|
|
Post by ay2013 on Oct 26, 2015 23:48:30 GMT -5
Based on this futures I'd hand down seeds as follows: 1-USC 2-Texas 3-Minnesota 4-Penn State 5-Florida 6-Ohio State 7-Nebraska 8-Kansas 9-Washington 10-UCLA 11-BYU 12-Kentucky 13-Wisconsin 14-Stanford 15-FSU 16-Missouri/Louisville I love these kind of discussions - here is what I think it will look like (same 16, Louisville instead of Missouri)
1. USC
8. UCLA
9. Ohio State
16. Louisville
2. Texas
7. Nebraska
10. Wisconsin
15. Florida State
3. Minnesota
6. Washington
11. Kansas
14. BYU
4. Penn State
5. Florida
12. Stanford
13. Kentucky
I was thinking of switching #11 and #12 - but could you imagine what people would say about a region that had Penn State, Florida, Kansas, and Kentucky, oh my would there be a lot of howling on that!!!
The problem I have bumping Washington that high is where the top 25/50 wins are going to come from. The Pac-12 is way down this year based on a paltry preseason conference record (comparatively speaking). UW just isn't given as many opportunities as some of those other teams for the rest of the season to play top 25 and top 50 teams. By seasons end Nebraska and Ohio State will have way more top 25 matches (and probably wins) and certainly more top 50 wins (kansas probably too). I just don't see it. |
|
|
|
Post by trollhunter on Oct 27, 2015 0:34:26 GMT -5
I believe you've mentioned in the past that these projections play out pretty accurately -- 90 percent, maybe? I'm thinking Minnesota must mess up a big chunk of that other ten percent, given that they started the season projected as a 10-10 team in conference play and they're now at 18-2? I don't think Minnesota's projected SOS has changed much since beginning of season. I think you are talking more about the Pablo component of Blues Future projection. The first Pablo of season is not super accurate and uses some of last years ranking. I'm sure Bofa will tell you more. |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Oct 27, 2015 2:31:51 GMT -5
Thanks for this bluepenquin. What would the +1/-1 for Hawaii be?
|
|
|
|
Post by dd2000 on Oct 27, 2015 8:08:48 GMT -5
That Texas bracket would be tough Why? It doesn't look distinctly more difficult than any other region. Agree. USC is tough. Minnesota is tough. Penn State will be tough. You can't tell me Penn State, Florida, and Stanford won't be ready to play good volleyball come tournament time. To me none of them stand out as being significantly harder than the other three. |
|
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) All-VolleyTalk 2nd Team 2023
Posts: 13,285 
Member is Online
|
Post by bluepenquin on Oct 27, 2015 8:30:19 GMT -5
I love these kind of discussions - here is what I think it will look like (same 16, Louisville instead of Missouri)
1. USC
8. UCLA
9. Ohio State
16. Louisville
2. Texas
7. Nebraska
10. Wisconsin
15. Florida State
3. Minnesota
6. Washington
11. Kansas
14. BYU
4. Penn State
5. Florida
12. Stanford
13. Kentucky
I was thinking of switching #11 and #12 - but could you imagine what people would say about a region that had Penn State, Florida, Kansas, and Kentucky, oh my would there be a lot of howling on that!!!
The problem I have bumping Washington that high is where the top 25/50 wins are going to come from. The Pac-12 is way down this year based on a paltry preseason conference record (comparatively speaking). UW just isn't given as many opportunities as some of those other teams for the rest of the season to play top 25 and top 50 teams. By seasons end Nebraska and Ohio State will have way more top 25 matches (and probably wins) and certainly more top 50 wins (kansas probably too). I just don't see it. I took some liberties on what I think (might) happen - and I am usually wrong. I am thinking if Washington goes 18-2 or 17-3 in conference, they are getting a 4th to 6th seed much like Penn State did last year getting a higher seed than their RPI. If they go 16-4, then probably not a top 8 seed. Wins against UCLA, USC, and Stanford (all at home) will be huge to their resume.
Question: What happens to Florida if they finish the SEC a full game behind Kentucky in the SEC, but still has an RPI in the top 3? I kind of think they will lose some seeding spots because of this?
|
|
|
|
Post by vbman100 on Oct 27, 2015 9:02:26 GMT -5
So what happens if UNI were to get to 12-6 in the MVC? They get up 10 spots and they are in at-large country by making the Valley finals UNI goes 12-6, then they will be at (approximately) #52. If they win their 1st two matches of the MVC (including Missouri State) and then lose to Wichita State in the finals, they will end up at #43. Do this with an 11-7 conference record and they will land at #51.
