RebKell's Junkie Boards
Board Junkies Forums
 
Log in Register FAQ Memberlist Search RebKell's Junkie Boards Forum Index

2018-2019 Bracketology
Goto page Previous  1, 2, 3, 4 ... 10, 11, 12  Next
 
Post new topic   Reply to topic    RebKell's Junkie Boards Forum Index » NCAA Women's Basketball - General Discussion
View previous topic :: View next topic  
Author Message
myrtle



Joined: 02 May 2008
Posts: 32326



Back to top
PostPosted: 01/28/19 7:53 pm    ::: Reply Reply with quote

pilight wrote:
The SEC continues to be overrepresented. They trail the BEast in both winning percentage and CRPI.


well he's down to 7 for the SEC and two of them are in the last four in, so there's hope it could become more reasonable.



_________________
For there is always light,
if only we’re brave enough to see it.
If only we’re brave enough to be it.
- Amanda Gorman
purduefanatic



Joined: 10 Aug 2011
Posts: 2819
Location: Indiana


Back to top
PostPosted: 01/28/19 9:51 pm    ::: Reply Reply with quote

myrtle wrote:
pilight wrote:
The SEC continues to be overrepresented. They trail the BEast in both winning percentage and CRPI.


well he's down to 7 for the SEC and two of them are in the last four in, so there's hope it could become more reasonable.


But then he also has 2 more in the Next Four Out: Auburn and LSU


patsweetpat



Joined: 14 Jul 2010
Posts: 2306
Location: Culver City, CA


Back to top
PostPosted: 01/28/19 10:30 pm    ::: Reply Reply with quote

My Bruins are on the board, baby!

If UCLA makes the tourney this year (*big* if, obvi), it'd be their 4th consecutive NCAA tourney bid... a streak that UCLAWBB has never previously accomplished in the history of its program.


Phil



Joined: 22 Oct 2011
Posts: 1258



Back to top
PostPosted: 01/29/19 11:25 am    ::: Reply Reply with quote

Charlie has Tennessee in and West Virginia out. I'll cut him some slack that he put up his latest before last night's game but still…

Tennessee's best win is a six point victory over Texas at Texas.

Last night, West Virginia had a six point victory over Texas at Texas.

Again, that was last night but that's not even West Virginia's best win.

West Virginia beat Iowa State, a team far better than any team Tennessee has beaten.

Tennessee's second best win is over Auburn, ranked number 51 by Massey and not likely to make the tournament based on their results so far.

If we call Tennessee's win over Texas a push with West Virginia's win over Texas, then look at other wins. Tennessee has Auburn.

West Virginia beat Oklahoma State, a team comparable to all quite as strong as Auburn, but they also beat Kansas State a team stronger than Auburn and Iowa State, who Charlie has in the top 16.

Even before the win over Texas, West Virginia had a better resume than Tennessee. After the win, far better.


Phil



Joined: 22 Oct 2011
Posts: 1258



Back to top
PostPosted: 01/29/19 11:36 am    ::: Reply Reply with quote

Massey ratings of:

Last four in
UCLA 35
North Carolina 39
Arkansas 42
Tennessee 46

and
First four out
West Virginia 29
Butler 34
Arizona 41
Georgia Tech 52

If it wasn't for Georgia Tech I'd wonder if he accidentally swapped the two groups.

Three of the first four out are well ahead of Tennessee. West Virginia and Butler are ahead of all four. Arizona would fit smack in the middle of them. Only Georgia Tech seems logically placed.

Yes, I know there's more to bracket apology then Massey ratings but something seems out of whack.


SpaceJunkie



Joined: 10 Sep 2012
Posts: 4241
Location: Minnesota


Back to top
PostPosted: 01/29/19 11:41 am    ::: Reply Reply with quote

Phil wrote:
Even before the win over Texas, West Virginia had a better resume than Tennessee. After the win, far better.


West Virginia got weighed down by the obscene number of cupcakes they play in non-conference, making their NC SOS terrible and weighing down their overall RPI/SOS, along with having lost to the two non-cupcake teams they actually faced, Iowa & Missouri. The committee seems to give favor to teams with good non-conference schedules/wins, not just teams that do all their winning in conference play.


Phil



Joined: 22 Oct 2011
Posts: 1258



Back to top
PostPosted: 01/29/19 12:10 pm    ::: Reply Reply with quote

I agree that West Virginia's scheduling makes me wince.

After all, they played Coppin State with the Massey rating of 713.

Compare that to Tennessee whose worst team is Florida A&M with the Massey rating of 703.

Seriously, I agree that the West Virginia scheduling is weak and they deserve to be dinged for it. But dinging them means moving them down from seven or an eight down to a 10 on an 11 not off the list.

