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

How bad is the Big 10 RPI this year?

 
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
CBiebel



Joined: 23 Dec 2004
Posts: 1054
Location: PA


Back to top
PostPosted: 03/08/17 5:00 am    ::: How bad is the Big 10 RPI this year? Reply Reply with quote

For all the talk about how weak UConn's conference is, the #1 Big 10 team (Maryland) is just 1 spot ahead of the #2 AAC team (Temple) in the RPI and the #2 Big 10 team (tOSU) is only 4 spots ahead of the #3 AAC team (USF).


purduefanatic



Joined: 10 Aug 2011
Posts: 2819
Location: Indiana


Back to top
PostPosted: 03/08/17 8:34 am    ::: Reply Reply with quote

Yes, this has been discussed in several threads and acknowledged by pretty much everyone on this board. The big problem this year was that the 14-team Big Ten had 4 teams that were not very good at all (Nebraska, Rutgers, Wisconsin and Illinois). Here are the current RPI's of the 2 leagues yet again:

AAC
1 - UConn
18 - Temple
29 - South Florida
76 - Tulane
95 - Central Florida
101 - SMU
144 - Tulsa
162 - Memphis
168 - East Carolina
175 - Houston
181 - Cincinnati

Big Ten
17 - Maryland
25 - Ohio State
42 - Michigan State
45 - Michigan
50 - Indiana
53 - Purdue
59 - Iowa
68 - Northwestern
71 - Minnesota
85 - Penn State

197 - Nebraska
207 - Illinois
211 - Wisconsin
226 - Rutgers

So, after looking at those numbers, the AAC has 5 Top 100 teams while the Big Ten has 10. To take it a step further, the AAC only has 3 Top 75 teams while the Big Ten has 9.

It has also been discussed about Maryland's very weak out of conference schedule and that Ohio State didn't win any of their tough out of conference games, with the exception of Syracuse (lost to South Carolina, Baylor, UConn and Miami).

We can take a look at the schedules of the AAC teams other than UConn but honestly, they all benefit from playing UConn twice (or even 3 times in some cases). Temple's big wins are DePaul, South Florida, Quinnipiac and Penn while they had a very bad loss to Hampton while also losing to Harvard & Florida. South Florida's only top 50 wins are in their conference (Temple twice) while having a bad loss to Long Beach State.

When we look at games vs UConn:

Maryland lost by 6, 87-81 at UConn
Ohio State lost by 19, 82-63 at home
Nebraska lost by 43, 84-41 at home

Temple lost by 28 at home (97-69) & by 45 (90-45) at UConn
South Florida lost by 65 at UConn (102-37), by 28 at home (96-6Cool and by 56 in the championship game, 100-44.

* Regarding Nebraska...they only had 7 wins on the season but 3 of them were over Top 50 teams (Colorado State, Indiana and Michigan State). Crazy.[/u]


Phil



Joined: 22 Oct 2011
Posts: 1255



Back to top
PostPosted: 03/08/17 10:57 am    ::: Reply Reply with quote

There is a whole lot of cherry picking going on.

Comparing conferences was tougher than some people realize. While it sounds like a simple question to ask whether conference A is better than conference B, that isn't even a well-defined question. What on earth does it mean? One possible answer is to compare the average strength of each conference. Another possibility is to imagine the two conferences playing each other with all teams playing all other teams and observing who would have more wins. This obviously has to be done hypothetically.

Some others might declare one conferences than another conference based on how many teams over some period of time go to the NCAA conference or how many wins they have, which is a legitimate thing to think about but is more dependent on the top end of the conference in the overall conference.

Conference comparisons are problematic when the conferences of the same size but when they are did a different size this becomes even more difficult.

However, I suggest that arbitrarily dropping the weakest and teams from the larger conference is a classic example of cherry picking.


How many teams are in the top N is a metric but it is a metric that mathematically favors the larger conference. Why not look at the proportion?

If you are going to look at teams in the top N, you have to choose a value for N. identifying values for which your preferred conference is dominant is the classic example of cherry picking.

Yes Big Ten has more in the top 100 and more in the top 75. 100 is a nice round number, but my guess is that almost nobody uses the top 75 for anything other than this particular exercise.

I won't pretend that the following is proof that the AAC is better than the big 10, merely that if one is going to cherry pick there are other things one can pick and they give different indications.

For example which conference has more teams in the top five?
Answer AAC

Which conference has more teams in the top 20?
Answer AAC

Which conference has more teams in the top 25?
Answer both have two teams

Which conference has more teams in the top 30?
Answer AAC

Again, I'm not suggesting that my selections mean that the AAC is better, but that you can get different indications for different plausible numbers.


purduefanatic



Joined: 10 Aug 2011
Posts: 2819
Location: Indiana


Back to top
PostPosted: 03/08/17 11:23 am    ::: Reply Reply with quote

Phil wrote:
There is a whole lot of cherry picking going on.

Comparing conferences was tougher than some people realize. While it sounds like a simple question to ask whether conference A is better than conference B, that isn't even a well-defined question. What on earth does it mean? One possible answer is to compare the average strength of each conference. Another possibility is to imagine the two conferences playing each other with all teams playing all other teams and observing who would have more wins. This obviously has to be done hypothetically.

Some others might declare one conferences than another conference based on how many teams over some period of time go to the NCAA conference or how many wins they have, which is a legitimate thing to think about but is more dependent on the top end of the conference in the overall conference.

