election-methods@mailman.electorama.com

Technical discussion of election methods

View all threads

Poll ballot

KM
Kristofer Munsterhjelm
Fri, May 10, 2024 7:15 PM

Here's my ballot so far. This turned out to be incredibly hard and I'm
not entirely satisfied with the order, but, on the other hand, I can't
think of a better one, and the deadline is in 10 hours, so better put
something out there.

The main problem, I think, is that so much is context dependent. Do you
want max quality and the voters aren't too strategic? Use RP or Schulze.
Is strategy a problem? Use Condorcet-IRV. Is complexity a problem?
Smith//Approval or Smith//Range. Does it have to look like an
elimination method? Benham, Raynaud or Baldwin. Etc.

As is often the case, the middle is the hardest to pin down. And as
always, no slight intended against the designers or proposers of methods
I've ranked low.

With all that said:

  1. Ranked Pairs

  2. Schulze

  3. Benham

  4. Woodall

  5. Minmax (wv)

Approval cutoff --> ============ <-- Approval cutoff

  1. Schwartz-Woodall

  2. Smith//Approval (implicit)

  3. Smith//Approval (explicit)

  4. Smith//Range

  5. Raynaud (Gross Loser Elimination)

  6. Baldwin

  7. Majority Judgement

(13th place is for methods I'm not familiar enough with.)
13. Double Defeat, Hare
13. Max Strength Transitive Beatpath
13. Margins-Sorted Minimum Losing Votes (equal-rated whole)
13. Margins-Sorted Approval

  1. Copeland//Borda

  2. Smith//DAC

  3. Approval + runoff

  4. STAR

  5. RCIPE

  6. Condorcet//Borda (Black)

  7. Approval

  8. IRV

  9. Plurality

-km

Here's my ballot so far. This turned out to be incredibly hard and I'm not entirely satisfied with the order, but, on the other hand, I can't think of a better one, and the deadline is in 10 hours, so better put *something* out there. The main problem, I think, is that so much is context dependent. Do you want max quality and the voters aren't too strategic? Use RP or Schulze. Is strategy a problem? Use Condorcet-IRV. Is complexity a problem? Smith//Approval or Smith//Range. Does it have to look like an elimination method? Benham, Raynaud or Baldwin. Etc. As is often the case, the middle is the hardest to pin down. And as always, no slight intended against the designers or proposers of methods I've ranked low. With all that said: 1. Ranked Pairs 2. Schulze 2. Benham 4. Woodall 5. Minmax (wv) Approval cutoff --> ============ <-- Approval cutoff 6. Schwartz-Woodall 7. Smith//Approval (implicit) 7. Smith//Approval (explicit) 9. Smith//Range 10. Raynaud (Gross Loser Elimination) 11. Baldwin 12. Majority Judgement (13th place is for methods I'm not familiar enough with.) 13. Double Defeat, Hare 13. Max Strength Transitive Beatpath 13. Margins-Sorted Minimum Losing Votes (equal-rated whole) 13. Margins-Sorted Approval 17. Copeland//Borda 18. Smith//DAC 19. Approval + runoff 20. STAR 21. RCIPE 22. Condorcet//Borda (Black) 23. Approval 24. IRV 24. Plurality -km