The RPI debate on MUScoop...year 7...will this FINALLY be the year it sinks in

[ChicosBailBonds]

So here we are again, and again people are shocked, surprised, stunned that teams with higher Ken Pom scores or higher Sagarin scores than RPI didn't make the NCAA tournament while teams with higher RPI scores and lower Ken Pom ratings.

It's like an annual pilgrammage.  I should just restate the old posts, but someone would accuse me of Hoopinlooping.

Solid Ken Pom numbers, but not great RPI numbers examples

SMU Ken Pom 32, RPI of 53...didn't get in
St. John's Ken Pom 38, RPI of 67...didn't get in
Louisiana Tech Ken Pom 35, RPI of 62...didn't get in
Utah Ken Pom of 36, RPI of 80...didn't get in
Florida State Ken Pom of 41, RPI of 54...didn't get in
Maryland Ken Pom of 43, RPI of 78....didn't get in
Clemson Ken Pom of 51, RPI of 79....didn't get in

Let's look at the reverse, good RPI but so-so Ken Pom

George Washington, RPI of 29, but a Ken Pom of 46...GOT IN
Colorado, RPI of 34, but a Ken Pom of 64...GOT IN
Dayton, RPI of 43 but a Ken Pom of 53....GOT IN
North Carolina State, RPI of 55, but a Ken Pom of 66....GOT IN

When are we going to learn, how much evidence do you want?  You can scream about how the RPI sucks and isn't a great system, but more often than not it directionally forecasts what the committee does.  There are exceptions, always will be, but when it comes down to it I don't know why annually so many of you are surprised. I  really don't get it.  It's the NCAAs system, they're going to use it.  Why is there this stupid argument every year that it isn't used, or not used much.  

[NersEllenson]

So here we are again, and again people are shocked, surprised, stunned that teams with higher Ken Pom scores or higher Sagarin scores than RPI didn't make the NCAA tournament while teams with higher RPI scores and lower Ken Pom ratings.

It's like an annual pilgrammage.  I should just restate the old posts, but someone would accuse me of Hoopinlooping.

Solid Ken Pom numbers, but not great RPI numbers examples

SMU Ken Pom 32, RPI of 53...didn't get in
St. John's Ken Pom 38, RPI of 67...didn't get in
Louisiana Tech Ken Pom 35, RPI of 62...didn't get in
Utah Ken Pom of 36, RPI of 80...didn't get in
Florida State Ken Pom of 41, RPI of 54...didn't get in
Maryland Ken Pom of 43, RPI of 78....didn't get in
Clemson Ken Pom of 51, RPI of 79....didn't get in

Let's look at the reverse, good RPI but so-so Ken Pom

George Washington, RPI of 29, but a Ken Pom of 46...GOT IN
Colorado, RPI of 34, but a Ken Pom of 64...GOT IN
Dayton, RPI of 43 but a Ken Pom of 53....GOT IN
North Carolina State, RPI of 55, but a Ken Pom of 66....GOT IN

When are we going to learn, how much evidence do you want?  You can scream about how the RPI sucks and isn't a great system, but more often than not it directionally forecasts what the committee does.  There are exceptions, always will be, but when it comes down to it I don't know why annually so many of you are surprised. I  really don't get it.  It's the NCAAs system, they're going to use it.  Why is there this stupid argument every year that it isn't used, or not used much.  

Thanks for putting that together...key words bolded.  NCAA System and stupid.  They go hand in hand generally.  Hopefully we can do a better job with our scheduling moving forward with all focus toward projected RPI rankings/historical of the cupcake options...

I don't disagree that RPI forecasts better who makes tournament, yet I strongly feel Pomroys rankings are much more reflective of actual team's strength...would suggest that Pomroy's rankings will fare better in tourney than RPI rankings of teams...would be interesting to contrast that dynamic.  You up for it Chicos?

