[EM] An unusual "absolving" PR method

[Kristofer Munsterhjelm]

On 2025-12-13 09:11, Etjon Basha wrote:
> Good evening Kristofer,
> 
> I've played with basically the same system myself, though I go with 
> quadratic voting at the individual level in the name of countability.
> 
> Anyway, i don't think there'd be anything stopping the rush to the 
> centre here: parties would just have their national profile set at 
> whichever ideological equilibrium they need to tend to by nation-wide 
> spots or polemic, and have the local candidates rush one over the other 
> to say "well, the boys at central are a bit rough, but their heart is in 
> the right place. We'll find reasonable solutions here".

Right, that's a good point. It's very dependent on strong central party 
discipline.

In the lack of such, maybe the following sketch is better: For each 
district, for each party, deweight voters who didn't count as voting for 
that party in the list election, and usse Condorcet on post-deweighted 
ballots to create an order of finish for the candidates of that party. 
Pick winners from that "generated list".

But then the question becomes: how strongly do you deweight? You'd end 
up with something that probably still fails the DPC - that's the cost of 
the simplicity - but a choice of weight factor makes it more complex.

-km