The RPI or Ratings Percentage Index is a metric used by the NCAA Selection Committee as an aid in choosing schools for the 34 at-large spots in the post-season tournament.   But starting with the 2013 season, a new RPI is being employed.

The purpose of the change in the RPI this season is to reduce the road disadvantage faced by northern teams that must play a large part of their early season schedules on the road while southern teams play mainly at home.  Now the RPI is  weighted for wins and losses, on the road and at home.  A road win is now worth 1.3 instead of 1.0 in determining the RPI rankings and a home win is 0.7 instead of 1.0.  Under this formula weighting, a home loss counts as 1.3 in reducing the RPI and a road loss will count as 0.7 in reducing the RPI.  A neutral site game will count as 1.0 and will not be weighted.  Bonuses that used to be part of the RPI formula have been eliminated as well.

Michael Conforto at 1st base (click to enlarge)

The RPI formula is an algorithm made up of three main components:

1 – Winning percentage – makes up 25% contribution to the formula.  Here’s where the big changes in the formula have taken place.  Under the old formula, if the winning percentage is 0.950 (OSU’s current 19-1 record), then 0.950 x 0.25 = 0.238 or 25% of the total RPI.  But now the winning percentage is adjusted for where the wins and losses take place – at home or on the road.  OSU’s adjusted winning percentage is 0.935 (18.7 wins and 1.3 losses).  Since the loss to San Francisco took place at home, the loss totals are adjusted upward to 1.3 from 1.

2 – Opponent’s winning percentage – makes up 50% of the formula.  Sometimes a team’s RPI can drop even when winning a game over a low RPI opponent because that opponent had a poor winning percentage.  Conversely, the RPI can rise even when losing to an opponent with a high winning percentage.  Playing a strong schedule is important for achieving a high RPI.

3 – Opponent’s, opponents winning percentage – this makes up the final 25% of the formula.  The RPI for a particular team can move up with a win over an opponent that has played other teams with high winning percentages.

To calculate the RPI, the opponent’s and opponent’s opponents winning percentage or sometimes referred to as the strength of schedule is determined.  In the current OSU example, the strength of schedule component is 0.551.  Thus, 0.75 x 0.551 = 0.413.  The RPI is then calculated as the sum of the adjusted winning percentage component and the strength of schedule (0.234 + 0.413 = 0.647).

Teams are ranked by the RPI values, and in this case the RPI ranking is 3 so only 2 teams had higher RPIs.  A common mistake made by fans and members of the media is to refer to the RPI rankings as the RPI, though it is the rankings and not the indexed number that is important in comparing potential candidates for post-season play.