<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<title></title>
</head>
<body>
James G-A,<br>
You wrote (Thu.Dec.9,04):<br>
<blockquote type="cite">
<pre wrap="">Am I correct in thinking that this meets the criteria mentioned above?
<span class="moz-txt-citetags">>></span>Does this seem like a sensible way to do IRV-completed Condorcet in
<span class="moz-txt-citetags">>></span>general?
<span class="moz-txt-citetags">>> </span>
<span class="moz-txt-citetags">>></span>
</pre>
<pre wrap=""><span class="moz-txt-citetags">></span>My answer to your first question is that it seems to me that it does,
<span class="moz-txt-citetags">></span>and it also seems to meet Minimal Defense.
<span class="moz-txt-citetags">></span>But it shares Condorcet (Winning Votes)'s zero-information random-fill
<span class="moz-txt-citetags">></span>incentive, which in my view is silly and unfair
<span class="moz-txt-citetags">></span>and therefore not really acceptable.
</pre>
<pre wrap=""><!---->
Personally, I don't find that very disturbing. Silly, maybe, but not a
serious problem. I'm guessing that most methods meeting the truncation
resistance/SFC criteria will have this problem. Do you think that the
truncation resistance criterion is misguided? I sort of go back and forth
on it... sometimes I think it's quite important, sometimes I don't. I do
think that it's more important that the random fill incentive, however.
</pre>
</blockquote>
<i><br>
truncation resistance</i>: Define the "sincere top set" as the smallest
subset <br>
<i> </i> of alternatives such that, for each alternative in the subset,
say <i>x</i>, and <br>
<i> </i> each alternative outside the subset, say <i>y</i>, the number
of voters who <br>
<i> </i> sincerely prefer <i>x</i> over <i>y</i> exceeds the number
who sincerely prefer <i>y</i> <br>
over <i>x</i>. If no voter votes the reverse of any sincere preference
regarding <br>
any pair of alternatives, and more than half of the voters rank
some <i>x</i> in <br>
the sincere top set over some <i>y</i> outside the sincere top set,
then <i>y</i> must <br>
not be elected. <br>
<blockquote type="cite">
<pre wrap="">(from Steve Eppley's MAM site)
</pre>
</blockquote>
CB: To answer your question, I think that the Truncation Resistance criterion
(as defined above) is desirable but not<br>
essential. If there is a random-fill incentive, then I think it is rendered
pretty meaningless.<br>
We know that is impossible to thwart sufficiently well-informed strategists,
but quite easy for a method to have no <br>
zero-information strategy; so why not at least achieve that?<br>
<br>
This version of IRV-completed Condorcet of yours,<br>
<blockquote type="cite">
<pre>1. Eliminate non-members of the minimal dominant set.
2. Eliminate all candidates who are pairwise-beaten by a full majority
UNLESS this doesn't leave anyone at all.
3. Hold an IRV tally between remaining candidates.</pre>
</blockquote>
(with your step 2) has a big and obvious random-fill incentive. Voters who
are sincerely indifferent regarding their lower<br>
choices should (at least) random-fill, because they might give one of their
favourites's opponents a "majority defeat" that<br>
that candidate might not otherwise have (which, if their favourite has
one, might save their favourite from being eliminated).<br>
(I suppose there is a small chance that it could backfire, by causing the
elimination of a candidate that otherwise would have<br>
lost to Favourite in the final runoff.)<br>
To me it is just obviously unfair that insincere random-fillers should have
more voting power than sincere truncators.<br>
If the method used is something like IRV or Margins, and the voters assured
(by election officials and all the media pundits)<br>
that without taking into account how others might vote the do best to just
vote their full sincere preferences, then I think most<br>
(or many) voters will just respond "that's fine" and do just that. <br>
If on the other hand, they are told that they would be mugs to not give
a full strict ordering regardless, then I think that some<br>
of these voters would think "I don't want to vote randomly, and I don't
think the result should be determined by people <br>
voting randomly. If , with 'zero-information' I can do better by random-filling,
then maybe I can make use of some information<br>
to do better still for my favourite." And so they are encouraged down the
path to outright order-reversal.<br>
So to sum up the main effects of your step 2: naive sincere truncators will
be unfairly disadvantaged, and if the voters are<br>
savvy then the result will tend to be randomized and there will be more
Burying strategising.<br>
<br>
I don't see why Condorcet, Truncation Resistance, Minimal Defense should
be incompatible with No Zero-Information<br>
Strategy. As far as I can tell, they are all met by Schwartz // SC-WMA
that I described in the "recommendations" thread<br>
in September this year.<br>
<br>
<blockquote type="cite">
<pre wrap="">My answer to your second question is "No". I assume we all agree
<span class="moz-txt-citetags">> </span>that two completely essential criteria that a method
<span class="moz-txt-citetags">></span>must meet are Woodall's "Mono-add-plump" and "Mono-append".
