On Thu, 1 Feb 2001, Joshua Bell wrote:
> (To reiterate: If a candidate has >50% of public support on an issue, his
> opponents disagreeing on that issue obviously have <50% support - they'd be
> foolish to broadcast this support as they would be alienating the majority
> of voters. This will tend to cause candidates to select issues with an exact
> 50-50 polarization of public opinion, and other issues will simply be
> ignored. This only leaves room for two candidates to argue an issue.)
You're assuming that stances on all issues are perfectly correlated.
If you have two issues, each of which has 50% support, but for which
that 50% is randomly distributed relative to each other, you can
have:
Pro A, Pro B (25% of popular support)
Pro A, Anti B (ditto)
Anti A, Pro B (ditto)
Anti A, Anti B (ditto)
This is admittedly extreme, but it's no more absurd than your assumption tha
any election is *in principle* rather than in practice based on a single
issue-cluster, where the public is either pro A,B,C,D.... or anti A,B,C,D....
with no issue-crossing, or that all politicians will agree to ignore
the *same* set of 50/50 split issues to collapse it down to a single
issue.
--
Andrea Leistra [EMAIL PROTECTED]
"If you can keep your head while all those about you are
losing theirs, perhaps you have misunderstood the situation."
-- Daniel Keys Moran, _The Long Run_
