<!DOCTYPE html>
<html>
  <head>
    <meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
  </head>
  <body>
    <p>Of course!  Thanks.  Pardon the brain fade.<br>
      <br>
      Normally Smith sets don't have two members, but your example
      didn't mention a third member or say how many candidates there are
      or how many<br>
      rating slots the ballots have.<br>
      <br>
      So maybe there is another candidate F (which is in the Smith set)
      and the ballots have 6 slots, so our Score ballot interpretation
      becomes<br>
      <br>
      D6, E5, A4, B2, F0.<br>
      <br>
      So then our calculation becomes  4+2+0/3 = 2.<br>
      <br>
      Only A is scored higher than 2, so the ballot only approves A.<br>
      <br>
      <blockquote type="cite">This is a nice technique, but it is not
        clone resistant. Adding Smith candidate clones can change the
        average and thus the approval cutoff would change. This seems
        unstable to me.</blockquote>
      <br>
      With ratings ballots I think only candidates that get the same
      rating on every ballot are considered clones.  And besides that I
      think that more than three candidates in the Smith set would be
      vanishingly rare.<br>
      <br>
      Chris<br>
      <br>
      <br>
      <br>
    </p>
    <div class="moz-cite-prefix">On 27/01/2024 5:40 am, Ted Stern wrote:<br>
    </div>
    <blockquote type="cite"
cite="mid:CAHGFzORefGEv8wXQ0035vCboT4Wi=f6-TEYaDOzicYQzf+pqcw@mail.gmail.com">
      <meta http-equiv="content-type" content="text/html; charset=UTF-8">
      <div dir="ltr">
        <div dir="ltr"><br>
        </div>
        <br>
        <div class="gmail_quote">
          <div dir="ltr" class="gmail_attr">On Mon, Jan 22, 2024 at
            7:07 PM C.Benham <<a href="mailto:cbenham@adam.com.au"
              moz-do-not-send="true" class="moz-txt-link-freetext">cbenham@adam.com.au</a>>
            wrote:<br>
          </div>
          <blockquote class="gmail_quote"
style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
            <div>
              <div lang="x-unicode">
                <div lang="x-unicode">
                  <p>Ted,<br>
                    <br>
                  </p>
                  <blockquote type="cite">...you didn't comment on
                    whether ballots with all Smith candidates below top
                    rating should have their ratings bumped up: i.e., D
                    > E > A > blank > B  (A and B in Smith)
                    would be recounted as A > blank > B. </blockquote>
                  <br>
                  I don't think I left anything ambiguous.<br>
                  <br>
                  Assuming in your example we are using say 5-slot
                  ratings ballots then we interpret it as a score ballot
                  thus:  D5, E4, A3, B0.  <br>
                  <br>
                  If A and B are in Smith then the average score of
                  candidates in the Smith set is  3+0/2 = 1.5.   Only A
                  is scored above 1.5 so only A is approved.<br>
                  <br>
                </div>
              </div>
            </div>
          </blockquote>
          <div><br>
          </div>
          <div>This is a nice technique, but it is not clone resistant.
            Adding Smith candidate clones can change the average and
            thus the approval cutoff would change. This seems unstable
            to me.</div>
          <div><br>
          </div>
          <div>I think a better technique is to either have an explicit
            cutoff which is lowered per ballot so that max(Smith score)
            is approved, or to make the cutoff such that candidates with
            max(Smith candidate score)/2 or greater per ballot are
            approved. I prefer the former.</div>
          <blockquote class="gmail_quote"
style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
            <div>
              <div lang="x-unicode">
                <div lang="x-unicode">
                  <blockquote type="cite">I also noticed that there were
                    cases where Smith//ASM(implicit) would get different
                    results (better, IMO) than Smith//Implicit-approval.
                    <div><br>
                    </div>
                  </blockquote>
                  <br>
                  What does "ASM" stand for?<br>
                </div>
              </div>
            </div>
          </blockquote>
          <div><br>
          </div>
          <div><a
              href="https://electowiki.org/wiki/Approval_Sorted_Margins"
              moz-do-not-send="true">Approval Sorted Margins</a></div>
          <div> </div>
          <blockquote class="gmail_quote"
style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
            <div>
              <div lang="x-unicode">
                <div lang="x-unicode"> <br>
                  Chris<br>
                  <br>
                  <br>
                  <div>On 23/01/2024 10:56 am, Ted Stern wrote:<br>
                  </div>
                  <blockquote type="cite">
                    <div dir="ltr">Chris:
                      <div><br>
                      </div>
                      <div>Thanks for the clarifications, though you
                        didn't comment on whether ballots with all Smith
                        candidates below top rating should have their
                        ratings bumped up: i.e., D > E > A >
                        blank > B  (A and B in Smith) would be
                        recounted as A > blank > B. I think this
                        makes the most sense as a voter whose favorites
                        are eliminated would want to ensure that their
                        highest ranked Smith candidate is counted as
                        approved.</div>
                      <div><br>
                      </div>
                      <div>In general I agree with your comments, though
                        I think Condorcet//Approval with all ranked
                        ballots approved is probably not optimal, and
                        Approval Sorted Margins with explicit approval
                        would be too complex for a public proposal. I'd
                        be happy with Condorcet//Top-ratings as a public
                        proposal.<br>
                      </div>
                      <div><br>
                      </div>
                      <div>Smith//Implicit-approval seems to perform
                        well in a number of situations, but not
                        appreciably better enough to make it worth the
                        effort of trying to get people to accept
                        something more complicated than
                        Condorcet/Top-ratings. I also noticed that there
                        were cases where Smith//ASM(implicit) would get
                        different results (better, IMO) than
                        Smith//Implicit-approval.</div>
                      <div><br>
                      </div>
                    </div>
