Ranked Choice and Most-Least Voting

Posted on by

Yves here. America’s simple two-party system means we are unfamiliar with coalition strategies and complex preference-registering systems. Some of them like ranked choice are starting to be adopted in the US. Our Terry Flynn digs into where two approaches, ranked choice and more-least, produce similar and divergent results.

By Terry Flynn, an NC regular who writes at Terry Flynn PhD and thinks hard and rigorously about matters like polling, voting, and other real-life areas where the interpretation of data matters yet is often done badly. Originally published at his website

I recently realised that two systems proposed as “PR-lite” or “a step towards full PR” can produce radically different outcomes IN REALITY and not just as a THEORETICAL CURIOSITY. The two are “single candidate ranked choice” and “most-least voting – MLV”, most notably when there are just three candidates.

Here’s the deal. Under ranked choice you must rank all three candidates, 1, 2 & 3 (most preferred to least preferred). Under most-least voting you indicate only the “most preferred” (rank 1) and least preferred (with 3 candidates, rank 3). The OBSERVED set of data should be the same. (I’m not going to get into the issue of why they might not – that gets into complex mathematics and I’ll do it another time).

For those who don’t want to get bogged down in the following discussion of the maths, here’s why the two systems can, given EXACTLY the same observed count data, give a different “winning candidate”. Ranked voting essentially tries to identify the (first or second best) candidate that the people-supporting-the-losing-3rd-party-candidate are “most happy with”. Under MLV, if both “first” and “second” preference candidates are diametrically opposite (and mutually hated) then NEITHER should necessarily be elected. The candidate who came a (very very) distant third can be elected if (s)he is NOT HATED by anyone. Essentially, if you polarise the electorate you are penalised. A “centrist” who hasn’t either “enthused” or “repelled” anyone will win under MLV.

I tended to think this was a “theoretical curiousity”. However, upon looking more closely at the 2016 Iowa Democratic Presidential primary I realised this ACTUALLY HAPPENED. Bernie Sanders and Hillary Clinton were essentially tied on about 49.5% each in terms of their “primary first preference vote”. Hillary had the edge, and the 3rd candidate, O’Malley dropped out (but too late so he got votes). Yet he was actually the key influencer, if either ranked voting or MLV had been used. Under ranked choice, either Hillary or Bernie would have won (determined by who the majority of O’Malley’s supporters put as second preference). Under MLV, and assuming that the “much talked about antipathy between Bernie and Hillary was real” then each candidate’s supporters would have put the other as “least preferred”. The “most-minus-least” counts would have been slightly negative for one and likely both candidates. O’Malley, on the other hand, would have obtained a small positive net most-minus-least vote (getting 1 to 2% of the vote, with few/no people putting him as “least preferred”). MLV simply subtracts the “least preferred” total from the “most preferred” total for each candidate giving a “net support rating”.

Under ranked choice voting either Hillary or Bernie would have won. Under MLV both would have been denied the win in favour of O’Malley, because he “pissed nobody off”.

Here’s the more detailed discussion.

Most-Least Voting (MLV) is a special case of a more general method of “stated preferences” called Best-Worst Scaling (BWS). Declaration of interest: I am a co-author on the definitive CUP textbook on BWS, was involved (along with its inventor) in much of the theoretical development and application in various fields (most notably health). HOWEVER I have had no involvement with the theory, parameterisation or application of MLV. Indeed, once I became aware of this method of voting, on checking the bibliography, it became clear that the authors were not actually aware of BWS and due to the “silo effect” in academia, had come up with it largely independently of what we had already done. Incidentally some of the Baltic States have used or do use MLV in certain instances so it isn’t just a “theoretical curiosity”.

OK, having got that out the way, what do I think of MLV? In short, I think it is worthy of serious consideration and wish we’d thought of it first! Like ranked choice voting with single member constituencies (something in use or proposed in various Anglo-Saxon countries like Australia, the UK and USA), it is not “proper” Proportional Representation (PR). However, it can be considered either as a nice compromise, or as a stepping stone to “full PR”. In terms of its similarities to ranked choice voting: suppose there are 5 candidates in your constituency. Under ranked choice, for the maths to not be horribly skewed and potentially very very gameable, you should be forced to rank all five, 1,2,3,4,5. The problem, known since the mid 1960s, is that people are good at “top” and “bottom” ranks but get very “random” and arbitrary “in the middle”. MLV exploits this. It only asks for top and bottom. Thus it may be considered to be the “minimum change to first-past-the-post – FPTP – possible” so as to “make things easy for people”. You only provide ONE extra piece of information – the candidate/Party you like least. If you do not provide both a MOST and a LEAST choice then your ballot is spoilt. This is IMPERATIVE for the maths to work, and for the system to be demonstrably “equitable”. (Most-minus-least vote totals must sum to zero.)

The common question is “Suppose there are only three candidates – aren’t ranked choice and MLV the same?” NO. See above for a real life example. Ranked choice MIGHT be unconstitutional in certain countries (if the mathematicians and lawyers got together because not everyone has the “same influence” mathematically).

So what is happening in practice?  The authors conclude that if the “FPTP winning” candidate espouses (say) a very extreme policy on (say) immigration or something, that all other parties abhor, then (s)he is likely to lose. All other parties “gang up” and place that candidate as “least”. Most-minus-least vote tally is net (highly?) negative. A more “moderate” candidate likely wins. Indeed, the authors claim that “centrists” likely prevail a lot of the time – though they might be an “O’Malley with 1% primary vote”. Though if a candidate would get a MAJORITY (and not just a PLURALITY) under FPTP, they’ll still win under MLV. So “majority” (non-coalition) governments still can happen – they’re just harder to achieve and “third parties” (etc) much more easily get a foothold. I happen to think that this “centrists rule” conclusion is a little simplistic when you move from a single dimension (left/right) to multidimensional space. Yes, maybe you get a candidate closest to the centroid across all dimensions but “how strongly people regard each dimension” can affect results. So, as they frustratingly say in academic papers, it’s “an empirical issue” as to what will happen. However, I will venture a conclusion that “extremists” will naturally get weeded out. Whilst some extremists might be generally considered bad (consider dictators who were first voted in via pluralities in 1930s Europe), others (painted as “extremists” by the MSM like a Sanders today or an Attlee or FDR of yesteryear) could be considered necessary and without them society would be much worse off. It gets necessarily subjective here…!

