[EM] Voter strategising ability

[Juho Laatu]

On 12 Jul 2014, at 22:34, Gervase Lam <gervase at madasafish.com> wrote:

> Given the above, I really find it hard to see a good proportion of
> voters doing the correct strategic calculations.

Yes, _in_typical_public_elections_ voters can reliably implement only some very simple startegies.

> I suppose to conclude all this, I'm just wondering if a voting method
> should handle the situation where voters carry out bad strategy!!?

Widespread bad startegic voting is probably catastrophic in all voting methods. The general rule should be that it makes sense to practically all voters to vote sincerely. (Or alternatively all to follow some easy strategy, as e.g. in Approval.)

> A variation to the above question is, can there be a voting method that
> can handle voting where each faction carries out their own strategic
> voting AND can handle voting where within each faction voters carry out
> their own strategic voting.  The latter may significantly be due to the
> fact the voters are floating voters who don't tow the candidate/party
> line.  Therefore, they would have different opinions about the other
> candidates.

Probably no method can cope with all this.

_In_typical_large_public_elections_, when the voting behaviour of some voter groups is expected to differ from others (i.e. I'm not talking about "same strategy to all" as in Approval), one would probably need a published (by party or media) strategy and large section of voters willing to follow a strategy that is given to them by some external source.

Maybe a good example of a working strategy might be a top two runoff election where the supporters of one candidate will not vote for their own candidate at the first round in order to guarantee that their second choice will make it to the final round. (This strategy can actually work also without public guidance (when voters know their TTR method well enough) and when multiple groups apply the same startegy for different candidates.)

An example of a non working strategy could be Condorcet where burials typically are not simple enough to master, to be working strategies in typical public elections. There are cases where we can see afterwards that some well controlled strategy could have worked, but I have still not seen any generic advices (to the voters or to the strategists that might advice the voters) on when and how people should vote strategically in Condorcet elections. I mean rules that could take any set of polls available before the election, and from those polls determine how certain voter groups should vote strategically, and the end result would be a significantly increased expected outcome in the election (when taking also the negative impacts of strategic voting (bad results and bad reputation) into account). (Also a proof of a catastrophe, after people voting rationally in a strategic way, would do.)

Well, all that was approximate. Are there other better examples of situations where strategic voting would work and improve the expected outcome in a more dramatic way (in typical large public elections), or of situations where they do not work?

> In this current climate, I think most voters would vote sincerely with
> practically any 'reasonable' voting method.

I think that's the general rule. Some simple and quite harmless strategies are possible, an in my TTR example above and as in Approval where all voters may be expected to follow a simple strategy. Note that in both those cases the strategies are not really malicious, since the outcome of the election is expated to be fair even after the strategies have been applied.

(There are also bad voting methods that can encourage voters to follow some strategies and make a mess of the election, but I guess we are talking about normal reasonably well working methods here.)

> Ideally, I think voters should always be voting sincerely as
> strategising (both for good and badly executed strategies) makes things
> complicated.  But Gibbard-Satterthwaite shows this can't happen.

I think Gibbard-Satterthwaite says that in theory there can always be probems. In practice life is luckily somewhat easier.

I think one should aim at voting methods where voters can feel safe when voting sincerely (or when voting according to some simple strategy asn in Approval, although I do not fancy Approval very much). I think there are methods that are good enoug to allow this in practically all typical large public elections (assuming that there are no societies that would e.g. use irrational strategies widely even when there are no rational strategies available).

> Just my stream of consciousness thoughts on all this...

That was a good approach to handling strategy related questions.

Juho