On the subject of the paradoxes of various voting systems; chapter 12 (Is
Democracy Mathematically Unsound?) of Paul Hoffmans [anti-apostrophe] book:
Archimedes Revenge. ISBN 0 14 012506 X, is an excellent place to start.
There are many references to other publications for the serious investigator
into games theory and other no-choice choices that can be made by the
unknowing, to the advantage of the knower (ignoti nulla cupido). If you have
not read it, do so now before you are sucked into schemes that may be
inimical to your interests.
-----Original Message-----
From: [EMAIL PROTECTED]
<[EMAIL PROTECTED]>
To: [EMAIL PROTECTED] <[EMAIL PROTECTED]>
Date: Saturday, 17 October 1998 5:42 PM
Subject: RE: another voting system
>[G]
>
>Please assign the candidates according to how you feel about them. Give
>them a rating of 1 (or 0) if you dont like them to 10 to give full
>support. [...]
>
>How is this counted? All the numbers are added together for each party.
>Preferencial voting would have to be scrapped in the senate elections.
>Quotas are formed by adding All of the numbers together.
>
>[AL]
>There is an extensive literature on electoral systems and the
>mathematics of voting systems. You might find it useful to do a search
>for web sites about these issues and provide a page of links for
>future inclusion on the web site.
>
>One of the interesting things about the "alternative vote" method
>used in Australian elections (whether with optional or compulsory
>preferences and assuming that voters are entitled to reject any
>candidates they choose) is that it simply does not work unless the
>candidates fall into two broad camps.
>
>There are numerous "paradoxes" among which are the fact that a candidate
>who would otherwise have won can be defeated as a result of getting
>increased support.
>
>Alexander Downing referred to this when he complained that the reason he
>might lose to the Democrat candidate was because the ALP "ran dead".
>
>If the ALP had polled ahead of the democrats in his seat as usual then
>he would have been elected with a comfortable margin as usual when the
>Democrat preferences were split between himself, the leading candidate
>and his ALP opponent running second. But because the ALP came third, and
>their preferences went overwhelmingly to the Democrats, he was almost
>defeated.
>
>In Phil Cleary's seat of Wills an Independent Labor candidate actually
>did defeat the ALP because the Liberals came third and their preferences
>went to the Independent Labor candidate rather than the ALP, so in a
>three way contest, Cleary was elected with a substantial majority
>against the leading ALP candidate.
>
>If between 4000 and 9000 ALP voters had voted Liberal instead, then even
>if they had put the ALP last, the Liberals would have come second
>instead of third, Cleary would have been eliminated and his preferences
>would have elected the ALP by a substantial majority against the Liberal
>candidate.
>
>Thus the ALP lost Wills because too many of their supporters voted for
>them instead of voting for their main opponent.
>
>These and similar absurdities should be widely publicized as part of
>discrediting the present electoral system, but in doing so we need to be
>careful to avoid distracting attention from the actual illegality of
>coercing electors to vote in favor of candidates they want to vote
>against, and to emphasis the need for PR.
>
>The method you propose has similar paradoxes though not quite as
>dramatic as the "alternative vote". For example if the strongest
>candidates are Blue and Orange, a voter who would actually rate Blue as
>8 and Orange as 4 would be better off marking Blue as 10 and Orange as
>1. This would add 9 votes to the difference between the scores of Blue
>and Orange and therefore maximize the chances of that voter's preferred
>candidate among the two winning, whereas a "sincere" vote would only add
>4 votes to the difference and might result in the less preferred Orange
>winning.
>
>----------------------------------------------------------------
>
>To unsubscribe from this mailing list send an email to
>[EMAIL PROTECTED] with unsubscribe as the subject.
>
>For help with this mailing list, look at
>http://www.neither.org/lists/public-list.htm
>
