|
Post by southie on Oct 17, 2017 18:22:36 GMT -5
I think the committee will scrutinize every possible detail.
For Nebraska, their loss against Northern Iowa (projected RPI #44) could be a black mark.
|
|
|
Post by gnu2vball on Oct 17, 2017 18:32:37 GMT -5
I think the committee will scrutinize every possible detail. For Nebraska, their loss against Northern Iowa (projected RPI #44) could be a black mark. Room for 8,500 fans might gray it up a bit.
|
|
|
Post by jayj79 on Oct 17, 2017 18:47:23 GMT -5
I think the committee will scrutinize every possible detail. For Nebraska, their loss against Northern Iowa (projected RPI #44) could be a black mark. UNI seems to be a black mark for quite a few top teams. Nebraska, USC, Iowa State, Missouri. What a coincidence.
|
|
|
Post by lionsfan on Oct 17, 2017 19:24:09 GMT -5
I think the committee will scrutinize every possible detail. For Nebraska, their loss against Northern Iowa (projected RPI #44) could be a black mark. Room for 8,500 fans might gray it up a bit. Yeah, if Nebraska is in the conversation at all I'd be shocked if they don't get a top 4 seed. I think they're a shoo-in to win the conference (or at worst tie), and that could be enough. The only potential loss they could have coming is at Michigan State.
|
|
|
Post by southie on Oct 17, 2017 20:00:15 GMT -5
I think the committee will scrutinize every possible detail. For Nebraska, their loss against Northern Iowa (projected RPI #44) could be a black mark. Room for 8,500 fans might gray it up a bit. That's definitely an intangible.
|
|
|
Post by southie on Oct 17, 2017 20:01:50 GMT -5
I think the committee will scrutinize every possible detail. For Nebraska, their loss against Northern Iowa (projected RPI #44) could be a black mark. UNI seems to be a black mark for quite a few top teams. Nebraska, USC, Iowa State, Missouri. What a coincidence. Other than Nebraska, those other teams you listed aren't likely to even be Top 16 hosts, let alone in the discussion for at Top 4 national seed.
|
|
|
Post by slxpress on Oct 17, 2017 20:09:10 GMT -5
Room for 8,500 fans might gray it up a bit. Yeah, if Nebraska is in the conversation at all I'd be shocked if they don't get a top 4 seed. I think they're a shoo-in to win the conference (or at worst tie), and that could be enough. The only potential loss they could have coming is at Michigan State. Well, first, it's possible they could be unexpectedly tripped up by several opponents. Agreed, not likely, but possible. Second, I'd consider a potential loss at Michigan State to be at least a 50/50 shot. Maybe I just have a higher opinion of Michigan State than others. Not to be dismissive of Nebraska, who we've seen is outstanding when they've got it going, but MSU did take a set in Lincoln, and they've now gone 18-1 sets at home, including the 3-1 victory over Wisconsin. I love Michigan State's service game. They have a bunch of good scorers, quality middle blockers. Nothing against Nebraska, but I'll be rooting on the Spartans in that match.
|
|
|
Post by gnu2vball on Oct 17, 2017 20:51:03 GMT -5
Room for 8,500 fans might gray it up a bit. That's definitely an intangible. Intangible it may be, but it's quite fungible.
|
|
|
Post by trollhunter on Oct 17, 2017 21:06:25 GMT -5
For teams on the bubble, they use Top 25 (of ultimate importance) and Top 50. For seeding, they use other teams under seeding consideration and Top 25 wins mostly. Losses are usually ignored, unless it's a H2H result or there is a critical mass of bad losses (3+ sub-100 losses seems to be a risk for at-large hopefuls) For Top 4 regional hosts, I'm not sure we have enough data yet - and the Committee may still be thinking through it. I guess you are answering my 1st question in the affirmative? The reason I asked is the Nitty Gritty criteria posted by ay doesn't specifically mention Top 25 wins as a factor. That's why I thought it had to fall under the "Significant Wins and Losses" category. The manual does not go into fine detail. Last years Nitty Gritty sheet is here: extra.ncaa.org/solutions/rpi/default.aspx
|
|
bluepenquin
Hall of Fame
4-Time VolleyTalk Poster of the Year (2019, 2018, 2017, 2016), All-VolleyTalk 1st Team (2021, 2020, 2019, 2018, 2017, 2016)
Posts: 12,447
|
Post by bluepenquin on Oct 17, 2017 21:43:08 GMT -5
I guess you are answering my 1st question in the affirmative? The reason I asked is the Nitty Gritty criteria posted by ay doesn't specifically mention Top 25 wins as a factor. That's why I thought it had to fall under the "Significant Wins and Losses" category. The manual does not go into fine detail. Last years Nitty Gritty sheet is here: extra.ncaa.org/solutions/rpi/default.aspxFascinating: I looked at the 11/26 Nitty Gritty used for the tournament selection. Kansas vs Texas - the big difference is the OOC RPI - Kansas #47 vs. Texas #3. Texas was just one better in RPI (4 vs. 5). But Kansas was better in T25, 26-50, road matches, last 10 games. Creighton vs. Kansas State. Creighton #17 RPI vs. #18 for K-State. Creighton better in last 10, road games, 26-51. K-State better in T25 (both were not good) and Conference RPI. And the big difference was OOC RPI - Creighton #44 and K-State #17. So K-State gets the seed instead of Creighton. So it looks like the committee valued OOC RPI? Until you compare Penn State and San Diego. Penn State's OOC RPI was #58 and San Diego was #14. San Diego has the much better RPI (16 vs. 26). PSU gets the #16 seed - San Diego... There just isn't any reconciliation of the Penn State seed other than they were vastly underrated by RPI. As one who has spent considerable time studying RPI - OOC RPI is among the dumbest of metrics in my opinion. It is essentially the RPI after the 4th week of the season - why would we want to increase the value (reward) matches played in the 1st 4 weeks of the year over the last 10 weeks? And RPI is already a bad metric - but it works much better with a larger sample size (like most things) - and 4 weeks is such a small and insignificant sample size. Oh well!
