Updated Bracketology 2.23.16

[CatsTopPac]

Some of them project out, which is why AZ is the 3 line in a couple of them, which could happen if they win out. Just FYI.

I see. Good to know. Seeing some of those 3s and just a couple other of the outlying seeding for other teams, I was a little confused how they justified some of them.

 

[mr.hockey4242]

Why the fuck does lunardi still have UNC as a 1 seed?

Their resume with barely no top 50 wins is more 3/4 seed than 1 seed.

 

[ericd7633]

Why the fuck does lunardi still have UNC as a 1 seed?

Their resume with barely no top 50 wins is more 3/4 seed than 1 seed.

Yeah I'm a little puzzled by that as well. I'd have them closer to a 3 seed than a 1 seed to be honest. They are only 3-5 against the top 50. I'd still have OU ahead of them OU is 8-5 against the top 50.

 

[CatsTopPac]

Why the fuck does lunardi still have UNC as a 1 seed?

Their resume with barely no top 50 wins is more 3/4 seed than 1 seed.

Yeah I'm a little puzzled by that as well. I'd have them closer to a 3 seed than a 1 seed to be honest. They are only 3-5 against the top 50. I'd still have OU ahead of them OU is 8-5 against the top 50.

UNC #1 NCAA College Basketball Rankings, Polls - ESPN

UNC #5 2016 Pomeroy College Basketball Ratings

UNC #5 NCAAB Sagarin - NCAAB Basketball - USA TODAY

 

[mr.hockey4242]

UNC #1 NCAA College Basketball Rankings, Polls - ESPN

UNC #5 2016 Pomeroy College Basketball Ratings

UNC #5 NCAAB Sagarin - NCAAB Basketball - USA TODAY

So?

 

[CatsTopPac]

So?

I'm thinking that's why Lunardi has them as a #1 seed. I'm not saying I do, only that it's probably what he's thinking.

 

[mr.hockey4242]

I'm thinking that's why Lunardi has them as a #1 seed. I'm not saying I do, only that it's probably what he's thinking.

But two of those are technically 2 seed rankings.

And none of those 3 are parts of a "resume" lol.

 

[CatsTopPac]

But two of those are technically 2 seed rankings.

And none of those 3 are parts of a "resume" lol.

Well, Lunardi uses BPI as a part of his resume, and UNC is #1. So that's going to factor in for him. And again, the committee takes BPI, Sagarin, and Kenpom into account. So you can put up their resume (RPI/SOS), but if the committee looks at more than that, then their "resume" isn't going to account for everything.

As #5 seeds for the other two, they are the first #2 and only one spot better puts them as a #1 seed. Looks like Lunardi is relying on BPI.

 

[mr.hockey4242]

Well, Lunardi uses BPI as a part of his resume, and UNC is #1. So that's going to factor in for him. And again, the committee takes BPI, Sagarin, and Kenpom into account. So you can put up their resume (RPI/SOS), but if the committee looks at more than that, then their "resume" isn't going to account for everything.

As #5 seeds for the other two, they are the first #2 and only one spot better puts them as a #1 seed. Looks like Lunardi is relying on BPI.

No the committee doesn't lol.

They don't use any of those. And they sure as hell don't use BPI a made up formula by a television network. Lol.

 

[CatsTopPac]

No the committee doesn't lol.

They don't use any of those. And they sure as hell don't use BPI a made up formula by a television network. Lol.

http://www.coloradoan.com/story/spo.../colorado-state-ncaa-tournament-rpi/24829067/

You guys really need to have a better understanding of what the committee considers. All you guys do is put up RPI (SOS) and record vs RPI rankings, and call it a day.

Men's Basketball Selections 101 - Selections

How do you not research this if you are going to spend so much time discussing it?
Absolutely the committee looks at it. And the BPI was officially developed to rate teams based on those above bullets not covered in RPI, etc., such as injuries, margin of victory, etc.

