will take over?
It was reported in early May that Pairwise would “soon be replaced” by the NPI, known as the NCAA Percentage Index, in determining the playoff field for Division I men’s and women’s college hockey.
Former Penn State men’s hockey insider Josh Bartosik gave fans constant updates about the Nittany Lions’ spot in those rankings.
A few other sports in the NCAA use power indexes for help with playoff selection, with only one other prominent example of this being used as an objective ranking.
What is the NPI? What does it replace? What are some examples of power indexes?
The NPI as a ranking system
This system is much like the RPI – a system that breaks down records against opponents, opponents’ records and opponents’ opponents’ records – in how it weighs wins and losses.
It is similar to the RPI in terms of how home and away wins are calculated, as a home win is considered 0.8 of a win and away wins are counted as 1.2.
One change that the NPI makes is that teams are no longer required to have a .500 win percentage or better to make the playoffs.
The NPI factors in team win percentage and strength of schedule with a 25/75 split to make the ratings.
Lastly, teams need a minimum of 12 wins, not for playoff contention but instead for the number required before they can throw out their ‘bad wins,’ which are any win that may hurt their strength of schedule from beating a poor opponent.
The NPI will be used to select 10 at-large bids across six conferences, and it will then rank the 16 teams in the field.
A simple breakdown of the Pairwise
Pairwise was the system in place for the Division I men’s and women’s college hockey playoffs from the 2013-14 season to the end of the 2024-25 season.
The system used three criteria in its ratings: RPI, record versus common opponents and head-to-head record.
RPI was one of many factors in the former system, weighing a team’s win percentage, average opponents’ win percentage and average opponents’ opponents’ win percentage at a 25-50-25 split.
Pairwise introduced a metric called the Quality Win Bonus, where a team would earn ‘bonus points’ in the RPI for beating a team in the top 20 in the RPI on a decreasing scale from .050 to .0025 on the 1-20 ranking.
How other selection committees have used power indexes
Different NCAA selection committees across different sports use power indexes to help as one of many deciding factors for playoff contention.
The most notable use of this in the past was when the Bowl Championship Series was used for the National Championship game, along with the New Year’s 6 bowl games.
The NCAA Division I men’s and women’s basketball tournament selection committees currently use NET, known as the NCAA Evaluation Tool, for help with selecting teams for the playoff field.
RPI is also one of many factors used in playoff consideration for the NCAA Division I Baseball Tournament selection committee. Despite this, Rutgers had 44 wins and an RPI of 43, and was left out of the field in 2022.
NPI vs. NPI: Comparing it to Division III football’s ranking system
The NCAA Division III Football selection committee draws the biggest parallel to the college hockey committees in terms of how it selects its at-large teams and seeds the field with its own NPI.
The Division III football NPI is called the NCAA Power Index, and it is an objective system that ranks all teams in that level just like hockey does.
Much like in hockey, the football NPI uses home win and away win factors, though those weights are different.
The football NPI has a Quality Win Base instead of a Quality Win Bonus, which is set at 54.00. The number of “competitive teams” included in this metric as opponents goes up to 100.
A team’s win percentage is 40 percent of the split, with strength of schedule at 60 percent.
The football NPI is used to select 13 at-large bids across 27 eligible conferences, and it then ranks the 40 teams in the field.
Bryan Portney is a second-year majoring in broadcast journalism. To contact him, email bep5295@psu.edu.
Credits
- Author
- Bryan Portney
- Photo
- Jim Rosvold