MVC has 6 really good records, so it would be very helpful to play a 1st round game as opposed to a bye. Going 2-1 in the tournament is way more valuable then going just 1-1. Looks like they have some chance of getting 3 teams if Wichita State loses. I don't think they can get 3 if Wichita State wins the tournament?
Could you do a +1/-1 for the MVC top 6 teams? With 5 in the top 60, and 3 on top of each other at 57,58,59, it would be interesting to see what just one win or loss would mean. That conference tournament, with only 6 teams, is going to have lot riding on it. It looks like these teams better beat the teams they are supposed to beat, or pull an upset to be safe with the committee: Baylor, Arizona, Michigan, Michigan State, Pacific, Miami (FL), Virginia, Pitt, Arkansas. Just one 'unexpected' loss and they drop to the 50s in RPI. |
|
|
|
Post by The Bofa on the Sofa on Oct 27, 2015 9:05:58 GMT -5
So what happens if UNI were to get to 12-6 in the MVC? They get up 10 spots and they are in at-large country by making the Valley finals UNI goes 12-6, then they will be at (approximately) #52. If they win their 1st two matches of the MVC (including Missouri State) and then lose to Wichita State in the finals, they will end up at #43. Do this with an 11-7 conference record and they will land at #51.
MVC has 6 really good records, so it would be very helpful to play a 1st round game as opposed to a bye. Going 2-1 in the tournament is way more valuable then going just 1-1. Looks like they have some chance of getting 3 teams if Wichita State loses. I don't think they can get 3 if Wichita State wins the tournament?
I think that the RPI is going to be less important than actually making the finals. Is the conference tournament a week before selection again? If so, that will make it matter a little more (since the RPI effect will already be included when the committee arrives). I don't worry too much about last weekend matches in terms of RPI, more about significant wins and losses. |
|
|
|
Post by The Bofa on the Sofa on Oct 27, 2015 9:14:27 GMT -5
I have never taken a position on the accuracy of any given team's projected wins for the season - or what their final RPI rank will be. I only claim that the SOS is most likely not going to change much after the 3rd or 4th week and the RPI rank will be reasonably accurate in projecting the final RPI rank for any given final wins and loss. In other words, the RPI Future would have been reasonably accurate in projecting what MN RPI rank would be after the 3rd week if they won 10 conference games, 14 conference games, or 18 conference games.
Hmm, must be confusing with some other projections. In any case, thanks for the hard work! The problem is, week 3 Pablo rankings actually do a pretty good job, and can accurately predict about 75% of the rest of the matches for the season. However, when there are still 24 matches left on the schedule, getting 3/4 right means that there are 6 matches that are going to be wrong. Add in variation in distributions and things can be very different. As the season progresses, the projection gets better, not so much because the rankings improve a lot (although they improve some), but because there are fewer matches to project. When you have to project a lot of matches, there is more variation. As BluePenguin notes, however, this is due to the actions of the teams themselves. The reason Minnesota improved it's projected RPI is because they started winning more. And so the difference between RPI Futures and Current RPI is that your current RPI is subject to change based on who you play in addition to how you do. If you want to move up in RPI futures, you have to win matches you weren't expected to win. It's extremely elegant. |
|
|
|
Post by The Bofa on the Sofa on Oct 27, 2015 9:16:51 GMT -5
I believe you've mentioned in the past that these projections play out pretty accurately -- 90 percent, maybe? I'm thinking Minnesota must mess up a big chunk of that other ten percent, given that they started the season projected as a 10-10 team in conference play and they're now at 18-2? I don't think Minnesota's projected SOS has changed much since beginning of season. I think you are talking more about the Pablo component of Blues Future projection. The first Pablo of season is not super accurate and uses some of last years ranking. As I mentioned above, it's not that it's not accurate, it's just that there are so many matches to project. Even if you always have a 75% prediction rate, it's a lot easier to predict the record of a team with 20 matches played and 12 to go than it is for a team with 24 matches left to play. |
|
|
|
Post by The Bofa on the Sofa on Oct 27, 2015 9:22:15 GMT -5
UNI goes 12-6, then they will be at (approximately) #52. If they win their 1st two matches of the MVC (including Missouri State) and then lose to Wichita State in the finals, they will end up at #43. Do this with an 11-7 conference record and they will land at #51.
