Should I vote for ranked-choice voting?

by philg

The first of my five mail-in ballots has arrived. 3 out of 7 candidates are Democrats running unopposed. The remaining 4 races are 100 percent guaranteed to be won by Democrats. A potential contest: Question 2 is whether to adopt ranked choice voting.

As someone whose political beliefs are most aligned with the libertarians, a last-choice party in a nation where people want a planned economy (my 2012 document after watching both the Republican and Democrat candidates promise that government would create jobs, ensure fair wages, etc.), is this for me? I could vote for a libertarian candidate and then also pick a second choice from a party that has a chance in a country whose citizens want government to cater to their every need? Yet in a Massachusetts general election it is almost inconceivable for a non-Democrat to win. So how can this have any practical effect?

The “Independent Women’s Law Center” opposes this question. We don’t know what people identifying with the remaining 50+ genders say. Wikipedia says that Estonia had something like this, but abandoned it in 2001. As government in Estonia is radically more efficient than here in the U.S., that’s a strike against the idea.

Readers: What do you say about this proposal?

Update, 10/16: a friend highlighted “The Ancient Greeks Teach Us The Perils of Ranked Choice Voting”, by a political science professor:

As this list [of supporters] makes clear, RCV supporters fall overwhelmingly into two (mostly overlapping) categories: Democrats and groups whose members vote heavily for Democratic candidates; and groups that (Libertarians aside) have practically no chance of winning elections even under RCV, except at the local level. Given Massachusetts’s status as a heavily Democratic state (the state’s congressional delegation consists solely of Democrats, who also have long held a substantial supermajority in the legislature), Democrats have little to fear from losing elections to Republicans as a result of RCV. Rather, they need to court members of left-liberal fringe groups, as well as public-employee unions, to ensure that they turn out to vote — knowing that those groups’ supporters would almost surely make the Democratic candidate, at worst, their second-choice candidate, further guaranteeing the defeat of any Republican contender.

32 thoughts on “Should I vote for ranked-choice voting?”

    • Alex: I agree with you on Question 1, but I also agree with opponents who say that this means the clever Russians will get hold of stupid Americans’ driving-related data! I do think there will be arguments about what “telematics” means.

    • @Philg: I think that’s probably true, but the Clever Russians can probably also find a way to get at the data directly from the manufacturers. There’s risk involved to be sure. For example, I use software called Forscan to read all the CAN network data / OBD-II data on my Escape Hybrid. I believe the software was written in Russia. No ill effects so far, and Forscan really lets you interrogate and control all the systems of the car. I use it to monitor the state of charge of the NiMH High Voltage battery and all the other systems. Otherwise I’m at the mercy of a Ford dealership for a lot of questions with this vehicle.

      Risks everywhere, but I think the automakers are just trying to totally monopolize access to the data for their own benefit, so that they can extract fees from independent repair shops and sell the data to advertisers, etc., which they have baked into their business models. Well, they can still do that, but I think they should have to ask for permission and individuals should have access to it also. I don’t buy the scare tactics.

      https://www.bostonherald.com/2020/09/25/letter-scare-tactics-have-nothing-to-do-with-car-repair/

      “In fact, the data covered by Question 1 is identical to the data that automakers have been sharing for years under Massachusetts’ existing right to repair law.”

      https://forscan.org/home.html

      In fact, I think you should be able to turn the telematics OFF. If the automakers don’t want to do that, then they should begin by being honest about the real reasons they want exclusive access to the data. They’re lying up front. Not a good way to deal with people.

    • Alex: Okay, you’ve convinced me that voting Yes on Question 1 is a good idea! Let’s see what other readers have to say about ranked choice. I do wonder what difference it makes in a one-party state in which it is rare to have even two candidates for a political office.

    • Well, our governor does call himself a “Republican”. But I’m not sure that someone from any other state would recognize the policies that he advocates as anything different than what a person calling him/her/zir/theirself a “Democrat” would advocate.

      Maybe people here keep voting Charlie Baker in so that they don’t have to pay a progressive income tax rate, actually work extra hours to pay reparations, give up our big houses so that public housing can be provided to everyone eligible (not put them on a 10-year waiting list or say “the waiting list is full”), etc. We can have a legislature that advocates for all manner of virtue and then the governor can be a convenient excuse for why we don’t do the things that we say we want to do.

