Meta election final list.

[David Marsay]

In response to:
From:          Charles Fiterman <cef at geodesic.com>
Subject:       Meta election final list.

I would like to comment on the list, add some criteria, and then 
comment on FPP versus alternatives.


My comments are:

1) that for each criterion one should have some 
guidance on interpretation and evaluation. 

2) most criteria have one variant in terms of how voters vote, and 
another in terms of their sincere preferences.

3) many combinations of criteria are impossible. it may therefore be 
necessary to rank a method by 'how far it goes towards a criterion', 
rather than absolutely. For example criterion met:
  a) for up to n candidates (the bigger n, the better)
  b) for n-D spatial voting
  c) in practice in some stated context.


Proposed criterion:
'Ability to dismiss tyrants':
If a majority rank X last, then X does not win.

(This is like 'the most unpopular candidates lose', but weaker. It 
was Democritus's motivation behind Greek democracy.)

(The next two criteria are based on Locke's desire for people to be 
able to vote for their true first preference.)

Proposed criterion:
'Honest first choices'
Voters never have a tactical reason not to rank 
their favourite candidate first.
(Nice, but often impossible to achieve)

Proposed criterion:
'Votes for minority parties are not wasted'
Suppose that the candidates are grouped into minor and major, with 
every major candidate having more first preferences than all the 
minor candidates together. Then Voters whose favourite is a minor 
candidate never have a tactical reason not to rank their favourite 
candidate first.
(Thus minor parties are not squeezed out all together, and could get 
to challenge the major parties. )

Proposed criterion:
'Moderate parties do not suffer from tactIcal voting (1)'
No voter whose first choice is CW ever has to tactically  vote 
another candidate over CW.
(We cannot avoid some other candidates having to vote tactically.)

(Now for an idea of my own)
Proposed criterion:
'Moderate parties do not suffer from tactical voting (2)'
The only voting tactics are moderate, in the sense that either 
(a) voters promote moderate candidates (e.g., the CW)
or (b) the result is more moderate.

I note that FPP does not meet any of these, while most ranking 
methods meet the first criterion (tyrants) and go a long way towards 
the rest.

It seems to me that the most common faults with FPP occur when there 
are two major candidates (neither with a majority) and a host of 
minor candidates. With FPP the minor candidates's supporters tend to 
vote tactically to make their votes count. 
According to the criteria of 'honesty', 'participation' and 
'accuracy' we need a method that lets them vote 'honestly', thus 
allowing minor parties to build support. Almost any ranking method 
does this, with a PR 'top-up' encouraging participation further.

I think that the case that we should use to choose between methods is 
where there are 3 major candidates, with spatial preferences.
--------------------------------------------------
Sorry, but apparently I have to do this. :-(
The views expressed above are entirely those of the writer
and do not represent the views, policy or understanding of
any other person or official body.