|
|
|
Post by jayj79 on Oct 17, 2017 22:35:01 GMT -5
As one who has spent considerable time studying RPI - OOC RPI is among the dumbest of metrics in my opinion. It is essentially the RPI after the 4th week of the season - why would we want to increase the value (reward) matches played in the 1st 4 weeks of the year over the last 10 weeks? And RPI is already a bad metric - but it works much better with a larger sample size (like most things) - and 4 weeks is such a small and insignificant sample size. Oh well! Schools have some control on who they schedule for OOC matches. Where as you don't have control over your conference schedule (or who your conference-mates play). Of course, sometimes you'll schedule opponents who you think will help your RPI, but they end up having disappointing seasons, but still.
|
|
bluepenquin
Hall of Fame
4-Time VolleyTalk Poster of the Year (2019, 2018, 2017, 2016), All-VolleyTalk 1st Team (2021, 2020, 2019, 2018, 2017, 2016)
Posts: 12,447
|
Post by bluepenquin on Oct 18, 2017 8:31:37 GMT -5
As one who has spent considerable time studying RPI - OOC RPI is among the dumbest of metrics in my opinion. It is essentially the RPI after the 4th week of the season - why would we want to increase the value (reward) matches played in the 1st 4 weeks of the year over the last 10 weeks? And RPI is already a bad metric - but it works much better with a larger sample size (like most things) - and 4 weeks is such a small and insignificant sample size. Oh well! Schools have some control on who they schedule for OOC matches. Where as you don't have control over your conference schedule (or who your conference-mates play). Of course, sometimes you'll schedule opponents who you think will help your RPI, but they end up having disappointing seasons, but still. Fine - then look at something that evaluates OOC SOS if we think that is important.
|
|
|
Post by n00b on Oct 18, 2017 10:47:25 GMT -5
Schools have some control on who they schedule for OOC matches. Where as you don't have control over your conference schedule (or who your conference-mates play). Of course, sometimes you'll schedule opponents who you think will help your RPI, but they end up having disappointing seasons, but still. Fine - then look at something that evaluates OOC SOS if we think that is important. OOC SOS is also on the team sheet that the committee uses... extra.ncaa.org/solutions/rpi/Stats%20Library/16%20VB%20Team%20Sheets%20thru%20Nov%2026.pdf
|
|
|
Post by n00b on Oct 18, 2017 10:53:47 GMT -5
Fascinating: I looked at the 11/26 Nitty Gritty used for the tournament selection. Kansas vs Texas - the big difference is the OOC RPI - Kansas #47 vs. Texas #3. Texas was just one better in RPI (4 vs. 5). But Kansas was better in T25, 26-50, road matches, last 10 games. Creighton vs. Kansas State. Creighton #17 RPI vs. #18 for K-State. Creighton better in last 10, road games, 26-51. K-State better in T25 (both were not good) and Conference RPI. And the big difference was OOC RPI - Creighton #44 and K-State #17. So K-State gets the seed instead of Creighton. So it looks like the committee valued OOC RPI? Until you compare Penn State and San Diego. Penn State's OOC RPI was #58 and San Diego was #14. San Diego has the much better RPI (16 vs. 26). PSU gets the #16 seed - San Diego... There just isn't any reconciliation of the Penn State seed other than they were vastly underrated by RPI. As one who has spent considerable time studying RPI - OOC RPI is among the dumbest of metrics in my opinion. It is essentially the RPI after the 4th week of the season - why would we want to increase the value (reward) matches played in the 1st 4 weeks of the year over the last 10 weeks? And RPI is already a bad metric - but it works much better with a larger sample size (like most things) - and 4 weeks is such a small and insignificant sample size. Oh well! Keep in mind that it is a collection of 8 people who vote on each and every seed ranking. Some people will value T25 wins more, some will value OOC SOS, some will penalize bad losses, etc. Then at the end of the day when all the voting is finished the committee chairperson uses the highlights of a team's resume to say this is why we chose them. That's why it looks like they change what they're looking for every year. Because every resume is looked at subjectively and then after the votes are complete, they just talk about whichever stat backs up the vote.
|
|
bluepenquin
Hall of Fame
4-Time VolleyTalk Poster of the Year (2019, 2018, 2017, 2016), All-VolleyTalk 1st Team (2021, 2020, 2019, 2018, 2017, 2016)
Posts: 12,447
|
Post by bluepenquin on Oct 18, 2017 12:43:38 GMT -5
Looking at some numbers:
38% chance that both Florida and Kentucky have an RPI in the top 4. 7% chance they end the season #1 and 2.
62% chance that the B1G has 2 teams in the top 4 in RPI and ~ 6% chance they end up with 3 teams. 34% chance that Penn State and Minnesota end up in the top 4 in RPI.
~ 90% chance that at least 5 teams from the PAC have an RPI 16 or better. 8% chance that they have 6 teams.
|
|