  3. Yes, you should vote for it. The reason third parties and independents don’t have a shot is because the system prevents them from growing their support. When they run, they are shamed as “spoilers,” and their supporters are told they have to vote for the “lesser of two evils.” That leads to a cycle of delegitimacy that is near-impossible to overcome. The only way to break that cycle is to enable voters to vote their true preference without wasting their vote.

  4. Thanks to Alex for the link to the Heritage Foundation paper. I’ll present the other side of the argument, since I disagree with that paper almost in its entirety!

    While ranked-choice is perhaps not the optimal system, I think it’s easily and clearly superior to the system we have now, and we should not let the perfect be the enemy of the much-better.

    The current system serves mostly to ensure that one of the two (or one, in the case of Massachusetts) major-party candidates will nearly always win, especially in races that are almost-equally split. Nobody will risk “wasting” a vote on a minor-party candidate if it could result in the Great Satan being elected. They’ll hold their nose, vote for the lesser-of-two-evils major-party candidate, and the minor-party candidate that people actually prefer will continue to poll at ~0.

    Where third-party candidates manage to get elected despite this, it’s often because they gained their name recognition while in the two-party system, then broke away. In any event, I think we can all agree that it’s extremely rare.

    In an information-theory sense, a ranked-choice-style system provides far more bits of information from each voter that can be used to try to optimise for overall electoral preference. There’s far more “signal” in a ranked-choice ballot.

    All of the “Cautionary Examples” in the Heritage paper sound to me like superior outcomes:

    > The liberal Labor Party won the Australian House despite receiving only “38 percent of first-place votes on the initial ballot, while the second-place Liberal-National coalition [the center right choice] captured 43 percent” of first-place votes.
    >
    > In other words, more voters wanted a center-right government than a left-wing government, but ranked choice made sure that did not happen.

    The authors are assuming facts not in evidence.

    First, Australians don’t vote for a party, they vote for a Member of Parliament (just as Americans vote for a Congressman, not a party). One cannot assume that a vote for an MP is necessarily an endorsement of their party. It may be related to local politics, or an endorsement of that candidate’s personal policies / persona, or simply a dislike for the opponents.

    Second, their statement ignores the fact that the Liberal-National Coalition is a diverse urban/rural coalition of often-competing interests. There are sometimes multiple Coalition candidates on a single ballot (e.g. a Liberal party candidate, a National party candidate, and even one or more allied yet-more-minor-party candidates). Although some combination of Coalition candidates may have collected 43% of the first-place votes in their example, there’s no evidence that any of them could’ve been elected one-on-one against Labor. If that were the case, then one would expect the voters to have ranked the L-N candidates consecutively at the top of the ballot, and thus one of the L-Ns would’ve been elected — but that didn’t happen.

    This suggests that the correct outcome was probably reached, or at least there’s no evidence that it wasn’t.

    > Or consider the mayor’s race in Oakland, California, in 2010, in which the candidate that received the most first-place votes lost the election to “a candidate on the strength of nearly 25,000 second- and third-place votes” after nine rounds of redistribution of the votes.

    They try to make this sound like a miscarriage of justice bordering on electoral fraud, and use scare-italics for “nine rounds of redistribution” — but somehow they never get around to actually explaining why we should think this is bad.

    The 25,000 second- and third-place votes helping someone win is a demonstration of the system working exactly as it should, and in a manner quite superior to the standard US system. Those voters’ first-place candidate wasn’t electable, so instead of their votes being wasted, they got to be counted for their next-most-electable choice. How is this a bad thing?

    If first-place votes were what mattered (i.e. the system in most of the US), then people would have voted differently (see below about game theory).

    But in the Oakland mayoral race, the percentage of first-place votes didn’t matter to the outcome, which gave people the freedom to express their actual preferences without fear that their vote would be “wasted” on an unelectable candidate.

    The Heritage argument sounds to me like the same one Democrats make when their candidate wins the popular vote but loses the electoral college, and it’s no more relevant here.

    If the voting system worked differently, then candidates and voters would behave differently. You can’t take a result achieved under one system to try to predict the outcome under a different system, because the system itself alters everyone’s behaviour.

