RPI is something you hear about a lot in college basketball. That talk intensifies as the postseason draws near. Most college basketball fans and bettors — even casual ones — have heard of RPI. They may even have a vague sense of what it is. Few people, though, really understand what it is and why it is significant. That’s where we come in.

First, the basics. RPI stands for Ratings Percentage Index. It has been used in college basketball since 1981, and is used in other sports — most notably college baseball. It is not used as a strict criterion in basketball, but rather as a guide used by the selection committee when filling the NCAA Tournament field. The amount they are used is a question of debate each year, and discussion around the RPI of teams that get in and are left out is a source of controversy every season. The NCAA releases their RPI calculations for every team once a week starting in January, but more up-to-date information is now available on several websites.

RPI is a formula that aims to assess the strength of schedule of a team, and how they have fared against that schedule. There are three parts. The first is the win percentage of the team. Next is the winning percentage of the team’s opponents. Finally, you have the winning percentage of the opponent’s opponents. The first and third factors account for a quarter of the RPI, while the opponent’s win percentage makes up half of the RPI.

For a long time there were concerns that the win percentage didn’t account for the location of games. In 2004 a change to the formula was made to compensate for that. It only impacts the team’s win percentage, though, and not anything to do with the opponents. Since then, a home win now only counts as 0.6 wins for the purposes of calculation. A neutral-site win is still counted as one win, while a road win counts as 1.4 wins. That’s one big reason why road wins are so valued in college basketball.

As with anything to do with applying formulas to sports, there is all sorts of criticism centered on the RPI. One of the complaints is that the margin of victory does not get factored in. A team that wins by a point at the buzzer gets just as much credit as one that dominates their game and wins by 20 points.

The lack of the margin of victory in the calculation limits the usefulness of the RPI for bettors. Without knowing how teams win you can’t rely on the RPI as a meaningful predictor of who will win, or especially whether they are likely to cover the spread or not. Bettors ultimately really don’t care too much about which team wins the game — unless they are betting the moneyline, that is.

Another big criticism is that the RPI tends to favor major conference teams. The biggest reason for this is that major conference teams have much more control over their nonconference schedule than minor conference teams do. That makes it easier for them to schedule games that they are going to win, and that will build a RPI foundation early in the season that gives them an edge right to the end. When most or all of the teams in a major conference are able to pick and choose their nonconference schedule like that the RPI effect is magnified in conference play because the early season success is shared amongst the teams in the conference as they play each other. Some minor conferences have been able to effectively overcome this obstacle by having their best teams aggressively schedule in nonconference play.

Despite the issues with the RPI as it relates to college basketball handicapping there are a couple ways that it can still be useful for bettors who understand it.

The first is that it can provide a general sense of how a team you are unfamiliar with stacks up. You don’t want to make a judgment about a team based just on their RPI, but by looking at the RPI you can get a basic sense of what caliber they are. You probably want to assess elite teams differently than weak teams that will be looking to cover massive spreads, so the RPI can be your first clue on how to approach a team.

More significantly, the RPI can give you a sense of how the public is going to view the team.

The RPI gets mentioned frequently, so a team with a high RPI is going to get respect from public bettors, and a low RPI will make it hard for a team to get respect. This can be particularly helpful for bettors when the relative ranking of the RPI doesn’t seem to accurately represent how good the team is, or how ready they are to play at this time.

A team with a high RPI may just be benefiting from playing in a tough conference or playing a soft nonconference schedule. They may not actually be particularly strong, or injuries or team changes could mean that they aren’t playing now like they were when they laid the foundation for their RPI. Because the RPI is a full-season statistic it is very slow to reflect changes in the way a team is playing. If the RPI is too high or too low for a team then you could potentially find nice value because the public is judging the team by an inaccurate indicator.

In general, though, RPI is a statistic that bettors are probably better served to ignore. If strength of schedule is something you want to consider then there are other statistics that do a much better job of giving a clear picture of how strong a team really is. Leave the RPI talk to the media, and spend your time instead learning about those stats and how to use them.