Legal Beagle: Voting in an STV election
115 Responses
First ←Older Page 1 2 3 4 5 Newer→ Last
-
Tim McKenzie, in reply to
Slightly too late to edit my post:
I've just noticed that, according to Wikipedia, The Condorcet criterion is incompatible with later-no-harm. If that's right, then I guess we can't both be satisfied, but I'm still not persuaded that later-no-harm is more important than having majority support for the winner, wherever possible. And in multiple-winner STV elections, later-no-harm seems likely to harm voters' preferences over the sets of possible winners.
-
Graeme Edgeler, in reply to
And in multiple-winner STV elections, later-no-harm seems likely to harm voters’ preferences over the sets of possible winners.
Tactical voting is possible in all fair electoral systems with more than two candidates.
-
Some more reading material here: https://dunedinstadium.wordpress.com/2010/08/26/in-defence-of-stv-2/ just after the post, and before the 34 responses.
-
Steve Todd, in reply to
Tim—
We don’t want single-seat STV (M-PV; AV; IRV; etc.) to satisfy the Condorcet criterion, because, “The Condorcet criterion is … incompatible with the later-no-harm criterion”. (I've just noticed you've seen that (on page 2), but here goes, anyway.)
The feature of STV, that later preferences cannot harm earlier preferences, is what gives the voters the confidence to rank-order the candidates in their true and genuine order of preference. By doing so, they can be satisfied that the outcome represents their true, collective, wish – the correct candidate has been elected.
If voters were to be told, before they vote, that the votes will be counted in such a way that it is quite possible that their second preference for B might well help to defeat their first preference for A, they will likely not express second and later preferences, and we’ll be more or less back to the unlamented FPP again. They would certainly have no confidence in such a system, that’s for sure.
In almost all public elections by majority-preferential voting, the winner *is* the Condorcet winner. Does anyone doubt that Kerry Prendergast was the Condorcet winner in 2004 and 2007? Of course not.
What about The Great Cliff-Hanger Election of 2010? Celia won the head-to-head over Kerry, so we know that Kerry wasn’t the Condorcet winner. Does anyone seriously think Jack Yan, who received 5,817 first-preferences out 53,369 votes cast, might actually have beaten Celia in a head-to-count? Of course not.
What about 2013? We know John Morrison was not the Condorcet winner, because Celia beat him in a head-to-head count, too. Again, what about Jack Yan? This time, he received 8,105 first-preferences out of 56,254 votes cast. Would he really have beaten Celia in a head-to-head count? Again, of course not.
In the run-up to the first Flag referendum, Eric Crampton, Head of Research at The New Zealand Initiative, was hoping the result would be close. He wanted to get his hands on the preference data in the hope of showing that the Condorcet winner did not win. (I had him on about that at the time, in a lead letter to the Editor of the Dominion Post.)
Once the results were in, he flagged that idea. Black Lockwood beat Red Lockwood head-to-head and was the clear Condorcet winner. Does anyone seriously think Red Peak, which received 122,152 first-preferences out of 1,393,615 valid votes cast, would have beaten Black Lockwood in a head-to-head count? Of course not.
Giving up later-no-harm is too high a price to pay to ensure that, in those very few public elections where the Condorcet winner would not otherwise be elected, the Condorcet winners gets to be elected anyway.
As far as I’m aware, the methods that comply with the Condorcet criterion, listed at the link you gave, are not used in public elections anywhere in the world. No doubt that is because they all violate later-no-harm.
So, Rich’s contention was very far from being dubious, as you put it. Quite the opposite, in fact.
-
Steve Todd, in reply to
My understanding is that with single-member electorates it is not proportional at all, and with multi-member it is somewhat locally proportional.
Moz, we call all STV elections in NZ, whether single- or multi-seat, Single Transferable Voting, rather than M-PV and STV, respectively. That is because, back in 2000, DIA officials thought the distinction would confuse the public. I say no more, other than to say, I was angry.
-
Steve Todd, in reply to
Paul, I'm responding to your post where you say you're an STV fan.
See my four responses here: https://dunedinstadium.wordpress.com/2010/10/09/local-govt-online-for-results/#comments starting with the one at 10.37 a.m. on October 21 [2010].
Both Bev Butler and yourself have nothing to worry about. Very briefly, a 43-vote difference between candidates, in an NZ STV (Meek's method) election, where 32,820 votes were cast, is huge. That is not in any way close, as it might be in an FPP election.