    Long story short, the authors provide no evidence whatsoever that this outcome doesn’t represent the collective will of the voters, or that an identical result wouldn’t have occurred in a head-to-head race, or a delayed-runoff.

    > However, the Maine Secretary of State, Matt Dunlop, “exhausted” or threw out a total of 14,076 ballots of voters who had not ranked all of the candidates.

    Sounds good to me!

    The system in most Australian states requires that you number all the boxes or your ballot is discarded in its entirety. I strongly favour the so-called “Optional Preference” system employed by Maine in which your vote is counted for all the boxes you number, and your vote is discarded only if all of those candidates are out.

    This is no different from you deciding to stay home on runoff day, if all of the people whom you support have been eliminated. (Or skipping the general, if your preferred primary candidate isn’t on the ballot.)

    More to the point, the current prevailing voting system in the US could be described as a degraded ranked-choice system in which voters are allowed to number only one box, after which their ballot is exhausted. How is that better? Again: it provides much less “signal”.

    > Second- or third-choice votes should not matter in America; they do not provide the mandate that ensures that the representatives in a republic have the confidence and support of a majority of the public in the legitimacy of their decisions.

    I expect better than “argument by assertion” from Heritage.

    If my first-choice is eliminated, why does that mean any less of a mandate for my preferred second choice? Particularly when the system is designed to get me to reveal my true preferences regardless of electability.

    Furthermore, this argument applies just as well to their preferred “delayed runoff” voting scheme. It’s still second- and third-choice candidates being elected, just a few weeks later. So what? That’s the reality of representative democracy sometimes.

    …

    Last but not least, they decry how ranked-choice voting requires people to engage in “game theory” to work out the most strategic way of voting in light of the potential down-ballot match-ups:

    > … and forcing average American voters into the world of mixed strategy game theory, where they are forced to try to predict the probability that particular candidates that they favor or do not favor will survive multiple rounds of vote tabulation.

    What they fail to mention is that today’s system is already plagued by this problem. Supporters of any candidate other than the major-party offerings are constantly forced to weigh whether their true preference has any chance of being elected under the current system. Since the answer is almost always “no”, they strategically cast a vote that does not represent their actual preference.

    In what way can that possibly be better than allowing voters to express their actual preferences?

    > The answer to this gimmickry is runoff elections. In the normal electoral process in the vast majority of states, there is a runoff election several weeks after a general election in which no candidate won a majority of the vote.

    The problem with this is that it still contains the worst flaws of the current system, without actually solving any of the purported issues they have with RCV.

    Why? Because in practice you can only have one runoff. You can’t eliminate one candidate at a time and hold election after election until you get down to a final pair. We don’t have the time, money, or voter patience to hold n-1 votes each cycle.

    Thus a runoff of the type that Heritage proposes still has the same problem of how to strategically vote in the first election, based on your prediction of who will make it to the runoff.

    One example is where there’s a strong incumbent who’s guaranteed to make it to the second round, but in a highly-fragmented race where nobody can get 50% on the first ballot. In this case it can actually be strategic to vote for someone you hate in the first round, to try to engineer the weakest possible opponent for the incumbent. This is an improvement!?

    …

    In my opinion, implementing RCV-style voting systems in as many states as possible — particularly states where citizens can bypass the political class and place binding initiatives on the ballot — is perhaps our only hope of breaking the dysfunctional two-party duopoly that is serving our country so poorly.

    Do I think it will actually help anything? Probably not, no. But making it easier to elect candidates not in the major parties, however marginally, certainly can’t hurt.

    The defence rests.

    • I wrote a long post that went to moderation, and I hope if the moderators approve it they can fix the formatting because it looked awful in the moderation preview, don’t know why that happened. I think what Oakland “proved” was that the RCV scheme to steal elections by declaring the loser the winner worked, which is why they want to expand it.

    • Asserting that an election was stolen doesn’t make it so. What evidence exists that the result of the election doesn’t reflect the (explicitly expressed!) preferences of the voters?

      You seem to be saying that the only legitimate winner of an election is whomever gets a plurality of first-round votes, but that’s an arbitrary definition. That may be tradition, but there’s nothing about that particular rule set that makes it philosophically (or for that matter mathematically) optimal.

      Let’s perform a hypothetical election for mayor of a small town with only 100 eligible voters, whom we gather in the school gymnasium.

