I think the best way to describe it is 50% of all votes cast *after transfers*. If your ballot becomes exhausted because all your listed preferences are eliminated, then you are not transferred, you don't cast a vote in subsequent rounds, and you aren't part of the group of which the winner needs half.
AV supporters have IMO misled on this point - in particular exaggerating the importance that it would have even if it were true. OK, so an AV winner does have over 50% of the vote, after discounting all the people who didn't vote at all, and all the people who didn't express an opinion between the winner and the second-placed candidate.
Of course we think, "so what?". At the point where I'm expressing a preference for the BNP over the National Front (or vice-versa) on the basis that I think that one of them is very slightly less a bunch of racists and criminals than the other, I don't in any sense *support* the winner, and if pro-AV people are going to interpret that preference as "support", then I'm hardly going to want to express a preference. I suppose I'll just take whatever racists I'm served.
Anyway, it's actually expected that under an AV system, many or most voters don't need to express a second preference, because elections are reasonably predictable. Once you've listed either of the candidates who (eventually) makes the final round, there's no point listing anyone below them, because there is no possibility that your ballot will go any further than that candidate. I expect that in many constituencies it will be pretty clear who the last two candidates will be. If you don't know who the last two will be (and after all elections are only *reasonably* predictable, not entirely predictable), or if you want to give an ultimately futile show of support to someone else, then you list two or more preferences.
Consider the London mayoral election. Conducted with a special dumbed-down AV system three times, but fundamentally if you wanted to have any influence on who won, literally all that mattered on the day was how you relatively ranked Boris and Ken. Most of the first-choice votes went to one of the two of them, suggesting that most London voters don't need AV on their own account. They might want in future want to express a first choice for a candidate expected to come third or worse, of course.
Last time out, 83% of voters used their second preference. I can't predict whether the ability to express a third preference, or general antipathy toward AV, would depress this. Or whether it was inflated in the first place by things like the Lab-Green second preference pact, a perception unique that the supplementary preference (uniquely) is an important "show of support" for someone you don't really want to win, or perhaps even by Londoners not really understanding the system.
![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png)
timeasmymeasure
no subject
Date: 2011-04-18 09:27 pm (UTC)AV supporters have IMO misled on this point - in particular exaggerating the importance that it would have even if it were true. OK, so an AV winner does have over 50% of the vote, after discounting all the people who didn't vote at all, and all the people who didn't express an opinion between the winner and the second-placed candidate.
Of course we think, "so what?". At the point where I'm expressing a preference for the BNP over the National Front (or vice-versa) on the basis that I think that one of them is very slightly less a bunch of racists and criminals than the other, I don't in any sense *support* the winner, and if pro-AV people are going to interpret that preference as "support", then I'm hardly going to want to express a preference. I suppose I'll just take whatever racists I'm served.
Anyway, it's actually expected that under an AV system, many or most voters don't need to express a second preference, because elections are reasonably predictable. Once you've listed either of the candidates who (eventually) makes the final round, there's no point listing anyone below them, because there is no possibility that your ballot will go any further than that candidate. I expect that in many constituencies it will be pretty clear who the last two candidates will be. If you don't know who the last two will be (and after all elections are only *reasonably* predictable, not entirely predictable), or if you want to give an ultimately futile show of support to someone else, then you list two or more preferences.
Consider the London mayoral election. Conducted with a special dumbed-down AV system three times, but fundamentally if you wanted to have any influence on who won, literally all that mattered on the day was how you relatively ranked Boris and Ken. Most of the first-choice votes went to one of the two of them, suggesting that most London voters don't need AV on their own account. They might want in future want to express a first choice for a candidate expected to come third or worse, of course.
Last time out, 83% of voters used their second preference. I can't predict whether the ability to express a third preference, or general antipathy toward AV, would depress this. Or whether it was inflated in the first place by things like the Lab-Green second preference pact, a perception unique that the supplementary preference (uniquely) is an important "show of support" for someone you don't really want to win, or perhaps even by Londoners not really understanding the system.