I seriously think Charlie and other voters are remembering that Tennessee used to play an extremely aggressive nonconference schedule. They still have Notre Dame and Stanford on the schedule, so I think voters just assumed that they still have a strong nonconference schedule. but after a couple notable names they've got teams like Murray State, Presbyterian and ETSU. They played Stetson who has been good in some years but didn't have a single Division I win when they played Tennessee (and still gave them a scare).

I think team should be encouraged to play decent schedules but if you think West Virginia should play a decent schedule, shouldn't you think Tennessee should as well?


patsweetpat



Joined: 14 Jul 2010
Posts: 2306
Location: Culver City, CA


Back to top
PostPosted: 01/29/19 5:04 pm    ::: Reply Reply with quote

Phil wrote:
Yes, I know there's more to bracket apology then Massey ratings but something seems out of whack.


Does the committee even look at Massey ratings? I don't think it does.

ETA: Could be wrong, though! Frequently am!


willtalk



Joined: 13 Apr 2012
Posts: 1088
Location: NorCal


Back to top
PostPosted: 01/30/19 9:20 pm    ::: Reply Reply with quote

patsweetpat wrote:
Phil wrote:
Yes, I know there's more to bracket apology then Massey ratings but something seems out of whack.


Does the committee even look at Massey ratings? I don't think it does.

ETA: Could be wrong, though! Frequently am!
I read one of Cremes explanations for why he moved teams up or down. i seemed obvious to me that he didn't apply the same standards to all teams. He would just cherry pick info to justify his personal perspectives. I would think that if he was doing this constantly he could create a program to emphasize the aspect he wanted and eliminate the ones he felt not worth while. At least that way he could maintain a sense of consistency. I will admit that it is difficult to do so many teams, but thats what he gets paid for.

If I knew a programmer I believe I could set up a program to give a better perspective than the ones they presently use. I believe those programs were created a long time ago and have yet to be revised to take advantage of the increase in program speed. I would use aspects of what already exists in both RPI and Massy but also include more relevant variables.



_________________
No one one is ever as good as their best game, nor as bad as their worst.
patsweetpat



Joined: 14 Jul 2010
Posts: 2306
Location: Culver City, CA


Back to top
PostPosted: 01/30/19 9:58 pm    ::: Reply Reply with quote

willtalk wrote:
If I knew a programmer I believe I could set up a program to give a better perspective than the ones they presently use. I believe those programs were created a long time ago and have yet to be revised to take advantage of the increase in program speed. I would use aspects of what already exists in both RPI and Massy but also include more relevant variables.


Yes, I have no doubt that you *could* create a better system than the one used by the committee, using better indices than the one used by the committee. But Creme's project is (or at least purports to be) one of projecting the committee's choices, using the committee's tools. So if the committee doesn't use Massey (which, again, I don't think they do), then there's no real sense in anyone else using Massey to critique Creme's projections. Ditto some sort of more-relevant-Massey that you and a programmer could devise.


Phil



Joined: 22 Oct 2011
Posts: 1258



Back to top
PostPosted: 01/31/19 9:31 am    ::: Reply Reply with quote

Please don’t read too much into my references to Massey. I make reference to Massey mainly so that if I want to talk about the strength of some team, it is easier to refer to some metric, and Massey is easy to check. Plus, if I make an error, as I am wont to do, it is easy to correct me. Most of what I say, other than the rankings themselves, can be discerned from watching the games.
So, for example, when I say that WVU weakest opponent is Coppin State while Tennessee’s is Florida A&M, I could simply say that they are both abysmal, confident that any follower of basketball, using any measure including eye test, would agree.
I don’t assume that Crème or others actually consult the Massey rating, but I assume that they watch the games and base much of their assessment on the same things measured by Massey.
I may regret getting too technical but both humans and Massey use Bayesian analysis in their assessments; Massey does it formally, humans informally. (Most human voters do it without knowing what it means.) RPI does not use Bayesian concepts, which explains why early season RPI numbers are laughable, while Massey and human results are less laughable.
Massey does not incorporate injuries, so humans are more likely to assess that the UNC ND game has potential for an upset – Massey misses this. On the flip side, humans give too much weight to results that are too old, which is why humans over-estimate Tennessee and took longer to rank teams like Utah.


Phil



Joined: 22 Oct 2011
Posts: 1258



Back to top
PostPosted: 01/31/19 9:35 am    ::: Reply Reply with quote

willtalk wrote:
I would use aspects of what already exists in both RPI and Massy but also include more relevant variables.


I’d be very curious to know which variables you would incorporate that are not in Massey. In my post above, I mentioned that injuries are excluded from Massey. I can think of a few other things that might be relevant in a few games, but not a lot.