TL;DR: Arrow’s Impossibility Theorem still holds. MLV doesn’t solve all problems but it is attractive in addressing a lot of the most commonly made criticisms of voting systems used in the UK and USA. However, it isn’t the ONLY system that can address these criticisms – it is merely the “simplest” in terms of practicality and requiring “minimum extra effort by voters beyond what they do now”. Whether you “like it” depends on your “values”.

Print Friendly, PDF & Email

61 comments

  1. Mark Gisleson

    Most-Least Voting reminds me of a suggestion I read years ago, quite possibly at this site.

    “None of the Above” would cure a lot of our electoral woes and would put an end to the duopoly forcing bad candidates on us. It would also be humiliating for the rejected candidates, hopefully encouraging them to go away and — unlike Nixon — stay away.

    Reply
    1. MartyH

      None of the Above sends a much more clear a message than voting for a candidate with no path to victory or not voting at all.

      Reply
      1. edwin

        None of the Above also is the only reasonable method where being listed on the ballot is heavily restricted or practically impossible. It is possible that MLV might lead to changes to how difficult it is to form a new party and have your name on the ballot in some areas.

        MLV does not prevent the need to do strategic voting. I’m not sure how much that would undermine MLV.

        Reply
        1. Terry Flynn

          FWIW the inventor of BWS, my former boss, was obsessed with the problem that BWS was and is a “relative preference” system. His favourite anecdote is “choose best and worst from Hitler, Stalin and Mussolini”. Awful and nonsense choice. How do you introduce “none of them” in a way that works mathematically in the wider model?

          He never solved it in a way that satisfied our 3rd author, the “god” of mathematical psychology. I’m sure there are “practical ways” to do it…..but in terms of “getting the maths right”? Not to my knowledge.

          Reply
    2. hunkerdown

      Unfortunately, NOTA came out of the neoliberal (“lolbertarian”) movement. Therefore, reactionary ideas of absolute property can be found underneath the Alice Cooper “Eighteen” wrapper. In lolbert land, “NOTA” means switching off as much of the public sphere that does not elevate property over life, until such time as a supreme leader can be seated. Because of their previous work in “developing” that doctrine, they are likely placed and prepared to slip such purpose-defeating provisions into any such laws that cross their path. Because their value systems, their partial elite status, and their consumption entitlements depend on it, they would do everything in their power and then some to ensure a simple NOTA doesn’t come to pass, and the major parties that currently advance neoliberal doctrine would Katy bar the door, shoulder to shoulder with them.

      Second, elections are only held because they inure to the benefit of the Elect. If the elect, as a class, did not require the legitimacy elections provided, there would be no reason to spend the effort. NOTA on the ballot doesn’t merely send the message that we don’t need elites and we don’t owe elites. The possibility of rejecting elite rule, even for a moment, is a direct challenge to a property right, which quickly ramifies to destroy the historical moment of all kinds of entitlements, rights, properties, traditions, norms, guardrails, and claims of priority. The Puritan class system fears “masterless men” above all, and its elites wouldn’t offer people the option to vote themselves out of an odious debt. (“Waste is someone else’s paycheck.”) Even if NOTA made it onto every ballot by a citizen’s initiative landslide, could/would people faithfully count votes against their interests and privileges, or would they unconsciously or deliberately err in favor of the imaginary grand debt?

      So, while NOTA is a fine and necessary idea, it’s fundamentally incompatible with the order of (at the very least) US society and interferes with the ability of elites to constitute themselves as something transcendently aloof from the mass. Election strikes might be more visible, but also more risky these days now that the two parties take the ownership of “THEIR” democracy as a life or death matter.

      Reply
      1. Terry Flynn

        I have Australian citizenship as well as UK citizenship. Australian residents are OBLIGED to vote (so I don’t get to vote now I’m back in UK and no longer resident).

        I firmly agree with this. I think that those who dislike all options should be taught that “striking through their ballot paper” (spoiling it – I did it in UK last month in local elections – they’re all $&##) or having a NOTA option should be there.

        Any key policy options (like referenda) should get 50%+1 voters supporting it or fail.

        Reply
        1. Mark Gisleson

          In 2016 I left the top line (Presidential) unvoted. Then I noticed that the news media went to great lengths to not report on undervotes except as a bottom of the ballot issue.

          I’ve come to suspect that if you set yourself on fire in your polling place, it would be reported as a disposable lighter malfunction.

          Reply
    3. Michael McK

      If none of the above wins several times in a row do the incumbents get to stay in office till a new slate wins? I don’t like the sound of that. I assume those candidates who none of the above beat are excluded from running for (that?) office for some period of time. Sometimes I think sortition (random citizens plucked out like jury duty if I spelled it right) but that might only further entrench the power of lobbyists and the deep state.

      Reply
      1. EarlyGray

        > Sometimes I think sortition but that might only further entrench the power of lobbyists and the deep state.

        If NOTA wins a vote randomly selecting a citizen to serve sounds like it could be a good solution, but yes, designing the system to make sure that they aren’t influenced too much by lobbyists is important.
        It’s easy to imagine that they would often feel that insufficiently informed and become susceptible to the powers of persuasion of vested interests.
        At a minimum, the randomly selected citizen should be limited to one term, barred from running again and adequately compensated (including pension) to lessen the risk of selling out. But I’m not sure that solves the lobbyist problem.

