<div dir="auto">Hi Kristofer,<div dir="auto"><br></div><div dir="auto">Sorry, I should have added that the inverse of the keep value will most likely only make sense if a very particular application of STV where instead of electing n winners from k candidates we "elect" K winners (i.e., we just redistribute votes among all candidates such as to leave each with exactly the same vote count). Only in this scenario does the keep value truly measure support.</div><div dir="auto"><br></div><div dir="auto">Regards,</div></div><br><div class="gmail_quote gmail_quote_container"><div dir="ltr" class="gmail_attr">On Thu, 12 Feb 2026, 12:55 am 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">On 2026-02-11 05:22, Etjon Basha wrote:<br>
> Hi Kristofer,<br>
> <br>
> Might the inverse of the keep value under a Warren or Meek STV count <br>
> suit as well?<br>
<br>
That's an interesting thought; they might, but I'd have to think more <br>
about how to represent scores for multiwinner, what they would mean, and <br>
how keep values would tally with it.<br>
<br>
For instance, I imagine using keep values would be better at showing a <br>
distinction between the winners than between losers (who have no surplus <br>
to redistribute as they don't reach the quota).<br>
<br>
The combinatorial Condorcet view (CPO-STV, Schulze STV) would be to give <br>
each *assembly outcome* a different score, but that's very hard to <br>
interpret.<br>
<br>
In any case, straightforward scores (one per candidate) would always <br>
lose some context because, unlike a single-winner method where closer to <br>
some favored point in space is better (median voter, or some wing <br>
position for center squeeze methods), what one candidate's optimum <br>
location is depends on what other candidates were elected.<br>
<br>
-km<br>
</blockquote></div>