Alternative alternative voting systems, part six: Net Voting

What is Net Voting? No, it’s not Internet voting, intriguing though the prospect of instant online popular democracy might be. Net Voting is another single member-constituency voting system I’ve dreamt up. I’m not sure that it’s totally original to me, and other systems may incorporate its central principle, which is the possibility of negative votes alongside positive ones.

This is how it works:

  • People can vote both for and against multiple candidates – but not for and against the same candidate, obviously
  • The number of individual electors’ positive votes must always be one greater than the total of their negative votes. E.g. if they vote against one candidate, they must vote for two candidates; if they vote for three candidates, they must vote against two; etc. Of course, people can still opt to vote in favour of just a single candidate, in which case they cannot exercise a negative vote
  • The option of voting in favour of multiple candidates creates the possibility that more than one candidate will obtain the positive votes of a majority of electors. If this is the case, the winner is the majority candidate whose net vote is highest. The net vote is that candidate’s positive votes minus their negative votes: the margin between the number of voters who definitely support a candidate and those who definitely don’t want that candidate to win
  • If only one candidate obtains a majority of positive votes, that candidate automatically wins in that they are supported by a majority and their net vote will by definition also be positive
  • If no candidate obtains a majority of positive votes, the winner is the candidate with the highest net vote.

Net voting addresses, among other things, the issue of how to prevent the election of candidates and governments that are highly unpopular amongst a substantial minority or even majority of the electorate. For instance, the Thatcher governments in the 1980s were all elected on the votes not only of a minority of British people as a whole but also a minority or small majority of voters in many constituencies returning a Conservative MP. In addition, many of those who did not vote Conservative during this period were bitterly opposed to Thatcher’s policies: so she was not only positively supported by a mere minority, but she was arguably strongly opposed by the majority. The majority will was clearly being thwarted.

An extreme illustration of the dangers of untrammelled popular majorities was presented to me by a supporter of range voting [link]. In this hypothetical range voting election, the results were as follows:

75 votes: A=9, B=8, C=0

25 votes: B=9, C=4, A=0

The number of votes here represents the percentage of voters indicating the different scores (from 0 to 9) for each of the three candidates (A, B and C). If you have a system that allows the candidate securing a majority of first preferences to win (in this case, the first preference is indicated by the highest score awarded to each candidate), then A is the clear winner (with 75% of first preferences), even though B obtains a higher overall score. Now, what if A is a Nazi that wants to kill the Jews, whose incompetent candidate is C, while B is also pro-Gentile / anti-Jewish but less extreme? Under a majority system, the Nazi is elected despite the intense opposition of the minority. Under a pure range voting system, B would have won.

If this result is transposed into my Net Voting system, let’s say that only 26% of those supporting A also decide to vote for B, enabling them to express a negative vote against C. Meanwhile, the 25% of voters who are Jewish vote for both C and B in order to vote against A. Accordingly, A gains 75% of positive votes, B obtains 51% and C obtains 25%. Under a pure majority system, A would win. However, under Net Voting, of the two candidates that have won a majority of positive support (A and B), the winner is the one with the highest net vote, which in this instance is B: B’s net vote is 51%, while A’s is only 50% and C’s is -1%. Therefore, the strong opposition of the Jews has overridden the majority support gained by the Nazis.

Net voting seems to me, then, to incorporate an important safeguard against the so-called ‘tyranny of the majority’ or even that of the plurality. It doesn’t provide fail-safe guarantees against the election of someone like Hitler but it makes this much harder. Net voting strongly promotes the most generally acceptable candidate, which in the above example is B: supported by 51% of the electorate, whereas A is supported by 75% but justifiably detested and feared by the other 25%. The importance of this type of safeguard can be illustrated by imagining a smaller group of people comprising, say, ten individuals making a democratic decision about an important course of action. Six people vote for a particular action, but three are strongly opposed to this decision. However, they are outvoted. Fair enough, that’s democracy, you might say. But the consequence could well be that the sub-group of three become permanently disaffected and start quietly sabotaging the actions of the group, rather than being committed team members. Scale that up to the level of a whole national community, and one wonders how much ‘winner takes all’ electoral systems contribute to social and political divisions in that the view of the ‘majority’ (or, under FPTP, the plurality) prevails over the strong disagreement of significant minorities.