More:

The RPI rating is often considered a factor in selecting and seeding the final few teams in the tournament field. However, the NCAA selection committee in 2015 said the RPI is no longer relevant only utilized for grouping the teams into groups such as top 50 and top 100 teams to value the wins and losses and not as a factor for selection.[3] Additionally, the committee officially considers other computer rankings such as ESPN's BPI, Sagarin and Pomeroy Ratings which use additional factors considered by the committee such as injured players in the case of the BPI. Additionally, committee members consider how teams do on the road and at neutral courts, strength of conference and schedule, non-conference strength of schedule, record against other selected tournament teams, and other extenuating factors. Finally, the "eye test" is often quoted by pundits as something the committee uses, however ncaa.org's sparse description of the selection process doesn't officially mention the "eye test".[4] For instance, in 2016 Oklahoma athletic director Joe Castiglione, the NCAA selection committee's chair, said that the stark contrast in Syracuse's performance in 2015-2016 with Jim Boeheim present versus absent was considered the same as missing a key player during the slump.[5]

 

[mr.hockey4242]

RPI no longer important, NCAA tournament makes example of CSU

You guys really need to have a better understanding of what the committee considers. All you guys do is put up RPI (SOS) and record vs RPI rankings, and call it a day.

Men's Basketball Selections 101 - Selections

How do you not research this if you are going to spend so much time discussing it?
Absolutely the committee looks at it. And the BPI was officially developed to rate teams based on those above bullets not covered in RPI, etc., such as injuries, margin of victory, etc.

More:

The RPI rating is often considered a factor in selecting and seeding the final few teams in the tournament field. However, the NCAA selection committee in 2015 said the RPI is no longer relevant only utilized for grouping the teams into groups such as top 50 and top 100 teams to value the wins and losses and not as a factor for selection.[3] Additionally, the committee officially considers other computer rankings such as ESPN's BPI, Sagarin and Pomeroy Ratings which use additional factors considered by the committee such as injured players in the case of the BPI. Additionally, committee members consider how teams do on the road and at neutral courts, strength of conference and schedule, non-conference strength of schedule, record against other selected tournament teams, and other extenuating factors. Finally, the "eye test" is often quoted by pundits as something the committee uses, however ncaa.org's sparse description of the selection process doesn't officially mention the "eye test".[4] For instance, in 2016 Oklahoma athletic director Joe Castiglione, the NCAA selection committee's chair, said that the stark contrast in Syracuse's performance in 2015-2016 with Jim Boeheim present versus absent was considered the same as missing a key player during the slump.[5]

Dude it's never included on a resume for a reason

 

[ericd7633]

Ok so using the procedures in that link this is UNC's resume:

RPI: 9
W/L: 24-6
OOC W/L: 11-2
OOC RPI: 19
Conf. W/L: 13-4
R/N W/L: 5-5
Injuries - Paige missed the UNI game
BPI: 1
KenPom: 5
Sagarin: 5
Record against teams in "consideration": 3-0
Record against the field: 3-5

Still not deserving of a 1 seed IMO.

Also, using a mid major, and referencing an article from a Colorado newspaper, really doesn't dispel the notion the committee doesn't use RPI. They also only had two top 50 RPI wins. Mid major's with higher RPI's have been left out before.

 

[CatsTopPac]

Dude it's never included on a resume for a reason

Ok so using the procedures in that link this is UNC's resume:

RPI: 9
W/L: 24-6
OOC W/L: 11-2
OOC RPI: 19
Conf. W/L: 13-4
R/N W/L: 5-5
Injuries - Paige missed the UNI game
BPI: 1
KenPom: 5
Sagarin: 5
Record against teams in "consideration": 3-0
Record against the field: 3-5

Still not deserving of a 1 seed IMO.

Also, using a mid major, and referencing an article from a Colorado newspaper, really doesn't dispel the notion the committee doesn't use RPI. They also only had two top 50 RPI wins. Mid major's with higher RPI's have been left out before.

I'm just telling both of you that per the ncaa selection committee selection criteria from their own official website, that their resume (what they look at as the criteria for their selections) contains so much more than RPI, SOS, and the w/l vs the RPI rankings. So when you guys have these conversations, or post your brackets, or whatever, those bullets are specifically what the committee has said, 'Put this on our website under Selection Criteria, so that anyone who wants to know, or is trying to figure out what we think, or mad because they didn't get in or the seed they wanted, this is what we consider as our criteria.'

Different committee members weigh those various criteria differently. But that is everything that they take into account. And for the most part, they take the mean seeds in all the committee's seedings, to make out the official bracket. But regardless, it is way more to consider all of those factors, plus what comprises the other rankings besides RPI shows, for them to make their decisions. So when you guys keep showing these resumes (whether anyone else also shows them, or comes to different conclusions that you do), that the selection committee has publicly stated specifically for everyone to see and understand what they use.