</pre>
<pre wrap=""><!---->
Er, "completely essential"? "must meet"? "all agree"??? I don't know
about that assumption, Chris...
</pre>
<blockquote type="cite">
<pre wrap=""><span class="moz-txt-citetags">></span>using IRV (aka AV)
<span class="moz-txt-citetags">> </span>to complete Condorcet by eliminating and then ignoring
<span class="moz-txt-citetags">></span>the not-allowed-to-win candidates not in the "top tier" creates a
<span class="moz-txt-citetags">></span>method that fails both Mono-add-plump and Mono-append.
<span class="moz-txt-citetags">></span>
</pre>
<blockquote type="cite">
<pre wrap=""><span class="moz-txt-citetags">>></span>abcd 10
<span class="moz-txt-citetags">>></span>bcda 6
<span class="moz-txt-citetags">>></span>c 2
<span class="moz-txt-citetags">>></span>dcab 5
<span class="moz-txt-citetags">>></span>
<span class="moz-txt-citetags">>></span>All the candidates are in the top tier, and the AV winner is a. But
<span class="moz-txt-citetags">>></span>if you add two extra ballots that plump for a, or append a to the two
<span class="moz-txt-citetags">>></span>c ballots, then the CNTT becomes {a,b,c}, and if you delete d from all
<span class="moz-txt-citetags">>></span>the ballots before applying AV then c wins.
</pre>
</blockquote>
</blockquote>
<pre wrap=""><!---->
My question is whether it is likely that voters will frequently be able
to exploit this strategically, and whether their efforts to exploit it are
likely to seriously distort their reported preference rankings. If not,
then I suggest that the problem is not very severe.
</pre>
</blockquote>
CB: I am astonished that you are not bothered by failures of mono-add-plump
and mono-append. These criteria<br>
are very easy to meet (much more than mono-raise, aka regular Monotonicity)
and to me their failures are completely<br>
absurd . You must at least agree that it is a huge potential marketing problem
in the face of detractors.<br>
<br>
<blockquote type="cite">
<pre wrap="">CB: In his example, in both the "before" and "after" cases all the
<span class="moz-txt-citetags">></span>candidates have a "full majority" pairwise loss.
<span class="moz-txt-citetags">></span>(You don't spell it out, but I assume "full majority" means more than
<span class="moz-txt-citetags">></span>half those ballots that distinguish between any of the
<span class="moz-txt-citetags">></span>Schwartz-set members.)
</pre>
<pre wrap=""><!---->
I was just thinking more than half of the valid vote, to keep things
simple.
</pre>
</blockquote>
CB: I think that would open the door to IIA-like absurdity. Say, in a
race between a,b,c,d, a wins and d is ignored<br>
by all the voters. Then two or three extra ballots are found and admitted,
and they vote for d and ignore abc. This changes<br>
the winner from a to b .<br>
<br>
Regarding Woodall's "CNTT, AV" (Condorcet (Net) Top Tier, Alternative
Vote) I forgot a small detail.<br>
(BTW, the Condorcet (net) top tier is another name for the Smith set; and
the Alternative Vote is another name for IRV).<br>
Woodall writes that if none of the CNTT candidates have any first-preferences,
"then this is equivalent to choosing one at<br>
random (not necessarily with equal probability). If this is considered unsatisfactory,
then one should first exclude all the candidates<br>
in the Condorcet bottom tier, and repeat until some candidate in.. [CNTT]..
has a first-preference vote, before applying AV as<br>
above."<br>
<br>
<br>
Chris Benham<br>
<br>
<br>
<blockquote type="cite">
<pre>
</pre>
</blockquote>
<br>
<br>
<blockquote type="cite">
<pre wrap="">
</pre>
</blockquote>
</body>
</html>