                    <br>
                    <div class="gmail_quote">
                      <div dir="ltr" class="gmail_attr">On Fri, Jan 19,
                        2024 at 4:05 PM C.Benham <<a
                          href="mailto:cbenham@adam.com.au"
                          target="_blank" moz-do-not-send="true"
                          class="moz-txt-link-freetext">cbenham@adam.com.au</a>>
                        wrote:<br>
                      </div>
                      <blockquote class="gmail_quote"
style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
                        <div>
                          <p> </p>
                          <blockquote type="cite">How is the average
                            calculated?</blockquote>
                          <br>
                          We interpret the ratings ballots as score
                          ballots, giving zero points for the bottom
                          rating (which is default for unrated),<br>
                          1 point for the next highest, 2 points for the
                          next above that and so on.  <br>
                          <br>
                          Then for any given ballot we add up the scores
                          of the candidates in the Smith set and divide
                          that by the number of candidates<br>
                          in the Smith set and interpret that ballot as
                          approving those Smith set candidates it scores
                          higher than that average score.<br>
                          <p>That simulates the best approval strategy
                            if the voters only know which candidates are
                            in the Smith set.<br>
                            <br>
                          </p>
                          <blockquote type="cite">What advantage does
                            Approval Sorted Margins have over
                            Smith//Implicit-Approval?</blockquote>
                          <br>
                          Do you mean Approval Sorted Margins using
                          ranking ballots with an explicit approval
                          cutoff?   <br>
                          <br>
                          Assuming yes, it uses a more expressive
                          ballot, it is less vulnerable to Defection
                          strategy, and burial strategies are more<br>
                          likely to have no effect rather than backfire.<br>
                          <br>
                          In the method I proposed, omitting the ASM
                          step and just electing the candidate with the
                          highest approval score (derived<br>
                          as specified) would I concede make for a
                          simpler method that is nearly as good.<br>
                          <br>
                          I worry a bit that with all methods that begin
                          with eliminating or disqualifying all
                          candidates who aren't in the Smith set or <br>
                          just "elect the CW if there is one", over time
                          if there is never a top cycle then the
                          top-cycle resolution method could stop<br>
                          being taken seriously.<br>
                          <br>
                          An attractive feature of ASM is that it is a
                          Condorcet method that a lot of the time would
                          work fine without anyone needing<br>
                          to know if there is top cycle or not.<br>
                          <br>
                          If the Approval order is  A>B>C  and A
                          pairwise beats B and B pairwise beats C no-one
                          needs to enquire about the pairwise<br>
                          result between A and C.<br>
                          <br>
                          If we want something super simple to explain
                          and sell, then  Condorcet//Top Ratings and
                          Condorcet//Approval (voted above bottom)<br>
                          are both not bad and much better than STAR.<br>
                          <br>
                          Chris Benham<br>
                          <br>
                          <p><br>
                            <br>
                          </p>
                          <div>On 18/01/2024 10:13 am, Ted Stern wrote:<br>
                          </div>
                          <blockquote type="cite">
                            <div dir="ltr">
                              <div dir="ltr"><br>
                              </div>
                              <br>
                              <div class="gmail_quote">
                                <div dir="ltr" class="gmail_attr">On
                                  Tue, Jan 16, 2024 at 7:27 AM C.Benham
                                  <<a
                                    href="mailto:cbenham@adam.com.au"
                                    target="_blank"
                                    moz-do-not-send="true"
                                    class="moz-txt-link-freetext">cbenham@adam.com.au</a>>
                                  wrote:<br>
                                </div>
                                <blockquote class="gmail_quote"
style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
                                  <div>
                                    <p>Ted,<br>
                                      <br>
                                    </p>
                                    <blockquote type="cite">
                                      <pre
style="white-space:pre-wrap;color:rgb(0,0,0);font-style:normal;font-variant-ligatures:normal;font-variant-caps:normal;font-weight:400;letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;word-spacing:0px;text-decoration-style:initial;text-decoration-color:initial"> 3. Otherwise, drop ballots that don't contain ranks above last for
      any member of the Smith Set.</pre>
                                    </blockquote>
                                    <p>Why not simply drop all ballots
                                      that make no distinction among
                                      members of the Smith set?<br>
                                      <br>
                                    </p>
                                    <blockquote type="cite">
                                      <pre
style="white-space:pre-wrap;color:rgb(0,0,0);font-style:normal;font-variant-ligatures:normal;font-variant-caps:normal;font-weight:400;letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;word-spacing:0px;text-decoration-style:initial;text-decoration-color:initial">I believe it passes LNHelp.</pre>
                                    </blockquote>
                                    <br>
                                    Douglas Woodall showed some time ago
                                    that Condorcet and LNHelp are