We can say that we don't like criteria X,Y, and Z, or that we think A, B, and C are more important, but the committee looks at it all. Some members completely agree with you, and some completely disagree, but they look at all of it.

All I'm doing is telling you what criteria are, and how they lead to different results than just the SOS, RPI, and W/L vs RPI. If you want to actually know what the committee uses, and how all the various factors that they choose will play out on Selection Sunday, you should know what that criteria is, and probably incorporate that into your bracketology. I'm not trying to say this because I'm an AZ fan, or because I'm a UNC fan. I just see what you are using, and it's limited compared to what the committee has stated that they use for their criteria.

That's it guys, that's all I'm saying. Take it or leave it, but just know it going forward that only putting up a portion of the overall criteria is limiting, and not reflective of what the committee looks at, if you want to look at the same criteria as the committee.

 

[ericd7633]

I'm just telling both of you that per the ncaa selection committee selection criteria from their own official website, that their resume (what they look at as the criteria for their selections) contains so much more than RPI, SOS, and the w/l vs the RPI rankings. So when you guys have these conversations, or post your brackets, or whatever, those bullets are specifically what the committee has said, 'Put this on our website under Selection Criteria, so that anyone who wants to know, or is trying to figure out what we think, or mad because they didn't get in or the seed they wanted, this is what we consider as our criteria.'

Different committee members weigh those various criteria differently. But that is everything that they take into account. And for the most part, they take the mean seeds in all the committee's seedings, to make out the official bracket. But regardless, it is way more to consider all of those factors, plus what comprises the other rankings besides RPI shows, for them to make their decisions. So when you guys keep showing these resumes (whether anyone else also shows them, or comes to different conclusions that you do), that the selection committee has publicly stated specifically for everyone to see and understand what they use.

We can say that we don't like criteria X,Y, and Z, or that we think A, B, and C are more important, but the committee looks at it all. Some members completely agree with you, and some completely disagree, but they look at all of it.

All I'm doing is telling you what criteria are, and how they lead to different results than just the SOS, RPI, and W/L vs RPI. If you want to actually know what the committee uses, and how all the various factors that they choose will play out on Selection Sunday, you should know what that criteria is, and probably incorporate that into your bracketology. I'm not trying to say this because I'm an AZ fan, or because I'm a UNC fan. I just see what you are using, and it's limited compared to what the committee has stated that they use for their criteria.

That's it guys, that's all I'm saying. Take it or leave it, but just know it going forward that only putting up a portion of the overall criteria is limiting, and not reflective of what the committee looks at, if you want to look at the same criteria as the committee.

That's fine. I think the biggest thing they look at is quality wins, wins vs the field, and wins vs teams "in consideration". Those are easily reflective on the team sheets they use to evaluate teams, along with SoS. Although SoS is mainly used when determining which bubble teams get in, specifically Non Conf. SoS.

 

[CatsTopPac]

That's fine. I think the biggest thing they look at is quality wins, wins vs the field, and wins vs teams "in consideration". Those are easily reflective on the team sheets they use to evaluate teams, along with SoS. Although SoS is mainly used when determining which bubble teams get in, specifically Non Conf. SoS.

You are basically saying that you think they go on RPI, SOSs, and W/L vs RPI rankings. What makes you say that? How do you know that the majority of the committee weighs those factors the most? Is it because those are the measures they put up on TV for a team's resume? Bottom line is that we have no idea. The CSU article specifically showed that it's not reliable. CSU was in according to those factors, and they got snubbed. If you guys want to put all of the criteria up for each team and then go on to justify which measures are the greatest indicators per team, then that's fine. But using the same small set of indicators across the board as the most important is just guessing that the committee uses them across the board too. When you look at a team like UNC, and they look like a 3/4 seed according to what you use, but they get a 1/2, then it's because the measures that you looked at were the worst of theirs, and you glossed over the others that others used for criteria.

 

[ericd7633]

You are basically saying that you think they go on RPI, SOSs, and W/L vs RPI rankings. What makes you say that? How do you know that the majority of the committee weighs those factors the most? Is it because those are the measures they put up on TV for a team's resume? Bottom line is that we have no idea. The CSU article specifically showed that it's not reliable. CSU was in according to those factors, and they got snubbed. If you guys want to put all of the criteria up for each team and then go on to justify which measures are the greatest indicators per team, then that's fine. But using the same small set of indicators across the board as the most important is just guessing that the committee uses them across the board too. When you look at a team like UNC, and they look like a 3/4 seed according to what you use, but they get a 1/2, then it's because the measures that you looked at were the worst of theirs, and you glossed over the others that others used for criteria.

