NCAA Bracket 2011: How Much Should RPI Factor in the Selection Process?

SchmolikCorrespondent IIMarch 14, 2011

GREENSBORO, NC - MARCH 10:  Head coach Seth Greenberg of the Virginia Tech Hokies reacts during the second half of the game against the Georgia Tech Yellow Jackets in the first round of the 2011 ACC men's basketball tournament at the Greensboro Coliseum on March 10, 2011 in Greensboro, North Carolina.  (Photo by Streeter Lecka/Getty Images)
Streeter Lecka/Getty Images

Hello, college basketball fans!

As a self-proclaimed bracketologist, I nitpick and criticize the NCAA selection process more than many NCAA Tournament fans. So I want to discuss the use of RPI in the selection process.

The RPI was supposedly created to put some objectivity in a subjective process. Without it, how in the world can you compare well over 300 teams, from 31 conferences, that have automatic bids in the tournament?

The RPI formula is pretty basic, 25 percent of a team's win/loss record, 50 percent of the record of the teams' opponents and 25 percent of the records of the teams a team's opponent played. Calculating it would obviously be very difficult, unless you have a software program to calculate it, but it certainly would be doable.

Recently, the RPI was altered to account for home and road games. The RPI now includes 0.6 and 1.4 factors. A road loss and a home win counts as 0.6 of a win or loss but a home loss or a road win counts as 1.4 of a win or loss. It was designed to encourage teams to play more road and/or neutral games, as opposed to loading up on home games.

However, since the RPI formula was changed, it seems that the RPI is less of a factor in the selection of NCAA at-large bids. While it never was or never was intended to be the sole factor in determining teams, I think it is a mistake to go further away from it. In one of the first few seasons since they changed the formula, a team had an RPI of 21 and was left out of the Big Dance.

Apparently, it seems like the public does not value the RPI that much either.

According to the numbers I saw on Sunday from CBS, a team that had an RPI of 35 (Harvard) was left out of the Big Dance. Almost no one had a problem with them being left out. Another team that got in, UAB, had an RPI of 31 and many people thought they were one of the most, if not most, undeserving teams chosen. I think someone on ESPN actually accused UAB of "gaming the system".

Meanwhile, two of the teams that many people thought got screwed? Virginia Tech and Colorado. RPI's of these teams: Virginia Tech had an RPI of 61, Colorado? 65. Another snub? Alabama, whose RPI was 73 (CBS had it has 76, I saw on TV it was 80).

I don't think the Selection Committee should use the RPI 100 percent and choose at-large teams based entirely on RPI. But I think it should be used more than I feel it is.

If people are saying a team with an RPI of 61 got screwed and a team with an RPI of 31 is a joke of an NCAA team, I think we are heading down a slippery slope. Eventually, the committee decisions will be less objective and more subjective.

For everyone who does criticize the RPI, let me just remind you of an important fact. One of the most used reasons to justify a team should or shouldn't be in the field is "quality wins" (the more, the better). Then again, what are all of you using to determine who counts as a quality win? That would be ...the RPI!

The fallacy of the argument is that any team that beat my Illini can be credited with a quality win, since our RPI was 45, but of course we shouldn't get credit for that rating. UAB's 31 RPI rating means absolutely nothing to most people, but if anyone beats UAB, we say "good job". Am I the only one who thinks that is backward logic?

Also, the "quality win" logic creates a bigger gap between the "power conferences" and the "mid majors". This year, the Big East had 11 teams in the NCAA tourney, including 10 teams with RPI's in the top 40. Look at every Big East team and they all have a lot of Top 50 or even Top 25 wins. But they played a lot of games.

By contrast, teams like Utah State, Missouri State, Harvard and UAB aren't even allowed to play that many games because their conferences aren't that good. How many teams were in the Top 50 RPI from the MVC? One. Missouri State can't play itself.

You can argue "they can always play these teams in non conference play". Can they? It's much easier for the big boys to schedule and to play more home games, which lead to more wins. How many power conference teams came to play Utah State or Missouri State? Probably not many.

It's bad enough to me that the RPI itself favors the power conferences (better strength of schedule), but when a mid major team like UAB actually has a high RPI, the media slams them and say they are "gaming the system".

My argument? Why can't Virginia Tech game the same system? Why can't Colorado game the system? How come only Conference USA teams take advantage of the RPI?

I think everyone in college basketball knows the system (schedule well). You would think it would be easier for a Colorado or a Virginia Tech to get quality opponents. If I know Virginia Tech will play Duke and North Carolina (depending on the year one or both possibly twice), I'd have more reason to schedule them than UAB, who probably won't play anyone else.

You would also think that Virginia Tech would have learned to schedule. They've had an above .500 ACC record three of the last four seasons. Has their RPI been above 50 in any of those seasons?

The problem is either them or the ACC in general. This season, only two teams in the ACC had RPI's in the Top 50 (of course Duke and North Carolina). The Ivy League and the Horizon League had two teams in the Top 50.

Supposedly the ACC is one of the best conferences in basketball. Forget the mid majors, the SEC and Pac-10 each had more teams in the Top 50 RPI than the ACC. Florida State had a record of 11-5 in the ACC and beat Duke, but not even they have an RPI in the Top 50. Maybe the problem is the ACC in general and they should stop scheduling the "Little Sisters of the Poor".

Another by product of the "quality win" only criteria. One team that I think was a joke of a bid, but no one seems to be complaining about was USC. Their RPI was the highest of any team (69), higher than Virginia Tech or Colorado.

USC had five Top 50 wins. They also lost six games to teams with RPI's below 100, including two HOME losses and three losses to teams with RPI's below 200! USC wants us to say their five good wins are important but let's forget about the other six bad losses.

Shouldn't a team be judged on their complete "body of work," instead of just the games that they are their best? Is USC the team that beat Texas or the team that lost to Bradley?

I think the correct answer is they're somewhere in between.

I'm not saying the RPI is perfect, because it's not. But I would rather judge teams on their entire season, rather than the four or five games each team wants us to concentrate on. USC's best five games are better than UAB's best five games, Harvard's best five games and Missouri State's best five games. But all three teams and many others had the better entire season, at least according to the RPI.

Just remember, if you don't have an RPI, how do determine what a quality win is? If the RPI formula stinks, do you have a better one? Do you want to judge over 300 teams with no formula at all? Good luck with that.

I think the RPI is better than those computer rankings the BCS uses. Aren't some of the formulas not even released to the public? They are judging you using criteria that no one knows about. Everyone in college basketball knows how the RPI works and if you are "gaming the system" what's wronGoog with that?