Bracketology (11/12)

[c4ndlelight]

Nov 16, 2024 20:41:19 GMT -5 GatorsChomp said:

Nov 16, 2024 14:54:24 GMT -5 trojansc said:

The reason why we are hyper-RPI (data) obsessed is because the committee has made us that way - not the other way around. If you could have a 22 RPI and miss the tournament, or a 68 RPI and make the tournament, things might be different. But through years of their decisions - we know how they select teams, and RPI is just soooooooo important.

The committee's process is not exactly robust or thorough - it's pretty simple. But even when dumbed down, they find ways to make extremely questionable decision.

Like, one of the complaints is that they don't even take the Final RPI into account. How in the world do you put such an emphasis on RPI and then NOT see that Tennessee and Ole Miss fell into the high 50s of RPI on the last week of the season.

Correct me if I'm wrong but I was under the impression that it was like a pretty well known thing that the last week of play doesn't impact the seedings for the tournament ? Like isn't that a thing?

Matchups before Thanksgiving matter. After? Probably not - unless it’s a noteworthy result (significant upset). In such case it will be over-reacted to by the committee. But something like a matchup between Top 10 teams or a 15 v 25 might as well not have been scheduled, unless two teams are in a toss up position and the committee can slot winner of match in one spot and the loser in another.

Conferences would be smart to schedule RPI bombs for their midlevel teams on the last Saturday.

[trojansc]

Nov 16, 2024 20:41:19 GMT -5 GatorsChomp said:

Correct me if I'm wrong but I was under the impression that it was like a pretty well known thing that the last week of play doesn't impact the seedings for the tournament ? Like isn't that a thing?

Are you asking if like that's an official thing? It's definitely not.

Unofficially? There is a good amount of evidence that supports the idea that the committee mostly ignores a lot of last-week stuff. Though there is also evidence that the committee will overreact to last-minute losses too and too harshly punish teams, it's random to predict.

     
In the case of Stanford vs. San Diego - Regional hosting. The committee chair said that Stanford's extra top 25 wins gave them an edge. Those "extra" top 25 wins only came on the very last day when USC slotted in front of BYU for a Top 25 spot. San Diego actually held the edge of Top 25 wins until the week concluded. So, they must have been paying attention to the RPI in the last week that season, at least in certain things THEY were looking for, because they also ignored Tennessee's loss to Alabama in the last week that season and didn't notice (or didn't care) that their RPI fell into the high 50's.     

[c4ndlelight]

Nov 16, 2024 20:46:08 GMT -5 trojansc said:

Nov 16, 2024 20:41:19 GMT -5 GatorsChomp said:

Correct me if I'm wrong but I was under the impression that it was like a pretty well known thing that the last week of play doesn't impact the seedings for the tournament ? Like isn't that a thing?

Are you asking if like that's an official thing? It's definitely not.

Unofficially? There is a good amount of evidence that supports the idea that the committee mostly ignores a lot of last-week stuff. Though there is also evidence that the committee will overreact to last-minute losses too and too harshly punish teams, it's random to predict.

     
In the case of Stanford vs. San Diego - Regional hosting. The committee chair said that Stanford's extra top 25 wins gave them an edge. Those "extra" top 25 wins only came on the very last day when USC slotted in front of BYU for a Top 25 spot. San Diego actually held the edge of Top 25 wins until the week concluded. So, they must have been paying attention to the RPI in the last week that season, at least in certain things THEY were looking for, because they also ignored Tennessee's loss to Alabama in the last week that season and didn't notice (or didn't care) that their RPI fell into the high 50's.     

That was a post-facto explanation re: USD v Stanford

[trojansc]

Nov 16, 2024 20:48:45 GMT -5 c4ndlelight said:

That was a post-facto explanation re: USD v Stanford

That's true - it might not have mattered. But the committee has also been quoted before using old RPI figures (the week before). So, I'm sure there is a high level of variance of how much attention they pay or consideration they give to the Final RPI - though I imagine it's not a ton in general.

[Kearney Kingston]

I understand that the ncaa committee is looking now at the KPI not the RPI. What is the difference?

[maigrey]

Nov 16, 2024 21:03:22 GMT -5 Kearney Kingston said:

I understand that the ncaa committee is looking now at the KPI not the RPI. What is the difference?