A more formal term for the most generally acceptable candidate, favoured by Net Voting, is the ‘Condorcet candidate’ [link]: the candidate who would beat all others in one-to-one contests with them. However, in my example, the winner under Net Voting (B) does not coincide with the Condorcet winner, who is in fact A. The cause of this divergence is that Condorcet elections, where all the implicit pair-wise contests in the election are counted separately (e.g. A versus B, A versus C and B versus C), generally express a preference for one candidate over another in the form of the numerical value 1: if you prefer A to B, this is counted as one vote for A and zero votes for B – and the same for all the other preferences [check this]. However, in Net Voting, if you vote positively for two candidates and negatively for another, this is expressed as a numerical difference of 2: one vote for each of your favoured candidates against minus-one vote for your rejected candidate. Accordingly, the act of rejecting a candidate is given double the weight and power.

With this proviso in mind, Net Voting can be defined as a modified and crude form of Condorcet voting, which should mostly pick the Condorcet winner but guards against Condorcet winners that are significantly unpopular among a minority of voters. An example of the Condorcet winner coinciding with the Net Voting winner would, say, be a seat where 40% of voters support the Conservative candidate and 40% the Labour candidate, while the Lib Dem candidate is the first choice of only 15% of voters. In order to undermine their opponents, three-quarters of both Conservative and Labour voters also vote Lib Dem so as to vote against Labour and Conservative respectively. This results in a victory for the Lib Dems, with 75% of positive votes and no negative votes, equating to a net vote of 75%. Meanwhile, the Conservative and Labour candidates gain a net vote of 10% each.

Breaking this result down, you could say that whereas only 15% of voters are happier with a Lib Dem victory than with any other outcome (those whose first preference was Lib Dem), a full 75% of voters are happier to some extent with a Lib Dem victory than the number of those who oppose it (which, in my example, is zero). By contrast, if, say, the Labour candidate were allowed to claim victory merely because the largest plurality (e.g. 40%) regarded that candidate as their first preference (even though Net Voting technically doesn’t differentiate between first and subsequent preferences), then while 40% of voters would be happier with this outcome than with any other possible result, only 10% more voters would be happy with this result compared with the number of voters that had rejected it outright.

So, where there is a need to determine a winner from more than one candidate obtaining a majority of positive votes, or from all of the candidates in the absence of a majority for any of them, Net Voting finds the candidate for whom the difference between the number of voters who are satisfied (to some extent) by their election and the number who are downright dissatisfied with it is greatest. It’s an index of net voter satisfaction in, or approval for, a particular party or candidate; and it makes it hard for divisive candidates (both popular and unpopular with significant sections of the population) to be elected.

As I indicated, the Condorcet winner, in this example, would probably also be the Lib Dem, as the Lib Dem candidate would almost certainly command the support of most Conservative or Labour supporters in a straight run-off against Labour or Conservative respectively. However, in a real election, the voting patterns would be more complex than in my example: the Lib Dem candidate would also be the recipient of negative votes, perhaps from Labour supporters who might also vote Green in order to vote against the Lib Dem, or from Tory voters adding a second vote for UKIP. Similarly, many Lib Dem voters would exercise a second vote (e.g. for Labour, Conservative or Green) combined with a negative vote for another party. This would make elections much more ‘contestable’ and unpredictable, and would mean that, in many seats (those where no party would gain a majority of positive votes from its core supporters alone), the winning candidate would rely on support from voters favouring other parties in order to win, as they’d need those additional votes to counteract negative votes from other quarters, while benefiting from the negative votes against their opponents deriving from such two-party supporters.

These characteristics of Net Voting could have enormous positive benefits for democracy and democratic participation in England and the UK as a whole. The major parties have been used to calling all the shots, given the tendency of the FPTP system to hand them sizeable majorities on the basis of a minority share of the popular vote. Under AV, however, those parties would need to canvass the votes of supporters of other parties. This would mean Labour and the Tories would have to address the concerns not only of voters whose views are more to the centre than their core support (e.g. Lib Dem voters), but also those who are more to the left and right (e.g. the Greens and UKIP respectively). Indeed, Net Voting would encourage the formation of local and national electoral alliances and coalitions between parties, with the aim of persuading voters to vote for blocks of two or three parties at the same time as voting against blocks of one or two other parties. Voters would of course be perfectly at liberty to ignore the parties’ efforts to enlist their endorsement of particular alliances. Indeed, the various combinations of parties selected by voters could be recorded separately and would give a strong indication to the parties about which coalitions voters would actually prefer to see, if coalition government was required.