        Reply
        1. Terry Flynn

          I totally get all your points. I don’t pretend to have answers here. I just feel very strongly that everyone should vote – even to say “NOTA” or “I don’t understand” or whatever. If nothing else it shows who is (dis)engaged and whether there is a problem concerning understanding of the issue(s) being voted upon.

          We need desperately to distinguish between “content and doesn’t vote” and “frustrated and doesn’t vote because they think their vote is worthless”.

          Reply
          1. philnc

            Adding NOTA to the ballot could be a first, positive, step towards deligitimatizing the duopoly. Of course we’d have to fight like hell for it, and the question is whether a “make the effort to go out and vote, but cast a substatially unmarked ballot” strategy would be more efficient. Of course either way you need to bypass the corporate media filter to mobilize a significant number of voters, and then inform them of the results. No matter what you do, overcoming the vice-like grip the mass media have on most minds is the key challenge. If someone figures out a serious strategy for accomplishing _that_, I’m ready to sign up.

            Reply
  2. PlutoniumKun

    Thanks Terry, very interesting.

    One of the problems I think in assessing any voting system is that the type of system changes the type of candidates chosen, and may alter peoples behaviour in voting. Here in Ireland, we have PR with Single Transferable Vote and multiple seat constituencies which significantly changes the dynamics compared to the otherwise fairly similar UK parliamentary system. It definitely creates an electoral system where elected reps are more attuned to voters needs, or put another way, it encourages parish pump politics. There is far more scope for both political parties and electorates to game their voting in all sorts of way.

    One thing thats hard to assess is how deeply voters think about how to work their vote. Some people are very attuned to the vagaries of the system – for example, by giving their first vote to the weaker of their favoured candidates to help them get past the first cull. Others just run down the list quite randomly. It can be quite intriguing running through vote transfers to try to see just what people were thinking. Sometimes, its baffling, sometimes you can see clear strategies at work.

    Reply
    1. Terry Flynn

      Thanks PK. Yes, the type of system does exactly what you say. I am “taking as given” a desire expressed in the UK, USA, and Australia, to have single-member constituencies (unlike Ireland). The irony is that we in UK have multi-member constituencies in other contexts (the now defunct European Elections) and we were fine about it. However, we seem to have some “sacrosanct belief” that everyone should have a SINGLE member of parliament (at Westminster) they can go to about an issue. USA people are the same with the House of Representatives; Australians are in their House of Representatives (though not the Senate curiously, but they probably inherited the British idea that “the 2nd chamber doesn’t really matter”).

      The “gaming” of the system by voters rapidly emerges following any new new system – or, as you note, it gets worriesome when some voters learn how to game it whilst others seem to utilise some “strategy” that is difficult to distinguish from complete randomness. This is why BWS (and MLV) can be advantageous – Helson (1964) was the original in showing that “just choosing top and bottom” minimises potential for the voter to mess about. Though, as you imply, the mere ACT of introducing MLV (or something based on the more general method of BWS) can influence WHO gets through any “primary” to get to the ballot.

      This is why I wonder if MLV might be something best suited for primaries. Yves and other people in parts of the USA might be best placed to speculate on “what might happen” in key Democrat and Republican primaries if MLV were in operation. For example. I am FIRMLY of the believe that Trump would NEVER have won the Republican nomination in 2016. He had ALL the other candidates against him and his “core support” was not >50% across key states. His most-minus-least total votes would have been negative. One of the members of “the blob” (Bush?) would have won. It’s less predictable what would have happened on the Democrat side.

      Reply
      1. PlutoniumKun

        One thing I’d note in relation to some systems ‘favouring centrists’ is that in Ireland it tends to favour the soft left as they are many peoples ‘least worst’ option. So, for example, the Greens consistently tend to end up with more elected prepresentatives than would appear obvious from their first vote (except when they are in power, when the opposite happens, as they attract generalised hatred from all sides). A particular sign of the way Irish politics works is that votes rarely transfer between the two main centre right parties, they hate each other so much, most of their voters prefer to transfer to the left or to independents.

        Reply
        1. Terry Flynn

          Thanks. I had kinda thought this from looking at the stages of voting in Irish elections….and believed it might be good. However, I fully acknowledge that Green Parties in some countries have not “behaved well”. So I acknowledge my potential “gap in information” here so if I’m wrong please let me know.

          I guess my desire – based on what my Irish relatives told me over the years – is that I want Fianna Fáil and Fine Gael humbled so that “proper issues” can be discussed” rather than WTF Eire did in a civil war 100 year ago.

          Electing “soft-left” might be a good step forward.

          Reply
          1. R

            I think Ireland or France might be an example of where MLV could favour the extremist. Imagine (hypothetically) Le Pen vs Macron, each running a right-wing ego party that hates the other. They would polarise the right and Jean-Luc Melenchon would slip through the middle! (I would pay money to see that). Something similar might happen with the Fianna Fail / Fine Gail double-act and Sinn Fein.

            As an historical aside, England was quite happy with multi-member constituencies for much of its parliamentary history. Boroughs (urban) had two members; counties (rural) had two, three or four; and county boroughs (a hybrid). These were only rationallised into single member constituencies by the end of the 1th century. There were also multi-member University constituencies throughout the UK, which gave their graduates an additional vote for a university MP (imagine that now, the PMC wet dream!). These lasted until 1950 in England and – I had not realised – until 1969 in Northern Ireland; what is better known is that they continue in Ireland for the Senate. Carson was a Dublin Universities MP and A P Herbert an Oxford Universities MP.

            https://en.wikipedia.org/wiki/United_Kingdom_constituencies
            https://en.wikipedia.org/wiki/University_constituency

            I learnt in checking the references that although plural voting was largely suppressed by the 1948 Labour government, it still remains for the City of London local elections….