Conference comparisons are problematic when the conferences of the same size but when they are did a different size this becomes even more difficult.

However, I suggest that arbitrarily dropping the weakest and teams from the larger conference is a classic example of cherry picking.


How many teams are in the top N is a metric but it is a metric that mathematically favors the larger conference. Why not look at the proportion?

If you are going to look at teams in the top N, you have to choose a value for N. identifying values for which your preferred conference is dominant is the classic example of cherry picking.

Yes Big Ten has more in the top 100 and more in the top 75. 100 is a nice round number, but my guess is that almost nobody uses the top 75 for anything other than this particular exercise.

I won't pretend that the following is proof that the AAC is better than the big 10, merely that if one is going to cherry pick there are other things one can pick and they give different indications.

For example which conference has more teams in the top five?
Answer AAC

Which conference has more teams in the top 20?
Answer AAC

Which conference has more teams in the top 25?
Answer both have two teams

Which conference has more teams in the top 30?
Answer AAC

Again, I'm not suggesting that my selections mean that the AAC is better, but that you can get different indications for different plausible numbers.


LOL!!! Thank you for all that. My response was to the Notre Dame fan that cherry picked his information. I didn't suggest dropping the last teams from the Big Ten but merely pointed out that when 4 of the 14 teams in your conference are that bad, it acts as an anchor to the rest of the league.

Oh, and to your point about proportion...one would also surmise that a bigger conference would also have a bigger chance of having more teams outside the top 100 (or some other random number you want to attach) correct? Well, the AAC would have 6 (more than half of the membership) while the Big Ten would have 4 (below 30%).

Anyway, I think we all know that the Big Ten is down quite a bit this year. It has been discussed throughout a few threads and I have no clue why another thread was started regarding the same thing yet again but with only very limited, cherry picked numbers as well.

Rolling Eyes Rolling Eyes Rolling Eyes Rolling Eyes Rolling Eyes Rolling Eyes Rolling Eyes


linkster



Joined: 27 Jul 2012
Posts: 5408



Back to top
PostPosted: 03/08/17 2:22 pm    ::: Reply Reply with quote

Evidence aside, what is revealing is that any P-5 conference is even being compared to a grossly top-heavy mid-major.


5thmantheme



Joined: 11 Apr 2016
Posts: 540



Back to top
PostPosted: 03/08/17 3:47 pm    ::: Reply Reply with quote

Sagarin's got it like this today :

1 PAC12 = 86.23 86.74 ( 1) 12 86.47 ( 1)
2 BIG12 = 85.95 86.58 ( 2) 10 85.92 ( 2)
3 A C C = 83.59 83.91 ( 4) 15 83.67 ( 4)
4 S E C = 83.49 84.52 ( 3) 14 83.86 ( 3)
5 BIG10 = 79.87 79.56 ( 5) 14 79.88 ( 5)
6 B.EAST = 74.65 74.54 ( 7) 10 74.65 ( 6)
7 AAC = 73.99 75.92 ( 6) 11 73.79 ( 7)
8 W.COAST = 69.99 70.26 ( Cool 10 70.10 ( Cool
9 MID-AM = 69.58 69.01 ( 9) 12 69.38 ( 9)


CBiebel



Joined: 23 Dec 2004
Posts: 1054
Location: PA


Back to top
PostPosted: 03/09/17 10:30 pm    ::: Reply Reply with quote

purduefanatic wrote:


LOL!!! Thank you for all that. My response was to the Notre Dame fan that cherry picked his information. I didn't suggest dropping the last teams from the Big Ten but merely pointed out that when 4 of the 14 teams in your conference are that bad, it acts as an anchor to the rest of the league.

Oh, and to your point about proportion...one would also surmise that a bigger conference would also have a bigger chance of having more teams outside the top 100 (or some other random number you want to attach) correct? Well, the AAC would have 6 (more than half of the membership) while the Big Ten would have 4 (below 30%).

Anyway, I think we all know that the Big Ten is down quite a bit this year. It has been discussed throughout a few threads and I have no clue why another thread was started regarding the same thing yet again but with only very limited, cherry picked numbers as well.

Rolling Eyes Rolling Eyes Rolling Eyes Rolling Eyes Rolling Eyes Rolling Eyes Rolling Eyes


Well, first of all, I hadn't really been reading the other threads that much.

Second, I just came across this when, ironically, I was trying to defend the Big 10 on another board when someone there said that "the AAC should be a Power 5 conference" after looking at the conference rankings. I figured "Well, that's probably just UConn at the top," went to check out the next few top teams in each conference, and was actually surprised at the #2 AAC team and #1 Big 10 team being very close, and the #3 AAC team and #2 Big 10 team being close, so even when you took out UConn they top 2 in each conference was even. The rest wasn't so much cherry picking as laziness on my part.

Maryland's scheduling didn't help, either. For their OOC, the average opponent RPI was 167, which was below both Temple (129.7 ave. for OOC opponents) and USF (160.6) and that was with Maryland benefiting from playing UConn and Temple dragged down by Rutgers. Maryland's scheduling of UConn and Louisville wasn't enough to counter the really bad teams on their schedule.

And as a long time ND fan, I'm well aware of the drag a few teams at the bottom of the conference can have on RPI. The Big East had Providence and the pre-Barnes-Arico SJU, both of whom would be in the 300s. What Barnes-Arico did at SJU was amazing.


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
Page 1 of 1

 
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