Net Voting would also drive the development of multi-party politics, in that it would push up the vote for parties such as UKIP and the Greens as voters who have previously chosen other parties felt emboldened to add these minor parties to their list of positive votes, thereby also enabling them to express negative votes for parties they oppose. You could argue that this gives minor parties an unfair advantage; but then, they’re just as likely to be the object of negative votes from other voters. So while Net Voting pushes up minor parties’ share of positive votes, their net vote would remain lower than that of the major parties in most cases – although, if negative voting became a prominent feature of a particular election campaign, there would be a chance for UKIP or Green candidates to win a majority of positive support and to be elected. While parties more to the right and left of the political mainstream might stand more of a chance of success, Net Voting would prevent the election of ‘extremist’ parties (such as the BNP) because these parties would doubtless be the recipients of large numbers of negative votes, as in my above example of the Nazi party.

Why link positive and negative votes in the way I have in the rules for Net Voting: ‘NV [negative vote] = PV [positive vote] – 1’? Couldn’t you just say that, for every positive or negative vote, voters would need to make one counter-balancing negative or positive vote (NV = PV)? Or why not have a free for all, in which voters could express as many positive and negative votes as they wish? Well, the answer is obvious: if you had a system of one negative vote for every positive vote, this would result in the positive votes for parties on either side of the political spectrum (e.g. Conservative and Labour) simply cancelling each other out, to the benefit of centre parties. If, on the other hand, you make it a condition of negative voting that you must vote positively for another candidate (additional to your most preferred candidate), this obviates the temptation for voters to express negative votes frivolously or destructively in a way that merely expresses anger or distaste for every party other than the one the voter mainly supports. And this latter behaviour would surely also be encouraged by allowing a free for all of negative voting.

Instead, if you have to choose another candidate to vote for as a condition of voting against a candidate, this should make voters weigh up their options much more carefully, both in terms of how their vote expresses what they actually feel and its likely implications. For example, a Labour voter thinking of also voting for the Lib Dem candidate in order to vote against the Tory candidate would have to weigh up which of the possible risks were more acceptable: the election of the Lib Dem candidate, instead of the Labour candidate, as a result of Labour supporters voting Lib Dem and anti-Tory; or the election of the Conservative or even the Lib Dem by only voting Labour, whereas Conservative supporters were voting for the Lib Dem in order to vote against Labour. You could argue this is a form of zero sum gain: Labour supporters are in danger of getting an undesirable outcome, from their point of view, whether they effectively vote for it (by voting Lib Dem and anti-Tory) or not (by only voting Labour). However, a Lib Dem or Tory MP will only be the outcome from the election if the ratio of those who endorse that result to those who reject it is high enough, and also if enough people oppose the election of the Labour candidate to prevent it.

Another reason for adopting the rule of ‘NV = PV-1’ is that the total of net votes will always be the same as the total number of people who turned out to vote, meaning that all the net votes will add up to 100%. This is not just arithmetically neat, but it also ensures that most net-vote totals will be positive values, whereas if voters were allowed to express an indefinite number of negative votes, they’d mostly or even all be negative. An outcome such as that would mean that the winner would appear to lack democratic legitimacy. Limiting the number of negative votes each voter can exercise to one fewer than the total of their positive votes ensures that the winner will be given the backing of a significantly higher number of voters than the total of those who oppose them.

Is it not unfair, you might ask, for Labour and Conservative voters to feel compelled to also vote Lib Dem in order to counteract Conservative and Labour voters who are voting against their parties? However, this sort of situation, and the possibility of Lib Dem candidates winning as a result of it, would tend to arise mainly in marginal constituencies where the Conservatives and Labour are closely matched. And these are precisely the situations in which the Lib Dem would be the Condorcet winner: where the Lib Dem would beat the Labour candidate and the Tory candidate in straight one-to-one run-offs with each of them. So this is fair to the electorate as a whole, in that the system picks a winner that is acceptable (to some degree) to the majority, rather than only to a large minority. But it should be borne in mind that any party gaining a majority of positive votes will win, so long as their net vote is higher than any other parties obtaining a majority. And, as I stated above, it’s also the case that Labour or Conservative candidates could benefit from Net Voting as much as lose out, in that many Lib Dem and other voters would add a positive vote for them in order to vote against another party. Similarly, it’s by no means self-evident that Tory and Labour voters would always pick the Lib Dems as their additional positive vote: they could just as easily opt for parties such as UKIP or the Greens.