[ChicosBailBonds]

So sure, I'm on board, but what system are you going to use that is truly fair?  I can make an effective argument that Ken Pom, because it is efficiency based system is a good predictor of who might win a game, but not an overall ranker of teams.  People might disagree, but that is one of the arguments against it.  Same for Sagarin, which I prefer the most.  The problem with these other systems is they factor in margin of victory.  Now, truth be told I have no issue with this.  A team barely beating someone by one point while another team beats that same team by 35 should matter.  The NCAA and the coaches do not want a system that rewards teams for running up the score, so that immediately becomes problematic for those using margin of victory.  If they put in a diminishing returns where anything above a 15 point win or some agreed upon number doesn't "earn" additional bonus to one's rating, then I think that is a start. 

I agree the RPI is flawed....no debate, but they are ALL flawed.  That's the problem. As long as this is the NCAAs system and their tournament, it is going to be used.  I just hope FINALLY this year people here come to grips with it.  I feel like I bang my head each year with Jay Bee, Lenny, etc, about how much the thing is used.  Doesn't matter if two former chairs of the committee flat out tell me it is used and I share it here....it is ignored.  Doesn't matter if the chairs say it publicly.  Doesn't matter if the data year in and year out on most matters point this out.  smh

[forgetful]

So here we are again, and again people are shocked, surprised, stunned that teams with higher Ken Pom scores or higher Sagarin scores than RPI didn't make the NCAA tournament while teams with higher RPI scores and lower Ken Pom ratings.

It's like an annual pilgrammage.  I should just restate the old posts, but someone would accuse me of Hoopinlooping.

Solid Ken Pom numbers, but not great RPI numbers examples

SMU Ken Pom 32, RPI of 53...didn't get in
St. John's Ken Pom 38, RPI of 67...didn't get in
Louisiana Tech Ken Pom 35, RPI of 62...didn't get in
Utah Ken Pom of 36, RPI of 80...didn't get in
Florida State Ken Pom of 41, RPI of 54...didn't get in
Maryland Ken Pom of 43, RPI of 78....didn't get in
Clemson Ken Pom of 51, RPI of 79....didn't get in

Let's look at the reverse, good RPI but so-so Ken Pom

George Washington, RPI of 29, but a Ken Pom of 46...GOT IN
Colorado, RPI of 34, but a Ken Pom of 64...GOT IN
Dayton, RPI of 43 but a Ken Pom of 53....GOT IN
North Carolina State, RPI of 55, but a Ken Pom of 66....GOT IN

When are we going to learn, how much evidence do you want?  You can scream about how the RPI sucks and isn't a great system, but more often than not it directionally forecasts what the committee does.  There are exceptions, always will be, but when it comes down to it I don't know why annually so many of you are surprised. I  really don't get it.  It's the NCAAs system, they're going to use it.  Why is there this stupid argument every year that it isn't used, or not used much.  

I agree with Ners.  Key words NcAAs system and stupid.  The list only shows that the committee frankly isn't paying attention to what the teams actually did and the RPI is waying too heavily.  

Best example is SMU.  The coaches and media think they are a top 25 team in the country.  3-4 vs. the RPI top 25.  4-5 against the RPI top 50....not in the tournament because of an RPI of 55...frankly idiotic.

6 A10 teams in, I'm not sure if any of them would have been able to beat out Providence for 3rd in our conference.  It's time the system is changed.

[ChicosBailBonds]

Fine...stupid and the NCAA owns it....I've been saying that for years now, and people still want to ignore it year in and year out.  This year was no exception.  Complain about the stupidity or the flaws is fine, what is worse is actually believing they don't use it or rely on it less and less each year...which is expressed here time and again.  What's the Albert Einstein quote.....

[NersEllenson]

I agree with Ners.  Key words NcAAs system and stupid.  The list only shows that the committee frankly isn't paying attention to what the teams actually did and the RPI is waying too heavily.  

Best example is SMU.  The coaches and media think they are a top 25 team in the country.  3-4 vs. the RPI top 25.  4-5 against the RPI top 50....not in the tournament because of an RPI of 55...frankly idiotic.

6 A10 teams in, I'm not sure if any of them would have been able to beat out Providence for 3rd in our conference.  It's time the system is changed.

