<html>
<head>
<style><!--
.hmmessage P
{
margin:0px;
padding:0px
}
body.hmmessage
{
font-size: 10pt;
font-family:Tahoma
}
--></style>
</head>
<body class='hmmessage'><div dir='ltr'>
<pre><br>Kevin--<br><br>You wrote:<br><br>ER-IRV(whole) doesn't satisfy FBC. You may need to demote your favorite in order<br>to get a preferable elimination order.<br> <br>[endquote]<br><br>How? Say that there's a particular candidate whom you need to have win. <br><br>You can give him a vote by downrating your favorite in order to get hir eliminated soon,<br>so that your ballot will give a vote to the compromise. But you could also just give the<br>compromise an immediate vote, by ranking hir in 1st place. Why would you need to do otherwise<br>in order to help hir win?<br><br>You wrote:<br><br>Also, I do plan at some point to find an FBC failure of MMPO//MMPO//MMPO//etc.<br>I'm thinking it may require four candidates. My idea is that by lowering your favorite,<br>you knock your favorite out of a tie, so that only two candidates remain. But if there<br>were a three-way tie with no other candidates, MMPO would be stuck. So there<br>must be a fourth candidate <a href="http://lists.electorama.com/pipermail/election-methods-electorama.com/2011-November/028892.html#" style="color: green; text-decoration: none; border-bottom: 3px double;" id="_GPLITA_1">available</a> to be eliminated and allow the method to <br>proceed.<br> <br>[endquote]<br><br>Yes, when you mentioned that, it occurred to me that there could be such a scenario.<br><br>Anyway, I've abandoned MMPO because of its 3P failure.<br><br>You wrote:<br><br>Maybe I can reword slightly what Jameson said:<br>If sincere scenario (plus specific assumptions of voter sincerity) "S" can give rise to<br>cast ballots B, and a criterion C says S must have outcome O, then any method is <br>going to have to select outcome O whenever the cast ballots are B. The criterion in<br>effect is requiring this, even when the scenario isn't S.<br><br>[endquote]<br><br>Rewording it in English would be preferable.<br><br>But let me try to answer anyway:<br><br>"S" represents a scenario in which some voters are stipulated to vote sincerely, correct?<br><br>Yes, if something has been stipulated about their preferences, and they're stipulated to vote sincerely<br>then yes, that implies something definite about their votes.<br><br>And the requirement requires that those votes produce (or don't produce) some particular outcome.<br><br>But what makes you think that any voting system must produce the same outcome, given the same <br>criterion-premise? Big claims like that require verification. Surely different methods can give<br>different outcomes, with the same ballots, no?<br><br>Or are you saying that the criterion requires that? Of course it does. Given a certain premise, the<br>criterion requires something about the outcome. All criteria require that. That's the nature of criteria.<br><br>That's true with votes-only criteria, and it's also true with my criteria. What's the problem with that?<br><br>I suggest that your error is in not realizing that the premises of my criteria indirectly stipulate <br>something about votes.<br><br>As I said in the posting that you're replying to (and already in this one too), <br>if a voter is stipulated to have certain preferences, and to vote sincerely, then<br>that says something about how s/he votes. That's what you're missing. My criteria (indirectly)<br>stipulate how people vote, just as do votes-only criteria.<br><br>What do you mean when you say "even when the scenario isn't S? If "S" represents a scenario <br>meeting the stipulations of the criterion's premise, then the criterion only makes a requirement<br>when those stipulations are met. It says nothing about any other scenario. That's true of all<br>criteria, not just mine.<br><br>No criterion, including mine, says anything about what should happen when the "scenario" described<br>by its premise-stipulagions doesn't happen.<br><br>You're might well be right about MMPO not meeting FBC. <br><br>You _weren't_ right Bucklin(= whole simultaneous)<br>not meeting FBC, because you were thinking that FBC requires that no one need to <br>vote someone equal to their favorite. Actually, FBC requires only that<br>no one need to vote someone _over_ their favorite.<br><br>You suggested that there could be a scenario in IRV(= whole) where you need to get your favorite eliminated<br>quickly to help a compromise. Why? You can give a vote to the compromise immediately just by<br>ranking hir in 1st place, along with your favorite. You can't give more than one vote to a candidate.<br><br>Returning to my criteria, which stipulate preferences and relationships between voting and<br>preferences, <br><br>You said:<br><br>It's like how SFC is supposed to protect sincere CWs, but in practice it ends up <br>also defending some candidates who just look like they *might* be sincere CWs.<br><br>[endquote]<br><br>Irrelevant. Maybe you mean that, in an actual election, from the election-results, votes and winner, we<br>can't tell whether my criteria's premises have been met. So what. How often do we use criteria by<br>applying them to actual election results? Criteria are about the question "Could this happen?" Whether or<br>not we can tell whether it happened in a particular actual election is irrelevant. With any criteria, not just<br>mine, we judge a method by whether, given the criterion's premise, some particular outcome could or must happen.<br>Period.<br> <br>Yes, SFC protects sincere CWs. Do SFC complying methods protect other candidates who don't<br>have a majority defeat (and, in that way "look like sincere CWs)? Sure. What's the problem<br>with that?<br> <br>Mike Ossipoff<br><br></pre> </div></body>
</html>