[EM] Random Ballot / Smith

[Kristofer Munsterhjelm]

On 2025-09-22 13:23, robert bristow-johnson via Election-Methods wrote:
> 
> I really think that cycles will be extremely rare.  I think Kristofer, 
> you're correct that the only two RCV elections in the US that, has they 
> been Condorcet would be a cycle, were Minneapolis city councilor Ward 2 
> in 2021 and Oakland school board commissioner District 4 in 2022.  I 
> think David McCune has documented both.  There is a recent multiwinner 
> RCV election in Moab Utah that I am told would have no CW if it were 
> single winner (but I dunno if it was a cycle or that the CW came in 3rd).

I agree that cycles are currently very rare. The question is whether 
they would remain rare as the political landscape changes; and that 
could go either way. The example I mentioned could happen, but then 
again, that could signal to candidates that there's a demand for a clean 
candidate with a good policy, as that candidate would then become a CW 
and win; or debates could help society figure out which issue or quality 
it considers most important.

Perhaps strategic cycles would become common -- the parties in New York 
certainly tried to game STV in the thirties -- and strategic resistance 
is a must-have. Or perhaps not.

It's hard to tell. Personally I think C,Plurality is a bit too risky and 
C,TTR (or Smith, or Copeland,) is considerably better. But that's not 
definite proof :-)

So jurisdictions should probably just try whatever Condorcet method that 
can be enacted - and then we'll hope it doesn't blow up in our faces 
IRV-style.

-km