It absolutely blows my mind SMU did not get in...worst snub I can recall in years.  Perhaps some at NCAA don't care for Larry Brown for some reason.  Something just reeks of really wrong with SMU not getting in..

[Benny B]

It absolutely blows my mind SMU did not get in...worst snub I can recall in years.  Perhaps some at NCAA don't care for Larry Brown for some reason.  Something just reeks of really wrong with SMU not getting in..

The committee has been penalizing teams for weak scheduling over the past few years (and rewarding teams with good OOC schedules)... the SMU snub should only be a surprise to those not paying attention.

[ChicosBailBonds]

SMU losing to South Florida (231) and Temple (176) and then Houston the other day (143) did not look good "visually".  Perception I'm sure came into play.  That, and the NCAA isn't wild about Larry Brown either.

[forgetful]

SMU losing to South Florida (231) and Temple (176) and then Houston the other day (143) did not look good "visually".  Perception I'm sure came into play.  That, and the NCAA isn't wild about Larry Brown either.

That shouldn't come in to play.  If it does, the committee needs to be replaced.

Also, the always say the entire body of work...clearly that isn't the case as the whole basketball community (coaches and media) think SMU is one of the top 25 teams in the country.

[ChicosBailBonds]

That shouldn't come in to play.  If it does, the committee needs to be replaced.

Also, the always say the entire body of work...clearly that isn't the case as the whole basketball community (coaches and media) think SMU is one of the top 25 teams in the country.

I agree, but I remember when UNLV one year was ranked in the polls 25th and didn't get in. 

RPI is the stone cold lock on this one and the cover they need not to put them in.....if you want to know who gets in, follow the RPI as it usually tells the story as it has for many many years.

[amen426]

So here we are again, and again people are shocked, surprised, stunned that teams with higher Ken Pom scores or higher Sagarin scores than RPI didn't make the NCAA tournament while teams with higher RPI scores and lower Ken Pom ratings.

It's like an annual pilgrammage.  I should just restate the old posts, but someone would accuse me of Hoopinlooping.

Solid Ken Pom numbers, but not great RPI numbers examples

SMU Ken Pom 32, RPI of 53...didn't get in
St. John's Ken Pom 38, RPI of 67...didn't get in
Louisiana Tech Ken Pom 35, RPI of 62...didn't get in
Utah Ken Pom of 36, RPI of 80...didn't get in
Florida State Ken Pom of 41, RPI of 54...didn't get in
Maryland Ken Pom of 43, RPI of 78....didn't get in
Clemson Ken Pom of 51, RPI of 79....didn't get in

Let's look at the reverse, good RPI but so-so Ken Pom

George Washington, RPI of 29, but a Ken Pom of 46...GOT IN
Colorado, RPI of 34, but a Ken Pom of 64...GOT IN
Dayton, RPI of 43 but a Ken Pom of 53....GOT IN
North Carolina State, RPI of 55, but a Ken Pom of 66....GOT IN

When are we going to learn, how much evidence do you want?  You can scream about how the RPI sucks and isn't a great system, but more often than not it directionally forecasts what the committee does.  There are exceptions, always will be, but when it comes down to it I don't know why annually so many of you are surprised. I  really don't get it.  It's the NCAAs system, they're going to use it.  Why is there this stupid argument every year that it isn't used, or not used much.  

I especially like how you chose to list NC State as a team with a "good RPI" of 55. While SMU (RPI: 53) and Florida state (RPI: 54) are listed in your category of teams with lower RPIs.

[ChicosBailBonds]

I especially like how you chose to list NC State as a team with a "good RPI" of 55. While SMU (RPI: 53) and Florida state (RPI: 54) are listed in your category of teams with lower RPIs.

Fair point....it was to show the discrepancy of Ken Pom and the RPI...under the Ken Pom number, NC State had NO SHOT to get in.  With the RPI number, they were likely the last or second to last in. 

The differences between Ken Pom and the RPI was the overall point....you can do the same exercise with Sagarin. 

