The RPI - Is it relevant?

Throughout the past years, the Ratings Percentage Index has been a major topic of discussion in college basketball, especially come March. It's used to organize teams based on their quality of victories, losses, strength of schedule, and give a general indicator as to how strong a team is. No other system of numbers or statistics has been discovered that has trumped the RPI.

In terms of it's importance to the selection committee, I believe they use it more frequently than what they admit. They claim they use it as a tool to organize the teams and nothing more. If that were the case then why do we find ourselves year after year breaking down a team's numbers based on this exact formula? If it was so worthless then why are analysts and so-called "experts" breaking down teams based on their record against the RPI Top 50 and their actual percentage index?

Because it is so simple to use, the RPI can be very effective. As qualifying for the NCAA Tournament has always been about the "body of work" of a team attempting to make a case, the RPI makes it easier to discern against teams on the bubble and the difference in the two. It rates your wins, your losses, and is able to rate the quality of conference said team competes in. It helps in the argument of whether Team A or Team B should be in the tournament and who is more deserving.

The main argument against the RPI has always been that one should judge a team based on how they play on the court rather than what their numbers looked like. However, I've always found these statements to be non-mutually exclusive. If a team is truly a good team, their numbers will reflect this, as it means they're accumulating wins against quality opponents. I've always found this statement to be a little odd and easy to de-bunk due to how the numbers are actually calculated.

The other argument against the RPI is how good wins can decrease in value, and bad losses can look better and better as the year goes along. I understand this argument, as it's true. The RPI of a team fluctuates on a daily basis. For the most part it is a legitimate concern when looking at a teams resume. However, if a good win at the time begins to look worse, it usually means that said team has been losing frequently due to it's falling RPI. Wouldn't this mean that the team wasn't very good in the first place? Therefore the win looking worse is justified based on the fact that they didn't turn out to be quite as good as advertised by early numbers, usually inflated by a poor early season schedule (see teams like Clemson last year at 17-0, beating up on low level D-I teams to build up early confidence, who we found out was the definition a pretender).

When looking at it from an opposite standpoint, bad losses looking better and worthless wins looking better, this is definitely a weakness. The simplest example of this is Indiana's win over Kentucky in early December. This was a Kentucky team who, quite honestly, wasn't in the top 150 teams in the country in November/December. Their non-conference RPI was a massive 204 before mounting a comeback in SEC play that has catapulted them to 49th overall. Is Kentucky the same team now that they were then? Absolutely not, therefore it's worthless to count this as a "good" win.

I believe the RPI is effective in being able to compare teams simply based on what they have accomplished during the year. It has its faults, but it works. Those who continue to discount its importance/effectiveness seem to be the same ones who compare teams using these same numbers to support their numbers. Of course you can and should still have your own opinions based on what you've seen the teams do on the court, but what they do on the court directly affects these numbers, so how can it be as worthless as some say?