What is RPI in college basketball? Explaining the sports rating system

The rating percentage index (RPI) is a sports rating system used to help the selection and seeding of teams that will compete in the NCAA Division I men's basketball tournament. It is based on the team's wins and losses and the team's strength of schedule.

The NCAA selectors used this system from 1981-2018 until the league replaced it with the NCAA Evaluation Tool (NET) beginning with the 2018-19 season. Let's take a look at how RPI is calculated and how it is different from NET.

---

How is RPI calculated?

The RPI consists of the following: 25% of the team's winning percentage, 50% of the opponents' winning percentage and 25% of the winning percentage of those opponents' opponents. The formula to get the RPI is as follows:

RPI = (WP * 0.25) + (OWP * 0.50) + (OOWP * 0.25)

WP - Winning Percentage

OWP - Opponents' Winning Percentage

OOWP - Opponents' Opponents' Winning Percentage

To get the WP, the team's total wins are divided by the number of games played during the season.

As for the strength of schedule, the opponents' winning percentage and the winning percentage of the opponent's opponents are calculated.

The strength of schedule represents 75% of the RPI and the remaining makes up the team's winning percentage.

In 2004, officials of the NCAA Division I men's basketball tournament decided to update the WP factor of the RPI to take note of the differences in home, away and neutral games. A victory at home is equivalent to 0.6 win while a win at the road is equal to 1.4 wins.

On the flipside, a home loss is 1.4 losses and if the loss is on the road, the team gets 0.6 loss. A game held in a neutral court counts as one win or one loss.

The 2004 calculation was made after research which showed home teams win about two-thirds of the scheduled matches in Division I basketball. This revision, however, applies only to the WP factor.

Meanwhile, the OWP is taken from the average WPs for each of the team's opponents granted that all games against the team are removed from the equation. On the other hand, the OOWP is obtained from the average of each opponent's OWP.

In August 2018, the NCAA announced it would use the NET as its rating system, replacing the RPI.

---

Difference between RPI and NET

The Houston Cougars placed second in NET behind UConn in the season-ending rankings.

The RPI uses a formula that heavily relies on the strength of schedule. The NET, meanwhile, is based on a two-component metric — Team Value Index and Adjusted Net Efficiency Rating — to calculate the teams heading to March Madness.

The RPI only uses the win/loss percentage, meaning that other statistical metrics aren't a factor in this system. The heavy regard to the OWP meant that losing to a team with a low RPI would not affect your RPI rating while winning over a squad with a bad RPI could negatively impact your RPI.

In the 2020-21 season, the NCAA Division I Men's Basketball Committee simplified the NET tool and adopted the TVI and the adjusted net efficiency rating.

According to the NCAA website, TVI rewards teams for beating quality opponents, especially if they are away from home. The adjusted efficiency, on the other hand, is calculated via team’s net efficiency, adjusted for strength of opponent and location (home/away/neutral) across all games played.

Unlike RPI, the NET rankings factor in the game results, strength of schedule, location, net offensive and defensive efficiency and the quality of wins and losses.

The overall and non-conference strength of schedule has also been amended to gauge the difficulty of beating opponents in one game. The SOS factors in the opponent's strength and site of each game, wherein a difficulty score is given to every match.

The total of the overall and non-conference strength of schedule will gain a better measure of the SOS as it calculates the results of all games in an overall expected win percentage against the team's schedule.