So, how does Net Voting stack up in relation to the criteria I have been using to assess different voting systems in this series of posts? The first criterion is: Does every vote count, and is every vote counted? On this, I would give Net Voting four points out of five. In Net Voting, every vote, positive or negative, really does have the potential to directly influence the result, since this is determined on the basis both of positive vote share and net vote taking into account negative votes. The facility to vote against candidates is quite empowering to voters. In simple majority or plurality systems, if you don’t like the leading party in your constituency, there’s nothing much you can do about it, and whether you actually vote or not can often make little impact to the end result. This is not the case with Net Voting: if you don’t like a party or candidate, you can vote against them as well as in favour of other parties, and seriously threaten the incumbent’s of being re-elected. Equally, it becomes more worthwhile voting for smaller parties under Net Voting, because they have a greater chance of being elected. This is because supporters of the larger parties will also vote for the minor parties they endorse to some degree, since this enables them to vote against their main rivals.

In addition, the results of elections held using Net Voting would give quite a detailed break-down of the anatomy of political feeling in any given constituency, and across England and the other countries of the UK. You could record all the different combinations of voting: how many voters had voted for combinations of particular parties and against which others – so you’d have a detailed sense, for instance, of how sympathetic towards UKIP were Tory voters in a particular seat, and how strongly opposed to the coalition government were voters in another, and how this translated in terms of anti-Tory and anti-Lib Dem voting, etc. So all the votes would be counted and would count, i.e. matter. The reasons why I would not give Net Voting a perfect five for this criterion is that it doesn’t distinguish between higher- and lower-preference positive votes; and there’s still an element of pointlessness about voting for particular parties in particular seats: if the level of support those parties enjoy simply is not high enough to secure any representation, then a vote for them could still be viewed as a wasted vote.

In terms of criterion No. 2 – Is the system proportional? – I’d give Net Voting three out of five. As a single member-constituency system it is by definition not proportional in its design. However, I believe it would produce more proportional results nationwide, at the same time as redefining what constitutes proportionality. That is to say, the proportionality of voting systems – or at least, of preferential systems, which is what FPTP, AV and STV all are, in effect – tends to be understood using a rather crude model for the meaning of a person’s vote: that it expresses unqualified positive support for a candidate or party, and that this support is exclusive, i.e. it precludes support for other candidates at the same time. On that basis, simple percentages of support for the parties can be arrived at, against which the proportionality of the system is assessed.

Net Voting breaks this model in two ways:

  1. The fact that you can vote for more than one candidate means that, rather like Approval Voting [link], the number of votes won by each candidate is not an ‘absolute’ measure of the extent of support for that party. In my previous example, the kind of assumptions that are behind preferential voting systems would lead one to suppose that 40% of voters primarily ‘support’ Labour rather than any other party, 40% ‘support’ the Conservatives only, and 15% exclusively ‘support’ the Lib Dems. By contrast, in Net Voting, the percentage of positive votes won by each candidate expresses the fact that x number of voters are willing to give some degree of backing to that candidate, ranging from luke-warm to passionate support. So it’s a measure of approval rather than absolute support.
  2. The fact you can also vote against candidates creates an additional mode of expression in voting. That is, in preferential systems, you either vote for a candidate or do not vote for a candidate; i.e. there is either the presence of a vote (with a numerical value of 1) or the absence of a vote (0). To this, Net Voting adds the presence of an anti-vote or negative vote (-1). So whereas preferential systems, and assessments of proportionality that are based on the results they produce, effectively consider the absence of a vote for a candidate as if it represented the absence of any support for or opposition to that candidate, Net Voting encourages voters to make the meaning of that absence of a positive vote more explicit: whether it equates to opposition (-1) or ‘acquiescence’ (0). I use this latter term, because if you can’t be bothered to vote either for or against a candidate, then you are effectively acquiescing in their election if they are successful.

As a result of these factors, in Net Voting, the winner – except in cases where only one candidate secures a majority of positive votes – is the candidate for whom the ratio of qualified support (or approval) to definite opposition is highest; and this can be affected not only by the number of votes both for and against a candidate, but also by the number of non-votes or abstentions in relation to that candidate. For example, a candidate winning 41% of positive votes but garnering 30% of negative votes and, effectively, 29% of abstentions will do less well than another candidate gaining 40% of positive votes, 20% of negative votes and 40% of abstentions.