MVC has 6 really good records, so it would be very helpful to play a 1st round game as opposed to a bye. Going 2-1 in the tournament is way more valuable then going just 1-1. Looks like they have some chance of getting 3 teams if Wichita State loses. I don't think they can get 3 if Wichita State wins the tournament?
Could you do a +1/-1 for the MVC top 6 teams? With 5 in the top 60, and 3 on top of each other at 57,58,59, it would be interesting to see what just one win or loss would mean. That conference tournament, with only 6 teams, is going to have lot riding on it. It looks like these teams better beat the teams they are supposed to beat, or pull an upset to be safe with the committee: Baylor, Arizona, Michigan, Michigan State, Pacific, Miami (FL), Virginia, Pitt, Arkansas. Just one 'unexpected' loss and they drop to the 50s in RPI. Just a general comment: Comments like yours here just shows how far we have come in our approach to understanding tournament selection. We've so gotten away from the conference breakdowns of "who's in, who's on the bubble, etc". That we can sit here now and see that, wow, the MVC tournament has a lot riding on it (and you are absolutely right, it does) is really cool. Or here are the teams that are going to need to win as expected puts a new take on the "bubble" I really, really like the way this is going. Good on BluePenguin for starting this, and good on y'all for working with it. As I said in one of the other "tournament prediction" threads, if you don't start with BluePenguin's RPI futures, you aren't doing to get it. |
|
|
|
Post by spikerthemovie on Oct 27, 2015 11:58:06 GMT -5
Hmm, must be confusing with some other projections. In any case, thanks for the hard work! The problem is, week 3 Pablo rankings actually do a pretty good job, and can accurately predict about 75% of the rest of the matches for the season. However, when there are still 24 matches left on the schedule, getting 3/4 right means that there are 6 matches that are going to be wrong. Add in variation in distributions and things can be very different. As the season progresses, the projection gets better, not so much because the rankings improve a lot (although they improve some), but because there are fewer matches to project. When you have to project a lot of matches, there is more variation. As BluePenguin notes, however, this is due to the actions of the teams themselves. The reason Minnesota improved it's projected RPI is because they started winning more. And so the difference between RPI Futures and Current RPI is that your current RPI is subject to change based on who you play in addition to how you do. If you want to move up in RPI futures, you have to win matches you weren't expected to win. It's extremely elegant. That's the point I was (inelegantly) trying to make: Minnesota is a team that is "screwing things up" quite a bit by winning significantly more matches than projected. |
|
|
|
Post by The Bofa on the Sofa on Oct 27, 2015 12:01:51 GMT -5
The problem is, week 3 Pablo rankings actually do a pretty good job, and can accurately predict about 75% of the rest of the matches for the season. However, when there are still 24 matches left on the schedule, getting 3/4 right means that there are 6 matches that are going to be wrong. Add in variation in distributions and things can be very different. As the season progresses, the projection gets better, not so much because the rankings improve a lot (although they improve some), but because there are fewer matches to project. When you have to project a lot of matches, there is more variation. As BluePenguin notes, however, this is due to the actions of the teams themselves. The reason Minnesota improved it's projected RPI is because they started winning more. And so the difference between RPI Futures and Current RPI is that your current RPI is subject to change based on who you play in addition to how you do. If you want to move up in RPI futures, you have to win matches you weren't expected to win. It's extremely elegant. That's the point I was (inelegantly) trying to make: Minnesota is a team that is "screwing things up" quite a bit by winning significantly more matches than projected. There are always teams like that. It's not screwing anything up. It's what the system is. |
