Top 25 discussion

[jagluski]

Another ranked team loses(although not an upset)

#2 Wash U 83, #7 Brandeis 66

[jagluski]

Another(although, this is what happens in the UAA when 2 ranked teams play).

#20 Chicago loses at home to #18 NYU 83-71


NYU sweeps what Gordon referred to earlier this week as potentially the toughest road trip in D3.

[feces monkey]

That's true about the lack of transaprency, but that is my choice and it stems from experiences I have had voting on other polls. I don't want the voters subjected to lobbying from coaches and fans, so I am the front man for the poll and I take the heat.

I am, by no means, advocating transparency. While the lack of transparency is peripheral to my intended point, it is essential for this site. However, while writing to the poll in general, it is impossible (without transparency) not to refer to the pollsters as a single entity.

[Ralph Turner]

That's true about the lack of transaprency, but that is my choice and it stems from experiences I have had voting on other polls. I don't want the voters subjected to lobbying from coaches and fans, so I am the front man for the poll and I take the heat.

I am, by no means, advocating transparency. While the lack of transparency is peripheral to my intended point, it is essential for this site. However, while writing to the poll in general, it is impossible (without transparency) not to refer to the pollsters as a single entity.

monkey, the transparency issue doesn't hold because the "controlling" authority is the editor/owner in each of the various polls to which you have access.

For example,

WBCA--the organization.
Massey--Massey himself
D3Hoops--Pat Coleman
Posters' Pool--Mr Ypsi is accumulating the data.

From that, you can make an assessment of the validity of the polls.  Whom do you believe?  One possible answer would be none of them.

The Week 13 poll will be the most interesting.  Let's do a side-by-side at the end and see how close each was.  In the "quality-rich" brackets, we can use some common sense.

[jagluski]

New Poll Out.  Deservedly so, NYU is the biggest gainer of the week.

[andrewf]

George Fox drops 10 spots by losing by 1 in overtime to a top 25 team?  Seems harsh.

[Ralph Turner]

George Fox drops 10 spots by losing by 1 in overtime to a top 25 team?  Seems harsh.

Andrew, actually I think that George Fox got what they deserved.

At #4, a pure solid unanimous #4 should expect to receive 25 votes worth 22 points each or 550 votes.  Their Week #7 vote total was 477 which is more consistent with #7. (25 votes times 19 points for 7th place = 475 votes.)

They fell about 6 votes on each of the 25 ballots or about 150 votes.  Week in and week out, you will see a team lose 5-6 slots on the ballot for a loss to a weaker opponent, and I believe that #4 should always defeat a #25 in-conference foe.

477 minus 150 = 327.  They received 321 votes for 14th place in week #8.

There is much ambivalence in several ranges of this poll.  A unanimous 14th place should receive 300 points.  Brandeis is a "unanimous" 12th getting the "perfect" 350 votes.

Let's start at the top.

Southern Maine and Scranton are solid at #1 and #2.  So. Maine received 613 of the unanimous 625.  Scaranton received 593 of the "perfect 600".  A unanimous 3rd place would have had 575 votes.  Wash StL,  at 510 votes, is not getting the love you would even expect for a unanimous 5th place (525 votes).  If Southern Maine and Scranton are on opposite sides of the bracket (as the Handbook suggests), aren't they everyone's guesses as the Finalists?

This next group of teams are clustered together:  (The unanmious vote total is in parentheses.)  They kinda look like good candidates for Elite 8 depending if they are bracketed that way.  (The women's Handbook is specific about the regional matchups.   See Below.)


3 Washington U.           14-2 510  (575)
4 Mary Washington (2) 16-0 505  (550)
5 DePauw                     17-1 492  (525)
6 Bowdoin                     13-2 471  (500)
7 Hope                          16-1 465  (475)

The third group of teams is much further back.   These are solid "mid 25" teams.  They look like Sweet 16's, don't they!

8 Baldwin-Wallace        15-2 395   (450)
9 New York University   15-1 382   (425)
10 Messiah                    16-1 379  (400)
11 Wheaton (Ill.)           14-2 360  (375)
12 Brandeis                   12-2 350   (350)
13 McMurry                    16-0 344   (325)
14 George Fox               15-2 321   (300)
15 Randolph-Macon       13-2 293   (275)
16 Capital                      15-2 286   (250)
17 Salem State              13-1 259    (225)
18 Calvin                        14-2 224   (200)

I place the next teams into another group.

19 Mississippi College     13-1 133   (175)
20 Hardin-Simmons         13-2 130   (150)
21 Pacific Lutheran          14-2 103   (125)
22 Trinity (Texas)             14-2   77   (100)
23 Rochester                    13-3  66    (75)
24 Chicago                        13-3 57    (50)
25 Wilmington                    12-5 43   (25)
26 Wash & Jeff                            41
27 Moravian                                38
28 Muhlenberg                            30
T-29 UW-Oshkosh                       29
T-29 Lawrence                             29

That is the way I see this week's poll.

Appendix E in the 2006 Women's Handbook shows the Sectional Finals brackets as below.

http://www.ncaa.org/library/handbooks/basketball/2006/2006_d3_w_basketball_handbook.pdf


Left side                                                      Right Side

Atl vs. NE                                                     E vs. MA
Atl vs. NE                                                     E vs. MA

S vs. W                                                        GL vs. C
S vs. W                                                        GL vs. C

[bbald eagle]

Ralph -

Yours is one of the most interesting/insightful posts I've read in a while.

Thanks.

[Mr. Ypsi]

Ralph,

I posted a 'statistical anomaly' about the power of consistency in voters' opinions (as opposed to how high SOMEONE ranks them) on the men's top 25 board.

In the Posters' Poll (men), Wartburg placed 9th even though NOT ONE voter placed them higher than 10th!  Wartburg got placed between 10th and 16th by EVERY voter; many teams below them got votes higher than 10th, but also votes WAY lower than 16th (in some cases, being left off ballots entirely).

Teams not coming even close to the theoretical vote totals that their spot would call for must have consistently placed close to that spot, while the teams beneath them had SOME voters placing them much lower.

What I can't answer is whether that makes them more or less resistant to falling after a loss.  More stable: nearly everyone thinks they are CLOSE to that spot, even if ranked a BIT too high.  Less stable: with a broader spread of rankings, some will think they are still worthy of a high spot, others didn't think they belonged there anyway; with consistency, they may ALL feel the team should fall by 3-4 spots.

[Ralph Turner]

Mr Ypsi, I thought your presentation of the poll balloting was quite useful for all of us who have pored over those ballots.

I hope each of my compadres tried as diligently to present a fair and serious ballot as I.  Your latitude with the Posters' Poll permits us to look at this question in ways that Pat Coleman probably cannot.  I am sure that the "r" value of a comparison of the Posters' Poll and the D3Hoops Top 25 is relatively high, especially near the top.

Thanks for conducting the Poll!

[Mr. Ypsi]

Ralph,

Thanks for your kind words.  I've enjoyed doing the PP.

I feel pretty confident that all the PP voters are conscientious and knowledgable - I would not otherwise spend my time compiling their votes!  That is not to say that they are necessarily AS knowledgable and conscientious as Pat's voters - this is NOT a contest - they're the REAL poll, we're the PP!

Since this is a commentary about MEN's rankings, I will take any further comments over to the PP board.  (Though I HOPE I may have given some insights to the women's rankings as well.)

[feces monkey]

monkey, the transparency issue doesn't hold because the "controlling" authority is the editor/owner in each of the various polls to which you have access.

For example,

WBCA--the organization.
Massey--Massey himself
D3Hoops--Pat Coleman
Posters' Pool--Mr Ypsi is accumulating the data.

From that, you can make an assessment of the validity of the polls.  Whom do you believe?  One possible answer would be none of them.

The Week 13 poll will be the most interesting.  Let's do a side-by-side at the end and see how close each was.  In the "quality-rich" brackets, we can use some common sense.

Ralph, thanks for the reply.

I guess the "controlling" comment strikes me as problematic. Although you implied as such, I would like to think (regardless of the issues I have with it) that the d3hoop.com poll isn't 'controlled' by Pat Coleman. Assumably, PC is just the tabulator of the votes from his selected constituents. I must say, if he somehow massages (i.e. controlled) the ballots, I would be supremely disappointed. And mildly surprised.

Contrary to what I posted before, your comments, in general, necessitate the need for transparency. To paraphrase Pat Coleman, "it is not his choice" for non-transparency, but simply a need to keep his pollsters privacy because of the possiblity of "lobbying." You, on the other hand, seem to feel that it is PC's poll to "control." This raises questions (not to say you both are of one mind on the issue).

I agree that the week 13 poll will most likely be indicative of the tournament. D3hoops.com has a long tradition of nailing the tournament teams, both in the polls and bracket-wise. However, my original point still stands: the d3hoops.com poll is not as fluid -- does not take into consideration week-to-week results -- as it should be.