                                    incompatible.  I can't find<br>
                                    his proof, but it says so here:<br>
                                    <br>
                                    <a
href="https://en.wikipedia.org/wiki/Later-no-help_criterion"
                                      target="_blank"
                                      moz-do-not-send="true"
                                      class="moz-txt-link-freetext">https://en.wikipedia.org/wiki/Later-no-help_criterion</a><br>
                                    <br>
                                    <blockquote type="cite"><span
style="color:rgb(32,33,34);font-family:sans-serif;font-size:14px;font-style:normal;font-variant-ligatures:normal;font-variant-caps:normal;font-weight:400;letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;word-spacing:0px;white-space:normal;background-color:rgb(255,255,255);text-decoration-style:initial;text-decoration-color:initial;float:none;display:inline"><span> </span>The<span> </span></span><a
href="https://en.wikipedia.org/wiki/Condorcet_criterion"
                                        title="Condorcet criterion"
style="text-decoration:none;color:rgb(51,102,204);background:none rgb(255,255,255);font-family:sans-serif;font-size:14px;font-style:normal;font-variant-ligatures:normal;font-variant-caps:normal;font-weight:400;letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;word-spacing:0px;white-space:normal"
                                        target="_blank"
                                        moz-do-not-send="true">Condorcet
                                        criterion</a><span
style="color:rgb(32,33,34);font-family:sans-serif;font-size:14px;font-style:normal;font-variant-ligatures:normal;font-variant-caps:normal;font-weight:400;letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;word-spacing:0px;white-space:normal;background-color:rgb(255,255,255);text-decoration-style:initial;text-decoration-color:initial;float:none;display:inline"><span> </span>is
                                        incompatible with later-no-help.</span></blockquote>
                                    <br>
                                    From your post again:<br>
                                    <blockquote type="cite">
                                      <pre
style="white-space:pre-wrap;color:rgb(0,0,0);font-style:normal;font-variant-ligatures:normal;font-variant-caps:normal;font-weight:400;letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;word-spacing:0px;text-decoration-style:initial;text-decoration-color:initial">It probably fails Participation ..</pre>
                                    </blockquote>
                                    <br>
                                    It has been known (for a longer
                                    time) that Condorcet and
                                    Participation are incompatible.<br>
                                    <br>
                                    So since we know for sure that your
                                    method meets Condorcet, we also know
                                    that it doesn't meet <br>
                                    Later-no-Help or Participation.<br>
                                    <br>
                                    Using a multi-slot ratings ballot
                                    for a Condorcet method of similar
                                    complexity I like:<br>
                                    <br>
                                    *Eliminate  all candidates not in
                                    the Smith set.<br>
                                    <br>
                                    Interpret each ballot as giving 
                                    approval to those remaining
                                    candidates they rate above average
                                    (mean <br>
                                    of the ratings given to Smith-set
                                    members).<br>
                                    <br>
                                    Now, using these approvals, elect
                                    the Margins-Sorted Approval winner.*<br>
                                  </div>
                                </blockquote>
                                <div><br>
                                </div>
                                <div>It seems to me that
                                  Smith//Implicit-Approval or
                                  Smith//implicit-approval-sorted-margins
                                  would be affected by a couple of
                                  factors:</div>
                                <div>
                                  <ul>
                                    <li>How is the average calculated?
                                      Do you normalize scores? In other
                                      words, if a ballot has non-Smith
                                      candidates in the first, say,
                                      three ranks, do you up-rank the
                                      Smith candidate scores on that
                                      ballot by three? Also, if there
                                      are ranks below the top that
                                      contain only non-Smith candidates,
                                      do you collapse those ranks or
                                      leave the relative rank spacing on
                                      the ballot between Smith
                                      candidates untouched?</li>
                                    <li>Approving Smith Candidates with
                                      scores above the mean has
                                      similarities to Median Ratings. It
                                      would be more similar and probably
                                      more stable to use the trimmed
                                      mean -- drop the top and bottom
                                      25% of scores. This would give you
                                      an average score closer to the
                                      median. </li>
                                  </ul>
                                  <div>What advantage does Approval
                                    Sorted Margins have over
                                    Smith//Implicit-Approval? I like ASM
                                    but fear it is probably too complex
                                    for any advantage it gives you.</div>
                                </div>
                                <blockquote class="gmail_quote"
style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
                                  <div> <br>
                                    <blockquote type="cite">
                                      <pre