Now, what if the level of support for the first of these two candidates is more intense and committed than that for the second, who has benefited more from supporters of other parties also giving him / her their vote in order to vote against the first candidate? Net Voting does blur the distinction between candidates who are voters’ primary and secondary choice, in this sense. But the whole point, precisely, is that just as the support for that candidate is passionate, so is the opposition against them, so the second candidate wins because he / she is less divisive of opinion. Net Voting records that extra measure of opposition towards candidates.

Besides, I would dispute that you could prove that, in my earlier example, the 40% of people voting Labour or Tory do so out of unqualified and primary backing for those parties, whereas 60% of the 75% who voted Lib Dem do so merely for tactical reasons, and not because they ‘really’ support (or approve of) the Lib Dems to some extent. In the same way, people who vote exclusively for a particular party under FPTP, or who would make sure that they listed that party as a high enough preference to ensure that their vote for that party was counted (in AV), also do so out a mixture of reasons, including:

  • tactical considerations (e.g. ‘vote Labour to ensure the Tory doesn’t get in’)
  • because they feel the seat and the government is in safer hands with that candidate and party
  • because they feel their personal interests would or might be better protected by that party
  • and / or because the candidate or party in question is the only vaguely acceptable one.

So the winners of FPTP, AV and even STV elections are also, as it were, not always the party that voters definitely support – to the exclusion of other parties – but are often the ‘merely OK’ option to which voters feel they have no alternative. Net Voting is a means, in effect, to extend the franchise of the merely OK to other parties whom the broad mass of voters might like to give a chance, but who are put off from doing so by the restrictive rules of the FPTP and AV systems, where only the party that people supposedly support exclusively or definitely ‘more than others’ can prevail, irrespective of how much negative feeling may also be directed towards that party by other voters. Net Voting prefers the rule of the merely OK party that can command mass support and a certain level of acquiescence to the rule of the supposedly strongly preferred party that does provoke mass discontent.

Therefore, the winner of a Net Voting election in any constituency would be the consensus candidate, who is possibly but not necessarily the same as the one most people ‘support’ as their primary choice – however high-quality that support might or might not be. In seats where there are presently Labour or Conservative majorities or strong pluralities, the winner would almost always continue to be the Labour or Conservative candidate. The result would be different, as I said, mainly in more marginal constituencies, where the strongest alternatives to the main parties (who would garner large amounts of both positive and negative votes) would emerge victorious: this would be mostly the Lib Dem, but also some UKIP and Green candidates might win through. Nationwide, the results would form an aggregate of consensus opinion across England and the UK, and would in fact form something more like what we would recognise as a proportionally representative parliament, with fewer Labour and Conservative MPs than the present parliament, more Lib Dems, and some Greens and UKIP. However, this parliament would be ‘proportional’ – to the extent that it would be – because it would be a reflection of the consensus view across different parts of England and the UK, not because it reflected the percentage of people’s votes for their ‘first choice’ or ‘preferred’ candidates, a choice whose basis is in reality quite complex and mixed.

As for criterion No. 3, Does the system foster accountability?, Net Voting scores a strong four out of five, in my view. If voters can vote against their MPs at the same time as for multiple alternatives, this should really put MPs on their toes and make them work hard to represent all their constituents, whatever their political views. Net Voting doesn’t score five out of five here because it is a single-member system, which means there will still be quite a lot of ‘safe seats’ (where one party is dominant), resulting in parliaments that are less representative of the electorate’s views at a nationwide level.

In terms of my fourth criterion, Does the voting system allow voters to express the full range of their political and personal preferences, and send a message to politicians?, Net Voting performs strongly here, too: four out of five. Voters can express their support for multiple parties, which other systems do not allow, and they can also express their anger and strong opposition towards other parties, particularly those of the incumbent MP or government. The different combinations of positive and negative votes, if recorded, would also speak eloquently of voters’ wishes for particular parties to work together, or not as the case may be; and they would indicate the extent of disapproval and dislike for certain parties in a way that voting systems that allow only positive votes could not. Again, it’s not five out of five against this criterion, because Net Voting does place structural limits on the number of positive and negative votes people can register; and some voters, for instance, would be deterred from voting against a particular party if they didn’t wish to register a second positive vote for a party other than their most preferred choice. But then, that decision does also effectively express an order of preference – positive vote(s) for one or more candidate(s); no vote or abstention for others; negative vote(s), or not, for others – so that the system does allow voters to express a range of views and feelings.

