<html><head></head><body><div class="ydpe982185cyahoo-style-wrap" style="font-family:Helvetica Neue, Helvetica, Arial, sans-serif;font-size:13px;"><div></div>
        <div dir="ltr" data-setdir="false">I also like proportional representation, but there are many different elections for many different things, and there will always be a need for single-winner methods. Because of that I'm not sure it's necessary to make the same point in every discussion about single-winner methods, especially specific discussions about solving specific problems (e.g. burial) rather than more general discussions.</div><div dir="ltr" data-setdir="false"><br></div><div dir="ltr" data-setdir="false">Toby</div><div><br></div>
        
        </div><div id="ydp7c69d6eyahoo_quoted_7702962143" class="ydp7c69d6eyahoo_quoted">
            <div style="font-family:'Helvetica Neue', Helvetica, Arial, sans-serif;font-size:13px;color:#26282a;">
                
                <div>
                    On Friday, 13 October 2023 at 19:18:06 BST, Richard Lung <voting@ukscientists.com> wrote:
                </div>
                <div><br></div>
                <div><br></div>
                <div><div id="ydp7c69d6eyiv6777086580"><div>
    <p><br clear="none">
    </p>
    <p>"Cycles" (in the paper-scissors-rock sense) are a problem of the 
      (under-candidated) single member systems own making. They rapidly
      disappear with a representative sample of candidates
      proportionally elected to large constituencies. The problem is not
      the 'pesky' cycles, it is the pesky single member system. Unless
      the politics in political science is to dictate to the science, it
      is up to academics to point out, as hundreds of American political
      scientists have, in conjunction with The New York Times, I
      believe, this requirement of a quota-preferential method.</p>
    <p>Remedies to the single member system are cosmetic. They cannot
      possibly please more than half the population, whatever you do --
      and probably a good deal less. UK monopolistic elections are a
      minorocracy not a democracy, and that is probably a fair
      indication of the US state of affairs.<br clear="none">
    </p>
    <p>Time to move on from the ancient Greek conception of democracy,
      as to elect a tyrant, unconditionally -- making Britain what
      Hailsham called an "elective dictatorship." Which shares some of
      the all too evident failings of any dictatorship, elected or
      otherwise. This should be a spur to avoid another Vietnam war or
      second Iraq war, which even W. may deplore, in his heart.<br clear="none">
    </p>
    <p>Regards,</p>
    <p>Richard Lung.<br clear="none">
    </p>
    <p><br clear="none">
    </p>
    <div id="ydp7c69d6eyiv6777086580yqt79040" class="ydp7c69d6eyiv6777086580yqt8789344907"><div class="ydp7c69d6eyiv6777086580moz-cite-prefix">On 13/10/2023 18:11, Forest Simmons
      wrote:<br clear="none">
    </div>
    <blockquote type="cite">
      <div>
        <div>Dear EM List Friends,
          <div><br clear="none">
          </div>
          <div>We need your feedback on this draft of a
            proposal before we submit a version of it to the voting
            reform community at large.</div>
          <br clear="none">
          <div class="ydp7c69d6eyiv6777086580gmail_quote">
            <div dir="ltr" class="ydp7c69d6eyiv6777086580gmail_attr">---------- Forwarded
              message ---------<br clear="none">
              From: <strong class="ydp7c69d6eyiv6777086580gmail_sendername">Forest
                Simmons</strong> <span><<a shape="rect" href="mailto:forest.simmons21@gmail.com" class="ydp7c69d6eyiv6777086580moz-txt-link-freetext" rel="nofollow" target="_blank">forest.simmons21@gmail.com</a>></span><br clear="none">
              Date: Thu, Oct 12, 2023, 5:35 PM<br clear="none">
              Subject: Duncan Proposal Draft<br clear="none">
              To: Michael Ossipoff <<a shape="rect" href="mailto:email9648742@gmail.com" class="ydp7c69d6eyiv6777086580moz-txt-link-freetext" rel="nofollow" target="_blank">email9648742@gmail.com</a>><br clear="none">
            </div>
            <br clear="none">
            <br clear="none">
            <div>Michael Christened our new Q&D burial
              resistant method "Duncan" after Duncan Black who
              popularized the idea of using  Borda's Method as a
              fallback "completion" when the ballots fail to 
              unambiguously reveal the sincere "Condorcet" pairbeats-all
              candidate.
              <div><br clear="none">
              </div>
              <div>Our Duncan method has the same form as
                Black's in that the official version directly specifies
                electing the unambiguous Condorcet Candidate when there
                is one, and falls back to another procedure that relies
                on Borda Scores, otherwise.</div>
              <div><br clear="none">
              </div>
              <div>It should be emphasized that in both cases
                the fall back Borda based expedient is rarely needed.
                For that reason some misguided voting reform advocates
                have cavalierly opined that any decisive completion/
                fallback method would be plenty adequate to supplement
                the Condorcet Criterion requirement.</div>
              <div><br clear="none">
              </div>
              <div>However, this casual attitude ignores the 
                feedback aspects of voting systems in that various
                voting methods vary in the degree that they encourage or
                discourage the creation of artificial beat cycles that
                subvert/ hide the Condorcet Candidate from view,
                bringing the completion method into greater prominence
                in a potentially unstable cycle.</div>
              <div><br clear="none">
              </div>
              <div>Unfortunately most of the extant methods
                fall into this "positive" feedback category, including
                Borda itself.  Some less sensitive methods like
                Approval  and IRV/RCV have a built in "friction" that
                dampens the feedback; but as systems engineers know, the
                high performance components are the ones that need the
                addition of some carefully engineered negative feedback
                "circuit" to stabilize the system as a whole.</div>
              <div><br clear="none">
              </div>
              <div>In our Condorcet Completion context, our
                use of the Borda Count scores is carefully designed with
                that stabilizing influence in mind: adventurous
                strategists who are aware of this feature, when acting
                rationally will be deterred from creating these cycles
                that come back to bite them. Those not aware will find
                out when their ploys backfire or otherwise disappoint
                them.</div>
              <div><br clear="none">
              </div>
              <div>How do these pesky cycles arise so easily
                in Borda and other rank based methods?</div>
              <div><br clear="none">
              </div>
              <div>Suppose that your personal preference
                schedule for the alphabetized candidates looks like ...</div>
              <div><br clear="none">
              </div>
              <div>A>C>X>Y>Z, and that C is the
                Condorcet Candidate projected to win the election if
                nobody acts nefariously.</div>
              <div><br clear="none">
              </div>
              <div>You, and like minded friends, get the idea
                to insincerely move your second choice to the bottom of
                your ballot (so it now reads A>X>Y>Z>C) ...
