<div dir="ltr">This is great, thanks for your work on this km!<div><br></div><div>I'm a bit confused on the IRV results—doesn't IRV already elect from the resistant set?</div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Tue, Jul 23, 2024 at 4:09 PM Kristofer Munsterhjelm <<a href="mailto:km-elmet@munsterhjelm.no">km-elmet@munsterhjelm.no</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">While I was away from the list, I implemented VSE and OUF simulators for <br>
my voting tool, quadelect. OUF is my term for the utility measure James <br>
Green-Armytage used in his papers; it's short for "Optimal Utility <br>
Fraction", i.e. the fraction of the time the method in question elects <br>
the highest utility winner.<br>
<br>
(One benefit to OUF is that it converges faster and is more <br>
discriminating than VSE in the spatial model setting I used. But purists <br>
would probably prefer VSE.)<br>
<br>
And then I thought I would check how much quality is lost by insisting <br>
on resistant set compliance. In line with JGA's Smith-IRV results, my <br>
simulations seem to indicate that there's not much of a loss. Which is <br>
fortunate if we want something that's highly strategy resistant.<br>
<br>
Here are the results:<br>
<br>
OUF:<br>
<br>
Number of voters: 99, number of candidates: 3<br>
Gaussian spatial distribution, 8 dimensions, sigma = 1<br>
<br>
Ext-Minmax 0.950199<br>
Resistant,Ext-Minmax 0.949857<br>
Plurality 0.918477<br>
IRV 0.945945<br>
Smith,IRV 0.949666<br>
Resistant,Eurovision 0.950432<br>
<br>
Relative resistant set compliance cost: 0.035% (Minmax)<br>
<br>
Number of voters: 99, number of candidates: 6<br>
Gaussian spatial distribution, 8 dimensions, sigma = 1<br>
<br>
Ext-Minmax 0.923995<br>
Resistant,Ext-Minmax 0.923015<br>
Plurality 0.813871<br>
IRV 0.9066<br>
Smith,IRV 0.921964<br>
Eurovision 0.937177<br>
Resistant,Eurovision 0.927298<br>
<br>
Relative resistant set compliance cost: 0.1% (Minmax), 1.05% (Eurovision)<br>
<br>
VSE:<br>
<br>
Number of voters: 99, number of candidates: 6<br>
Gaussian spatial distribution, 8 dimensions, sigma = 1<br>
<br>
Ext-Minmax 0.994717<br>
Resistant,Ext-Minmax 0.994538<br>
Plurality 0.965466<br>
IRV 0.991495<br>
Smith,IRV 0.994306<br>
Eurovision 0.996406<br>
Resistant,Eurovision 0.995076<br>
<br>
Relative resistant set compliance cost: 0.02% (Minmax), 0.13% (Eurovision)<br>
<br>
The Eurovision method is a weighted positional method that gives the <br>
highest ranked candidate 12 points, then 10, then 8 down to zero: (12, <br>
10, 8, 7, ..., 1, 0, 0, 0...)<br>
<br>
So for methods that already pass majority, it doesn't seem that we lose <br>
much in terms of utility. The main costs of resistant set compliance are <br>
instead, in my opinion, that we lose monotonicity and summability, and <br>
that calculating the set is quite complex.<br>
<br>
-km<br>
----<br>
Election-Methods mailing list - see <a href="https://electorama.com/em" rel="noreferrer" target="_blank">https://electorama.com/em</a> for list info<br>
</blockquote></div>