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 Nov 17, 2014 13:32:00 GMT -5
[quote author=" bluepenquin" source="/post/1222280/thread" timestamp="1416233162 One other thing to look at for your Futures is that some of the B1G teams may get the Schedule Bonus. With the unbalanced schedule, it won't be everyone, but with Purdue, Michigan, Michigan State, Minnesota, and Northwestern hovering around 50 rpi, it could easily happen. [/quote] I thought I checked this and there were no other teams that would qualify. However, it appears that Michigan would get the Sechdule Bonus with 16 of their 30 matches against the top 50. Michigan State has 13 of 31, Illinois has 14 of 31, Minnesota has 11 of 31, Northwestern has 12 of 31, and Purdue has 9 of 32 come up short. This is where Rutgers and Maryland could have a net negative impact on the other's RPI.
|
|
|
Post by trollhunter on Nov 17, 2014 16:22:35 GMT -5
Blue,
I wasn't clear on this. I'm saying that all those teams may end up being within the top 50, gaining some of the other B1G teams 2-4 more top 50 matches for schedule bonus.
For instance right now Wisconsin is listed with 11 of 26 top 50 matches. 3 of their last 4 are also top 50 matches putting them near bubble with 14 of 30 matches within top 50 (if Michigan makes top 50 Wisky will have 16 top 50 matches). Illinois currently appears to have 15 of 27 with and could finish 17 of 31 versus top 50. PSU will miss cutoff. Nebraska will be on Schedule Bonus bubble. Ohio State will miss. Michigan and Michigan State will probably make it.
Whether a couple of the teams listed previously as near the top 50 cutoff actually end up above or below 50 rpi will determine if some of B1G teams gets the Schedule Bonus.
|
|
|
Post by trollhunter on Nov 17, 2014 16:46:10 GMT -5
So Washington State original ncaa basic rpi rank was 117 this week and they jumped up to 110 in adjusted rpi with bonus and penalties applied. ncaastats.figstats.net/volleyball-rpi.cgi had Washington State at rpi 118 with .5370 score. Can anyone confirm these numbers with RKPI? I'm guessing that this site is including a team transitioning to DI perhaps. In any case, all of the teams Washington State leaped over are getting some sort of bonus or penalty, so can't use them to determine Schedule Bonus this week.
|
|
|
Post by n00b on Nov 17, 2014 16:56:54 GMT -5
The scheduling Bonus/Penalty only takes into account non-conference scheduling.
|
|
|
Post by ay2013 on Nov 17, 2014 17:12:23 GMT -5
The scheduling Bonus/Penalty only takes into account non-conference scheduling. IMO, the "scheduling" bonus is crap. The RPI itself ALREADY takes into consideration the NON Conference SOS. In fact, it's a rather important piece of a teams' overall RPI, so why should a team get additional bonuses for something the first calculation already shows? I can understand the wins against top 25 bonus, because the original calculation doesn't take that into consideration, but the scheduling bonus seems wholly redundant.
|
|
|
Post by The Bofa on the Sofa on Nov 17, 2014 17:16:20 GMT -5
The scheduling Bonus/Penalty only takes into account non-conference scheduling. IMO, the "scheduling" bonus is crap. The RPI itself ALREADY takes into consideration the NON Conference SOS. In fact, it's a rather important piece of a teams' overall RPI, so why should a team get additional bonuses for something the first calculation already shows? I can understand the wins against top 25 bonus, because the original calculation doesn't take that into consideration, but the scheduling bonus seems wholly redundant. Bah, I think the "good win" bonus and the bad loss penalties are also dumb, because the the committee also considers good wins and bad losses, so they effectively get double counted. So a team gets an RPI penalty for a bad loss that drops their RPI from 40 to 46. Then the committee says, "Oh, they are 46, they are on the bubble but they had a bad loss, so let's hold that against them." However, the reason they are on the bubble is because they have been penalized for their bad loss already. it's double counting.
|
|
|
Post by ay2013 on Nov 17, 2014 17:23:11 GMT -5
IMO, the "scheduling" bonus is crap. The RPI itself ALREADY takes into consideration the NON Conference SOS. In fact, it's a rather important piece of a teams' overall RPI, so why should a team get additional bonuses for something the first calculation already shows? I can understand the wins against top 25 bonus, because the original calculation doesn't take that into consideration, but the scheduling bonus seems wholly redundant. Bah, I think the "good win" bonus and the bad loss penalties are also dumb, because the the committee also considers good wins and bad losses, so they effectively get double counted. So a team gets an RPI penalty for a bad loss that drops their RPI from 40 to 46. Then the committee says, "Oh, they are 46, they are on the bubble but they had a bad loss, so let's hold that against them." However, the reason they are on the bubble is because they have been penalized for their bad loss already. it's double counting. When it comes to selection, I can agree with that. What I'm saying, however, is that the pure RPI itself includes (to a significant degree) the SOS of non conference teams, so why should those same teams get even extra points in the revised RPI? I do get that the committee sees redundant information but they, at least, can be subjective about the information they are given and how it's applied. I recall UNI got the 5 seed in 2010 with a mere 2-1 record against top 25 teams....needles to say, there were a number of teams with far better records against rpi top 25 teams but were seeded lower.