I have seen how electionz.com process the votes, and it is practically error-proof. When the votes are scanned onto the screens, the DEOs go down the list and type in the number beside each hand-written preference. If there's any doubt (say, whether a number is a 1 or a 7, or a 5 or a 6, etc.), the electoral officer, or perhaps someone acting in that capacity (probably Warwick Lampp or Steve Kilpatrick) is called over and a decision is made.
It is not possible for so many mistakes to have been made, e.g., Fliss Butcher receiving preferences that Bev Butler should have got, that Butler was unsuccessful because of them.
I tried to get DIA officials to recommend that the Regulations be amended so that full preference data could be made publicly available, so people could run the data using their own STV programs, to confirm the results (also necessitating the public availability of the source code), but they weren't interested.
-
Steve Todd, in reply to
But it’s a judgment call as to what particular criteria you want a voting system to meet. It cannot meet them all.
Agreed, but it's hard to imagine an informed electorate giving up later-no-harm in preference to guaranteeing the election of the Condorcet candidate 100% of the time instead of 99.9% of the time.
I really would like to dispose of this ‘Condorcet criterion’ nonsense, once and for all (I hope).
If a system is devised that ensures that “everyone’s second choice candidate” (otherwise known as the Condorcet candidate, or winner) will be elected, then that system will enable the ‘later preferences cannot harm earlier preferences’ rule to be violated. If given an informed choice, most people would not want their lower choices to have an influence in an election while their first choice remains a contender.
In an election, the Condorcet candidate (if there is one) is the candidate who, when compared two by two with all other candidates, is preferred to each other candidate by more than half the voters. Whereas there would almost never be a candidate who was “everyone’s second choice”, there is often a Condorcet candidate, being a candidate who is preferred to all other candidates by a majority of voters. (All four Wellington mayoral elections under STV have seen the Condorcet candidate elected.)
At first glance, it may seem reasonable for people to want to discard an election method that does not elect the Condorcet candidate (if there is one), but consider the following highly artificial example where there are five candidates and votes—
101 AE.. 100 BE.. 99 CE.. 98 DE.. 10 E..
E is certainly the Condorcet candidate and, in spite of a poor showing on first preferences, really a very strong candidate in that if any one of A, B, C, or D were to withdraw before the count (for a single seat), E would win without question.
Can it really be said, though, that E deserves to win this election simply because he or she is “everyone’s second choice” - the Condorcet candidate? If we ignore the second preferences and treat this election as an FPP election, we would still not know that E was “everyone’s second choice”, and A would win with 101/408 votes (less than 25%). Being a PV election, however (in which E is excluded first), the winner would at least be elected with an absolute majority of the votes remaining in the election (101/201 votes).
What if this election were for four seats? Surely it would be absurd not to elect A, B, C and D. Regardless of how many seats there are, if the election method were such that second choices could be considered while the first choice was still a contender, possibly leading to the defeat of the first choice, then, as I said upthread, the voters would not indicate second and subsequent choices, and we would effectively be back to FPP again.
The question has to be asked, would voters want an election method that enables a candidate with as few as, say, 10% of first preference votes, to win an election because lower choices were able to assist a second candidate defeat a third candidate and the lower choices of the third candidate then helped the second candidate to defeat their first choice candidate? I would venture to say “No”.
Most Condorcet candidates are, in fact, generally popular candidates in that they will have attracted a range of preferences, including a healthy share of first preferences. This means they are not likely to be excluded early on, and therefore will remain in a position to receive many of the ‘second (and subsequent) choices’ of those other candidates that are excluded from the election. Most Condorcet candidates do, in fact, come through to win the election, although, as the above extreme example shows, it is not guaranteed, as you have pointed out.
In my view, it is really only in small, private, elections (or, of course, in artificially-constructed examples) that Condorcet candidates fail to be elected. In large, public, elections, they would almost always be the successful candidate. It really isn’t a problem, well except for public choice theorists (usually situated somewhere in US academia), and perhaps for Tim.
As you say, no election method is perfect, but M-PV and STV are far superior to the FPP systems most local authorities in New Zealand use; systems which make no effort to ensure, as STV does, that as many voters as possible are fairly and equally represented on their local councils.
