Condorcet's mathematical built-in bias

[Mike Ossipoff]

DEMOREP1 at aol.com writes:
> 
> Second or additional choices in a real election are no more random than first
> choices.

Is that supposed to have something to do with Condorcet's method?

Or are you talking about simulations? If so, then 2nd choices aren't
random in real life, only in your simulation.

Say we have an election between a progressive, a Republican
& a Democrat. Is it a coin-toss whether it's the Democrat or the
Republican who will be the 2nd choice of those who vote
the progressive in 1st place?

No, Demprep, if you want simulations that are at all realistic,
then I suggest the spatial studies that I described in a
previous posting.

However,  I'm not saying your simulations haven't been of any
use. The fact that there was nearly always a beats-all winner,
even when rankings were totally random, means that natural circular
ties will be rare, however chaotic the conditions.


Mike






> 
> .-
> 


--