      There are three candidates. On the first ballot, the votes are cast:
      – Candidate A: 31 votes
      – Candidate B: 40 votes
      – Candidate C: 29 votes

      With nobody enjoying 50%+1 support, they send candidate C home and vote again.

      On the second ballot, everyone who preferred candidate C puts their support behind candidate A. “If candidate C is out,” they all say, “then we definitely want candidate A. Anyone but B!” And the result is:
      – Candidate A: 60 votes
      – Candidate B: 40 votes

      In a typical American voting system, candidate B wins on the first ballot and takes office with a 40% mandate, and 60% of the town are unhappy with the outcome.

      In our small-town runoff example, candidate A wins and goes to work with a 60% mandate.

      Would you say that this election was stolen?

      If you say yes, it was stolen, then in what way? The manner in which votes would be cast and counted were agreed in advance. No rules were broken. There was no skullduggery. Candidate A clearly enjoyed broader support. The voters were crystal clear that, in a two-man race, they prefer A over B by a 60-to-40 margin.

      If you say no, it wasn’t stolen, then how is the Oakland election any different? The only difference I see is that the runoff happened in a computer instead of a gymnasium.

    • If the town wants to do a runoff election and candidate A wins in the end, that’s OK.

      > In a typical American voting system, candidate B wins on the first ballot and takes office with a 40% mandate, and 60% of the town are unhappy with the outcome.

      And that’s OK too. You might not think that’s an optimal outcome, but politics is a game of give and take, isn’t it? What it means for Candidate B in practice is that yes, they won the election but they enter office understanding that they didn’t win a majority, they only won a plurality, and 60% of the people are skeptical of them. Fine. This means they will take office, but they should know that it’s a tentative prospect and reasonable people would say that they should govern with that in mind, because in the next election A and C will get together and vote them out, or will otherwise protest their policies through checks and balances. But they won fair and square.

      Cobbling together a synthetic 50+1 majority through RCV makes the process through which one’s ballot actually translates into a “vote” very obscure. A runoff election is clear: two candidates with nonmajority face off for Round Two and one of them wins, clearly. RCV doesn’t do that. As the Tufts paper points out very clearly, it means that nobody knows when they leave the polling place which candidate will get their vote. It puts everything into a tabulating machine and a winner emerges at the end, but nobody can really tell how that result was achieved. This is very wrong.

    • > You seem to be saying that the only legitimate winner of an election is whomever gets a plurality of first-round votes, but that’s an arbitrary definition.

      No, it’s isn’t. It’s pretty basic, but it’s not arbitrary. It’s what *you* call “arbitrary” but to most people in the world, whoever gets the most votes won the election. If it’s arbitrary, let’s think of other arbitrary definitions we could use:

      “The winner of an election is someone who died.”
      “The winner of an election is someone who got less votes than anyone else.”
      “The winner of an election is someone in a place where nobody could vote.”

      Those are arbitrary definitions, too. I’ve read that in North Korea, Kim Jong-Un always gets 100% of the vote in their extremely ranked choice voting scheme.

      https://www.independent.co.uk/news/world/asia/kim-jong-un-wins-100-vote-north-korean-elections-9180814.html

    • > And that’s OK too. You might not think that’s an optimal outcome, but politics is a game of give and take, isn’t it?

      Absolutely. If the rules of the election are that a single-ballot plurality wins, then I completely agree that that’s a fair outcome.

      …but I wasn’t the one trying to argue that elections were being stolen!

      > As the Tufts paper points out very clearly, it means that nobody knows when they leave the polling place which candidate will get their vote.

      Only by a particular definition of what a vote means. This argument is again starting from the perspective that the current system is somehow holy, instead of just one set of rules among many possible options.

      The reality is that voters in an RCV system leave the polling place knowing exactly how their vote will be assigned, because the rules are neither mysterious nor complex, and the voter expresses their preferences in a simple and clear manner.

      It does raise the bar for voting, by limiting it to people who can count. Perhaps even as high as 20 or 30 in an extreme case. I don’t think this is a bad thing.

      > Cobbling together a synthetic 50+1 majority through RCV makes the process through which one’s ballot actually translates into a “vote” very obscure. A runoff election is clear: two candidates with nonmajority face off for Round Two and one of them wins, clearly. RCV doesn’t do that.

