Strategic Voting as "Prisoner's Dilemma"

In game theory there is a well-known scenario commonly known as a prisoner's dilemma. The basic form is as follows (from the Wikipedia entry):

In its "classical" form, the prisoner's dilemma (PD) is presented as follows:

Two suspects are arrested by the police. The police have insufficient evidence for a conviction, and, having separated both prisoners, visit each of them to offer the same deal. If one testifies ("defects") for the prosecution against the other and the other remains silent, the betrayer goes free and the silent accomplice receives the full 10-year sentence. If both remain silent, both prisoners are sentenced to only six months in jail for a minor charge. If each betrays the other, each receives a five-year sentence. Each prisoner must choose to betray the other or to remain silent. Each one is assured that the other would not know about the betrayal before the end of the investigation. How should the prisoners act?

Strategic voting in the federal election shares some interesting similarities. As an example I'll look at the riding I live in, Ottawa West-Nepean. The two highest-polling contenders are Baird (Conservative) and Pratt (Liberal), with Baird in the lead. Far lower in the standings are Rivier (NDP) and Coates (Green). Now let's say I have a fervent wish to turf Baird from office and that I have leanings toward the Green Party (irrespective of the local candidate). Here are some conventional voting strategies that can be employed in the present instance:

• Vote for Coates: Call this a principle-based vote. However I strongly suspect she will get few votes and therefore my vote for her will not only not get her elected, it will leave Baird with his margin intact over Pratt. That is, my vote Coates leaves the contest between Baird and Pratt unchanged.
• Vote for Baird: Call this the unprincipled vote. Believing he'll be elected (based on the polls) I want to be on the winning side, even though I don't want him elected. This tactic is common in many places across the country in the belief that more benefits (pork) will come their way.
• Vote for Pratt: Call this the strategic vote: I prefer neither Baird nor Pratt, but by voting this way I increase the chance of defeating Baird since Pratt polls in second place by a wide margin.

Notice that in none of these approaches did I mention who I feel would be the best local representative. That is, the candidate who would best represent my local, not federal, interests. Typically there isn't often much to decide on that way since most elected MPs are only moderately responsive to local matters. Therefore I'll narrow the decision criteria to federal issues, and which are nearly always party aligned (enforced by the party whip).

To vote on principle (Coates) with any optimism requires that I not only believe that many other voters are also aligned with Greens on principle and that they will vote on principle. Like in the prisoner's dilemma, a mutually-beneficial outcome requires cooperation among non-communicating voters.

Unfortunately the election is far more complex a scenario: there are tens of thousands of potential 'prisoners' and there are several choices. There are also other competing principles such as the one for defeating Baird, which similarly requires voter cooperation. Not only that, many voters don't realize they're in a game of prisoner's dilemma and so do not consciously cooperate or defect.

I suspect that anyone in this position who looks at the problem this way will in the end throw up their hands in frustration and just vote for or against Baird. Therefore the game of prisoner's dilemma is lost, and the poll results hold through to the election itself and Baird is re-elected. Now it is true there are mathematically and computationally-proven optimum strategies for prisoner's dilemmas, but, unfortunately, they are entirely unreliable in the above election scenario; any one voter is unable to alter the momentum toward game loss solely using the ballot box. Regardless, some do try to think their way through it.

Note: In no way does any of the above truly represent my own voting intention. I simply picked a local and perhaps not so hypothetical example to illustrate the problems of strategic voting.