[EM] Some brief campaign argument

Martin Harper mcnh2 at cam.ac.uk
Mon Apr 16 18:47:11 PDT 2001

Here is my major problem with winning-votes methods in general - it's not
terribly subtle. :-)

1) Many voters are lazy, probably the majority. Lazy voters will trade-off
the amount of time they spending thinking about the election and how to
vote, as against the rewards that extra time will bring them. Being lazy
is rational when one considers the instrumental value of ones vote as
opposed to the value of time.

2) Define the "full-ranking" type of vote as being one which expresses a
full, sincere, ranking of all candidate. Note that filling out a
full-ranking requires much thought.

3) Define the "truncated" type of vote as being a full-ranking which has
been truncated at some point. This requires much less thought.

4) Define the "random" type of vote as a truncated vote which has had the
remainder of the candidates filled in at random. This also requires much
less thought, and around 2-3 more seconds in the voting cubicle than

5) In winning-votes, there is a general strategy advantage for voting
"random" rather than "truncated". This will be known, and mentioned in the
media, and is liable to become as well known (and as hated?) as the
"lesser of 2 evils" strategy in Plurality.

6) Lazy voters aren't lazy because they're slobs: they're lazy because
they are rational. As rational voters they will therefore vote "random"
rather than "truncated", because the extra time marking the paper is
*definately* worth it compared to the time already spent. There go all
your strategy garantees.

7) The human mind is notoriously disasterously bad at generating random
numbers. Therefore the "random" votes will not be evenly spread throughout
the probability space, but will be concentrated in certain patterns and
places. Candidates in random positions on the ballot paper, or with random
sounding names, will gain an unfair advantage. There goes your fairness to
all candidates.

Yes, I agree that SFC et al are a nice thing to have - but not at the
expense of encouraging voters to break truncations "randomly". Lazy
voters, I suggest, are *far* more common than strategic truncaters, and
are the more important problem.

More information about the Election-Methods mailing list