<div dir="ltr"><div>I forgot to mention that the method is efficiently summable over precincts:</div><div><br></div><div>Make a pairwise matrix where the off diagonal elements give the information for pairwise defeats, and the diagonal elements keep track of the first place totals.</div><div><br></div><div>In particular, if a certain ballot ranks the first three candidates equal top, then that ballot adds the fraction 1/3 to each of the first three diagonal elements of the summary matrix.<br></div><div><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
<br>
<br><br>
Message: 2<br>
Date: Thu, 6 Jun 2019 14:11:44 -0700<br>
From: Forest Simmons <<a href="mailto:fsimmons@pcc.edu" target="_blank">fsimmons@pcc.edu</a>><br>
To: EM <<a href="mailto:election-methods@lists.electorama.com" target="_blank">election-methods@lists.electorama.com</a>><br>
Subject: [EM] Improved Copeland (was "A New Spinoff of Our Recent<br>
Discussions")<br>
Message-ID:<br>
<CAP29onczeXL4533KDUN_ttziV2-0=<a href="mailto:AjCD26Ly60cHHR42pNZww@mail.gmail.com" target="_blank">AjCD26Ly60cHHR42pNZww@mail.gmail.com</a>><br>
Content-Type: text/plain; charset="utf-8"<br>
<br>
Here's another slightly simpler approach aimed at the lay voter:<br>
<br>
Tell the audience that the Condorcet ideal is a candidate that is not<br>
pairwise beaten by any other candidate.<br>
<br>
When that is not possible, it is natural to consider a candidate that is<br>
beaten pairwise by the fewest other candidates. This idea is the basis of<br>
the Copeland Method.<br>
<br>
There are two problems with the Copeland method: (1) It has a strong<br>
tendency to produce ties, and (2) More subtle problems created by cloning<br>
certain candidates to increase the number of defeats suffered by certain<br>
other candidates without increasing the number of defeats of the cloned<br>
candidates.<br>
<br>
Because of these two problems, Copeland is not considered a serious<br>
contender for use in public elections.<br>
<br>
But what if there were a simple modification of Copeland that would totally<br>
resolve these two problems in one fell swoop?<br>
<br>
There is; instead of counting the number of candidates that defeat<br>
candidate X, (and electing the candidate with the smallest count), we add<br>
up all of the first place votes of all of the candidates that defeat X, and<br>
elect the candidate with the smallest sum.<br>
<br>
This solves the first problem because in any moderate to large sized<br>
election, it would be rare for two candidates to have the same minimum sum.<br>
<br>
It also solves the second problem because if a candidate is cloned, the<br>
first place votes of the cloned candidate are divided up among the clones.<br>
<br>
[End of Introduction to Improved Copeland for the lay voter.]<br>
<br>
Now, as mentioned in my last post, it is more general to replace the phrase<br>
"first place votes" with "random candidate probabilities," i.e. benchmark<br>
lottery probabilities. Even other suitable lottery probabilities could be<br>
used.<br>
<br>
This method (at least under the top rank counts or benchmark lottery)<br>
always elects from Landau, since if X covers Y, then only a subset of the<br>
candidates that beat Y will beat X, yielding X a smaller probability sum<br>
than Y.<br>
<br>
Also since if candidate X is raised on a ballot it can only decrease the<br>
benchmark probability of any other candidate, and the set of candidates<br>
that now beat X will be a (possibly proper) subset of the candidates that<br>
did before raising X on the ballot; i.e. this method is monotonous (if not<br>
monogamous).<br>
<br>
And it seems tp satisfy the Chicken Defense criterion:<br>
<br>
49 C<br>
26 A>B<br>
25 B (sincere is B>A)<br>
<br>
C>A is the only pairwise defeat of A, so the A sum is 49.<br>
A>B is the only pairwise defeat of B, so the B sum is 26.<br>
B>C is the only pairwise defeat of C, so the C sum is 25.<br>
<br>
Candidate C (with the smallest sum) is elected, thus thwarting the<br>
threatened chicken attack<br>
<br>
What's not to like?<br>
<br>
Now think in terms of "Yee BoLson Diagrams":<br>
<br>
A candidate's score is the sum of the Dirichlet Cell probabilities (i.e.<br>
Voronoi Polygon probabilities). These are the Dirichlet/Voronoicells of<br>
the candidates that are closer to the center of the distribution than the<br>
given candidate. [the respective cells represent the voters that top rank<br>
the respective candidates.]<br>
<br>
So the winning candidate is the candidate for which the mass of cells<br>
closer to the center than the candidate has the smallest total probability.<br>
<br>
In the case of the standard centrally symmetric distribution used in Yee<br>
Bolson diagrams, the candidate closest to the center will be the winner<br>
with no defeats, so the "mass of defeating cells" will be empty.<br>
<br>
Not bad!<br>
<br>
Is it good enough and simple enough to propose?<br>
.<br>
Forest<br>
<br>
<br>
<br>
<br>************<br>
</blockquote></div></div></div>