-
Paul Campbell, in reply to
Both Bev Butler and yourself have nothing to worry about. Very briefly, a 43-vote difference between candidates, in an NZ STV (Meek’s method) election, where 32,820 votes were cast, is huge. That is not in any way close, as it might be in an FPP election.
I think it's really smaller than that as it only requires half that many mistakes to turn it around.
In general though I really think that votes should be counted, in public, in the city where they are cast, sending them elsewhere to be counted seems to me to be just plain wrong.
I tried to get DIA officials to recommend that the Regulations be amended so that full preference data could be made publicly available, so people could run the data using their own STV programs, to confirm the results (also necessitating the public availability of the source code), but they weren’t interested.
I agree, they should do this, appropriately anonymised
-
Steve Todd, in reply to
Tim, a voter may well want both A and B to win, but that is not the point. Under STV, every voter has one vote, which they express by giving their first preference to the candidate *for whom they vote*.
Second and subsequent preferences are contingency choices only, whereby the voter is saying (to the electoral officer / STV calculator), if my first preference candidate does not need all of my vote, or is excluded from the count, I want my vote to be transferred, in whole or in part, as the case may be, to my second preference candidate, to help elect that person, and so on.
So yes, STV does place a "heavy weight on your higher preferences", because that is what it is supposed to do - it is, after all, a single-vote system, not a multi-vote system.
See, at my first posting, above, page 3 of the paper at the third link, to see how a vote is used in an NZ STV election. A multi-candidate example can be seen here:
https://dunedinstadium.wordpress.com/2010/08/22/why-not-the-stv-voting-sytem/
Go to my reply to 'ro', at September 22 at 7.54 a.m., then click on the vote_example link at the conclusion of my comments. You will see how the amount of the vote that is kept by each candidate falls away very quickly, exactly as it should.
-
Steve Todd, in reply to
I think it’s really smaller than that as it only requires half that many mistakes to turn it around.
No, by the end of the count, those 43 votes (approx.) will be the difference between the 2,535.817272399 votes (across several thousand voting documents) for Colin Weatherall, and the 2,492.840680897 votes (also across several thousand voting documents) for Bev Butler.
Colin and Bev received 1,236 and 1,234 first preferences, respectively. The additional 1,300 votes and 1,258 votes, respectively, they each received, will be made up of tiny bits of votes, on perhaps as many as three or four thousand voting documents. There would have to be vote-entry errors on hundreds, perhaps over a thousand, voting documents, for those little bits to affect the count, through 58 iterations, in such a way as to elevate Butler above Weatherall. It's not just a matter of there being 22 errors in entering first-preferences for Butcher that should have gone to Butler.
But think about it. Would the DEOs give lots of Butler's '1's to Butcher, immediately above her on the voting documents? How would that happen? As previously explained, the DEOs carefully insert, beside the hand-written number, what that number is. It is not possible to see a hand-written '1' beside Butler's name, but insert a '1' beside Butcher's name. The DEO has moved on past Butcher.
Then, of course, if so many mistakes were made, affecting two candidates consecutively listed on the voting documents, then there would have been many other entry errors in respect of other candidates, too. But there weren't.
The computer technology needed to process votes electronically probably precludes every STV council having their own facilities. Those days are gone. electionz.com is doing Wellington's votes, too. I'm sure we will be posting our votes to Christchurch, as well. With OCR technology, the role of scrutineers is basically done away with. They were needed when the votes were literally hand-counted, and even when they were 'wanded' into the computer, but not now.
-
Steve Todd, in reply to
And it appears to limit the ability to change the alignment of councils - if there's a Green, Labour and National candidate with name recognition and good campaigns, they'll all get elected even though the Green might be on 40% and the Nat on 25%.
Rich--
Have a look at my response (and two replies) to Cr Simon Woolf, here:
https://www.facebook.com/OnslowWestern/posts/1069826773085165:0
My response is immediately under his post.
The problem you describe is known as electoral stasis. The answer, in my view, is to elect our council citywide (at-large).
-
Paul Campbell, in reply to
Second and subsequent preferences are contingency choices only, whereby the voter is saying (to the electoral officer / STV calculator), if my first preference candidate does not need all of my vote, or is excluded from the count, I want my vote to be transferred, in whole or in part, as the case may be, to my second preference candidate, to help elect that person, and so on.