|
|
|
Post by The Bofa on the Sofa on Nov 17, 2014 17:29:34 GMT -5
Bah, I think the "good win" bonus and the bad loss penalties are also dumb, because the the committee also considers good wins and bad losses, so they effectively get double counted. So a team gets an RPI penalty for a bad loss that drops their RPI from 40 to 46. Then the committee says, "Oh, they are 46, they are on the bubble but they had a bad loss, so let's hold that against them." However, the reason they are on the bubble is because they have been penalized for their bad loss already. it's double counting. When it comes to selection, I can agree with that. What I'm saying, however, is that the pure RPI itself includes (to a significant degree) the SOS of non conference teams, so why should those same teams get even extra points in the revised RPI? I do get that the committee sees redundant information but they, at least, can be subjective about the information they are given and how it's applied. But they AREN'T! That's the problem. They still apply a "good wins" and "bad losses" screen even though RPI already accounts for good wins and bad losses. Now they are going to still look at non-conference schedule even though RPI already double counts the non-conference schedule.
|
|
|
Post by c4ndlelight on Nov 17, 2014 17:31:06 GMT -5
When it comes to selection, I can agree with that. What I'm saying, however, is that the pure RPI itself includes (to a significant degree) the SOS of non conference teams, so why should those same teams get even extra points in the revised RPI? I do get that the committee sees redundant information but they, at least, can be subjective about the information they are given and how it's applied. But they AREN'T! That's the problem. They still apply a "good wins" and "bad losses" screen even though RPI already accounts for good wins and bad losses. Now they are going to still look at non-conference schedule even though RPI already double counts the non-conference schedule. Or we could just switch to a better rating system and forget the adjustments.
|
|
|
Post by The Bofa on the Sofa on Nov 17, 2014 17:39:20 GMT -5
But they AREN'T! That's the problem. They still apply a "good wins" and "bad losses" screen even though RPI already accounts for good wins and bad losses. Now they are going to still look at non-conference schedule even though RPI already double counts the non-conference schedule. Or we could just switch to a better rating system and forget the adjustments. I'm happy to base it completely on Pablo. No committee needed.
|
|
|
Post by scubasteve on Nov 17, 2014 17:48:32 GMT -5
Is this what the expected final rpi teams would be projected to have if everyone wins and loses to whom they should relative to their rpi/Pablo, or is this just for this week?
|
|
|
Post by The Bofa on the Sofa on Nov 17, 2014 17:56:47 GMT -5
Is this what the expected final rpi teams would be projected to have if everyone wins and loses to whom they should relative to their rpi/Pablo, or is this just for this week? Projected final RPI based on projected records based on Pablo
|
|
|
Post by BeachbytheBay on Nov 17, 2014 17:57:31 GMT -5
Or we could just switch to a better rating system and forget the adjustments. I'm happy to base it completely on Pablo. No committee needed. unfortunately, the problem with using strict Pablo (as accurate in predicting as it is) is a 'fundamental selection flaw', if you want to call it that, that doesn't necessarily reward wins as much as it does points - and so runs counter to a some 'bottom line' results that occur to repeat a broken record, the compounding use of RPI when it also used to determine 'ratings sheets' with top 25 W/L, top 50 W/L, top 100, and/or 'bad losses' is significant flaw that is made. so long as committee member choose to use RPI 'tunnel vision', it's bound to have inequities in the selections this year, Michigan State, Alabama, LSU, UOP, Santa Clara, Miami, Northridge, even Hawaii (as well as Kansas/Iowa State for seeding) will be among the interesting teams to watch in how they fare in the selection/seeding as to how the committee treats those teams
|
|
|
Post by c4ndlelight on Nov 17, 2014 18:00:20 GMT -5
You forgot Miami. Won't have any Top 50 wins when the Committee meets.
|
|
|
Post by The Bofa on the Sofa on Nov 17, 2014 18:08:34 GMT -5
I'm happy to base it completely on Pablo. No committee needed. unfortunately, the problem with using strict Pablo (as accurate in predicting as it is) is a 'fundamental selection flaw', if you want to call it that, that doesn't necessarily reward wins as much as it does points - and so runs counter to a some 'bottom line' results that occur But that turns into an issue of the premise. Look at the discussions we've had about best teams. There are those with the "Winning is all that matters the end" approach. It completely begs the question, of course, of matters in what respect? Because as I've shown, if you are talking about how good teams are, winning really doesn't matter, and what matters there is how good the teams are. We've got enough data from other sports to put the burden of the "some teams know how to win" argument on those making the claim, and there's no reason to think that teams that lost while outscoring their opponents are any less than teams that win in that situation. But that presumes your focus on how good teams are. It doesn't have to be. If you start with the premise that wins are what it is about, then wins and losses are the best measure of that. There's no reason you have to start from that position, though.
|
|