I say we take the subjectivity of the human element and use the computers to come up with the at-large teams. Sure there would be some issues and a few deserving teams denied entry to the Dance but this eliminates human biases and ignorances. Perfect a formula and select the best teams based on that formula.

And yes, there'd still be autobids going to conference champions. I also realize computer generated rankings wouldn't be without flaws but again you eliminate biases and ignorances when comparing teams and leave it out of anyone's control.

(01-22-2023 04:00 AM)_C2_ Wrote:  I say we take the subjectivity of the human element and use the computers to come up with the at-large teams. Sure there would be some issues and a few deserving teams denied entry to the Dance but this eliminates human biases and ignorances. Perfect a formula and select the best teams based on that formula.

And yes, there'd still be autobids going to conference champions. I also realize computer generated rankings wouldn't be without flaws but again you eliminate biases and ignorances when comparing teams and leave it out of anyone's control.

If somebody could actually perfect a bias free formula, then why not? Sadly, no one has been able to do that. If computers could do that, they would all produce the same results, but no two computers produce the same results.

(01-22-2023 09:29 AM)ken d Wrote:  

(01-22-2023 04:00 AM)_C2_ Wrote:  I say we take the subjectivity of the human element and use the computers to come up with the at-large teams. Sure there would be some issues and a few deserving teams denied entry to the Dance but this eliminates human biases and ignorances. Perfect a formula and select the best teams based on that formula.

And yes, there'd still be autobids going to conference champions. I also realize computer generated rankings wouldn't be without flaws but again you eliminate biases and ignorances when comparing teams and leave it out of anyone's control.

If somebody could actually perfect a bias free formula, then why not? Sadly, no one has been able to do that. If computers could do that, they would all produce the same results, but no two computers produce the same results.

Right - the computer ranking is still built off of what a human subjectively deems to be the important inputs.

In any event, I have no issue with the NCAA Tournament committee (unlike the CFP committee). Basketball inherently has a lot more data points (nearly 3 times as many games, a lot more meaningful/instructive non-conference games, a lot more informative statistics about teams that can be compared that have legitimate value, etc.), so the basketball committee has a lot more solid information to work with beyond the “eye test” compared to football.

I like the system as is.

the more bids you have, the less it matters how the selections get made.

(01-22-2023 10:56 AM)Frank the Tank Wrote:  

(01-22-2023 09:29 AM)ken d Wrote:  

(01-22-2023 04:00 AM)_C2_ Wrote:  I say we take the subjectivity of the human element and use the computers to come up with the at-large teams. Sure there would be some issues and a few deserving teams denied entry to the Dance but this eliminates human biases and ignorances. Perfect a formula and select the best teams based on that formula.

And yes, there'd still be autobids going to conference champions. I also realize computer generated rankings wouldn't be without flaws but again you eliminate biases and ignorances when comparing teams and leave it out of anyone's control.

If somebody could actually perfect a bias free formula, then why not? Sadly, no one has been able to do that. If computers could do that, they would all produce the same results, but no two computers produce the same results.

Right - the computer ranking is still built off of what a human subjectively deems to be the important inputs.

In any event, I have no issue with the NCAA Tournament committee (unlike the CFP committee). Basketball inherently has a lot more data points (nearly 3 times as many games, a lot more meaningful/instructive non-conference games, a lot more informative statistics about teams that can be compared that have legitimate value, etc.), so the basketball committee has a lot more solid information to work with beyond the “eye test” compared to football.

One issue I have had with computer algorithms is that there are always factors they can't account for. For example, in football weather often plays a significant role, as do injuries. Should a team be given as much credit (or penalty) if they got to play an opponent when their star QB wasn't on the field? NET in hoops tries to value some wins more than others by the quadrant system. UNC got credit yesterday for a Quad 1 win over NC State. But how will that win be valued at the end of the year if NC State is no longer in the Top 30 because they lost another key player to injury?

I'm troubled by concerns that a team will get in (or be left out) because a member on the selection committee played for, or coached, or was AD at a particular school. Or because their school/conference isn't represented on the committee? Do we think so little of the people selected for that committee that we assume they will be biased, and that their bias will unduly sway the rest of the committee?

My only problem with the selection committee is that I think they can only hurt themselves when they try to justify their choices to the media. I get why they have to submit to those interviews. I'd just prefer the process was more like a papal enclave. I don't want to know how the sausage was made. Just give me the results.

Not against the idea but I don't think it would do away with people complaining about snubs

(01-22-2023 02:39 PM)WhoseHouse? Wrote:  Not against the idea but I don't think it would do away with people complaining about snubs

I don't believe we will ever do away with people complaining about being left out.

(01-22-2023 02:58 PM)ken d Wrote:  

(01-22-2023 02:39 PM)WhoseHouse? Wrote:  Not against the idea but I don't think it would do away with people complaining about snubs

I don't believe we will ever do away with people complaining about being left out.

Yah I think thats a pretty safe bet

It's kind of a mystery what the NCAA committee considers, but RPI is supposedly a big thing they look at.

But I don't really have that much of a problem with the committee for the basketball tournament. Nobody except for the fans of the "bubble teams" that get passed over really get that upset about the committee's decisions. And, really, there are 68 NCAA tournament teams-if a team didn't want to get passed over by the committee, then they should have had a better season and not been a bubble team in the first place.

