3 ms·

I'd rather see approval voting than ranked choice. The outcomes are roughly the same, but the system is easier and less disputable. Which is very much a featu

I'd rather see approval voting than ranked choice. The outcomes are roughly the same, but the system is easier and less disputable. Which is very much a feature in the US; remember hanging chads in the 2000 race? If they can dispute that, then they can certainly dispute any unclear ranking marks. And also because ranked choice usually comes along with instant-runoff here in the US, which is possibly the only system with worse outcomes than first-past-the-post. http://zesty.ca/voting/sim/ http://zesty.ca/voting/sim/ Note that the two weird graphs are always plurality, aka first past the post, and Hare, aka instant runoff. Also note that there's no obvious difference in any of the graphs between approval and Condorcet, even though Condorcet is ranked.

jakelazaroff 6y ago
I would actually still prefer IRV to FPTP. The center candidate's boundary can be unpredictable, but the other boundaries are in basically the same place as they are for approval and Condorcet. FPTP shifts all the boundaries and drastically reduces or eliminates center candidates' winning regions. (I'd also prefer it to Borda, which drastically enlarges center candidates' winning regions). That said, I agree that approval seems to strike the best balance between simplicity and outcome, with the caveat that where possible we should also have proportional representation rather than winner-take-all districts, e.g. https://www.fairvote.org/fair_representation https://www.fairvote.org/fair_representation
- jdmichal 6y ago
  Agreed on all accounts. I'm mostly just disappointed that, when the chance comes to actually affect change, we seem to mostly just be moving from one provably broken system (FPTP) to another (IRV). While at the same time increasing complexity more than necessary to achieve a better outcome. But even with IRV, I would be happy I could finally throw some votes at third-party candidates whose platform I align with, without throwing away my entire vote.

jakelazaroff 6y ago

I would actually still prefer IRV to FPTP. The center candidate's boundary can be unpredictable, but the other boundaries are in basically the same place as they are for approval and Condorcet. FPTP shifts all the boundaries and drastically reduces or eliminates center candidates' winning regions. (I'd also prefer it to Borda, which drastically enlarges center candidates' winning regions). That said, I agree that approval seems to strike the best balance between simplicity and outcome, with the caveat that where possible we should also have proportional representation rather than winner-take-all districts, e.g. https://www.fairvote.org/fair_representation https://www.fairvote.org/fair_representation

jdmichal 6y ago
Agreed on all accounts. I'm mostly just disappointed that, when the chance comes to actually affect change, we seem to mostly just be moving from one provably broken system (FPTP) to another (IRV). While at the same time increasing complexity more than necessary to achieve a better outcome. But even with IRV, I would be happy I could finally throw some votes at third-party candidates whose platform I align with, without throwing away my entire vote.