            Reply
          2. PlutoniumKun

            I should clarify my comment above as I realise it overstated the failure for FF and FG to cross-vote. It does happen to a significant degree, but to a large extent it reflects personality rather than politics. For example, a FG voter might vote 1,2,3 for FG candidates, but then give a no.4 to a particular FF candidate because ‘he seems decent enough’ or ‘I want to support the female candidates’ or whatever (the latter is one reason why Irish political parties are so keen to run younger female candidates on their slate as they know they pick up residual transfers like this). It also explains why personal likability is extremely important in Irish elections, its a transfer catcher.

            And on the latter point this is another feature that is often ignored in the discussion of electoral systems – the party choice of candidate may be influenced by the system. Certainly in Ireland there is a strong emphasis on picking candidates who are not seen as divisive or ‘party people’ as they don’t get transfers. There is a strong preference for ‘everyman/woman’ types. Contrast the UK where ‘he’s a dick but he comes across well on TV’ or ‘everyone hates her, but we need her so put in her a safe seat’ can be viable candidate selection strategies.

            Incidentally, a lot of esteemed number crunchers like to say that transfers are not so important in Ireland as people think, as the majority of final votes finish up pretty much following the pattern of the first preferences. But this ignores the fact that many first preferences are in fact strategic votes, there is no guarantee they’d be replicated under another system. In many constituencies, parties actively instruct supporters who to give their first preferences to in order to game the system as well as possible.

            Just to give a well known example of the above, in the 1980’s FF was led by the egotistical, charismatic Charles Haughey, while the intellectual low key (but very popular) Garret Fitzgerald led FG. Both were in Dublin constituencies. Haughey was obsessed with getting more first preferences than Fitzgerald, so he refused to vote manage. The result was that he usually got a massive first preference, leaving his running mates so low down that they couldn’t recover with his transfers. This invariably cost FF a seat at the expense of FG or Labour. Fitzgerald on the other hand actively told supporters to give him a No.2 in favour of his running mate, which allowed him to win FG an extra seat.

            Incidentally, the new found success of Sinn Fein is presenting them with a new challenge – managing votes to maximise the number of seats, and for the first time this means doing informal deals with other left wing groups (and occasionally FF). It remains to be seen how good they will be at this, as they’ve historically suffered by not getting transfers. But their focus on articulate younger female candidates is paying off in the long term, plus they are slowly strangling the traditional hard left, leaving the field clear.

            Reply
            1. Terry Flynn

              Thank you! I was politically aware from a very young age. Even back then I was puzzled by the success of FG given what the UK media said. They either didn’t understand the Irish voting system or decided to deliberately misrepresent what had been going on. Even my Irish relatives were puzzled by GF’s success. They tended to “hate Haughey but be vaguely sympathetic to FF” so found themselves in a weird situation.

              I think, ironically, Thatcher had one of her better moments in talking to GF and initiating a process that ultimately led to the GF agreement. I do remember that lots of rightwingers here thought she had gone bonkers by talking to him. Maybe she recognised a fellow “clever political operative”? Or maybe the Brighton bomb just scared her……

              Reply
              1. PlutoniumKun

                GF was a very canny operator in his low key, bumbling way. A few years back the official papers of that period in the 1980’s were released and there were many interesting vignettes. There is a story that Thatcher, Fitzgerald and Mitterand were seated at an EU dinner. Fitzgerald and Mitterand were engaged in deep conversation in french, leaving Thatcher furious. She was convinced they were plotting against the UK, but apparently they were just discussing their favourite 19th Century catholic French novelists (this would be very typical Fitzgerald thing, he famously read airline timetables for fun and was a dedicated Francophile). Fitzgerald also openly described himself as the most left wing member of his party and never bothered hide the fact that he only joined because of his family history, he was much more sympathetic to Labour.

                I think Fitzgerald simply confused Thatcher – he was everything that she was suspicious of – Irish, catholic, an intellectual, deeply modest and having a simple sort of integrity (he was a very unusual successful politician). She could handle Haughey as she knew and understood his type, but I think Fitzgerald simply did his groundwork behind the scenes and left Thatcher so isolated that she had no choice but to go against her instinct and sign up to the peace deal. But I don’t think Thatcher ever had anything but the most superficial understanding of what she was signing up for.

                Reply
                1. Terry Flynn

                  haha that’s an amazing anecdote. Love that kind of thing. I have loads about the family of the then Margaret Roberts but due to libel laws I can’t repeat them.

                  Hint: my paternal grandmother’s family were the “other” Roberts family in Grantham. My great-grandfather forbade ANY of his daughters to visit the other Roberts grocer shop unless he was with them. My grandmother’s younger sister was same age as “Margaret Roberts” – girls grammar school so seated by surname. They ended up as friends and the future-thatcher visited our family home. Great-grandpa not amused. Grandmother, a dyed-in-the-wool tory never liked Thatcher and me as a kid/teenager never knew why……til later…..that is why the residents of Grantham refuse to fund a statue of her.

                  Reply
                  1. PlutoniumKun

                    Oh wow, that is intriguing (I’ve heard suggestions of this before, and the explanations are probably libellous).

                    Reply
      2. Tom Bradford

        Under New Zealand’s hybrid PR system half the seats in Parliament go to members representing territorial constituencies elected on an FPTP basis by that constituency in the ‘usual’ way, while the other half are members drawn from a ranked list offered by political parties in accordance with the share of a ‘party vote’ they get from a second vote on the voting paper.

        The results seem to show that electors don’t simply vote on a party basis – giving their party vote for their preferred party and their constituency vote to the person standing for that party. There are smaller parties, like the Greens or with a religious or more extreme political bent, which simply can’t afford to fight for every constituency seat and would never win more than one or two anyway, but which are given a meaningful voice by the system and do get enough support through the party vote to get a say in parliament – often a big say if their few seats are needed to build a coalition government as is often the case. Yet you also retain the members whose political lives depend on them ‘fighting for’ the interests of the population of geographic entity to which they are tied.

        It is a compromise and as such has all the advantages and disadvantages of all its constituent systems. To me its biggest drawback is that it can, and has, put the choice of whether ‘left’ or ‘right’ can form the Government into the hands of a small minority, although that does at least tend to keep everyone more honest and as the Lib-Dems in the UK discovered when it had that power – rare under FPTP – and ‘blew it’, it can have severe repercussions.