Here I have to disagree – your second and subsequent preferences are more than contingency choices … for example if you and your neighbours give candidate A twice as many votes as they need to be elected then a full 1/2 of your vote is still live and will pass to your second choice, and will keep being passed down your preferences until it is used up …
In fact because the chances of a candidate getting exactly exactly the perfect amount to get elected it’s quite possible that tiny little bits of your single vote will get distributed all the way down your preference list (missing those candidates that get kicked out for being lowest) modulo the precision that the algorithm mandates.
-
It's interesting that STV is still used in so few local body elections, and only where strong centres of MMP electoral reform acivism were before 1993. As others have noted, one drawback is the perceived logistical challenge of calculating the flow of preferences and the end result. Is another the lack of substantive civil authority that New Zealand local government actually has, apart from the Auckland Council? After all, we are a fairly centralised nation, especially compared to Australia.
-
Tim McKenzie, in reply to
Tactical voting is possible in all fair electoral systems with more than two candidates.
This is true, but it only makes the later-no-harm criterion seem even more peculiar. Later-no-harm insists that a particular kind of tactical voting is unnecessary and impossible. But why is it more important to rule out that kind of tactical voting, while ignoring all the other possible kinds? And is it worth it if it means that a voter's preference between their two favourite candidates is always treated as more important than their preference for their second-favourite candidate over their least-favourite candidate? Why not insist instead that the system should ensure that a voter's best strategy (or at least one of their best-equal strategies) always involves ranking their favourite candidate first? STV certainly doesn't have that property, and I wouldn't be surprised if it was incompatible with later-no-harm.
Steve Todd:
consider the following highly artificial example where there are five candidates and votes—
101 AE.. 100 BE.. 99 CE.. 98 DE.. 10 E..
I really don't think this example helps your case. Under STV, one of A, B, C, or D will win (depending on whether, and which, subsequent preferences are expressed). Without much loss of generality, suppose A is the winner. Do you really want to insist that this would be the correct result, even though more than three quarters of all the voters would have preferred E to win? Would B's supporters really be saying to themselves "Oh, well; at least my preference for E over A didn't harm B's chances of winning"? If pre-election polls indicated that this was likely to happen, would the supporters of B, C, and D really have had "the confidence to rank-order the candidates in their true and genuine order of preference", or would they have decided to tactically give their first preferences to E, in order to deny A the unpopular victory over E?
Paul Campbell:
it’s quite possible that tiny little bits of your single vote will get distributed all the way down your preference list
Only if the eventual first runner-up is your lowest preference (or you didn't give them a ranking at all). Because the eventual first runner-up is neither elected nor excluded, preferences for them are never passed on to other candidates. Once, back when I was more enthusiastic about casting as much of a vote as possible, I ranked all two-dozen or so candidates in a health board election. My first preference was, in the end, the first runner-up, so all my work of deciding how to rank the remaining candidates was wasted. (And then, from memory, she was appointed to join the board anyway.)
-
Steve Todd, in reply to
Here I have to disagree – your second and subsequent preferences are more than contingency choices … for example if you and your neighbours give candidate A twice as many votes as they need to be elected then a full 1/2 of your vote is still live and will pass to your second choice, and will keep being passed down your preferences until it is used up …
Oh, Paul, that is basically the start of the “official” description of how you vote in an STV election. They most definitely are contingency choices.
What you are describing, is the effect the vote has on the election, at the end of an NZ STV count. At the beginning, when people are actually voting, they (theoretically, at least) say to themselves the passage that you disagree with.
If you look at the extended example I linked to in my reply to Tim, you will see (in column 5) that the first-preference candidate (A) kept none of the vote, the second-preference candidate (B) kept 47.72% of it, and the third-preference candidate (C) kept 36.73% of it, and on down the list.
When this vote was cast, the voter was voting for A. At the end of the count, it can be seen that with A having been excluded, the vote transferred to elected candidate B, the surplus portion of which passed over excluded candidate C and landed upon elected candidate D, who only needed 70.26% of it.
So, yes, the vote substantially contributed to the election of two candidates, and contributed in a minor way to the election of several others. But that is just a happy consequence of the way the votes in this election were cast (and the fact that they were counted by the far more efficient NZ STV method). It was not the voter’s original hope that it might, or intention that it would, be used in this way.
In this regard, each individual voter has no way of knowing how his or her “neighbours” are going to vote. All he or she knows is that s/he’s voting for A, with the expectation that if A doesn’t need all of it, or is excluded from the count, it will be transferred, in whole or in part, to B, and so on, until it hits a candidate whose (final) keep value is 1.0.
