I've just had my attention drawn to a very interesting article (linked above) about Tactical Voting and how it might work under AV. This blog entry is based on my response to it.
The article is a good one, in that it makes mathematical sense and is interesting to read. However the scenarios described are making a crucial assumption, that plenty of voters have a very good idea of what everyone else is going to do.
Even under FPTP where tactical voting is simply a matter of looking at the two most likely winners and choosing between them, denying your first choice any vote from you at all (which is what makes FPTP such a terrible system for more than two parties), some second guessing of how your fellow constituents are likely to vote is required.
But with the complexity of the AV situation there are far too many variables for most voters to consider and I think it would be too much for most voters, especially with AV being a new and unfamiliar system to most, to take the counter-intuitive action of putting the most disliked candidate first.
Most importantly, this situation can only really happen in a tight battle between 3 parties, and for this to be of use to aiding anyone to make a choice between FPTP and AV you have to take this type of scenario and examine it fully from the perspective of both systems.
With the uncertainty brought about by the complexities of AV and the variations from constituency to constituency, let's say there's approximately a 1/3 chance that putting your own party 1st instead of 2nd will lead to them losing when they might otherwise have won. But consider that under FPTP you don't even have that choice. You either put your party 1st, or not at all. If you put them 1st, in a tight 3-way marginal you're averaging a 1/3 chance of your party winning.
So to summarise that, 1/3 chance of losing under AV or 1/3 chance of winning under FPTP. It rather points to AV as a better option I think.
But if you really want to keep a party out under AV, then all you have to do is put them last. If everyone who wants to keep a party out puts that party last, then the remaining support for that party will either be 50% or it won't. If it is, then they've won fairly, and if it isn't then they won't win, and that's all there is to it.
OK, guilty as charged, that last paragraph is also built on an assumption. But my assumption is that most people will vote based on their preferences, which has to rank higher in likelihood than the assumption that plenty of voters will have an excellent idea of how everyone else is likely to vote.
I must admit it is fascinating to look at the voting mechanics and how you can tweak situations to show various surprising outcomes. But the reality is that everyone votes without the information required about everyone else's vote. I think the sensible option really is to stop umming and ahhing about what other people might do and just go with your preferences. Most people will do that anyway. It's really too much to expect voters to stop and think about the various permutations, even if they did have an idea somehow of how everyone else would be ranking everyone.
And at least AV lets you do that if you want to. FPTP often gives you a miserable choice between a tactical vote and a wasted vote. For people to have to deny their preferred candidate their vote just to be able to feel that they're using their vote wisely is undemocratic and has no place in our politics.
Tactical voting in FPTP is so prevalent because it's one of the simplest concepts to grasp even for the least politically minded of voters. The more complicated the tactics are in a voting system the fewer people will vote tactically. It's easy enough for a party to tell the would-be voters of another party to vote for them instead because of the way FPTP tactics work. At a push a party could hand out "how to vote" cards, outlining how to assign all the preferences from 2nd onwards under AV, not that they'd be that likely to succeed. However, even with the best research and the finest mathematicians in the land combined with world class marketing strategists and on-the-ground campaigners, I couldn't see any party successfully persuading just the right proportion of their voters to put the opposing party first and their own party second, if any at all.
And even if they could, if all parties were doing that then they'd cancel each other out with second guessing, third guessing, fourth guessing, and beyond. It would turn into a complete lottery with politicians and voters alike saying "Forget it, just go with your preferences, it's too complicated for tactical voting to leave us anything other than a random outcome".
And this I believe is a real strength for AV. Instead of being afraid of complex tactics leading to unpredictable results, I see it as complex tactics discouraging voters from straying from their true preferences leading to more democratic results. With AV you can keep it simple, put your first choice first, put the most disliked candidate last, rank everyone else in between based on your preference and let democracy do the rest.
No comments:
Post a Comment