style="white-space:pre-wrap;color:rgb(0,0,0);font-style:normal;font-variant-ligatures:normal;font-variant-caps:normal;font-weight:400;letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;word-spacing:0px;text-decoration-style:initial;text-decoration-color:initial">Has Smith//Median Rating been proposed before?</pre>
                                    </blockquote>
                                    Not that I know of.<br>
                                    <br>
                                    Chris Benham<br>
                                    <br>
                                    <br>
                                    <br>
                                    <blockquote type="cite">
                                      <h1
style="color:rgb(0,0,0);font-family:"Times New Roman";font-style:normal;font-variant-ligatures:normal;font-variant-caps:normal;letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;word-spacing:0px;white-space:normal;text-decoration-style:initial;text-decoration-color:initial"><br>
                                      </h1>
                                      <b
style="color:rgb(0,0,0);font-family:"Times New Roman";font-size:medium;font-style:normal;font-variant-ligatures:normal;font-variant-caps:normal;letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;word-spacing:0px;white-space:normal;text-decoration-style:initial;text-decoration-color:initial">Ted
                                        Stern</b><span
style="color:rgb(0,0,0);font-family:"Times New Roman";font-size:medium;font-style:normal;font-variant-ligatures:normal;font-variant-caps:normal;font-weight:400;letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;word-spacing:0px;white-space:normal;background-color:rgb(255,255,255);text-decoration-style:initial;text-decoration-color:initial;float:none;display:inline"><span> </span></span><a
href="mailto:election-methods%40lists.electorama.com?Subject=Re%3A%20%5BEM%5D%20Pairwise%20Median%20Rating&In-Reply-To=%3CCAHGFzOSm%3Deni2SuD5YRMrYBu4Gn9%2BYQ2NrqC_sXG8QFPrcVApQ%40mail.gmail.com%3E"
title="[EM] Pairwise Median Rating"
style="font-family:"Times New Roman";font-size:medium;font-style:normal;font-variant-ligatures:normal;font-variant-caps:normal;font-weight:400;letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;word-spacing:0px;white-space:normal"
                                        target="_blank"
                                        moz-do-not-send="true">dodecatheon
                                        at gmail.com</a><br
style="color:rgb(0,0,0);font-family:"Times New Roman";font-size:medium;font-style:normal;font-variant-ligatures:normal;font-variant-caps:normal;font-weight:400;letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;word-spacing:0px;white-space:normal;text-decoration-style:initial;text-decoration-color:initial">
                                      <i
style="color:rgb(0,0,0);font-family:"Times New Roman";font-size:medium;font-variant-ligatures:normal;font-variant-caps:normal;font-weight:400;letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;word-spacing:0px;white-space:normal;text-decoration-style:initial;text-decoration-color:initial">Tue
                                        Jan 2 15:12:26 PST 2024</i><span
style="color:rgb(0,0,0);font-family:"Times New Roman";font-size:medium;font-style:normal;font-variant-ligatures:normal;font-variant-caps:normal;font-weight:400;letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;word-spacing:0px;white-space:normal;background-color:rgb(255,255,255);text-decoration-style:initial;text-decoration-color:initial;float:none;display:inline"></span>
                                      <ul
style="color:rgb(0,0,0);font-family:"Times New Roman";font-size:medium;font-style:normal;font-variant-ligatures:normal;font-variant-caps:normal;font-weight:400;letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;word-spacing:0px;white-space:normal;text-decoration-style:initial;text-decoration-color:initial">
                                        <li>Previous message (by
                                          thread):<span> </span><a
href="http://lists.electorama.com/pipermail/election-methods-electorama.com/2024-January/005215.html"
                                            target="_blank"
                                            moz-do-not-send="true">[EM]
                                            [Game Theory] Iterated
                                            Prisoners' Dilemma as a
                                            voting method metric</a></li>
                                        <li>Next message (by thread):<span> </span><a
href="http://lists.electorama.com/pipermail/election-methods-electorama.com/2024-January/005218.html"
                                            target="_blank"
                                            moz-do-not-send="true">[EM]
                                            Pairwise Median Rating</a></li>
                                        <li><b>Messages sorted by:</b><span> </span><a
href="http://lists.electorama.com/pipermail/election-methods-electorama.com/2024-January/date.html#5216"
                                            target="_blank"
                                            moz-do-not-send="true">[
                                            date ]</a><span> </span><a
href="http://lists.electorama.com/pipermail/election-methods-electorama.com/2024-January/thread.html#5216"
                                            target="_blank"
                                            moz-do-not-send="true">[
                                            thread ]</a><span> </span><a
href="http://lists.electorama.com/pipermail/election-methods-electorama.com/2024-January/subject.html#5216"
                                            target="_blank"
                                            moz-do-not-send="true">[
                                            subject ]</a><span> </span><a
href="http://lists.electorama.com/pipermail/election-methods-electorama.com/2024-January/author.html#5216"
                                            target="_blank"
                                            moz-do-not-send="true">[
                                            author ]</a></li>
                                      </ul>
                                      <hr
style="color:rgb(0,0,0);font-family:"Times New Roman";font-size:medium;font-style:normal;font-variant-ligatures:normal;font-variant-caps:normal;font-weight:400;letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;word-spacing:0px;white-space:normal;text-decoration-style:initial;text-decoration-color:initial">
                                      <pre
style="white-space:pre-wrap;color:rgb(0,0,0);font-style:normal;font-variant-ligatures:normal;font-variant-caps:normal;font-weight:400;letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;word-spacing:0px;text-decoration-style:initial;text-decoration-color:initial">Continuing my search for a summable voting method that discourages burial
and defection, I've come across this hybrid of Condorcet and median ratings
that acts like Smith/Approval with an automatic approval cutoff. I'm
calling it Pairwise Median Rating (PMR), but it could also be described as
Smith//MR//Pairwise//MRScore:

