The unofficial online home for fans of the UAB Blazers. Part of Collegiate Sports Nation (CSN).

I was looking at realtimerpi the other day, and as I saw the list of top 10 teams, I noticed something. A lot of people like to bash the RPI, and say it is manipulated, and shouldn't be used anymore. While looking through the rankings, I saw that 11 of the top 15 RPI teams(Duke, Louisville, Miami, Kansas, Florida, Indiana, MSU, OSU, Marquette, Syracuse, and Arizona) are still in the tournament. It is also a bit odd that of the 4 teams that lost that were in the top 15, only Georgetown was from a "power conference".

Then again, the remaining 5 teams are RPI 21(Michigan), 37(WSU), 46(La Salle) 47(Oregon), and 93(Florida Gulf Coast).

I understand that a lot of the new metrics have new/different value for the committee to use. I also understand that most of these teams were highly seeded in friendly environments, but I think this still indicates that the RPI does have more value than some people think.

I like the RPI over the BPI or other rankings.

(03-27-2013 08:03 AM)blazers9911 Wrote:  I was looking at realtimerpi the other day, and as I saw the list of top 10 teams, I noticed something. A lot of people like to bash the RPI, and say it is manipulated, and shouldn't be used anymore. While looking through the rankings, I saw that 11 of the top 15 RPI teams(Duke, Louisville, Miami, Kansas, Florida, Indiana, MSU, OSU, Marquette, Syracuse, and Arizona) are still in the tournament. It is also a bit odd that of the 4 teams that lost that were in the top 15, only Georgetown was from a "power conference".

Then again, the remaining 5 teams are RPI 21(Michigan), 37(WSU), 46(La Salle) 47(Oregon), and 93(Florida Gulf Coast).

I understand that a lot of the new metrics have new/different value for the committee to use. I also understand that most of these teams were highly seeded in friendly environments, but I think this still indicates that the RPI does have more value than some people think.

I agree that the RPI has value, but the seeding in friendly environments is a bigger reason than any other. It's similar to the talking heads this weekend with the incessant fraudulent argument about how the Big 10 is "proving" it's the best conference. The argument is completely self-perpetuating. The Big 10 was viewed as very strong. Thus, its teams got high seeds and played in hyper-friendly environments. But the highest seed any Big 10 team actually beat was a 6 (Memphis and over-seeded UCLA). No Big 10 team except Minnesota beat a higher-seeded team, and that was viewed as such as great win that Minnesota promptly fired its coach. And that win was offset by the fact that Wisconsin lost to a 12 seed. Indiana and Ohio State barely advanced against two double-digit seeds. Only two Big 10 teams really performed well. Yet, because it has four teams in the Sweet 16, it's viewed as the best conference.

And it may be, but that isn't why. The Big 10 teams all won games they should, have not yet won any games they shouldn't, and really didn't impress against lower-quality opponents. The system set them up to easily advance. But, of course, that's what the system is designed to do. It just doesn't really prove anything. I guess my point is you could randomly put 15 other teams in the top of the RPI rankings, and they would likely have done just as well if they were also given high seeds and friendly environments.

That doesn't mean your point that the RPI has value is wrong; in fact, I agree with the premise. Again, the tournament is designed to reward the best teams with easier roads. And all those teams seem like really solid Sweet 16 teams (save maybe Marquette). I just think it's circular logic to say the fact that those teams remain proves the RPI's accuracy given the way the tournament is structured. If the top 15 had played the second 15, my guess is there'd be a pretty even split.

I agree with a lot of what you are saying. Personally, I believe the Big Ten is the best conference top to bottom this year, but that doesn't always mean you have the best singular team in your conference. The bottom line is every team in the country has flaws this year, and most have been exposed at one point or another. And I do agree that the friendly environments tend to have a lot to do with teams winning, especially as the competition gets closer and closer. But the system is set up that way.

For what it's worth, Louisville came out of Lexington, Oregon out of San Jose, Michigan State out of Auburn Hills, Duke out of Philly, Kansas out of Kansas City, Michigan out of Auburn Hills, Florida out of Austin, Florida Gulf Coast out of Philly, Indiana out of Dayton, Syracuse out of San Jose, Marquette out of Lexington, Miami out of Austin, Wichita State out of Salt Lake City, La Salle out of Kansas City, Arizona out of Salt Lake City, and Ohio State out of Dayton.

Duike, Florida, Florida Gulf Coast, Syracuse, Miami, Wichita State, and La Salle were the geographic oddities you could point out, and some of those weren't in a hostile environment, just more of a neutral environment.

And just for fun, the top 15 did play the second 15 some. UNC lost to Kansas. Colorado State lost to Louisville. Belmont lost to Arizona. Butler lost to Marquette. All 4 of those games went to the higher ranked RPI team.

And I can see how you would say the logic is a bit circular. I wasn't really trying to say the teams were there because they were the top RPI teams, I was just pointing out that the RPI does have some validity, and I think you agree with that.

And for comparison's sake, Kenpom also had 11 of the top 15 teams in the Sweet 16 in his rankings.(oddly, both rankings had Arizona in the top 15, indicating they may have been underseeded) The only real oddball in his ranking was Florida Gulf Coast at 92, and then La Salle was next lowest at 47. All the other teams were in his top 31.

I think they are all good, valid tools at this point. If you look at a team that has an RPI in the 20's and a Kenpom rank in the 50's, then I would say it would require a more in depth look. And vice versa.

Things are always changing, I was just merely trying to point out I don't think the RPI should be left for dead at this point, like some people seem to think.

Regarding FGCU, their KenPom rating was 144 at the end of the regular season. It improved 20 points (124) after winning the Atlantic Sun Conference Tournament. It went to 107 after beating Georgetown, and 92 after beating San Diego State.

KenPom has Florida favored by 19 points over FGCU. Of course, KenPom had FGCU as a huge underdog in the past two games as well.

I don't have a problem with the RPI, I just find KenPom to be a lot more sophisticated. I'm glad all data sets are considered in making selections.