Where did you see that? I don't think that is true.

[Kearney Kingston]

I believe I heard that from a college coach in an interview recently.

[avid 2.0]

its not KPI over RPI.

Its KPI and RPI.

but there was no (big) evidence last year that they used KPI to make any of their decisions

[bbg95]

Nov 16, 2024 14:54:24 GMT -5 trojansc said:

Nov 15, 2024 18:11:12 GMT -5 buttery said:

I think it’s moot because we all spend way too much time looking at the data and numbers than the committee seems too. And that isn’t a testament to the committee being bad (though they’ve had their handful of questionable decisions over the years), but more a testament to how much time people put into analyzing the RPI Futures, quality wins etc. on this board. vollleytalk’s culture is an extremely data driven one and is impressive among the college sport landscape.

The reason why we are hyper-RPI (data) obsessed is because the committee has made us that way - not the other way around. If you could have a 22 RPI and miss the tournament, or a 68 RPI and make the tournament, things might be different. But through years of their decisions - we know how they select teams, and RPI is just soooooooo important.

The committee's process is not exactly robust or thorough - it's pretty simple. But even when dumbed down, they find ways to make extremely questionable decision.

Like, one of the complaints is that they don't even take the Final RPI into account. How in the world do you put such an emphasis on RPI and then NOT see that Tennessee and Ole Miss fell into the high 50s of RPI on the last week of the season.

Is there a reason for this other than laziness?

[bluepenquin]

Nov 16, 2024 21:31:02 GMT -5 bbg95 said:

Nov 16, 2024 14:54:24 GMT -5 trojansc said:

The reason why we are hyper-RPI (data) obsessed is because the committee has made us that way - not the other way around. If you could have a 22 RPI and miss the tournament, or a 68 RPI and make the tournament, things might be different. But through years of their decisions - we know how they select teams, and RPI is just soooooooo important.

The committee's process is not exactly robust or thorough - it's pretty simple. But even when dumbed down, they find ways to make extremely questionable decision.

Like, one of the complaints is that they don't even take the Final RPI into account. How in the world do you put such an emphasis on RPI and then NOT see that Tennessee and Ole Miss fell into the high 50s of RPI on the last week of the season.

Is there a reason for this other than laziness?

Nope.

There is no reason they couldn't wait until the last match is played and work through the bracket.  Stay up all night if you have to, but there is more than enough time to do all of this on Sunday.

[bluepenquin]

re: reason's given by the chairman for what they did.

The committee has several members (7-10?) that have to come up with a consensus for everything. From what I can tell - these people are pretty knowledgeable of the teams, well aware of the metrics used to make the selections and fill out the bracket. But there is no way they are going to agree on most of this. Selections and seeding end up being compromises where the defining reasons for the selection could be varied and not consistent for other selections. Then the the committee chair is asked on why they did something - has to give a reason that may not be a consensus reason and most likely not consistent with other decisions made.

I would take all reasons given by the chairman with a grain of salt and not apply this across all the decisions made by the committee - they are making all kinds of compromises.

[bbg95]

Nov 16, 2024 20:46:08 GMT -5 trojansc said:

Nov 16, 2024 20:41:19 GMT -5 GatorsChomp said:

Correct me if I'm wrong but I was under the impression that it was like a pretty well known thing that the last week of play doesn't impact the seedings for the tournament ? Like isn't that a thing?

Are you asking if like that's an official thing? It's definitely not.

Unofficially? There is a good amount of evidence that supports the idea that the committee mostly ignores a lot of last-week stuff. Though there is also evidence that the committee will overreact to last-minute losses too and too harshly punish teams, it's random to predict.
     
In the case of Stanford vs. San Diego - Regional hosting. The committee chair said that Stanford's extra top 25 wins gave them an edge. Those "extra" top 25 wins only came on the very last day when USC slotted in front of BYU for a Top 25 spot. San Diego actually held the edge of Top 25 wins until the week concluded. So, they must have been paying attention to the RPI in the last week that season, at least in certain things THEY were looking for, because they also ignored Tennessee's loss to Alabama in the last week that season and didn't notice (or didn't care) that their RPI fell into the high 50's.     

Yeah, seems like they were just looking for a reason to give it to Stanford instead of San Diego.