   1. Equal Ranking and ranking gap allowed (essentially a ratings method
   with rank inferred). For purposes of this discussion, assume 6 slots (5
   ranks above rejection).
   2. In rank notation for this method, '>>' refers to a gap. So 'A >> B'
   means A gets top rank while B gets 3rd place. Similarly '>>>' means a gap
   of two slots: 'A>>>B' means A is top ranked while B is in 4th place.
   3. [Smith]
      1. Compute the pairwise preference array
      2. The winner is the candidate who defeats each other candidate
      pairwise.
      3. Otherwise, drop ballots that don't contain ranks above last for
      any member of the Smith Set.
   4. [Median Rating]
      1. Set the MR threshold to top rank.
      2. While no Smith candidate has a majority of undropped ballots at or
      above the threshold, set the threshold to the next lower rank,
until there
      is no lower rank.
      3. The winner is the single candidate that has a majority of
      undropped ballots at or above the threshold.
   5. [Pairwise]
   1. Otherwise, if more than one candidate passes the threshold, look for
      a pairwise beats-all candidate among candidates meeting the MR threshold.
      (i.e. Condorcet on just the MR threshold set).
      2.  If there is one, you have a winner.
   6. [MR Score]
   1. Otherwise, the winner is the Smith set candidate with the largest
      number of ballots at or above the Median Rating threshold (their MRscore).