-
Steve Todd, in reply to
Later-no-harm insists that a particular kind of tactical voting is unnecessary and impossible.
Does it. Why not explain how?
But why is it more important to rule out that kind of tactical voting, while ignoring all the other possible kinds?
As previously stated, it gives voters the confidence to rank-order the candidates in their true order of preference. When that is done, the outcome represents the true, collective, wishes of the voters, when taken as a whole. Surely, that is a good thing.
And is it worth it if it means that a voter’s preference between their two favourite candidates is always treated as more important than their preference for their second-favourite candidate over their least-favourite candidate?
Yes it is. Why would it not be?
Why not insist instead that the system should ensure that a voter’s best strategy (or at least one of their best-equal strategies) always involves ranking their favourite candidate first? STV certainly doesn’t have that property, and I wouldn’t be surprised if it was incompatible with later-no-harm.
That is exactly what STV does. As there is no incentive for voters not to vote their true preferences, it logically follows that STV most certainly must have the ‘rank-the-most-favoured-candidate-first’ property.
What a nonsensical statement. What proof do have that STV “certainly doesn’t have that property?”
-
Steve Todd, in reply to
Do you really want to insist that this would be the correct result, even though more than three quarters of all the voters would have preferred E to win?
Yes, I do want to insist that. If more than three-quarters of voters preferred E to win, then why didn’t they vote for E. The simple fact is only 10 voters wanted E to win, the other 398 didn’t. They all voted for candidates other than E – “You have one vote, which you express by giving your first preference to the candidate *for whom you vote*. The argument I laid out, stands. Second and later preferences are contingency choices only, etc., etc.
I put up that voting pattern to show that, even though E is the Condorcet candidate, s/he cannot realistically, in a public election, be regarded as the correct winner. Yet, here you are, effectively arguing that s/he is. Astonishing!
If pre-election polls indicated that this was likely to happen, would the supporters of B, C, and D really have had “the confidence to rank-order the candidates in their true and genuine order of preference”, or would they have decided to tactically give their first preferences to E, in order to deny A the unpopular victory over E?
And here we get to the nub of the issue. People like you (e.g. Doron and Kronick, et al) argue your case, using carefully constructed artificial voting patterns. You argue points that would never be in the minds of voters when they vote. You point out so-called anomalies, that no voter, not having knowledge of precisely how other people are going to vote, could never take advantage of in real, public elections.
The polling you refer to, above, is “artificial”. Polling in STV jurisdictions has never unearthed a candidate that was everyone’s second choice, because they don’t exist in real life, and never will. Had such a candidate ever been identified, the voters in the jurisdiction concerned would certainly have been told about it.
No-one is ever everyone’s second choice, just as no-one is ever everyone’s first choice, or third, or fourth choice. In the real world, voters just aren’t like that.
The best we can hope for is two-candidate-preferred (2CP) polling, such as that which I suggest here: http://wellington.scoop.co.nz/?p=91188#comment-949241. (See my second comment, 6 August, at 20:44 hrs.) Unfortunately, after 12 years, it would seem we are no closer to having such polling than we were in 2004.
No election method used for public elections anywhere in the world sets out to find, and elect a compromise candidate. The purpose of elections is to enable people to elect their representatives, not to arbitrarily discard candidates who “stand for something” in favour of *compromise* candidates. Any method that elected “everyone’s second choice” in a real, public election, would almost certainly be thrown out at the first opportunity.
It astonishes me that you continue to pick away at a perceived fault of STV (which is used in public elections in many places around the world), which can only be remedied by accepting another fault, but you don’t argue for *your* preferred system (which, whatever it might be, we know is not used in public elections anywhere), explaining why it is better than STV. How about you do so.
Of course, you’re not going to do that – you might expose yourself to ridicule – so how about at least explaining to us why you are “still not persuaded that later-no-harm is more important than having majority support for the winner, wherever possible.” Explain to us why it is worthwhile to give up later-no-harm so that the majority winner – let’s call it the Condorcet winner – is elected 100% of the time, instead of 99.9% of the time.
Then, while you’re at it, explain to us what you mean by this elitist clap-trap, “And in multiple-winner STV elections, later-no-harm seems likely to harm voters’ preferences over the sets of possible winners.”