      What is unclear or obscure about it? Is there any evidence that anyone involved in this discussion doesn’t understand the way in which votes would be counted?

      If you accept that a runoff is clear, then in what way is RCV less clear? It’s exactly the same as a runoff, except that instead of removing all but two candidates in one fell swoop (creating the game-theory complications discussed previously), we remove only one at a time and iterate.

      > It puts everything into a tabulating machine and a winner emerges at the end, but nobody can really tell how that result was achieved. This is very wrong.

      I don’t see how RCV is any less at the mercy of tabulating machines than a single-round plurality result.

      In both cases, the rules are agreed in advance of votes being cast. In both cases, we’re trusting that local election officials (properly monitored) will collect and report votes accurately. In both cases, we’re trusting that central tabulators (properly monitored) will combine the local results accurately. In both cases, there should be a paper trail of actual ballots (something we sadly lack in many jurisdictions under our current system), which can be reviewed and re-counted if there’s any doubt about the veracity of the original process. What’s the difference?

      No citizen can personally verify the election results, regardless of the system we use. In any system, the citizens at large are trusting a large bureaucratic machinery to perform correctly, and (perhaps moreso) trusting in the mutual distrust of competitors to minimise the risk of cheating. In even a medium-sized state of several million people, what’s the alternative?

    • > Only by a particular definition of what a vote means. This argument is again starting from the perspective that the current system is somehow holy, instead of just one set of rules among many possible options.

      Thank you for admitting that RCV fundamentally changes the definition of a vote. If people want to think that’s unholy, there’s nothing I can do to stop them.

      >The reality is that voters in an RCV system leave the polling place knowing exactly how their vote will be assigned, because the rules are neither mysterious nor complex, and the voter expresses their preferences in a simple and clear manner.

      They are? How many people have read the Act in Massachusetts well enough to understand it? It may not be mysterious to you, because you’re obviously well above Grade 14, but I would like to know if you think people should debate these things on election night:

      Section 2A. As used in this section and section 2B, unless the context otherwise indicates, the following terms have the following
      meanings:

      “Active preference” means a vote assigned to the highest continuing ranking on a continuing ballot.

      “Batch elimination” means the simultaneous defeat of multiple candidates because, with respect to the candidates, one of the following applies:
      (a) The candidate cannot be elected because the candidate’s active preference total in a round of ranked-choice voting tabulation plus the total of all continuing ballots that could possibly be transferred to the candidate in future rounds from candidates with fewer active preferences or an equal number of active preferences is not enough to surpass the candidate with the next-higher active preference total in the round; or
      (b) The candidate has a lower active preference total than a
      candidate described in subparagraph (a).

      “Concluded ballot” means a ballot that does not rank any continuing candidate, contains an overvote at the highest continuing ranking, or contains two or more sequential skipped rankings before its highest continuing ranking.

      “Continuing ballot” means a ballot that is not a concluded ballot.

      “Continuing candidate” means a candidate who has not been defeated.

      “Highest continuing ranking” means the continuing candidate with the highest ranking on a voter’s ballot.

      “Last-place candidate” means (a) the candidate with the lowest active preference total in a round of the ranked-choice voting tabulation, or (b) a candidate that is defeated in batch elimination.

      “Overvote” means a circumstance in which a voter ranks more than one candidate at the same ranking.

      “Ranking” means the number assigned on a ballot by a voter to a candidate to express the voter’s preference for that candidate. Ranking number one is the highest ranking, ranking number two is the next-highest ranking and so on.

      “Round” means an instance of the sequence of voting tabulation steps established in subsection 1 of section 2B.

      “Skipped ranking” means a ranking not assigned on a ballot by a voter to any candidate, in the circumstance where a lower ranking is assigned on a ballot by a voter to a candidate.

      > If you accept that a runoff is clear, then in what way is RCV less clear? It’s exactly the same as a runoff, except that instead of removing all but two candidates in one fell swoop (creating the game-theory complications discussed previously), we remove only one at a time and iterate.

      That’s not true at all. It’s not like a runoff in any sense, because the rankings that people make on their ballots are not known to each other, and it’s the tabulating machine that does the counting. In a runoff election, you two boxes, and the voter chooses one or the other, and the results are very clear.

