RPI

[cycloneworld]

I posted something similar in the bracketology thread but thought it deserved its own. Is it just me, or does it seem the RPI rankings are vastly overrated (but still used heavily in the media)? Take a look at this year's RPI rankings (from ESPN).

NCAA College Basketball RPI Rankings - ESPN

Everyone wants "quality wins" and Top 25 and Top 50 wins but digging into the RPI rankings, they don't seem to matter nearly as much. For example

Who has a better resume?

Team A
Record: 17-8
SOS: 4
Top 25: 1-4
Top 50: 2-6
Top 100: 9-8


Team B
Record: 24-4
SOS: 104
Top 25: 0-1
Top 50: 0-2
Top 100: 2-3

Team C
Record: 18-8
SOS: 77
Top 25: 2-3
Top 50: 2-5
Top 100: 6-6


It seems to me that Team C and Team A have similar resumes (except for SOS, which I'll go into in the next post) and that Team B has no quality wins. But:

Team A - Oklahoma - RPI 18
Team B - Middle Tennessee State - RPI 23
Team C - Iowa State - RPI 44

It seems to me that Top 50 wins should be the most important factor when determining the field, but so many people talk about RPI.

[cycloneworld]

Looking at strength of schedule, how is Oklahoma #4 in the nation? Here is their non-con schedule:

ULM
Texas-Arlington
UTEP
Gonzaga
West Virginia
Oral Roberts
NW State
@ Arkansas
Texas A&M
Steven F Austin
Ohio
Texas A&M - CC

Sure they played Gonzaga (but they got beat by 25), they also lost at Arkansas, and at home to Steven F Austin. Yet somehow they have the 4th best SOS? To me, it doesn't matter if you play a tough schedule if you lose the tough games...

[Cyclonepride]

Trying to figure out our 4 losses in the 100+. TT, UT, maybe Iowa? And who?

[GMackey32]

100+ for ISU should be 12-2, not 8-4. 8-4 is the L12.

[bosco]

I wouldn't say it is vastly overated. It has it's limitations and shouldn't be the sole gauge of a tournament team but it does have it's uses. It basically takes in account who you played and if you won. Using the RPI in conjunction with the BPI can give you a better idea of how good a team in is. The BPI takes in account the actual performance in the game not just the outcome. For example in ESPN's BPI they currently have ISU ranked 33, MTS 44, and OU 38.

NCAA College Basketball BPI Rankings - ESPN

[cycloneworld]

Trying to figure out our 4 losses in the 100+. TT, UT, maybe Iowa? And who?

I misread a column (edited the post), we don't have 4 100+ losses.

If you click on each team in the ESPN link, you can see everyone's wins and losses by category. Our losses:

101-150 - Texas
150+ - Texas Tech

Iowa is 79.

[CtownCyclone]

Trying to figure out our 4 losses in the 100+. TT, UT, maybe Iowa? And who?

I can only find TT and UT in the 100+ range. Even Iowa is inside 100 now.

NCAA College Basketball RPI Rankings - ESPN

[cycloneworld]

100+ for ISU should be 12-2, not 8-4. 8-4 is the L12.

Yep, fixed.

[Cincyclone]

Since top 50 wins are probably the most important criteria in the selection process I'm not sure I see your point.

[clonedude]

Looking at strength of schedule, how is Oklahoma #4 in the nation? Here is their non-con schedule:

ULM
Texas-Arlington
UTEP
Gonzaga
West Virginia
Oral Roberts
NW State
@ Arkansas
Texas A&M
Steven F Austin
Ohio
Texas A&M - CC

Sure they played Gonzaga (but they got beat by 25), they also lost at Arkansas, and at home to Steven F Austin. Yet somehow they have the 4th best SOS? To me, it doesn't matter if you play a tough schedule if you lose the tough games...

Just for clarification.... West Virginia is a conference game. Yeah, I'm still trying to get used to it too. Doesn't seem right.

[cyclonepower]

Trying to figure out our 4 losses in the 100+. TT, UT, maybe Iowa? And who?

We only have 2 top 100+ losses, Texas and Texas Tech...(Although since Tech is top 200 do you count it as 2??) Iowa is in the high 70's so they aren't close to being a bad loss...Also of not Fl Gulf Coat is hovering at 101, if they could sneak up a 2 spots they are a top 100 win...

[cycloneworld]

I wouldn't say it is vastly overated. It has it's limitations and shouldn't be the sole gauge of a tournament team but it does have it's uses. It basically takes in account who you played and if you won. Using the RPI in conjunction with the BPI can give you a better idea of how good a team in is. The BPI takes in account the actual performance in the game not just the outcome. For example in ESPN's BPI they currently have ISU ranked 33, MTS 44, and OU 38.

NCAA College Basketball BPI Rankings - ESPN

That seems much more reasonable...yet 90% of the media's discussion uses RPI.

[cyclonepower]

Just for clarification.... West Virginia is a conference game. Yeah, I'm still trying to get used to it too. Doesn't seem right.

Actually that was in a tournament so technically it was a non-con game, they played them 3 times this year.

[cycloneworld]

Just for clarification.... West Virginia is a conference game. Yeah, I'm still trying to get used to it too. Doesn't seem right.

Not for Oklahoma. They had a previously scheduled non-con game with WV before they switched. So that game was actually a non-conference game. And then they play then 2 more times in conference.

[GMackey32]

Just for clarification.... West Virginia is a conference game. Yeah, I'm still trying to get used to it too. Doesn't seem right.

It was counted in their non-con because of weird scheduling. They'll play them no less than 3 times this year.