-
Graeme Edgeler, in reply to
What proof do have that STV “certainly doesn’t have that property?”
The Gibbard–Satterthwaite theorem?
-
But, like nonmonotonicity, the no-show paradox, etc., G-S can only be demonstrated with artificial examples. STV's so-called defects are not properties, if that's the right word, that ordinary voters in large public elections can in any way take advantage of.
I'm obviously biased, but, although they're all very interesting, they're ultimately meaningless when it comes to public elections.
-
Graeme Edgeler, in reply to
STV’s so-called defects are not properties, if that’s the right word, that ordinary voters in large public elections can in any way take advantage of.
I’m absolutely fine with STV, and what anomalies remain don't really concern me either.
-
Steve Todd, in reply to
I’m absolutely fine with STV, and what anomalies remain don’t really concern me either.
Graeme, rest assured, I have never doubted that you are, at the very least, very kindly disposed towards STV. You wouldn’t do what you do here, and make the supportive comments about STV here and elsewhere, that you do, if you weren’t. I, for one, am extremely grateful to you.
We know that all electoral systems have to suffer from some anomaly or other, but STV does have the following advantages that no other electoral system used in public elections, that I know of, has—
(A) The number of ‘wasted’ votes in an election (i.e., which do not contribute to the election of any candidate) is kept to a minimum.
(B) As far as possible the opinions of each voter are taken equally into account.
(C) There is no incentive for a voter to vote in any way other than according to his or her actual preference.Happily, NZ STV, with its iterative feedback mechanism (and the consequent reduction in the quota as additional votes become non-transferable, i.e. drop out of the election, as voters, in effect, exercise a delayed abstention), fulfils these conditions, particularly condition (B), much better than traditional STV methods.
In fact, I consider NZ STV is about as far as we can go in the refinement-manageability-public understanding trade-off in the single transferable vote. I think the notion that STV, with further advancements, being used in public elections, will have to wait until everyone has PhDs in pure mathematics and political science!
-
izogi, in reply to
Consequently from (C) and at least when compared with something like FPP, I'd add (D): There is a much lessened incentive for potential candidates (and their parties) to conspire to reduce voters' choice by withdrawing candidacy for fear of splitting the vote against a mutually disliked opponent.
-
Brent Jackson, in reply to
In fact, I consider NZ STV is about as far as we can go in the refinement-manageability-public understanding trade-off in the single transferable vote
The one change I would like to see is a mechanism to reduce invalid ballots, which would also allow a person to rank the candidates they want at the top, and the ones they don’t want at the bottom, without requiring them to rank every candidate.
Something like, upon reaching 2 or more candidates with the same ranking, the remaining vote should be split evenly between them. So if someone mistakenly writes two 14s and no 16, then the votes split to those ranked 14. If one doesn't require it, it goes to the other, and if neither require it, it continues on to 15.
In addition, any unranked candidates are assumed to have the lowest ranking that has not been assigned to any candidate.
For example : With 40 candidates, and 1,2,3, 39, and 40 assigned on the voting paper, then the other 35 candidates are given a ranking of 38. If 1,2,3 no longer require all of the vote, then the remainder is spread over all the other candidates, except those that were ranked 39 and 40.
-
Tim McKenzie, in reply to
Later-no-harm insists that a particular kind of tactical voting is unnecessary and impossible.
Does it. Why not explain how?
I worded that badly. It would have been better for me to say "Later-no-harm insists that a particular kind of tactical voting should be unnecessary and impossible". The kind of tactical voting later-no-harm insists on ruling out is the kind where you can help (or avoid harming) one candidate by refraining from expressing your true preferences among less-preferred candidates. It ignores the kind of tactical voting where you have to give a favoured candidate a lower ranking in order to give them a better chance of winning. Why is ruling out the first kind of tactical voting more important than ruling out the second?
And is it worth it if it means that a voter’s preference between their two favourite candidates is always treated as more important than their preference for their second-favourite candidate over their least-favourite candidate?
Yes it is. Why would it not be?
Because if there are dozens of candidates, then it seems very likely that many voters will have a relatively weak preference between their two favourite candidates, compared to their preference that their least-favourite candidate doesn't beat their second-favourite candidate. If the system assumes that the strengths of voters' preferences are significantly different from the actual strengths of those preferences, then those are ideal conditions for encouraging voters to cast tactical votes, in order to better achieve their actual preferences.