        Its far from perfect but IMHO is a vast improvement over the two-party system in the UK and the US that seems the be inevitable under FPTP.

        Reply
        1. Mikkel

          Hey Tom, this isn’t quite accurate although the way you described it would appear to be how it’s structured

          You are right that everyone gets two votes – a FPTP for their local constituency and a party vote. However the seats *aren’t* split. The total number of seats is allocated based solely on the party vote so it is still a proportional system*

          The difference is that the winner of a constituency is guaranteed to be in parliament and they have local responsibilities, while parties have free rein to select whomever they want to fill the rest of the seats.

          So for example, if a party wins 1/3 if the vote they get 40 seats no matter what. If they win 10 constituencies then they select 30 people from the list, whereas if they won 30 then they would only select 10.

          This means there is no* difference between a regionally concentrated of dispersed party.

          I keep putting * because there is a slight wrinkle. You need 5% to get into parliament unless you win a constituency; then you get your full percentage. So a party that gets 4.5% is either out of parliament or wins around 5 seats. Similarly, winners are guaranteed in, so a small party can have one or two more MPs than its overall percentage – which mostly comes into play around special Māori wards.

          It’s an interesting solution to distributing power based on national preferences while respecting local relationships

          Reply
    2. Hayek's Heelbiter

      PK,
      Please explain me to also why the Brits tolerate a system where a candidate can “stand” (not “run”, – this might explain the sclerotic nature of British politics) for an office but not only do not live in their constituency but also rarely actually visit it.

      Reply
  3. Mark K

    Two thoughts:

    1. Regarding Yves’ introductory comment that “America’s simple two-party system means we are unfamiliar with coalition strategies and complex preference-registering systems.” This is definitely true in terms of general elections, but less so for primaries, where there are routinely multiple candidates. It would be interesting to know how the 2020 Democratic presidential primaries would have turned out under ranked choice voting, or under most-least voting.

    2. I happened to click through the link to Terry Flynn’s original posting of this article on his website. The next article under it, “The 2021 Notts Labour Collapse – Both Simple and Complex” is fascinating, and well worth a listing in NC’s links. It’s a more formal write-up of a comment Terry made on May 7th’s Links (the comment at 5:50 pm — Terry made several comments that day.)

    Reply
    1. Terry Flynn

      Thanks for shout out. Although, being a Brit, I “naturally” am very interested in point 2, you’ll see I’ve posted elsewhere in the thread about point 1.

      The bottom line is that I am VERY interested in what American experts like Yves would say about how MLV might have played out in primaries in 2016 and 2020. I’m simply not knowledgable enough. I don’t say this as a “task” (against NC rules) but if any regulars can easily answer such questions…… :-)

      Reply
  4. Altandmain

    Bottom line is that the first past the post system has to go.

    It entrenches 2 parties that don’t represent people.

    PR with Single Transferable Vote combined with Swiss style referendums would be my preference. Lobbying laws should be hyper strict and there should a very tight leash on corporations, PACs, lobbyists, and other agents of rich people.

    Otherwise we get a “managed democracy” which is the status quo.

    Reply
    1. Fazal Majid

      I’m assuming you are in the UK. They had a referendum in 2011 for replacing FPTP with Alternative Vote, and it was rejected by 2/3 of voters. I know AV is not perfect, but it’s still way better than the current system, unless you are a Tory or Labour apparatchik, but sad to say the British people blew it, and to a large extent they deserve what they are getting (Scotland, Wales and NI also overwhelmingly voted against).

      Reply
      1. Terry Flynn

        Caveat: I was living in Sydney when the UK referendum happened. However, I was pretty horrified at how bad the “anti FPTP” case was. It seemed muddled and without a clear message.

        I’d have proposed MLV instead of AV. I’d have said two things:
        (1) All you have to do beyond the current system is state the candidate you hate most.
        (2) ALL votes, positive and negative, get equal weight; continue to choose your preferred party as “most preferred”, but if you think there is a candidate who should under no circumstances represent you, put them as “least preferred”.

        End of story. Of course large amounts of cash went into the campaigns. However, I think all/most non-Tory voters on the ground would have turned out and caused the referendum to pass. It WOULD NOT – as I say above – be “proper PR” – but neither was AV! And I think the “message” would have been simpler and would have passed. We’d now be arguing (with coalition govts) as to whether MLV is “the final step” or a “step towards full PR”.

        Reply
  5. Tom Doak

    A new system that favors centrists? Where do I sign up? /s

    Seriously, the example of Martin O’Malley becoming the Democratic nominee in 2016 is mind-boggling, but of course, had this system been in place, the primaries would have been engineered to make sure the centrist back-up candidate was someone the Democratic machine wanted. But it would have saved Obama from having to pull out the long knives in 2020, as so many would have united to disqualify Bernie.

    Reply
    1. Terry Flynn

      This is EXACTLY why I don’t go “ra ra ra MLV!” I’m merely pointing out how a small change to voting could lead to radically different candidates winning. As I was VERY careful to say, whether you think that is RIGHT, depends on a bunch of YOUR VALUES!

      THAT is what Arrow’s Impossibility Theorem says! I say all this as a firm supporter of Bernie.

      Reply
  6. Terry Flynn

    One final comment about why MLV might be PROFOUNDLY uncomfortable with people in the UK/USA who are used to “someone who at least got a high vote in terms of first preferences”. MLV could elect the BOTTOM candidate. Are Brits and Yanks ready for that?

    I’m not answering as I don’t know. I do know that centuries of tradition have established that “a high primary vote matters”. This could doom alternative systems like MLV and AV. It requires a lot of national debate.

    Reply
    1. Mikkel

      Yeah I think that cultural orientation has a big influence on what voters will accept…as well as how parties function.

