<div dir="ltr"><br><div class="gmail_extra"><div class="gmail_quote">2014/1/6 Dick Burkhart <span dir="ltr"><<a href="mailto:dickburkhart@comcast.net" target="_blank" onclick="window.open('https://mail.google.com/mail/?view=cm&tf=1&to=dickburkhart@comcast.net&cc=&bcc=&su=&body=','_blank');return false;">dickburkhart@comcast.net</a>></span><br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Except that the restricted Borda that I've described is the best general<br>
purpose voting method as far as I can tell. Sure you can do somewhat better<br>
in particular circumstances, such as when you want true proportionality, or<br>
when you have non-partisan judges, etc.<br></blockquote><div><br></div><div>You haven't defined "best". And I'm very skeptical you can do so in such a way that Borda is best. Without strategy, Score is better than Borda. With strategy, Borda is one of the worst possible systems; worse even than plurality or random ballot, as with simple first-order strategy Borda guarantees a nonentity will win. Your "restricted" Borda, insofar as I understand it, is slightly better than that worst case, but only by means of taking the edge off the system and making it more like plurality. Yech.</div>
<div><br></div><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<br>
BTW, when I've run tests on actual sets of votes from Glasgow, most of the<br>
time Borda, Condorcet, and Approval give the same result!<br></blockquote><div><br></div><div>Using actual votes ignores system-specific strategy, which is Borda's worst defect (though not the only one.)</div><div><br>
</div><div><br></div></div></div></div>