                not to be "nefarious" so much as to just increase the
                winning chances of your favorite A.</div>
              <div><br clear="none">
              </div>
              <div>Could this work?</div>
              <div><br clear="none">
              </div>
              <div>Yes, under Black's method if your friends
                follow your lead, this "nurial" of C under the "busses"
                X, Y, and Z, could easily subvert one or more of C's
                pairwise victories over X,Y, and Z, into defeats of C by
                them, thereby hiding C's identity of sincere Universal
                "pairbeater" status to just one more member of a
                "beatcycle" of the form A beats X beats Y beats Z beats
                C beats A.</div>
              <div><br clear="none">
              </div>
              <div>Note that the buried candidate C still
                beats the buriers' favorite, A ... because lowering C 
                does not decrease the number of ballots that support C
                over A ... which is how easily and innocently beatcycles
                like this can be created in Condorcet style elections
                ... at least in the absence of negative feedback from
                the cycle resolution fallback method.</div>
              <div><br clear="none">
              </div>
              <div>In traditional Black that fallback method
                is Borda. Does that fix the problem? ... or does it
                exacerbate it.</div>
              <div><br clear="none">
              </div>
              <div>Well ... the same burial that put C at
                disadvantage in the pairwise contests with X thru Z,
                also lowered C's Borda score by 3 counts per ballot, and
                raised</div>
              <div> the Borda score of each of X thru Z to
                the tune of one count per ballot.</div>
              <div><br clear="none">
              </div>
              <div>The likely outcome is that C will end up
                with the lowest score, and come in last in the finish
                order.</div>
              <div><br clear="none">
              </div>
              <div>By way of contrast, under our new Duncan
                method, the most likely winner is X, and the least
                likely winner is A, the burier faction's favorite ...
                thus disappointing the burier faction supporters ...
                teaching them that if they try to outsmart new Duncan
                with insincere ballot rankings, they are apt to end up
                helping elect their third (or later) choice instead of
                their first choice or their second choice ... the one
                that they so cleverly buried (however innocently or
                without malice).</div>
              <div><br clear="none">
              </div>
              <div>Too many dabblers in voting method reform
                (as well as most professionals) are unaware of these
                dynamics.</div>
              <div><br clear="none">
              </div>
              <div>But now, with your new understanding, you,
                at least, can become part of the solution.</div>
              <div><br clear="none">
              </div>
              <div>Duncan Definition:</div>
              <div><br clear="none">
              </div>
              <div>In the vast majority of the cases ...
                those in which the pairwise counts of the ballots
                unambiguously identify the candidate that pairbeats each
                of the others ... elect that candidate.</div>
              <div><br clear="none">
              </div>
              <div>Otherwise, elect the highest score
                candidate that pairbeats every candidate with lower
                score.</div>
              <div><br clear="none">
              </div>
              <div>[Nominally "score" = Borda Count, though
                STAR Voting scores, for example, could also serve]</div>
              <div><br clear="none">
              </div>
              <div>How does this Duncan fallback procedure
                work to prevent A from getting elected in our scenario
                regarding A thru Z?</div>
              <div><br clear="none">
              </div>
              <div>Well, could A pairbeat every lower score
                candidate? In particular, could A pairbeat C, which is
                now at the bottom of the Borda score pile ... certainly
                lower than A ...?</div>
              <div><br clear="none">
              </div>
              <div>Well, remember that "C beats A" was the
                last step in the beatcycle created by A's friends.</div>
              <div><br clear="none">
              </div>
              <div>So A does not pairbeat every lower score
                candidate, and therefore cannot win.</div>
              <div><br clear="none">
              </div>
              <div>New Duncan is burial resistant.</div>
              <div><br clear="none">
              </div>
              <div>Next time ... more examples and insights
                ...</div>
              <div><br clear="none">
              </div>
              <div>fws</div>
              <div><br clear="none">
              </div>
              <div><br clear="none">
              </div>
              <div><br clear="none">
              </div>
            </div>
          </div>
        </div>
      </div>
      <br clear="none">
      <fieldset class="ydp7c69d6eyiv6777086580moz-mime-attachment-header"></fieldset>
      <pre class="ydp7c69d6eyiv6777086580moz-quote-pre">----
Election-Methods mailing list - see <a shape="rect" href="https://electorama.com/em" class="ydp7c69d6eyiv6777086580moz-txt-link-freetext" rel="nofollow" target="_blank">https://electorama.com/em</a> for list info
</pre>
    </blockquote></div>
  </div></div><div class="ydp7c69d6eyqt8789344907" id="ydp7c69d6eyqt43151">----<br clear="none">Election-Methods mailing list - see <a shape="rect" href="https://electorama.com/em" rel="nofollow" target="_blank">https://electorama.com/em</a> for list info<br clear="none"></div></div>
            </div>
        </div></body></html>