ArtBest23



Joined: 02 Jul 2013
Posts: 14550



Back to top
PostPosted: 01/31/19 11:00 am    ::: Reply Reply with quote

I know nobody seems to like the RPI, but it is, in fact, the measure created by the NCAA for selection purposes and it - not Massey or Sagarin or anyone else - is the metric still being used by the Women's Committee.

So if you're talking about selection projections, it seems to me you ought to be using RPI.

It was intentional by the NCAA not to consider MOV which is central to Massey and others. The men are moving away from that but not the women yet as I understand it.


Coyotes



Joined: 28 Jan 2018
Posts: 1467



Back to top
PostPosted: 01/31/19 3:36 pm    ::: Reply Reply with quote

By RPI:

Last Four IN:

Tennessee: 53
UCLA: 41
North Carolina: 34
Arkansas: 50

First Four Out:

West Virginia: 62
Butler: 42
Arizona: 65 (No team with an RPI worse than 60 and a losing conference record has gotten an at large bid).
Georgia Tech: 38

It seems to me that this makes it more clear cut, with Georgia Tech and Butler having more of an argument.


Looking at RPI, there are a bunch of teams that could cause bubbles to burst.

Drake (15), Gonzaga (13) or BYU (26), Central Michigan (25), Buffalo (30), The South Dakota's (32 and 33), or even New Mexico (44) in the MWC. If these teams don't win their conference tournaments....it could get dicey for some stronger teams.


patsweetpat



Joined: 14 Jul 2010
Posts: 2306
Location: Culver City, CA


Back to top
PostPosted: 01/31/19 4:54 pm    ::: Reply Reply with quote

Phil wrote:

I may regret getting too technical but both humans and Massey use Bayesian analysis in their assessments...


NERD!!!!!!!!!!


Ex-Ref



Joined: 04 Oct 2009
Posts: 8835



Back to top
PostPosted: 01/31/19 10:26 pm    ::: Reply Reply with quote

Watching the Vandy/TAM game. DA just said that Charlie Creme has updated his bracketology and now has UConn on the 2 line.

It would suck to be the 1 seed that had to go to Albany!

(I'm thinking that it could be ND.)



_________________
"Women are judged on their success, men on their potential. It’s time we started believing in the potential of women." —Muffet McGraw

“Thank you for showing the fellas that you've got more balls than them,” Haley said, to cheers from the crowd.
BD22



Joined: 27 Nov 2016
Posts: 70
Location: East


Back to top
PostPosted: 01/31/19 10:31 pm    ::: Reply Reply with quote

I'm a bit confused by why UConn would be a two with their losses being at Baylor and at Louisville (two top 4 teams) while Notre Dame remains a 1 with losses at home to UConn and on the road to UNC.

Do you all think ND's wins are that much better than UConn's? Just curious on folk's thoughts.


pilight



Joined: 23 Sep 2004
Posts: 66773
Location: Where the action is


Back to top
PostPosted: 01/31/19 10:44 pm    ::: Reply Reply with quote

BD22 wrote:
I'm a bit confused by why UConn would be a two with their losses being at Baylor and at Louisville (two top 4 teams) while Notre Dame remains a 1 with losses at home to UConn and on the road to UNC.

Do you all think ND's wins are that much better than UConn's? Just curious on folk's thoughts.


Notre Dame's wins are vastly superior to UConn's. The Hussies have only one win against a currently ranked team.



_________________
Let us not deceive ourselves. Our educational institutions have proven to be no bastions of democracy.
Ex-Ref



Joined: 04 Oct 2009
Posts: 8835



Back to top
PostPosted: 01/31/19 10:44 pm    ::: Reply Reply with quote

BD22 wrote:
I'm a bit confused by why UConn would be a two with their losses being at Baylor and at Louisville (two top 4 teams) while Notre Dame remains a 1 with losses at home to UConn and on the road to UNC.

Do you all think ND's wins are that much better than UConn's? Just curious on folk's thoughts.


I just pulled up the linked bracket from ESPN. It is dated 1/28/19. So either Deb doesn't know what she's talking about or she spoke out of turn.

Anyway, to answer your question, I don't know what CC thinks. He has NC State as the auto bid from the ACC. But he has them on the 2 line with Baylor. Baylor, I would think is the overall #1 seed at this point. That would make NC State the overall #8 seed. Supposedly ahead of Louisville and ND that he has on the 1 line. And he has ND in Greensboro, which is why he moved NC State out of that regional.

It's a mess. At least until the end of the season when speculation is over and we know where teams actually stand.



