After looking at the profiles and rankings of many bubble teams over the weekend, only one thing was made clear about the selection process: the RPI needs to go.
It does for college basketball the exact same thing that the BCS does for college football. It merely plugs numbers into a formula and spits out results, who should and who should not play for a national title.
But the difference is, we don't need computers for the NCAA tournament.
65 teams get a chance to go all the way, and could a computer really tell us who numbers 65 and 66 are?
In football, it makes sense. With only two teams playing for the title, the computers are needed to avoid bias. But for the tournament, I don't think many people are going to fix the selection process so that they get in as a 13 seed.
That being said, let's move on to some interesting RPI cases this year. First, let's compare three teams.
Team A: 18-13 (9-10 in conference). RPI: 58. SOS: 33
Team B: 18-14 (9-11). RPI: 38. SOS: 2
Team C: 19-12 (9-10). RPI: 82. SOS: 87
Team C beat Team B twice (home and away) and split with team A (won at home).
Team A is Oregon, a nine seed. Team B is Arizona, a 10 seed. Team C is Arizona State, left out.
Without the computers, Arizona State clearly has the best profile of those three teams.
The Sun Devils demolished Xavier (a third seed) by 22 points. In the current system, teams get more credit for losing to a good team, as Arizona did multiple times, than beating a pretty good team.
I agree that close losses to No. 1 seeds should be taken into account. But a team's profile shouldn't be based on the fact that they almost beat Memphis or Kansas.
Second, I need someone to explain this for me. On February 11th, Kentucky was ranked 87th in the RPI. They then went on the road to Vanderbilt, a good but not great team.
They lost by 41.
When the RPI was re-published the next week, Kentucky had risen 16 spots. They lost to a pretty good team by 41 points, and jumped 16 spots. Any system that causes that to happen is severely flawed.
Cases like these jumped out everywhere this weekend. Check out profiles of teams that got in and teams that didn't. Almost all of the numbers are the same; except for the RPI.
So while the committee may say the RPI isn't that big of a deal, it clearly is. It's time to get rid of it. We simply don't need it.