Replaces RPI/Pairwise For Selection, Seeding of National Tournament
CHN Staff Report
The NCAA Men’s Ice Committee formally approved the move to NPI (NCAA Percentage Index) for use in selecting and seeding the NCAA Tournament. It replaces the Pairwise system, which, in conjunction with the old RPI (Ratings Percentage Index) as the main component, was in place for over 30 years, in various forms.
CHN first reported on this in May.
The NPI is similar in many ways, and the ultimately differences will be slight. In a study of the last four seasons, the same teams made the NCAAs, but the seeding order was different in some cases.
The NPI — like RPI — is a mathematical system that attempts to factor in a team’s strength of schedule to their overall winning percentage, and comes up with a ranking of all teams nationally.
In the past, RPI was just one component of the Pairwise, albeit in recent years, a heavily dominant one as most other criteria were eliminated. In the last few years, besides RPI, the Record vs. Common Opponents and Head-to-Head record were still components of the Pairwise.
Though the NPI now stands alone, it will continue to be adjusted for various factors, just as the RPI was.
For example, the home/road weighting will remain 1.2/0.8 — i.e. a home win will only be worth 0.8 of a win, while a road win is 1.2 of a win. However, this weighting will no longer pertain to conference tournament games, a welcome change for many coaches who believed it was penalizing a team for earning home ice for its conference playoffs.
The overtime factor has been tweaked slightly. Before an OT win was worth two-thirds, now it will be 0.6.
The Quality Win Bonus will be tweaked slightly, but will effectively remain the same.
The requirement that a team has a winning percentage of .500 or better to qualify for the NCAAs, has been removed.
CHN is working on a new primer that will break down all of the particulars. This is the Pairwise Primer that explains the system as it was. The system for creating the seedings from the NPI will remain the same as it was with the Pairwise.
The biggest way the NPI differs from the Pairwise is that it’s a singular component, thus there is no need for the system of “comparisons” that previously existed. In calculating the a team’s strength of schedule in the new system, it does so recursively, more like the way KRACH works, rather than the old RPI way, which was to take a certain percentage of opponents’ winning percentage, and opponents opponents’ winning percentage.Â
With NPI, a team’s winning percentage is 25 percent of the calculation, and its strength of schedule is 75 percent. That’s another thing that can be adjusted — for example, other sports are using 20/80.
There is a final factor, which is “minimum wins,” which is 12. This can be confusing, because it doesn’t mean a team needs 12 wins to qualify. It is related to the concept of throwing out “bad wins” from the NPI. In the past, any win that actually had the affect of lowering that team’s RPI (because their opponent was very poor and thus the poor strength of schedule overshadowed the positive on a team’s winning percentage), were tossed out of the RPI calculation. Now, a team must count at least 12 wins, regardless of whether those wins lowered their NPI. Once they’ve counted at least 12, a team can throw out its remaining “bad wins.”