[EM] IRV vs Condorcet

[Adam Haas Tarr]

Markus wrote:

>Gore and Nader are no clones in your example. They would have been clones
>only when this example had looked as follows:
>
>Bush 49%
>Gore>Nader 24%
>Nader>Gore 27%

Fair enough: I guess my point is that Nader is also not an irrelevant 
alternative, neither in the strict sense nor even in the local sense.

>Whether independence from clones is violated has nothing to do with
>whether "winning votes" or "margins" are being used.

Yes, both beatpath and ranked pairs (Schulze and Tideman) are clone-independent 
regardless of whether you evaluate defeat strength using winning votes or 
margins.

-Adam