If you'd like to follow along, fill out two brackets: in bracket A, always select the lower seeded team (i.e. #8 beats #9, etc.) all the way to the FF, then let RPI take over. In bracket B, the better RPI (the NCAA's version) governs who wins and who loses.
In the 2011 bracket, assuming the team with the better RPI always wins, there will be seven upsets (i.e. the lower seeded team loses). Two of these games involve BYU, which is understandable given the circumstances of their recent "dismissal." Two involve #9 seeds upsetting #8 seeds, again not surprising since these games are always a toss-up. One game has SDSU upsetting Duke; however they are #3 and #4, respectively in RPI, so nothing monumental aside from the "smell test" of a mid-major beating Duke.
The remaining two upsets I would throw into the "significant" category and involve one team -- Utah State (RPI: 15), a 12-seed. They would pull two upsets, Kansas State (RPI: 23) and Buckytown (RPI: 16), before losing to Pitt.
I haven't repeated this exercise for previous years, but my notable observations:
1) Only two "RPI upsets" seems pretty low. It would seem to me that a soft bubble should produce many more RPI upsets. If I can get my hands on previous years' RPI numbers, I'll look at how this played out in last year's soft bubble.
2) My guess is that Utah St is the best RPI to earn a 12-seed ever. Maybe even the best RPI to get a double-digit seed ever. Plus, their resume is probably as good as it can be at a low/mid-major program. If they go to the Sweet 16, the world will watch ignorantly for Wild Bill to dress up as Cinderella... what he should dress up as is The Beast (as in Beauty and the...) because that's what USU is in the RPI - a beastly 15.
3) Pitt's RPI of 10 is probably the lowest RPI a #1 seed has ever received; then again, maybe not, but it has to be close. So much for the anti-Big East bias/conspiracy theories.
4) #2 and #3 above notwithstanding, one could conclude that all the talk (mine included) about RPI being over-hyped was certainly dispelled this year in the selection of Florida as a 2-seed, inclusion of UAB (RPI: 31), etc. However according to that logic, there's no way to explain Harvard's snub, Marquette's and USC's inclusion, and Michigan as an 8-seed.
5) Highest RPI teams to be snubbed: Harvard (25), Cleve-land St (41), MO State (42), St. Mary's (46), CO State (51). Lowest RPI at-large selections: USC (69), Marquette (64), Clemson (56), FL State (54), Michigan (50).
Quote from: Benny B on March 14, 2011, 03:46:25 PM
5) Highest RPI teams to be snubbed: Harvard (25), Cleve-land St (41), MO State (42), St. Mary's (46), CO State (51). Lowest RPI at-large selections: USC (69), Marquette (64), Clemson (56), FL State (54), Michigan (50).
This is why I don't think MU gets in if we lost to Providence. Our RPI would have dropped to 76. Not going to get a bid with that RPI.
As far as the other low RPI numbers that got "snubbed", in all but one case is their SOS. Utah State had a great RPI but a 121 SOS. I believe the committee punished them because of that. In effect, Utah State is able to "game" the system by getting 30 wins but they don't play anyone so their RPI is great but their SOS sucks.
Colorado State is the exception to this rule. 35 SOS and an RPI of 54. Seems the committee doesn't like the state of Colorado.
I blame Rocky_WarriorCleveland State, good RPI but SOS 106
Harvard, good RPI, but SOS 140
MO State SOS 127
St. Mary's SOS 101
I guess I shouldn't have sent them that nasty letter earlier in the year.