As for the fifth criterion, Does the system mitigate tactical voting?, the answer is both yes and no – so I’d give Net Voting three out of five. It does largely eliminate classical tactical voting: voting for a candidate who stands a chance of defeating another candidate you don’t like instead of voting for the candidate you actually prefer, who can’t win. With Net Voting, what you would probably do is vote for both your preferred candidate and your tactical choice, and vote against the candidate you don’t want; so you don’t have to surrender your real voting preference. You could also decide not to vote for the tactical candidate but for another candidate in addition to your preferred candidate, so as to vote against your rejected candidate. You’d have to weigh the risk of inadvertently voting in your tactical candidate instead of your preferred candidate (by the first of these courses of action) versus allowing your hated candidate to win by not voting for the one candidate who could have beaten them (by the second route). So Net Voting gives a number of tactical choices to voters, but they don’t have to make those choices if they don’t want to. But, it’s true, Net Voting in general could become very tactical, and voters could be confronted with bewildering dilemmas trying to weigh up the different possible outcomes. Hence, I score it in the middle here.

Finally, No. 6: how easy is the system for voters to understand, trust and use effectively? Well, Net Voting could initially seem rather complex and confusing to voters, who’d have to be educated about the fact that the number of your positive votes must always be one more than the number of your negative votes; about why certain results produced certain winners and losers; and about weighing up the different potential outcomes in the way I’ve just described. But once voters got the hang of Net Voting, I feel sure they’d take to it with gusto, enjoying the fact they can give governments, MPs and parties a slap in the face with negative votes, and the fact that the combination of positive and negative votes allows them to express quite a lot about the different parties they like and dislike. And ultimately, it’s not that difficult to understand that you must have one positive vote more than the total of negative votes. Hence, I’d give Net Voting three out of five for this criterion.

The table below shows a comparison of Net Voting with the other systems I’ve analysed in relation to my six criteria:

Criterion FPTP AV AppV ARV TMPR AV+ NetV
Does every vote count?

3

2

3

4

4

3

4

Is the system proportional?

1

2

2

3

3

3

3

Does the system foster accountability?

3

3

4

4

4

4

4

Does the system let voters express their views?

1

2

3

4

4

3

4

Does the system mitigate tactical voting?

1

2

2

3

4

2

3

How user-friendly is the system?

3

2

4

3

3

3

3

Total scored out of a maximum of 30

12

13

18

21

22

18

21

Net Voting emerges as a perhaps surprisingly strong contender: equal to the range-voting ARV method and just behind the two member-constituency proportional system TMPR.

It’s definitely one that’s worth considering, and it would make elections very interesting, lively and unpredictable. Just the tonic English and British democracy require!

Advertisements

2 Responses

  1. “The number of individual electors’ positive votes must always be one greater than the total of their negative votes.”

    I imagine that this would just cause the parties to run extra dummy-candidates to sop up any excess positive or negative votes, not unlike the “teaming” problem in Borda.

    I think a rang system with a range of 3 (which, therefore, has the some number of options for each candidate, but valued at 0/1/2 instead of -1/0/+1), without the seemingly artificial requirement of a net-positive score, would work better, or at least as-well.

    After all, is a coerced majority (which this system still doesn’t guarantee) really better than no majority at all?

    • Yes, good point about the clone candidates. I might be being naive, but I don’t think English voters would take too well to that (fair play and all that) and would punish the parties if they did run dummy candidates. Certainly, the press would be ruthless. But yes, it’s a risk, which I ought to have acknowledged in the article.

      I think a range system involving points from 0 to 2 would promote completely different voter behaviour than Net Voting, which is not of itself a reason not to consider it. Net Voting encourages (you say ‘coerces’) people to vote for parties / candidates they support to some degree, and works rather more like Approval Voting, in that respect, than range voting. It’s arguable that that would lead to less voter disappointment at the result compared with a range voting system that could result in a candidate with the largest number of 2’s (in your example) still losing.

      Plus a range from 0 to 2 doesn’t deal with the issue of unpopular or downright dangerous candidates being voted in. If you used the range 0 to 2 in my example of the Nazi candidate (A) obtaining the highest score from 75% of voters, and the other Gentile candidate (B) obtaining the second-highest score from those 75% plus the highest score from the 25% of Jewish voters (whose candidate is C), the Nazi would win.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: