USDEBATE Archives

December 2012

USDEBATE@LIST.UVM.EDU

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
"Barnes, R Eric" <[log in to unmask]>
Reply To:
USA Debating in the WUDC Format <[log in to unmask]>
Date:
Sun, 16 Dec 2012 16:29:13 +0000
Content-Type:
text/plain
Parts/Attachments:
text/plain (403 lines)
Hi Shengwu,

Surely reconsideration is reasonable.  Indeed, I hope that it will happen and that the rules for breaking will change.  We have no disagreement there and nothing I said should be taken to be against reconsideration.

My second point was entirely about the TIMING of such reconsideration and the HASTE with which such reconsideration seems to be proceeding.  Although the vote to change the system last year was very close, the deliberations happened over more than one year's meetings of council, and those meetings were very long indeed.  We should take our time in reconsidering this issue, making sure that all representatives have adequate chance to attend a full meeting of council and to discuss it.

The parenthetical remark at the end of your e-mail is tantalizing, but too brief to be at all satisfying.  Can you say why you find it both arbitrary and incoherent?  I think it may be inferior to other systems, but I'm not at all sure that I'd call it either of incoherent (a very strong charge) or arbitrary.  Indeed, it is the lack of arbitrariness that is supposed to be its primary advantage.

Yours,
Eric

******************************************
Eric Barnes
Hobart and William Smith Colleges
Philosophy Department
Public Policy Program
Debate Coach
(315) 781-3182
[log in to unmask]<mailto:[log in to unmask]>

On Dec 16, 2012, at 10:15 AM, Shengwu wrote:

Hi Eric,

I've every respect for you, Andy, and Steve, and I'm thankful you've taken the time to examine the issue and write on it. I'd point out, though, that presumably the whole point of having a second ratifying vote at this WUDC was to allow the community a chance to change its mind and rescind a major change to the rules. Particularly since this change passed last year by the slimmest of margins, I don't think it's at all unreasonable of the American, or indeed, the world debate community to reconsider its beliefs on a controversial topic.


Cheers,
Shengwu

(Speaking in no capacity but my own, I also think that the current system is arbitrary and incoherent, and would cheer any hastening of its abolition.)

On Dec 16, 2012, at 10:58 PM, "Barnes, R Eric" <[log in to unmask]<mailto:[log in to unmask]>> wrote:

Hey Everyone,

I am not opposed to having a 10th round of prelims.  In fact, I think that an excellent argument can be made for doing this, but this is ultimately a side issue.

I do want to point out that sorting is only one of the functions of the break.  If this were the only function, then we would likely be justified in entirely eliminating (or at least radically reducing) the break.  The recent article in the Monash Debating Review (by me, Andy Hume and Steve Johnson) makes this point extensively, so I will quote the most relevant excerpt below.  I urge those making this decision to keep in mind the other functions of the break.  The functions of the break as an award in itself seems particularly important, as does the opportunity to evaluate the debaters in front of a large audience (which affects people's performance).  The main point is that deciding how to define the break is not just about sorting out the best teams.  The full article can be found here:  http://www.monashdebaters.com/downloads/MDR/MDR%20V9%202011.pdf

Also, if I may, it strikes me as very late in the day to be making any decisions regarding how to define the break in a WUDC that is starting in about a week.  I'm more than happy to reconsider the question for the long term, but it seems ridiculous to overturn the decision of last year's World's Council this late in the game.  This is not because I am a proponent of the system that was adopted; I have primarily advocated for a 48-team break.  The point is that we should respect the result of last year's vote, for at least this year (and perhaps more), until we can evaluate how it works and revisit this issue with adequate care.  Rushing policy decisions is generally a bad idea, unless it is necessary to prevent a grave injustice.  The 18+ break may not be your favorite policy, but it is certainly not an unreasonable proposal that we must abolish with great haste.

All the Best,
Eric



The Purpose of Elimination Rounds

What, then, is the function of elimination rounds at Worlds? The simple answer
seems to be that they are a device for sorting teams by quality. Of course, at the
end of preliminary rounds, we already have such a list (“the tab”), which we use
to decide who breaks. So, if the sole purpose of these rounds were sorting, then we
would need evidence that a single elimination format does a better job at sorting
than four or five more power-paired preliminary rounds.

We are not confident that elimination rounds do a better job at sorting teams for
three reasons. First, they start by largely throwing out a lot of relevant evidence
from the rounds that have already occurred. Second, if we assume that the
seeding for the single elimination bracket (i.e., the tab ordering) is not particularly
accurate—which we must assume, or else there is no good reason for additional
sorting—then the single elimination format does not even theoretically do a good
job of sorting anyone beyond the best two teams.*  Third, single elimination is a
poor sorting device for even the best team (especially four rounds of it) because
even excellent judging is imprecise and even excellent teams can get put in very
tough situations (e.g., by another team’s bizarre argumentation).

[* Note -- To see why, first consider why a single-elimination bracket in any two contestant event (e.g.,
tennis) will only be theoretically effective at sorting the single best player if we presume that the
initial seedings are not already accurate.  it is very possible that the two best players will meet
before the finals, so being in finals doesn’t reliably indicate that you are one of the two best players,
being in semi-finals doesn’t reliably indicate that you are one of the four best players, etc.  The
same problem exists with bP debate, except that each single elimination contest advances two
teams.  Of course, all this is true even if we grant that judge panels in elimination rounds never
make errors.  The problem is mathematical, not practical.]

If you think that none of this is a persuasive reason for abandoning single
elimination rounds, we agree. However, it does show that there are important
functions of the elimination rounds beyond sorting.

These other functions include: A) creating an exciting, high-stakes set of
increasingly high quality debates; B) providing a learning opportunity for
viewers; C) generating a celebratory culmination of the tournament; D) giving
an award in its own right for debaters who have performed well; and, E) providing
an opportunity to evaluate debaters’ performance in front of an audience. All of
these are good reasons to keep single elimination rounds, and indeed to expand
them.

Elimination rounds are very good at fulfilling function A. They bring an element
of uncertainty to the competition, like a sporting event in which top teams may
be humbled by underdogs before the final. Some contend that it is somehow
unfair for a skilled team to run the risk of being eliminated by a “wrong” decision.
Indeed, there are those who would even be happy to see the world champions
of debate selected on the basis of the team that topped the tab after 12 or 15
preliminary rounds of debate. But where would be the excitement in that?

Elimination rounds also provide what for many competitors will be a rare
opportunity to watch, listen to and learn from the best of their contemporaries
(B), and the value of function C is fairly self-explanatory. All three proposals
to expand the break will accomplish functions A, B and C to roughly the same
degree. The difference between the proposals primarily lies in five things: 1) the
practicality of their implementation; 2) the fairness of their implementation; 3)
how effective they are at sorting teams, such that the higher quality team are likely
to progress further in the tournament; 4) how well they perform function E,
evaluating debaters with an audience; and, 5) how well they accomplish function
D, distributing a valued award.

Let us briefly consider the function of the break as an award. There are very few
trophies given out at Worlds, but being able to say that one “broke at Worlds” is
in itself a significant intangible award. The two important most aspects of this
award’s value are its exclusivity and its fairness. Although giving and getting
awards (even intangible ones) is nice, as the percentage of debaters who break
increases, the prestige of this award decreases. This is surely one reason we can
all agree that breaking 128 out of 350 teams would be a mistake, even if it were
practical. But at the same time, breaking only 8 teams would be a mistake because
such a distribution is too stingy, even if it were a more reliable sorting method.
The point here is that there is some admittedly vague point that is the ideal
compromise between being too stingy and excessively devaluing the award of
breaking. The most objective means of estimating the prestige of breaking across
years is to calculate the percentage of participating teams that break. So, let us
look quickly at how these three proposals would have affected these percentages
over the last 10 years and with a couple hypothetical larger fields.

Basically, the lower the percentage of the break, the higher the prestige of the
award, but the fewer the people who get to enjoy it. The top two lines are
included to frame the perspective on what some see as the likely future growth
of the tournament. Obviously some of the older data is of limited usefulness,
since future tournaments are unlikely to have fewer than 200 teams, but they
do provide an important historical perspective. As noted above, in the 1990s
there were 32 teams in the break, with only about 150 teams in the field, making
the portion of teams breaking over 20%. But, history provides no authoritative
guidance on the correct percentage of breaking teams.

Year

Field

17s in 48

18+ teams

% in 32

% in 18+

% in 48

% in 64



500





6.4



9.6

12.8



450





7.1



10.7

14.2

2011

316

12

36

10.1

11.4

15.2

20.3

2010

388

2

46

8.2

11.8

12.4

16.5

2009

316

12

36

10.1

11.4

15.2

20.3

2008

396

2

46

8.1

11.6

12.1

16.2

2007

344

7

41

9.3

11.9

14.0

18.6

2006

324

11

37

9.9

11.4

14.8

19.8

2005

312

10

38

10.3

12.2

15.4

20.5

2004

307

13

35

10.4

11.4

15.6

20.8

2003

193

9 broke in 32

23

16.6

11.9

24.9

33.2

2002

228

6 broke in 32

26

14.0

11.4

21.1

28.1



******************************************
Eric Barnes
Hobart and William Smith Colleges
Philosophy Department
Public Policy Program
Debate Coach
(315) 781-3182
[log in to unmask]<mailto:[log in to unmask]><mailto:[log in to unmask]>

To see why, first consider why a single-elimination bracket in any two contestant event (e.g.,
tennis) will only be theoretically effective at sorting the single best player if we presume that the
initial seedings are not already accurate.  it is very possible that the two best players will meet
before the finals, so being in finals doesn’t reliably indicate that you are one of the two best players,
being in semi-finals doesn’t reliably indicate that you are one of the four best players, etc.  The
same problem exists with bP debate, except that each single elimination contest advances two
teams.  Of course, all this is true even if we grant that judge panels in elimination rounds never
make errors.  The problem is mathematical, not practical.

On Dec 15, 2012, at 9:18 PM, Shengwu Li wrote:

Hey James, Coulter,

I'd just like to voice support for James' idea, and to add that (having talked in the past to people who've run tab simulations) there's some computational evidence to suggest that having an extra prelim round at worlds will sort teams more effectively than having a partial double-octofinal, or a hexadecifinal, or whatever else the case may be.  So it may be a Pareto improvement; non-breaking teams get one extra round to debate in, and teams in contention for the octos are sorted more fairly than by an extra knockout round.


Cheers,
Shengwu

On Sun, Dec 16, 2012 at 3:15 AM, James Kilcup <[log in to unmask]<mailto:[log in to unmask]><mailto:[log in to unmask]>> wrote:
Hey Coulter,
Thanks for your solid work here, looking forward to seeing you in Berlin. I don't know how you feel about this but I've been talking with a lot of people about the benefits of moving future Worlds prelims from 9 rounds to 10 and I'd like your thoughts on taking that up at Worlds Council. The bullet point argument for it would be that the break becomes less statistically noisy and more fair because round 9 currently represents a 3 point swing in a format where closing opposition is twice as likely to pick up points out of the gate as opening gov. With a tenth round, we could reduce some of the noise, advance better teams, and increase fairness. It is also a round number which is a plus and all it would require is having 4 rounds on day 1, which wouldn't require much more on an organizational level. Plus, then we could debate the merits of a 20+ point break. Let me know what you think.
Thanks,
James

On Dec 15, 2012, at 10:24 AM, Coulter King <[log in to unmask]<mailto:[log in to unmask]><mailto:[log in to unmask]>> wrote:

Hi everyone,

I'm writing about two important Worlds Council issues, the first of which is particularly time-sensitive. Please read carefully if the Berlin break or Worlds Council issues in general are of concern to you:

1. Worlds Break - special session
Pending a vote at pre-Council, it is possible that there will be a vote to reconsider the break at a special session held after pre-Council (any changes would affect the break at Berlin). As you may be aware, the current break is for all teams on 18+ points to break; England has proposed a 48-team break instead, but the other commonly floated alternatives (a return to the 32-team break or a larger expansion to a 64-team break) also will be up for consideration if this session is held. For reference, schools voted to support the 18+ teams break proposal from Canada (and originally, Steve) at caucus last year.

It may be difficult to meet as a group in Berlin prior to pre-Council, so we'll conduct an online vote, as we did earlier in the year when determining the US vote for the 2014 host of Worlds. I know some people ran into trouble with the forum last time (and there's no sense in making you register for another site to vote), so please vote through the Google form by December 25th at 11:59pm, Eastern time:

https://docs.google.com/spreadsheet/viewform?formkey=dHYxRlNsc1VDLVYzSnl6cmNpanJFNVE6MQ#gid=0

Votes submitted after the deadline will not be counted, one vote per school, etc. For transparency, the results are viewable here: https://docs.google.com/spreadsheet/ccc?key=0Ao1UYdwsq4cmdHYxRlNsc1VDLVYzSnl6cmNpanJFNVE#gid=0

2. US Caucus in Berlin prior to Worlds Council
As was the case last year in Manila, we'll meet for a US caucus at some point during the tournament to discuss the major issues that will come up at Council and what the US position should be on each of them. I've been in touch with the Auckland bid about attending the beginning of caucus to answer questions and better inform our decision, and I plan to extend the same invitation to Malaysia. Please bring your questions about the two bids to that meeting, so that (assuming things work out), we can clarify anything with the two bids before our internal discussion.

It's impossible to know now when and where this meeting will take place right now, so I'm writing also to encourage you to listen for an announcement between rounds (which is when we and most other countries communicated information about caucus times and locations). Obviously there's a risk that not everyone who's interested in attending will be in the room / be listening at the time, so please try to spread the word to other US teams once we finalize a time and place, so that everyone who is interested in contributing to the discussion and voting on likely topics can attend.

Thanks, and I look forward to seeing you all in Berlin!

Best,

Coulter

ATOM RSS1 RSS2