On Aug 29, 2011, at 6:25 AM, Jameson Quinn wrote:
> Dave Ketchum wrote:
> > NOT true, for the vote, without the voter's vote, could be a tie - and
> > the voter's vote mattering.
>
> That notion of effect has several drawbacks:
>
> ...all of which merely serve to minimize its practical importance, not to
> assail its mathematical validity.
>
> ... But if we (this is my
> hope) can cogently demonstrate this failing to the experts in this
> list, especially in terms of the voting mechanisms they understand so
> well, then they will be more open to drawing the larger conclusions
> that seem so obvious to you and me, and I daresay others in this list.
>
> I've been trying to avoid entering this sub-thread, as I think it's mostly
> angels-on-pinheads stuff, but if you actually have a point, I suggest you
> make it, rather than portentiously musing on how it depends on a
> supposedly-proven, but still-debated claim.
>
The usual argument that I've seen is that the expected utility of casting a
vote (the utility of the result you favor, however you might measure that,
times the probability that your vote will be decisive) is so small (because the
probability is small) that the cost of casting the vote outweighs its utility.
The validity of the argument depends on the election, of course. In a small
enough voting body, it's not true. OTOH, it's obviously true for a US
presidential voter in California, who we can safely assert will never be
decisive in a presidential election. (And yet voters cast presidential votes in
California.)
----
Election-Methods mailing list - see http://electorama.com/em for list info