Because its emphasized on the team sheets they look at. The wins that is. And CSU was never a lock. They had TWO wins vs the field. Same reason teams like Missouri State and Hofstra were left out in 2006 because they had even a higher RPI than CSU did. As for UNC I said I lean them closer to the 3 line than the 1 line.

There's always going to be 1 or 2 outliers with the RPI. Just like Princeton sitting at 33 right now. They have no shot for an at large. And there's always going to be 1 or 2 outliers in the other metrics too.

Also, wins vs the field and "under consideration" has NOTHING to do with RPI.

 

[CatsTopPac]

Because its emphasized on the team sheets they look at. The wins that is. And CSU was never a lock. They had TWO wins vs the field. Same reason teams like Missouri State and Hofstra were left out in 2006 because they had even a higher RPI than CSU did. As for UNC I said I lean them closer to the 3 line than the 1 line.

There's always going to be 1 or 2 outliers with the RPI. Just like Princeton sitting at 33 right now. They have no shot for an at large. And there's always going to be 1 or 2 outliers in the other metrics too.

Also, wins vs the field and "under consideration" has NOTHING to do with RPI.

Eric, I'm not going to keep going around and around with you debating semantics. I asked you what you are basing your criteria on for your brackets. You specifically stated the following:

"I try and mirror what I think the committee will do and take my own personal biases out of the equation, because of that RPI, Non Conf. SOS/SOS and Top 25/Top 50 wins are weighted more heavily.

Arizona's resume:
W/L: 22-5
RPI: 27
SOS 128
Non Conf. SOS: 268
Vs. Top 25: 0-2
Vs. Top 50: 2-4"

All I have been doing is saying that if you are truly interested in mirroring the committee to develop your bracket, then you need more than looking at record, RPI, SOS, non-con SOS, and record vs the RPI top 25 and 50. All of that besides record is part and parcel of RPI. That's it. I know for a fact that the committee has clearly stated that RPI is only one of many criteria that they use. I told you that there are more than a dozen different criteria that the committee specifically looks at to decide. I told you that you are missing criteria that the committee looks at like key injuries, the other rankings (and the criteria that they use to develop their rankings), MOV, home, away, and neutral wins and losses, etc. You can do whatever you want. But if you are trying to mirror the committee, then your set of criteria is limited.

It's really that simple. If you want to mirror the committee, then I'm just letting you know that you have to bring in these other criteria. You, and other hoopsters didn't seem to know that the committee considers other criteria, so I listed them for you. I'm not going to go back and forth on this. Do whatever you want. You said you want to mirror the committee, then consider everything that the committee considers. If you don't, and you just want to consider those criteria above, than do whatever you want. But it's not going to be based on what the committee considers.

If you want to look at teams like the committee does, then your "resume" is going to be much longer, and you are going to weigh all of the criteria, accordingly. All I'm saying is that if your threads are just going to be based on discussions of W/L, RPI, SOSs, and W/L vs top 25/50, then you are not looking at everything the committee does. Take it or leave it, but that is fact. You can either expand your criteria, or you can keep it. I'm not trying to attack you, I'm not saying this just because you had AZ as a 5 seed, I'm not just arguing just to argue. I just noticed that your criteria is limited compared to what the committee considers. You said you want to mirror the committee, so I provided the committee criteria so you know. Do what you want.

 

[ericd7633]

Eric, I'm not going to keep going around and around with you debating semantics. I asked you what you are basing your criteria on for your brackets. You specifically stated the following:

"I try and mirror what I think the committee will do and take my own personal biases out of the equation, because of that RPI, Non Conf. SOS/SOS and Top 25/Top 50 wins are weighted more heavily.

Arizona's resume:
W/L: 22-5
RPI: 27
SOS 128
Non Conf. SOS: 268
Vs. Top 25: 0-2
Vs. Top 50: 2-4"

All I have been doing is saying that if you are truly interested in mirroring the committee to develop your bracket, then you need more than looking at record, RPI, SOS, non-con SOS, and record vs the RPI top 25 and 50. All of that besides record is part and parcel of RPI. That's it. I know for a fact that the committee has clearly stated that RPI is only one of many criteria that they use. I told you that there are more than a dozen different criteria that the committee specifically looks at to decide. I told you that you are missing criteria that the committee looks at like key injuries, the other rankings (and the criteria that they use to develop their rankings), MOV, home, away, and neutral wins and losses, etc. You can do whatever you want. But if you are trying to mirror the committee, then your set of criteria is limited.