      For example, Americans are taught to hold strong onto self constructed beliefs, try to win the war of ideas and then compromise if necessary. Thus directly electing officials and fetishizing bipartisanship makes complete sense.

      It was an adjustment for me moving to NZ (and having a Nordic wife) which has a culture that is more built on consensus rather than compromise. Generally speaking, when there is disagreement — particularly within the group — then it is expected that individuals gently adjust their underlying views rather than keep forcefully arguing. Compromise across groups isn’t as important, although it is acknowledged that group identity occurs on many scales. So proportional representation goes down easily.

      I’ve found many pluses and minuses to both cultures, as well as developing a skepticism that the differences are actually real when it comes down to it. However, it’s clear to me that Americans would tightly hang onto a “strongest leader” system rather than “least objectionable” one

      I think transferrable voting could potentially take hold but I’m not sure MLV ever could.

      Reply
  7. Sub-Boreal

    Despite referenda in multiple provinces during the past two decades, and a 2015 election promise by the winning Liberals to carry out electoral reform before the 2019 federal election, such initiatives have never gained sufficient traction in Canada to achieve implementation.

    Before Trudeau completely repudiated his promise, it was clear that he and his party favoured ranked voting as their preferred alternative to FPTP. This pointed out in a review of a 2017 book on Canadian electoral reform [ https://reviewcanada.ca/magazine/2017/05/why-trudeau-abandoned-electoral-reform/ ]:

    “Just as importantly, the Liberals certainly did not want a proportional system. It was never clear what Trudeau expected. There were indications that he was favourable to the idea of ranked ballots—the system whereby voters choose their favourites in descending order. It took little time for experts to predict, using past results and some imagination, that under such a system the Liberals would be guaranteed a place in government forever. It was a non-starter for the majority of non-Liberals on the committee.”

    It seems to be a general principle that in many countries the largest centrist party tends to be the least-hated alternative to a voter’s first choice, right across the political spectrum. Both this system and FPTP make it hard for minor parties with widely-dispersed support to elect representatives. Getting MPs elected by minor parties in national FPTP elections requires regional geographical concentration, favouring nationalist / separatist causes (cf. Scotland, Quebec).

    Reply
  8. ex-PFC Chuck

    Here in Minnesota, USA, there has been an active movement promoting RCV since the early 1990s. It was originally billed as “Instant Runoff Voting.” It first got serious public attention during the 1998 election for state offices, a contest most noted for the election to the governorship of Jesse Ventura of the Minnesota Independence Party. Ventura campaigned on the need for significant changes in the process of government, and his pitches resonated well with large sections of the populace. It helped that the two legacy party candidates were colorless mediocrities: Hubert Humphrey III, the sitting State Attorney General; and Norm Coleman, the mayor of St. Paul.
    The MIP ran a full slate of candidates for the state executive offices and its candidate for Secretary of State, whose name I sadly don’t now recall, ran a peripatetic campaign promoting “Instant Runoff Voting,” as the RCV concept was then called here. His message was well received, and although he didn’t win he did far better than all other MIP candidates for down-ballot state-wide offices. Ventura was a good governor in many ways but he did not pick his fights wisely. Instead of leveraging on the success of fellow slate member’s promotion of RCV, he instead bet his chips on converting the state to government by a unicameral Legislature by eliminating the state Senate. All to predictably, this went absolutely nowhere. Even if the MIP had elected some legislators, which they had not, there was no way the Senators would vote themselves out of jobs. Nor would the House members, half of whom could expect competition from recently defenestrated Senators. One of Minnesota’s bigger might-have-beens is where RCV would be here and in the rest of the country if Ventura had taken it up.

    Reply
    1. Terry Flynn

      As I have implied elsewhere – I am quite receptive to the use of MLV in what I’ll call “lower ballot” votes (State and county positions). This is not meant to demean those positions – merely to allow “opening up” of them to other parties and individuals, allow people who aren’t “mainstream beltway Democrat/Republican” to gain a foothold and get associated media exposure they wouldn’t normally be able to afford.

      I’m not espousing MLV as “THE solution”……merely as a possible way to help people “not in the party machines” to get exposure and allow the system to be shaken in a way that traditional ways (courts, major house of congress etc) have shown themselves unable to do.

      Reply
  9. Mikkel

    Terry, I don’t understand how MLV translates to power dynamics. Sure from a mathematical perspective it might make sense but actual governing requires mobilizing forces both across government and outside of it.

    In either FPTP or PR there is a direct relationship between votes and this mobilization power, but in MLV you could often have officials who functionally have no power base. This would surely lead to ineffective governance in an executive based system like the US. Has that been considered?

    Reply
    1. Terry Flynn

      That’s a good point….. It’s why I kinda think MLV might be best for primaries etc…. Having it as system for “primary power” could lead to problems you mention.

      Though shaking up the system with a party (green?) which had had some power. . . . Just not unbridled national power…. Might be what we need…. .

      Reply
    1. Terry Flynn

      Unfortunately rating scale (star systems or any numerical based scale) are terrible and another of my “issues”. Essentially, when you buy a can of beans at the store you NEVER go “4 out of 5 stars for that one….. 5 out of 5 for that one”. They are simply not how we make decisions. Choice modellers like me replicate the real-life decision-making context as closely as possible – or test a new context that is actually feasible or under consideration.

      Look up anything by Louviere and his co-authors for proof at how bad these “conjoint analyses” using numerical ratings are from the early 1980s onwards. When you buy beans you look at brand, ingredients, price etc and decide “buy/don’t buy”. It’s a DISCRETE choice – requiring a LIMITED DEPENDENT VARIABLE model, NOT a continuous outcome model. It should be analysed and understood as such. Same with all voting systems in industrialised countries. All that stats you learned regarding least squares and continuous outcomes like blood pressure or GDP? USELESS. WORSE than useless since you’ll instinctively believe heteroscedasticity is simply a “nuisance”. In logit/probit models it not “just” a nuisance but a fatal weakness. Which is why most of psephology is trash.