|
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) All-VolleyTalk 2nd Team 2023
Posts: 13,285
Member is Online
|
Post by bluepenquin on Oct 27, 2015 18:12:49 GMT -5
32. Wichita State (16 Wins-30) (17 Wins-26) (15 Wins-35)
48. Missouri State (12 Wins-48) (13 Wins-41) (11 Wins-55)
57. Southern Illinois (12 Wins-55) (13 Wins-48) (11 Wins-60)
58. Northern Iowa (11 Wins-59) (12 Wins-51) (10 Wins-67)
59. Illinois State (12 Wins-60) (13 Wins-55) (11 Wins-69)
71. Loyola-Chicago (12 Wins-69) (13 Wins-61) (11 Wins-85)
21. Creighton (17 Wins-21) (18 Wins-17) (16 Wins-25)
27. Marquette (14 Wins-27) (15 Wins-23) (13 Wins-32)
38. Villanova (15 Wins-36) (16 Wins-30) (14 Wins-40)
26. Hawaii (15 Wins-28) (16 Wins-24) (14 Wins-34)
51. Long Beach State (13 Wins-51) (14 Wins-45) (12 Wins-60)
33. Colorado State (18 Wins-29) (-) (17 Wins-35)
54. Boise State (14 Wins-55) (15 Wins-49) (13 Wins-61)
55. Wyoming (12 Wins-55) (13 Wins-48) (11 Wins-61)
15. Western Kentucky (16 Wins-14) (-) (15 Wins-19)
17. Arkansas State (15 Wins-19) (16 Wins-15) (14 Wins-25)
25. SMU (19 Wins-24) (-) (18 Wins-26)
31. Dayton (13 Wins-33) (14 Wins-27) (12 Wins-39)
37. Ohio (15 Wins-36) (-) (14 Wins-40)
41. James Madison (15 Wins-41) (16 Wins-35) (14 Wins-50)
42. Idaho State (14 Wins-42) (15 Wins-36) (13 Wins-50)
46. Northern Arizona (14 Wins-45) (15 Wins-36) (13 Wins-50)
60. Lipscomb (12 Wins-60) (13 Wins-51) (11 Wins-67)
61. Saint Louis (11 Wins-61) (12 Wins-51) (10 Wins-68)
62. Cleveland State (14 Wins-62) (15 Wins-56) (13 Wins-71)
64. Appalachian State (14 Wins-67) (15 Wins-58) (13 Wins-85)
65. UNC Wilmington (12 Wins-68) (13 Wins-61) (11 Wins-86)
|
|
|
|
Post by The Bofa on the Sofa on Oct 27, 2015 19:48:05 GMT -5
32. Wichita State (16 Wins-30) (17 Wins-26) (15 Wins-35) 48. Missouri State (12 Wins-48) (13 Wins-41) (11 Wins-55) 57. Southern Illinois (12 Wins-55) (13 Wins-48) (11 Wins-60) 58. Northern Iowa (11 Wins-59) (12 Wins-51) (10 Wins-67) 59. Illinois State (12 Wins-60) (13 Wins-55) (11 Wins-69) 71. Loyola-Chicago (12 Wins-69) (13 Wins-61) (11 Wins-85) Yeah, that is interesting. I'm seeing two bids here, possibly three. The problem is that for UNI or SIU to get a bid, they have to take out someone above them. Wich St could survive it, but Missouri State certainly can't. So UNI or SIU is going to have to upset Wich St to get a three bid league. That makes sense. Hawaii's RPI is already pretty much W/L saturated, so nothing they can do would have a big effect. |
|
|
|
Post by vbman100 on Oct 28, 2015 7:05:45 GMT -5
UNI goes 12-6, then they will be at (approximately) #52. If they win their 1st two matches of the MVC (including Missouri State) and then lose to Wichita State in the finals, they will end up at #43. Do this with an 11-7 conference record and they will land at #51.
MVC has 6 really good records, so it would be very helpful to play a 1st round game as opposed to a bye. Going 2-1 in the tournament is way more valuable then going just 1-1. Looks like they have some chance of getting 3 teams if Wichita State loses. I don't think they can get 3 if Wichita State wins the tournament?
I think that the RPI is going to be less important than actually making the finals. Is the conference tournament a week before selection again? If so, that will make it matter a little more (since the RPI effect will already be included when the committee arrives). I don't worry too much about last weekend matches in terms of RPI, more about significant wins and losses. 4 weeks of reg season play left, and the conf tourn is over Thanksgiving Day weekend Nov 26-28 (Thur, Fri, Sat). Selections are made on Sun Nov 29. Currently, Loyola sits at 2nd in the conf, and Mo State is 6th. A long way to go to settle things, and conference seeding is totally up for grabs. |
|