[LAZER]

I think most people around here understand the importance of RPI

[Benny B]

I think most people around here understand the importance of RPI

Agreed... as imperfect as it is (especially prior to mid-Feb), I think everyone realizes that if it's not the first number that appears on the committee's tourney resume, it's certainly the one with the largest font size.

[willie warrior]

Fair point....it was to show the discrepancy of Ken Pom and the RPI...under the Ken Pom number, NC State had NO SHOT to get in.  With the RPI number, they were likely the last or second to last in. 

The differences between Ken Pom and the RPI was the overall point....you can do the same exercise with Sagarin. 

SMU and Fla. State were probably hosed. We were not.

[Jay Bee]

Fine...stupid and the NCAA owns it....I've been saying that for years now, and people still want to ignore it year in and year out.  

Who are these people that "still want to ignore it year in and year out"? I think the VAST MAJORITY around here believe the RPI is stupid but recognize it's in front of the committee's face. Is someone saying MU should be in the NCAA tournament because their RPI isn't fair and some other metric, calculation or analysis says they should be in?

[ChicosBailBonds]

I think most people around here understand the importance of RPI

Some do, but you get stuff like this....including from Buzz Williams, who is a Ken Pom fan

Paraphrasing from the Buzz interview that Heisenburg listened to 'RPI is good for clustering in broad category but that's it.'   Actually Buzz, it is relied on by the committee, whether we like it or not.  I agree it is good for clustering, but to say "that's it" suggests he isn't paying attention to how it is used currently.

http://www.muscoop.com/index.php?topic=42249.msg575963#msg575963

I won't even get into all the comments by posters here over the years on this topic, but this one just in the last few months is an example.

"They (RPI) are worthless, when compared with multiple other rating systems, in determining a team's true strength. That doesn't mean the NCAA selection committee doesn't consider them - just not as much as in the past. Old habits, good and bad, die hard." 

[Jay Bee]

I agree, but I remember when UNLV one year was ranked in the polls 25th and didn't get in. 

RPI is the stone cold lock on this one and the cover they need not to put them in.....if you want to know who gets in, follow the RPI as it usually tells the story as it has for many many years.

Oh yes... you mean like how Iowa with an RPI of 56 got in... and was seeded ahead of Xavier (47 RPI)... while Minnesota had an RPI of 50 and an SOS (misnomer) of 9... didn't get in... but many models had Iowa as a much better team than Minnesota...

Maybe, just maybe, the committee isn't so goo goo ga ga over the RPI as you suggest.

No doubt, it's in front of their faces. Again, to what degree it makes certain decisions?... tough call. I think it depends on the individuals in the room at the time.

"stone cold lock", baby!

[The Equalizer]

Thanks for putting that together...key words bolded.  NCAA System and stupid.  They go hand in hand generally.  Hopefully we can do a better job with our scheduling moving forward with all focus toward projected RPI rankings/historical of the cupcake options...

I don't disagree that RPI forecasts better who makes tournament, yet I strongly feel Pomroys rankings are much more reflective of actual team's strength...would suggest that Pomroy's rankings will fare better in tourney than RPI rankings of teams...would be interesting to contrast that dynamic.  You up for it Chicos?

Teams know that the NCAA favors the RPI--a system that doesn't account for margin of victory.

If the the NCAA suddenly started using something like Ken Pom or Sagarin--where the margin of victory is considered--I predict that they would start coaching (and scheduling) completely different. 

Down five points with under seconds to go?  No longer will coaches go for the win--throw in the towel, because there's a greater chance FTs causing you to lose by 9 than closing the gap.

Or up by 20 with 5 minutes to go--time to give the bench some game time?  Nah--they might let the lead shrink to 10.  Instead, there will be incentive to play the starters and see if the lead can grow it to 25.

If you started to use a system where margin of victory was important--coaches would start to coach with that assumption.  The other systems are only perceived to be better because nobody is coaching to optimize Pomoroy or Sagaran.  If the NCAA ever did switch, coaches would make a whole lot of changes in their style reflective of that.