It's really that simple. If you want to mirror the committee, then I'm just letting you know that you have to bring in these other criteria. You, and other hoopsters didn't seem to know that the committee considers other criteria, so I listed them for you. I'm not going to go back and forth on this. Do whatever you want. You said you want to mirror the committee, then consider everything that the committee considers. If you don't, and you just want to consider those criteria above, than do whatever you want. But it's not going to be based on what the committee considers.

If you want to look at teams like the committee does, then your "resume" is going to be much longer, and you are going to weigh all of the criteria, accordingly. All I'm saying is that if your threads are just going to be based on discussions of W/L, RPI, SOSs, and W/L vs top 25/50, then you are not looking at everything the committee does. Take it or leave it, but that is fact. You can either expand your criteria, or you can keep it. I'm not trying to attack you, I'm not saying this just because you had AZ as a 5 seed, I'm not just arguing just to argue. I just noticed that your criteria is limited compared to what the committee considers. You said you want to mirror the committee, so I provided the committee criteria so you know. Do what you want.

And I'm telling you based on precedent set forth by the committee that top 25/50 wins, SOS, Non Conf SOS and ultimately record are more important than those other metrics. Again, based on seeding from previous years. That's what I try and mirror. RPI isn't the end all be all, neither are the other three metrics. Its not like the committee is going to average the 4 metrics and seed based on that anyway.

 

[CatsTopPac]

And I'm telling you based on precedent set forth by the committee that top 25/50 wins, SOS, Non Conf SOS and ultimately record are more important than those other metrics. Again, based on seeding from previous years. That's what I try and mirror. RPI isn't the end all be all, neither are the other three metrics. Its not like the committee is going to average the 4 metrics and seed based on that anyway.

What precedent? Previous years? That's all just factually untrue. Hear it straight from the horse's mouth.

What's it like to be a member of the tournament selection committee? Ask WCC commissioner Jamie Zaninovich

It's just not representative enough for many of the reasons I've outlined.

The Ratings Percentage Index Myth

It's common knowledge in CBB that RPI is just not as accurate as using all of the data combined. It's a simple baseline to assemble which teams are near the top, which are in the middle, and which are around the bottom. And then from there, everything else gets piled on, and they adjust all over based on the dozen or other/total criteria.

Hoop Thoughts: Dispelling Selection Sunday myths as tourney draws near

NCAA tournament 2012: Is it time to say RIP to the RPI?

Basketball RPI: Why it's a lousy way to pick teams for the NCAA Tournament.

Again, I'm done. You saying that the committee puts more weight on it that anything else is simply not true. It's statistically limiting given all of the other factors, variables, and information that RPI doesn't calculate. The committee completely recognizes that and specifically looks at everything to make selections. I'm not saying that RPI isn't considered, it obviously is, but its weight is outside the committee is widely known to be overstated.

 

[ericd7633]

What precedent? Previous years? That's all just factually untrue. Hear it straight from the horse's mouth.

What's it like to be a member of the tournament selection committee? Ask WCC commissioner Jamie Zaninovich

It's just not representative enough for many of the reasons I've outlined.

The Ratings Percentage Index Myth

It's common knowledge in CBB that RPI is just not as accurate as using all of the data combined. It's a simple baseline to assemble which teams are near the top, which are in the middle, and which are around the bottom. And then from there, everything else gets piled on, and they adjust all over based on the dozen or other/total criteria.

Hoop Thoughts: Dispelling Selection Sunday myths as tourney draws near

NCAA tournament 2012: Is it time to say RIP to the RPI?

Basketball RPI: Why it's a lousy way to pick teams for the NCAA Tournament.

Again, I'm done. You saying that the committee puts more weight on it that anything else is simply not true. It's statistically limiting given all of the other factors, variables, and information that RPI doesn't calculate. The committee completely recognizes that and specifically looks at everything to make selections. I'm not saying that RPI isn't considered, it obviously is, but its weight is outside the committee is widely known to be overstated.

We'll find out how they seed it come next Sunday. You think one way, I think another. And like I said RPI isn't the end all be all, never said it was, it is however, MY starting point, then I adjust accordingly. Been doing it like this for the past 5 years. One team to keep an eye out on is Gonzaga. Their metrics are top 30, their RPI is 62.