This method is essentially Smith//Approval(explicit) with the approval
cutoff automatically inferred via median ratings

Step Smith.3, dropping non-Smith-candidate-voting ballots, could be
considered optional, but by doing that, you ensure Immunity from Irrelevant
Ballots (IIB), aka the zero ballot problem that affects other Median Rating
/ Majority Judgment methods. In other words, the majority threshold is
unaffected by ballots that do not rank a viable candidate. It is possible
to do this summably if need be.

PMR either passes the Chicken Dilemma criterion without adjustment, or
there is a downranking strategy for defending against defection.

Consider the following examples from Chris Benham's post re MinLV(erw)
Sorted Margins (
<a
href="http://lists.electorama.com/pipermail/election-methods-electorama.com/2016-October/000599.html"
                                      target="_blank"
                                      moz-do-not-send="true"
                                      class="moz-txt-link-freetext">http://lists.electorama.com/pipermail/election-methods-electorama.com/2016-October/000599.html</a>
):

><i>* 46 A>B
</i>*>* 44 B>C (sincere is B or B>A)
*>* 05 C>A
*>* 05 C>B
*>>* A>B 51-49,    B>C  90-10,    C>A 54-46.
*

With sincere ballots, A is the Condorcet Winner (CW).  With B's defection,
there is a cycle, and there is no CW. The Smith set is {A, B, C}. The MR
threshold is 2nd place, and A and B both pass the threshold. A defeats B,
so A is the winner and B's defection/burial fails.

><i>* 25 A>B
</i>*>* 26 B>C
*>* 23 C>A
*>* 26 C
*>>* C>A  75-25,    A>B  48-26,   B>C  51-49*

C wins with PMR (MR threshold is first place). B would win with most
other Condorcet methods.

><i>* 35 A
</i>*>* 10 A=B
*>* 30 B>C  (sincere B > A)
*>* 25 C
*>>* C>A  55-45,     A>B  35-30 (10A=B not counted),   B>C 40-25.