I've had much more of a problem with the football committee. That's a 4 team playoff, so you can't tell the passed over teams that they should have had a better season like you can in basketball. I've always wondered why they didn't just have the teams who finished in the top 4 of the BCS standings make the playoff.

(01-22-2023 04:19 PM)Poster Wrote:  It's kind of a mystery what the NCAA committee considers, but RPI is supposedly a big thing they look at.

But I don't really have that much of a problem with the committee for the basketball tournament. Nobody except for the fans of the "bubble teams" that get passed over really get that upset about the committee's decisions. And, really, there are 68 NCAA tournament teams-if a team didn't want to get passed over by the committee, then they should have had a better season and not been a bubble team in the first place.

I've had much more of a problem with the football committee. That's a 4 team playoff, so you can't tell the passed over teams that they should have had a better season like you can in basketball. I've always wondered why they didn't just have the teams who finished in the top 4 of the BCS standings make the playoff.

During the 2018 offseason, the NCAA announced that the RPI would no longer be used in the selection process for the Division I men's basketball tournament. Effective immediately, it was replaced with the NCAA Evaluation Tool (NET).

Just found that after a quick google search.

(01-22-2023 02:58 PM)ken d Wrote:  

(01-22-2023 02:39 PM)WhoseHouse? Wrote:  Not against the idea but I don't think it would do away with people complaining about snubs

I don't believe we will ever do away with people complaining about being left out.

If the formula is public it would reduce complaints greatly.

No one heard Washington complain after they defeated 7-4 Wazzu by 3+ scores and Utah jumped them in the rankings after the Utes beat 1-10 Colorado and went to Vegas. Why? Because an algorithm predetermined that Utah would advance and not Washington.

You do want a formula that reduces the incentive of running up the score. Use scores to build strength of schedule and strength of schedule to determine strength of record (which is a good “deserve”-based ranking).

(01-22-2023 04:00 AM)_C2_ Wrote:  I say we take the subjectivity of the human element and use the computers to come up with the at-large teams. Sure there would be some issues and a few deserving teams denied entry to the Dance but this eliminates human biases and ignorances. Perfect a formula and select the best teams based on that formula.

And yes, there'd still be autobids going to conference champions. I also realize computer generated rankings wouldn't be without flaws but again you eliminate biases and ignorances when comparing teams and leave it out of anyone's control.

Why not simply use the NET rankings but uncap the margin of victory numbers? I think it's capped at 10 points or something. Close victories are worth less than blow out victories. Maybe it'll get the remaining smaller conferences to do neutral site tournaments instead of home field advantage for conference tournaments.

That also would maybe increase neutral site tournaments for smaller conferences. With hbcu schools in multiple conferences that's a easy tournament. We could have a West Coast classic with a team from: the Big Sky, WCC, Big West, WAC, MWC and 3 invited conferences that change yearly. That can be done regionally maybe in a regionally good recruiting state to try to get people watching or used as a recruiting venture.

(01-22-2023 01:15 PM)ken d Wrote:  

(01-22-2023 10:56 AM)Frank the Tank Wrote:  

(01-22-2023 09:29 AM)ken d Wrote:  

(01-22-2023 04:00 AM)_C2_ Wrote:  I say we take the subjectivity of the human element and use the computers to come up with the at-large teams. Sure there would be some issues and a few deserving teams denied entry to the Dance but this eliminates human biases and ignorances. Perfect a formula and select the best teams based on that formula.

And yes, there'd still be autobids going to conference champions. I also realize computer generated rankings wouldn't be without flaws but again you eliminate biases and ignorances when comparing teams and leave it out of anyone's control.

If somebody could actually perfect a bias free formula, then why not? Sadly, no one has been able to do that. If computers could do that, they would all produce the same results, but no two computers produce the same results.

Right - the computer ranking is still built off of what a human subjectively deems to be the important inputs.

In any event, I have no issue with the NCAA Tournament committee (unlike the CFP committee). Basketball inherently has a lot more data points (nearly 3 times as many games, a lot more meaningful/instructive non-conference games, a lot more informative statistics about teams that can be compared that have legitimate value, etc.), so the basketball committee has a lot more solid information to work with beyond the “eye test” compared to football.

One issue I have had with computer algorithms is that there are always factors they can't account for. For example, in football weather often plays a significant role, as do injuries. Should a team be given as much credit (or penalty) if they got to play an opponent when their star QB wasn't on the field? NET in hoops tries to value some wins more than others by the quadrant system. UNC got credit yesterday for a Quad 1 win over NC State. But how will that win be valued at the end of the year if NC State is no longer in the Top 30 because they lost another key player to injury?

I'm troubled by concerns that a team will get in (or be left out) because a member on the selection committee played for, or coached, or was AD at a particular school. Or because their school/conference isn't represented on the committee? Do we think so little of the people selected for that committee that we assume they will be biased, and that their bias will unduly sway the rest of the committee?

My only problem with the selection committee is that I think they can only hurt themselves when they try to justify their choices to the media. I get why they have to submit to those interviews. I'd just prefer the process was more like a papal enclave. I don't want to know how the sausage was made. Just give me the results.

ken d: I do not know all the requirements for members of the selection committee to the BB tournament, but I believe I read three or four years ago that if a member of the committee has a current attachment to a school about to be discussed, that member leaves the room during the discussion of that team.