[trojansc]

Nov 16, 2024 21:44:15 GMT -5 bluepenquin said:

re: reason's given by the chairman for what they did.

The committee has several members (7-10?) that have to come up with a consensus for everything. From what I can tell - these people are pretty knowledgeable of the teams, well aware of the metrics used to make the selections and fill out the bracket. But there is no way they are going to agree on most of this. Selections and seeding end up being compromises where the defining reasons for the selection could be varied and not consistent for other selections. Then the the committee chair is asked on why they did something - has to give a reason that may not be a consensus reason and most likely not consistent with other decisions made.

I would take all reasons given by the chairman with a grain of salt and not apply this across all the decisions made by the committee - they are making all kinds of compromises.

Compromising like when the Bowling Green committee member really wants their friend Ball State to get an at-large bid even though it makes no sense with the criteria. I don't know how much of it is actual criteria disagreement or people arguing with... "motives"

I think this was the best example of incompetence recently. Though I really, really disliked SFA last year - at least you can argue RPI. There was no legitimate argument for Ball State the year prior.

I also think the committee's laziness was emphasized last year when they wouldn't give a first 4 out.. they gave a first EIGHT out in no particular order...

[bluepenquin]

Nov 16, 2024 21:50:47 GMT -5 trojansc said:

Nov 16, 2024 21:44:15 GMT -5 bluepenquin said:

re: reason's given by the chairman for what they did.

The committee has several members (7-10?) that have to come up with a consensus for everything. From what I can tell - these people are pretty knowledgeable of the teams, well aware of the metrics used to make the selections and fill out the bracket. But there is no way they are going to agree on most of this. Selections and seeding end up being compromises where the defining reasons for the selection could be varied and not consistent for other selections. Then the the committee chair is asked on why they did something - has to give a reason that may not be a consensus reason and most likely not consistent with other decisions made.

I would take all reasons given by the chairman with a grain of salt and not apply this across all the decisions made by the committee - they are making all kinds of compromises.

Compromising like when the Bowling Green committee member really wants their friend Ball State to get an at-large bid even though it makes no sense with the criteria. I don't know how much of it is actual criteria disagreement or people arguing with... "motives"

I think this was the best example of incompetence recently. Though I really, really disliked SFA last year - at least you can argue RPI. There was no legitimate argument for Ball State the year prior.

I also think the committee's laziness was emphasized last year when they wouldn't give a first 4 out.. they gave a first EIGHT out in no particular order...

There could certainly be those kinds of compromises (but I would hope not).

You follow this WAY more than I do, but I generally get the sense that past committees spend way more time on the top 16 than the last 4 in or first 4 out.  And with regional host - they will spend a lot more on the top 4.  I think this is why we see (I think it is this way) more deviations from RPI for those first 16 then at the end?  

[bluepenquin]

Nov 16, 2024 21:47:09 GMT -5 bbg95 said:

Nov 16, 2024 20:46:08 GMT -5 trojansc said:

Are you asking if like that's an official thing? It's definitely not.

Unofficially? There is a good amount of evidence that supports the idea that the committee mostly ignores a lot of last-week stuff. Though there is also evidence that the committee will overreact to last-minute losses too and too harshly punish teams, it's random to predict.
     
In the case of Stanford vs. San Diego - Regional hosting. The committee chair said that Stanford's extra top 25 wins gave them an edge. Those "extra" top 25 wins only came on the very last day when USC slotted in front of BYU for a Top 25 spot. San Diego actually held the edge of Top 25 wins until the week concluded. So, they must have been paying attention to the RPI in the last week that season, at least in certain things THEY were looking for, because they also ignored Tennessee's loss to Alabama in the last week that season and didn't notice (or didn't care) that their RPI fell into the high 50's.     

Yeah, seems like they were just looking for a reason to give it to Stanford instead of San Diego.

I don't remember the details - but Trojansc wrote that the last weekend of games changed some T25 wins and RPI to Stanford's favor.  My thinking (and I have zero data to support this) - it was/is very easy to switch #4 and #5 at the last minute - and wait for all matches to play.  You could flip 12 and 13 or leave them the same and no one would know the difference.  You start switching around at 17 through 64 - and you all kinds of things to consider (especially 33-64) as you now have to rework optimal travel considerations.