*A wins with sincere voting. When B defects to try to win, which it
would do with most other Condorcet methods, B wins. With PMR, C wins,
an undesirable outcome for B.

Here is another example from Rob LeGrand
(<a href="https://www.cs.angelo.edu/~rlegrand/rbvote/calc.html"
                                      target="_blank"
                                      moz-do-not-send="true"
                                      class="moz-txt-link-freetext">https://www.cs.angelo.edu/~rlegrand/rbvote/calc.html</a>). It's not a
good example for chicken dilemma resistance, but it does demonstrate
differences from Schulze, MMPO, RP and Bucklin:

# example from method description page
 98:Abby>Cora>Erin>Dave>Brad
 64:Brad>Abby>Erin>Cora>Dave
 12:Brad>Abby>Erin>Dave>Cora
 98:Brad>Erin>Abby>Cora>Dave
 13:Brad>Erin>Abby>Dave>Cora
125:Brad>Erin>Dave>Abby>Cora
124:Cora>Abby>Erin>Dave>Brad
 76:Cora>Erin>Abby>Dave>Brad
 21:Dave>Abby>Brad>Erin>Cora
 30:Dave>Brad>Abby>Erin>Cora
 98:Dave>Brad>Erin>Cora>Abby
139:Dave>Cora>Abby>Brad>Erin
 23:Dave>Cora>Brad>Abby>Erin

The pairwise matrix:

against
Abby Brad Cora Dave Erin
for Abby  458 461 485 511
Brad 463  461 312 623
Cora 460 460  460 460
Dave 436 609 461  311
Erin 410 298 461 610

There is no Condorcet winner.  The Smith set is {Abby, Brad, Dave, Erin}.

Abby wins with Schulze, MMPO, while Brad wins with Ranked Pairs, Erin
wins with Bucklin.

In PMR, with a threshold of 3rd place, Abby, Brad, and Erin all pass
the threshold. Brad defeats Abby and Erin to win. But Brad's threshold
score of 484 is only slightly over the 50% mark of 460.5, so the Dave
voters hold the balance of power. Dave defeats Brad pairwise, so Dave
voters might not be as happy with a Brad victory, and Abby might be
able to persuade Dave voters to downrank Brad but not Abby. If
successful, Brad drops 44 points in MRScore and is no longer in the MR
threshold set. Abby defeats Erin, so Abby wins.

* 21:Dave>Abby>>Brad>Erin>Cora (Brad -> 4th place)*
 30:Dave>Brad>Abby>Erin>Cora
 98:Dave>Brad>Erin>Cora>Abby
139:Dave>Cora>Abby>Brad>Erin
* 23:Dave>Cora>>Brad>Abby>Erin (Brad -> 4th place)*

PMR passes Condorcet Winner, Condorcet Loser, IIB, and is cloneproof.
I believe it passes LNHelp. It probably fails Participation and IIA.
There are probably weird examples where changing one vote changes the
MR threshold. But overall, I think it has a good balance of incentive
to deter burial and deliberate cycles.

Has Smith//Median Rating been proposed before? It seems like a simple
modification to MR on its own.</pre>
                                    </blockquote>
                                    <br>
                                  </div>
                                  ----<br>
                                  Election-Methods mailing list - see <a
                                    href="https://electorama.com/em"
                                    rel="noreferrer" target="_blank"
                                    moz-do-not-send="true"
                                    class="moz-txt-link-freetext">https://electorama.com/em</a>
                                  for list info<br>
                                </blockquote>
                              </div>
                            </div>
                          </blockquote>
                        </div>
                      </blockquote>
                    </div>
                  </blockquote>
                </div>
              </div>
            </div>
            ----<br>
            Election-Methods mailing list - see <a
              href="https://electorama.com/em" rel="noreferrer"
              target="_blank" moz-do-not-send="true"
              class="moz-txt-link-freetext">https://electorama.com/em</a>
            for list info<br>
          </blockquote>
        </div>
      </div>
    </blockquote>
  </body>
</html>