      For numerical star systems to “work” mathematically and properly reflecting strength of preference, they must be “cardinal”. 4 stars must be twice as good as 2 stars. Adding 2 stars to 2 stars must equal 4 stars etc. Yet when you look at frequencies of choice these are violated…..virtually all the time. People are totally lousy with numerical scales. Again, Louviere references. That’s before you even get to heuristics – older Chinese avoid “4” because the character has connotations with death.

      TL;DR Any voting system requiring you to give “numerical outcomes” (as opposed to discrete choices such as “pick one” or “rank these with no ties”) should be assumed to be trash. Extraordinary claims (and cardinality is an EXTRAORDINARY claim – 60 years of research have shown this in the math psy literature) require extraordinary evidence. I have not, in 20 years of research, seen any such evidence. And I’m not the “person with the longest experience” – people like McFadden and Louviere are. I’m not dissing you – I know those voting schemes seem very intuitive. But burrow into the maths and they are truly horrid, unfortunately.

      Reply
  10. The Rev Kev

    I think that the real problem with voting in America is that every four years, Americans face their own Kobayashi Maru scenario. Last time it was a choice between Trump & Biden. The time before that was Hillary & Trump. I sometimes think that you could pick ten random names out of the Republican party membership rolls and the same for the Democrats and have a markedly superior set of candidates. In three years time I would not be surprised to see people like DeSantis, Pompeo, Manchin & Harris stepping forward as candidates so what sort of “choice” is that?

    https://memory-alpha.fandom.com/wiki/Kobayashi_Maru_scenario

    Reply
    1. Terry Flynn

      Sometimes I think I need to get out more. *every* *single* *comment* you have made with a sci-fi origin, I got straightaway ;-)

      But I’ll never apologise for knowing borg comments. I can make borg jokes (like when we all order the same at a restaurant) with ANYONE and they get it. Is there ANY baddie in the history of art or cinema that has achieved such recognition?

      Reply
      1. The Rev Kev

        Terry – The Borg are truly in a class of their own in cinema history. Death is one thing. Extinction of self and being ‘recycled’ for their own purposes is another. Frightening when you think about them for awhile. Greetings from a Tertiary Adjunct of Unimatrix 01.

        Reply
        1. Terry Flynn

          hehe I just wish the ST:TNG movies and Voyager had not happened and “neutered” The Borg via the introduction of the Queen etc. People on YT debate recent ST troubles but as far as I’m concerned ST ended with DS9 finale and all subsequent series/movies beyond ST6 etc are non-canon.

          But I digress and I won’t subvert further Yves’s kind reposting of my blog!

          Reply
  11. SimpleNow

    My fear about MLV is that, for instance, the right would sponsor an uncharismatic Nazi to soak up the leftist ‘no way!’ voters, keeping their more moderate candidate in the running. Similarly, the left would push a shrill Woke scolder to attract anti-left ‘L’ vote.

    Reply
    1. Terry Flynn

      But this kinda illustrates the point of the Dutch/Belgian researchers who proposed MLV. Either the “proper” right-winger is effective (in which case the “least” votes will eliminate him/her) leaving the “uncharismatic puppet” or the “proper” candidate IS the uncharismatic one and although not getting a large negative “net rating” is by definition not going to be an effective politician – certainly not a POPULIST!

      Similarly for the left. The “shrill woke scolder” will get enormous least vote and be eliminated, leaving the “proper” left-wing candidate or the “proper left” candidate gets eliminated via high least count and the “woke” candidate is the “left” candidate remaining in the final tally. If that happens and he/she/it is elected……then emigrate…..sorry, I couldn’t resist a tasteless joke, I’m British, gay and hate wokeness.

      Reply
  12. Pekka Oksa

    Finland is holding its quadrennial municipal elections this year (on right now, in fact). It’s a kind of PR system. Councillors are elected ‘at large’ for the whole district. In my city, Tampere, the council has 67 members. Each of the main parties offers many candidates; there are also small parties and independents. This year there are a total of 512 candidates. Each voter chooses one from the list.

    Reply
  13. T_Reg

    I think there’s a way to achieve the effects of best-worst voting within the bounds of ranked choice voting.

    1. Use ranked-choice ballots.
    2. Allow voters to select the rankings in any order (including, 1 for their favorite, last for the most disliked). The obvious alternative, requiring selections be made first to last, would make the selection of last much more difficult.
    3. Assign points to the candidates based on their position; (total number of candidates minus their position on the ballot). The winner is the person with the most points accumulated across all ballots. This would eliminate the problem of a candidate receiving the fewest number of 1st position votes, who receives the overwhelming majority of 2nd position votes, being eliminated right away.

    …. Voting machines would be quite helpful here, both in preventing multiple selections for one position, and for allocating points when there is no one selected for a position (obvious approach would be to equally distribute the total points involved to the unselected candidates). However, I’m convinced that a paper trail (and, preferably, a paper ballot reviewed by the voter) is an urgent need. So, a voting machine should print a completed ballot.

    This technique would be easy to apply to multiple-seat districts using the single transferrable vote method; the winners are those with the most points. (It also eliminates the problem of reallocating votes when a candidate has received the necessary number of votes).

    Adding None Of The Above becomes an easy addition, although what to do if NOTA wins remains a bit of a quandary.

    P.S. NO PARTY PRIMARIES ALLOWED!

    I’m curious whether something similar is being done anywhere.

    Reply
    1. Terry Flynn

      Plus – voting machines. Yves will have something to say about them (and me for that matter). You know you are coming across in a rather bad light? I won’t label you here but others will.

      Reply
  14. Terry Flynn

    I think there’s a way to achieve the effects of best-worst voting within the bounds of ranked choice voting

    Sorry – not possible. The likelihood function as to what is “best” and “worst” is fundamentally different.

