[EM] Why IRV is better than Condorcet

[Eric Gorr]

At 6:57 PM -0400 7/29/04, Warren Schudy wrote:
>Whether or not approval or
>condorcet or IRV produces the highest societal utility given honest voting
>is basically irrelevant for public elections, since voters do not, in
>practice, vote honestly in public elections.

Why make this assumption that that voters would not, in practice, 
vote honestly?

It seems a rather unreasonable assumption to make, at least with 
voting systems which are resistant to strategic manipulation.

With the better Condorcet methods I have yet to see a yet where 
strategic manipulation, without near perfect information (which 
requires accurate information beyond the ability of any pre-election 
poll to provide), would not have a risk of electing a candidate all 
sides consider poor.

However, if this perfect information was available, as it likely 
would be for small groups where all the voters knew everyone else and 
would how others would vote, it may be possible to vote in such a way 
as to guarantee an acceptable outcome. But, I would guess that such 
guarantees would be very hard to come by in practice. Since it is not 
reasonable to assume that just one side would attempt such 
manipulations, the risk of selecting the worst possible candidate 
would again likely be unacceptable.