16

About that counting system


 

The biggest single knock against STV, the one that the critics have had the most fun with, is the elaborate system for counting the ballots — the basis for complaints that the system is too complicated for voters to understand.

It’s not that complicated. You count up first choices. You eliminate the last place candidate, and redistribute his or her votes according to their second choices. You do the same thing on the second ballot, and the third, until you’ve elected the required number of members. It’s actually quite familiar, to anyone who’s watched a party leadership race. Only instead of holding multiple ballots to find out people’s second and third choices, it’s all done in one.

Is it only the losers’ votes that get redistributed? No: that’s the first wrinkle. You also redistribute the votes of candidates that have been elected (since we’re electing more than one in each riding): that is, once they pass the threshold number of votes needed to make it a mathematical certainty they’ll be among the winners, like a hockey team that’s clinched a playoff spot.

How do we know where that threshold is? Simple. In a one-member riding, it would obviously be 50% plus 1: with that many votes, there’s no way that anyone could finish ahead of you. In a two member riding, it would be 33% plus 1: again, there’s no way two other candidates could both have more votes than you. In three member riding, it’s 25% plus one, and so on.

Whatever number of members a riding elects, the threshold for election is the number of votes divided by one more than the number of members. Once a candidate crosses that threshold, he’s declared elected, and his votes are redistributed.

All of them? No: only his surplus votes — the number of votes in excess of the threshold. That’s the second wrinkle. Before redistributing his votes among the remaining candidate, they’re weighted by the proportion that is considered surplus — if the surplus is a tenth of the total, for example, they would only count a tenth as much.

And that’s it. That’s the Big Complexity the critics complain of: you redistribute votes from candidates as they are either eliminated or elected to those still in the race.

But of course, all the voters need to know is how to count to five. The precise intricacies of the counting system would be of more concern if this were the first time this system had ever been tried. But as it is already in use in many jurisdictions around the world, it rather puts to rest any fear that this is some sort of trick — that there is some nasty surprise lurking in the fine print.

Anyway, the Irish seem able to stumble through it somehow.


 

About that counting system

  1. Andrew, I'm really glad to see you writing so persuasively and clearly on this timely topic!

    I don't think I would ever have compared you to John Cleese before this….but now I can….

    http://www.youtube.com/watch?v=NSUKMa1cYHk

  2. Didn't Andrew post that a long while ago?

  3. Not to sound like an an STV-hater, because I am not (although I forgive anyone who reaches that conclusion based on my commentary the last few days)…

    What happens to the rejected-ballot rate when STV happens? Do you write your number in a circle (Are Roman Numerals OK? What about A, B, C…), or are there 5 columns of boxes per name, and you check off a single number one, a single number two…

    The Chief Electoral Officer in Quebec was not terribly curious about some spectacularly discrepant rejected ballot rates the last time Quebec held a referendum. And I will charitably leave it at that. Will this system permit more or less of a certain form of disenfranchising shenanigans?

  4. On my facebook page, I think. There's also a link to it in the post.

  5. Sorry for double posting then….as much as I love embedded posts which have easter egg qualities…I tend only follow them and figure out what they are if they confuse me.

    Yeah, John Cleese and PR viralled around facebook a few days ago (and Andrew's long article did too, even among my usually pretty lefty Green friends).

  6. Well. If the Irish can do it ……….

    Actually, in my wildest dreams I could never imagine that there would be anything emerging from the fevered brow of Mr. Coyne that I could ever agree with. But there it is.

    Another sleepless night.

  7. Andrew, I’m really glad to see you writing so persuasively and clearly on this timely topic!
    I don’t think I would ever have compared you to John Cleese before this….but now I can….
    http://www.youtube.com/watch?v=NSUKMa1cYHk

  8. Didn’t Andrew post that a long while ago?

    • On my facebook page, I think. There’s also a link to it in the post.

      • Sorry for double posting then….as much as I love embedded posts which have easter egg qualities…I tend only follow them and figure out what they are if they confuse me.

        Yeah, John Cleese and PR viralled around facebook a few days ago (and Andrew’s long article did too, even among my usually pretty lefty Green friends).

  9. Not to sound like an an STV-hater, because I am not (although I forgive anyone who reaches that conclusion based on my commentary the last few days)…

    What happens to the rejected-ballot rate when STV happens? Do you write your number in a circle (Are Roman Numerals OK? What about A, B, C…), or are there 5 columns of boxes per name, and you check off a single number one, a single number two…

    The Chief Electoral Officer in Quebec was not terribly curious about some spectacularly discrepant rejected ballot rates the last time Quebec held a referendum. And I will charitably leave it at that. Will this system permit more or less of a certain form of disenfranchising shenanigans?

    • Again, a hypothetical flaw has been identifed that simply does not actually occur in the jurisdictions where STV is in use. In the 2007 Irish election, only 0.1% of its ballots were spoiled.

  10. Well. If the Irish can do it ……….

    Actually, in my wildest dreams I could never imagine that there would be anything emerging from the fevered brow of Mr. Coyne that I could ever agree with. But there it is.
    Another sleepless night.

  11. Again, a hypothetical flaw has been identifed that simply does not actually occur in the jurisdictions where STV is in use. In the 2007 Irish election, only 0.1% of its ballots were spoiled.

  12. What a sad day to be a British Columbian! Thoroughly pissed by the results, but what could you expect? Worst case of communications I've seen in ages… starting from the god-awful naming of it to all of the marketing materials that focused almost exclusively on the complexities of the system instead of the benefits, with the net result being CONFUSION… even for people who 'more-or-less' understood it.

    And I think the NDP shares some of the blame. They should have really championed STV as a central part of their platform. If every Green and NDP vote were also a vote for STV, maybe we wouldn't have another four years of Gordo's raping and pillaging.

    Hopefully a more visionary province will take the lead because we blew it!

  13. What a sad day to be a British Columbian! Thoroughly pissed by the results, but what could you expect? Worst case of communications I’ve seen in ages… starting from the god-awful naming of it to all of the marketing materials that focused almost exclusively on the complexities of the system instead of the benefits, with the net result being CONFUSION… even for people who ‘more-or-less’ understood it.

    And I think the NDP shares some of the blame. They should have really championed STV as a central part of their platform. If every Green and NDP vote were also a vote for STV, maybe we wouldn’t have another four years of Gordo’s raping and pillaging.

    Hopefully a more visionary province will take the lead because we blew it!

Sign in to comment.