    . This would eliminate the problem of a candidate receiving the fewest number of 1st position votes, who receives the overwhelming majority of 2nd position votes, being eliminated right away

    And this is why you didn’t understand the post. I’m not claiming this isn’t a problem. ARROW’S IMPOSSIBILITY THEOREM SHOWS THERE WILL ALWAYS BE A PROBLEM. If you have a solution to Arrow please submit it to the Nobel Committee. I’m talking about 2nd best solutions. The real world.

    Assign points to the candidates based on their position; (total number of candidates minus their position on the ballot). The winner is the person with the most points accumulated across all ballots

    Please see my comment above regarding points. I’m perfectly open to a “Nobel” prize winner being wrong……but if you have the solution why are you posting on a blog rather than contacting Sweden? Sorry for the snark but just like the immunologists who are daily getting rubbish re covid, do you not think voting has not received 60+ years of checking all these “solutions”? Please give a worked example at the very least. I’ll give a constructive criticism of it with no snark.

    Reply
      1. Terry Flynn

        Try doing a citation search on Borda plus my other co-author on the book. It would have taken you 30 seconds. Stop quoting stuff you don’t understand. You appear as us Brits would say “like a twat”.

        Do you have any idea how frustrating it is after 20 years of PhD and postdoctoral research and experience and STILL recognising my omissions in highly technical stuff which makes me so angry when some random does a Google search and just pulls a long discredited theory out their arse claiming it disproves the work of the “god” of this area who I’ve tried to learn from and puts it forward as “the solution” from their armchair?

        You could have simply asked why borda methods do or don’t work. Instead you made a statement implying you had the solution.

        Reply
  15. Bijou

    Arrow’s Theorem only applies to generic voting. Fair results can be obtained if particulars are taken into account. When you only have a few candidates MLV is not what you’d go for. With a huge pool of eligible candidates, say 1000, all available for say 9 seats, then Cumulative vote tallying is ideal. It’s horses for course to get around Arrow. In other words, you select the most appropriate voting system for the size of the candidate pool and the seats being vied for.

    If one year only three candidates apply for election you might switch systems, if you’ve conditioned your voters to know why and how you’ve switched. To do so with minimum disruption you’d run a few rounds of dummy elections for practice — a one-time cost in effort. If people are too lazy to do this then just ONE of Arrow’s criteria can be chosen as lesser in importance to get a “fair enough” system.

    Also something polsci experts often fail to consider is degree of polarization. You don’t have to have just “like” vs “dislike”, you can have a Likert scale on degree of like/dislike, and use it to weight the votes, so that a polarizing candidate who is less polarizing than the other still has a chance to be ahead of the milquetoast centrist. I know, I know, requires fairly sophisticated voters, but worth a shot some time in experimental research trials.

    Reply
  16. Frederika

    San Francisco city government, supervisors, sheriff and district attorney are chosen by ranked choice voting. That, combined with district elections for supervisors, has resulted in a parade of ineffectual, sometimes dangerous, political mediocrities, a chaotic disaster, controlled by the Democratic County Central Committee.
    https://www.sfdemocrats.org/our-party/the-dccc

    If a voter fails to choose three candidates, their vote is thrown out. Some supervisors have been elected with less than 25% of the vote.

    Effects on the street: “In 2020, the San Francisco Police Department presented 6,333 felonies to newly elected district attorney Chesa Boudin’s office for prosecution. Since taking office Jan. 1, 2020 through March 1, 2021, Boudin has tried just 23 cases resulting in 16 convictions, including four assaults (three convictions); one auto burglary, one residential burglary, one gun felony (no conviction); three sexual assaults (two convictions); two robberies; seven mis- demeanor DUIs (four convictions); and one misdemeanor vehicular homicide, which he lost.”
    In 2019 during the same timeframe, Boudin’s predecessor, George Gascon, tried 294 cases and got 203 convictions.

    Prosecution and prevention of Gun violence? “Take the case of Zion Young. He was charged with 11 firearms felonies, which Boudin reduced to one misdemeanor and released him on an ankle monitor. Less than three months later, Young shot and killed 19-year-old Kelvin Chew, a student out for a walk in a botched robbery attempt.”
    http://www.capoliticalreview.com/capoliticalnewsandviews/chesa-boudin-by-the-numbers-the-un-da/

    Reply
  17. George Phillies

    While it is an interesting fad, there is no real guarantee that rigging elections to favor centrists will get you better government. As it happens, I am a Libertarian. Some of my ill-advised fellow party members argue vociferously for ranked choice voting or the like. I attempt to point out to them that RCV tends to guarantee that my party will never win elections, but the RCV faithful will not listen.

    Reply
  18. Jack Parsons

    This is mostly irrelevant. Our core problem is that gerrymandering allows party control to calcify. Nobody wants to fix this, because everybody wants to be the gerrymanderer.

    I have seen it claimed, via a game theory proof: if you are allowed to vote in neighboring districts as well as your own, gerrymandering stops working. It does make sense for state&federal assembly/congress districting, since we have become a very “non-local” society where actions in neighboring districts affect us.

    Reply
  19. Odysseus

    There’s a lot of talk about candidates and parties, but not a lot of talk about policy.

    One way to create significant momentum to deal with global climate change is to place high taxes onto fossil fuels. As Illinois recently demonstrated, this is highly unpopular.

    In either Ranked Choice or Most-Least systems, how do necessary but unpopular policies get enacted?

    Reply
    1. hunkerdown

      It’s an error, although one encouraged by our gentry classes, to conflate the system of elections with the system of government. Aside from the necessary operational and organizational contradictions, they can vary independently.

      If policies are necessary to elites, there is a 2/3 chance they will be enacted if introduced. If policies are not necessary to elites, there is a 0% chance they will be enacted if introduced. (Gilens and Page 2014) That is a feature of the liberal republican form of government, not a corruption of some ideal. “Unpopular” policies are never necessary except to preserve elite domination, which is unnecessary.

      Reply

Leave a Reply

Your email address will not be published. Required fields are marked *