_________________
"Women are judged on their success, men on their potential. It’s time we started believing in the potential of women." —Muffet McGraw

“Thank you for showing the fellas that you've got more balls than them,” Haley said, to cheers from the crowd.
Ex-Ref



Joined: 04 Oct 2009
Posts: 8835



Back to top
PostPosted: 01/31/19 10:56 pm    ::: Reply Reply with quote

I found this when I was looking for the ACC standings. It is an "ELO" ranking. I had never heard of it before. (It started with chess, which is why elochess is in the address.) Anyway, it has Louisville #1, ND #2, and UConn #6.


http://warrennolan.com/basketballw/2019/elochess





If you want to read more about ELO ranings.

https://en.wikipedia.org/wiki/Sports_rating_system



_________________
"Women are judged on their success, men on their potential. It’s time we started believing in the potential of women." —Muffet McGraw

“Thank you for showing the fellas that you've got more balls than them,” Haley said, to cheers from the crowd.
Phil



Joined: 22 Oct 2011
Posts: 1258



Back to top
PostPosted: 02/01/19 7:22 am    ::: Reply Reply with quote

I can’t disagree with Crème that UConn drops to a 2. That said, I would have picked MS State over Oregon as the 4th top seed. Perhaps he is putting too much weight on RPI, which has Oregon ahead of MS State. Or maybe he used the head to head result as a tiebreaker, which would be a better argument, although that’s yesterday’s news. Oregon hasn’t had too much competition recently, so Friday’s game against Utah may be revealing. I think Oregon will prevail, but the MOV will tell us more about the Ducks.
MS State took apart LSU Last night, which suggests they are now hitting on all cylinders.

[edit] Another argument for OR over MS State is that it provides a better option for the Portland site.




Last edited by Phil on 02/01/19 8:34 am; edited 1 time in total
Phil



Joined: 22 Oct 2011
Posts: 1258



Back to top
PostPosted: 02/01/19 7:25 am    ::: Reply Reply with quote

BD22 wrote:
I'm a bit confused by why UConn would be a two with their losses being at Baylor and at Louisville (two top 4 teams) while Notre Dame remains a 1 with losses at home to UConn and on the road to UNC.

Do you all think ND's wins are that much better than UConn's? Just curious on folk's thoughts.


There are a lot to choose from, but I'd say ND's beatdown of Louisville makes the case almost by itself.


Phil



Joined: 22 Oct 2011
Posts: 1258



Back to top
PostPosted: 02/01/19 8:46 am    ::: Reply Reply with quote

Ex-Ref wrote:





If you want to read more about ELO ranings.

https://en.wikipedia.org/wiki/Sports_rating_system


Thanks for providing that link, I had not read it before.

It is, unfortunately, very poorly done.


PickledGinger



Joined: 04 Oct 2013
Posts: 1350



Back to top
PostPosted: 02/02/19 1:29 am    ::: Reply Reply with quote

Phil wrote:
Ex-Ref wrote:


If you want to read more about ELO ranings.

https://en.wikipedia.org/wiki/Sports_rating_system


Thanks for providing that link, I had not read it before.

It is, unfortunately, very poorly done.


Wikis are created by unpaid volunteers. So, you want better, do it yourself. Otherwise, don't criticize. Cause it's kinda dickish.


calbearman76



Joined: 02 Nov 2009
Posts: 5152
Location: Carson City


Back to top
PostPosted: 02/02/19 5:09 pm    ::: Reply Reply with quote

I actually think the Wiki article is rather well done as a primer. It does not go into depth about individual systems but it does lay out most of the variables in developing a system.

The NCAA has their RPI system which is garbage. It is a simplistic formula that can be dramatically gamed based on schedule. The NCAA realizes that so they say they only use it as a tool, and instead use that as the basis for determining an initial ranking from which they can assess individual wins and losses. Even this introduces biases because games against certain teams are given greater value based on an inflated RPI. The best example this season is Central Florida who is currently ranked 15 despite their best win being over #39 Villanova and their next best win is over #63 Quinnipiac. UCF has only 1 game left against a team ranked in the top 80, UConn. Unless they win that game UCF will likely get a 9 seed in the NCAA tournament, but for teams that beat UCF (UConn, Syracuse, Central Michigan) they will be rewarded with a top 25 win.

The ELO system is a good one for evaluating wins and losses without pointspreads, although I have seen different ELO systems with different results for the same data.

To be clear in order to devise a usable system you must determine which factors to consider. I believe that games played later in the season should be given more weight, something that most systems do not. With 30 games including margin of victory is not generally necessary, but problems do exist for widely disparate schedules. Evaluating mid-majors (Gonzaga, Drake, South Dakota, Quinnipiac) is particularly difficult.


Display posts from previous:   
Post new topic   Reply to topic    RebKell's Junkie Boards Forum Index » NCAA Women's Basketball - General Discussion All times are GMT - 5 Hours
Goto page Previous  1, 2, 3, 4 ... 10, 11, 12  Next
Page 3 of 12

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum


Powered by phpBB 2.0.17 © 2001- 2004 phpBB Group
phpBB Template by Vjacheslav Trushkin