[sunny]

monkey, the transparency issue doesn't hold because the "controlling" authority is the editor/owner in each of the various polls to which you have access.

For example,

WBCA--the organization.
Massey--Massey himself
D3Hoops--Pat Coleman
Posters' Pool--Mr Ypsi is accumulating the data.

From that, you can make an assessment of the validity of the polls.  Whom do you believe?  One possible answer would be none of them.

The Week 13 poll will be the most interesting.  Let's do a side-by-side at the end and see how close each was.  In the "quality-rich" brackets, we can use some common sense.

Ralph, thanks for the reply.

I guess the "controlling" comment strikes me as problematic. Although you implied as such, I would like to think (regardless of the issues I have with it) that the d3hoop.com poll isn't 'controlled' by Pat Coleman. Assumably, PC is just the tabulator of the votes from his selected constituents. I must say, if he somehow massages (i.e. controlled) the ballots, I would be supremely disappointed. And mildly surprised.

Contrary to what I posted before, your comments, in general, necessitate the need for transparency. To paraphrase Pat Coleman, "it is not his choice" for non-transparency, but simply a need to keep his pollsters privacy because of the possiblity of "lobbying." You, on the other hand, seem to feel that it is PC's poll to "control." This raises questions (not to say you both are of one mind on the issue).

I agree that the week 13 poll will most likely be indicative of the tournament. D3hoops.com has a long tradition of nailing the tournament teams, both in the polls and bracket-wise. However, my original point still stands: the d3hoops.com poll is not as fluid -- does not take into consideration week-to-week results -- as it should be.

Monkey,

I believe Ralph meant "controlling authority" to mean the person/organization who decides who the voters are and how they are selected, what information is given to voters, when the ballots are due, etc.  Or, in the case of Massey, what formula is used.

[Ralph Turner]

monkey, the transparency issue doesn't hold because the "controlling" authority is the editor/owner in each of the various polls to which you have access.

For example,

WBCA--the organization.
Massey--Massey himself
D3Hoops--Pat Coleman
Posters' Pool--Mr Ypsi is accumulating the data.

From that, you can make an assessment of the validity of the polls.  Whom do you believe?  One possible answer would be none of them.

The Week 13 poll will be the most interesting.  Let's do a side-by-side at the end and see how close each was.  In the "quality-rich" brackets, we can use some common sense.

Ralph, thanks for the reply.

I guess the "controlling" comment strikes me as problematic. Although you implied as such, I would like to think (regardless of the issues I have with it) that the d3hoop.com poll isn't 'controlled' by Pat Coleman. Assumably, PC is just the tabulator of the votes from his selected constituents. I must say, if he somehow massages (i.e. controlled) the ballots, I would be supremely disappointed. And mildly surprised.

Contrary to what I posted before, your comments, in general, necessitate the need for transparency. To paraphrase Pat Coleman, "it is not his choice" for non-transparency, but simply a need to keep his pollsters privacy because of the possiblity of "lobbying." You, on the other hand, seem to feel that it is PC's poll to "control." This raises questions (not to say you both are of one mind on the issue).

I agree that the week 13 poll will most likely be indicative of the tournament. D3hoops.com has a long tradition of nailing the tournament teams, both in the polls and bracket-wise. However, my original point still stands: the d3hoops.com poll is not as fluid -- does not take into consideration week-to-week results -- as it should be.

Monkey,

I believe Ralph meant "controlling authority" to mean the person/organization who decides who the voters are and how they are selected, what information is given to voters, when the ballots are due, etc.  Or, in the case of Massey, what formula is used.

Sunny, thanks!  That's what I meant.  I am glad that Pat has shielded his posters from as much as possible.

The value that I see in Mr Ypsi's Posters' Poll is that we can explore some of the statistical nuances that may arise in some of the polls to help all of the posters to consider what happens in a poll.

I don't mind being "wrong"...it is hard to be "wrong" if you are conscientious and try to do better next week.  But it does allow the readers to work out in each person's mind what the polls are saying.

Pat will seek conscientious knowledgeable pollsters.

The WBCA pollsters are under peer pressure to cast a reasonable ballot.

Massey (or Sagarin or even our friend smedindy) is trying to improve the calculations.

I think that is enough to ensure that the integrity of the polls.

[JMW1973]

I'm really annoyed that the polls are forgetting about my alma mater, Cortland State. Try 14-2 and 9-0 in the SUNYAC and they have received NO VOTES in the polls. Even the men's team is getting some love at 16-2 and 8-1 in the SUNYAC. I think it's about time the pollsters woke up about the Lady Dragons!