[EM] Voter strategising ability

[Juho Laatu]

On 20 Jul 2014, at 22:48, Kristofer Munsterhjelm <km_elmet at t-online.de> wrote:

> Discussion about which kind of strategy is most likely to happen can go on forever without data. Even if there is data, it is quite easy and/or tempting to explain it away as not being representative of what would happen under an ordinary election. As long as that's possible, it's really hard to convince someone who is worried about burial not to be, or vice versa.

Unfortunately we don't have data from very many Condorcet elections. And those elections have been quite non-competitive. So we don't know very well what would happen (in different societies) in competitive Condorcet elections.

I however note also the absence of practical strategic voting advices for Condorcet methods. We have some rules of thumb for Approval voters, but I'm not aware of any for Condorcet voters. There are thus no generic rules "if ... then modify your vote ...". There are also no generic rules for parties or media, like "if ... then recommend all voters whose preferences are ... to modify their vote ...". I also have not seen any recommendations like "in that rea life Condorcet election voters could have improved the outcome of the election by using strategy: if ... then ...". That would of course also be insufficient since strategies should be applied before the election, not after the election. Also in EM list most used examples are of form "if all members of this group of voters would have modified their vote so that ...".

What would be a practical strategy rule would be one that takes few uncertain polls, one and/or two weeks before the election as input and then tell how to modify one's vote. That strategy could be either one that ordinary individual voters can follow by themselves, or one that some party or media strategists can publicly recommend for them to implement.

My conclusion is that until I see such recommendations, we are lacking evidence also on that side. (This also means that a theoretical vulnerability is not a practical vulnerability until proven so.)

My guess is that many societies (those ones that are not very eager to use whatever rational and irrational strategies, or maybe even dislike strategic plots) could use Condorcet methods with no problems with strategic voting (it would be just some background noise).

Juho