That is exactly what STV does. As there is no incentive for voters not to vote their true preferences, ...
That is, at least in voting theory, not true. And since you've already mentioned non-monotonicity, I assume you already know that. So you must mean that in practice there will never be such an incentive. But that isn't true, either. Wikipedia's article on the monotonicity criterion has a section on real-life monotonicity violations that mentions two elections in Australia: one in which it was shown after the election that monotonicity was violated (in that if somewhere between 31 and 321 Liberal supporters had switched their votes to Labor, then the Liberal candidate would have won), and one in which it was identified before the election that monotonicity might be violated, and that Labor supporters might be better off if a few of them put the Liberal candidate first on their preference list instead.
What proof do have that STV “certainly doesn’t have that property?”
I already explained how, in your own example, supporters of B, C, and D could be better off if they said their first preference was E, instead of voting according to their true preferences.
If more than three-quarters of voters preferred E to win, then why didn’t they vote for E.
Perhaps I should have been clearer. I meant that "more than three quarters of all the voters would have preferred E to win rather than A ". I didn't mean that more than three quarters of the voters had E's victory as their absolute highest preference.
I put up that voting pattern to show that, even though E is the Condorcet candidate, s/he cannot realistically, in a public election, be regarded as the correct winner.
Earlier, you said:
E is certainly the Condorcet candidate and, in spite of a poor showing on first preferences, really a very strong candidate in that if any one of A, B, C, or D were to withdraw before the count (for a single seat), E would win without question.
Why does the choice between A and E of the "correct winner" depend on whether or not D chooses to stand as a candidate?
-
Tim McKenzie, in reply to
I think I'm starting to understand at least part of the reason for our difference of opinion. Please correct me if I'm wrong, but I think you're thinking of a "vote" as something that a voter gives to a particular candidate, as in FPP. But unlike FPP, you're allowing the vote to be re-allocated to another candidate if the first one doesn't need it any more (because they've either been elected or eliminated). With this understanding of what a vote is, you quite reasonably want the candidate or candidates who end up with the most votes to win; STV is a reasonable way of achieving this.
But I'm thinking of a "vote" as the voter's entire list of preferences. It doesn't belong to any of the candidates; it belongs to the voter. With this understanding of what a vote is, I want the winner of the election to be the one that best satisfies all of the preferences of the voters. Arrow's theorem shows that it isn't always possible to do this in an unambiguous way, but I want a system that, wherever possible, chooses a winner who, over every other candidate, has the support of a majority of voters. STV doesn't do that. Other systems do.
People like you (e.g. Doron and Kronick, et al) argue your case, using carefully constructed artificial voting patterns.
I was careful to use your artificially constructed voting pattern to argue against your case.
It astonishes me that you continue to pick away at a perceived fault of STV (which is used in public elections in many places around the world), which can only be remedied by accepting another fault, but you don’t argue for *your* preferred system (which, whatever it might be, we know is not used in public elections anywhere), explaining why it is better than STV. How about you do so.
For single-winner elections (and multiple-winner elections if you don't care about proportionality), I prefer the Schulze method. It always selects the winner from a minimal non-empty set of candidates none of whom are beaten by any candidates outside the set. This is a stronger requirement than the Condorcet criterion, so if there is a candidate who has, over every other candidate, majority support, then that candidate will be elected. STV does not have this property, which is the main reason that the Schulze method is, according to my preferences, better than STV. Other people may reasonably disagree, if they don't care about the winner having majority support (wherever possible), but they have other (peculiar, to my mind) preferences about what kinds of tactical voting should be ruled out. The Schulze method also satisfies the monotonicity criterion, so it rules out the kind of tactical voting in which a voter's most-preferred candidate can be helped by that voter not giving their first preference to that candidate; STV does not have this property, in theory or in practice, as I mentioned earlier.
For multiple-winner elections in which proportionality is desired, I'm less settled about my preferences, and I'm aware that better systems may be developed over time. For now, I'm somewhat attracted by Schulze STV, but the computational complexity of its counting process puts me off. (Incidentally, the computational complexity of ordinary STV is another minor point against it, from my point of view, but it may well be better than Schulze STV on that score.)
I may not be as opposed to STV as you may think I am. I just think that, at least in the single-winner case, we could do better.
Post your response…
This topic is closed.