      Keep going…you’re doing your best to prove me right….

    • > No, it’s isn’t. It’s pretty basic, but it’s not arbitrary. It’s what *you* call “arbitrary” but to most people in the world, whoever gets the most votes won the election.

      Of course it’s arbitrary. Or if you prefer, discretionary. Most of society’s rules are arbitrary / discretionary. That doesn’t make them bad. There’s just nothing special about them, apart from tradition. Most of our rules have no fundamental grounding in mathematics or physical laws, or in most cases even in morality or philosophy. They cannot be proved or disproved as optimal. They’re just what we’ve mostly settled on for now, and we could just as easily settle on doing things differently. How we count votes is no different.

      And let’s also be clear: with ranked-choice voting, whomever gets the most votes wins the election. You may choose to define someone’s “vote” as being limited to their first-round preference, but that’s your definition (or if you prefer, the definition of the current system in most places). But there is nothing holy or sacrosanct about the first round. One-round voting didn’t come down the mountain on stone tablets. It’s not the speed of light, or e=mc².

      > If it’s arbitrary, let’s think of other arbitrary definitions we could use:

      What’s your point? That there are many arbitrary rules that we could choose for how to count votes? That some are better than others, and by examining them on their merits, we can choose rules that we think will produce better outcomes? I agree entirely! That’s actually my whole point.

    • I’m also glad to see you haven’t rested! Since you rested your defence earlier, I’ll a grant you a mulligan for the rest of commentary just to a be good sport. 🙂

      Don’t you agree that very few of the people voting for the ballot initiative (written at the 14th grade level, just as a summary and a condensation) will ever read or understand the actual wording of the law? What is your estimate for the percentage of voters who actually comprehend the law they’re voting for? And if they don’t, how will any of them ever know when a sentence gets changed and what that means?

    • > No citizen can personally verify the election results, regardless of the system we use.

      Wrong again. In my little town, we use paper ballots that are marked with an ink pen and then fed into a optical scanning machine that reads them and instantly – within 10 minutes after the election – gives an exact answer of how many people voted for whom. And those can be individually verified and recounted if necessary.

      The RCV scheme instead necessitates that the ballots all go to a centralized location where the ranked choices are “tabulated” and the winner is announced. This is also fundamentally different from how it is done now and had been done successfully for a long time.

    • Every city and town in Massachusetts votes on paper ballots, which can be audited and hand-counted later if necessary, and that would not change under ranked choice voting. And on election night, all the first choices on the ballots would be counted in the same way plurality ballots are counted today — in the case of your town, that would be by optical scan count.

      If no candidate has a majority of the first choices, then the instant runoff is needed. But you are incorrect that the ballots from your town themselves would be sent centrally. Copies of the ballot data files, spreadsheets known as “cast vote records,” would be. Your town’s ballots would remain would remain in your town, as they are today. Also, all of the data files — all the cast vote records — would be published publicly, just as they are in RCV jurisdictions around the country, for a level of transparency above and beyond that offered by plurality elections today.

    • > Most of our rules have no fundamental grounding in mathematics or physical laws.

      Tell that to a nuclear arms agreement inspector from the IAEA. Or anyone from the NIST. I know, it’s not the same as election law.

    • > The reality is that voters in an RCV system leave the polling place knowing exactly how their vote will be assigned, because the rules are neither mysterious nor complex, and the voter expresses their preferences in a simple and clear manner.

      Maybe you should take this up with the authors of the paper at Tufts University, because they said the exact opposite:

      “When voters leave the polling place they won’t actually know which candidate will get their vote: their top choice or their #7.”

      That’s what they said, not me. It could be that they’re wrong, but it’s right there in black and white.

    • > But you are incorrect that the ballots from your town themselves would be sent centrally.

      Again, that’s not what the analysts at Tufts said at all:

      “But anytime the front-runner has less than 50% of the first-round vote — and the multi-round ranked choice counting procedures take effect
      — all of the ballots would need to be moved to a central location so that rankings from each individual precinct could be combined and tallied. Alternatively, electronic records containing the ranking information from all ballots could be gathered in a central place.”

      https://tischcollege.tufts.edu/sites/default/files/cSPA_ranked_choice_voting.pdf

      If you want to convince someone, convince them, not me.

  5. I talked to three people this afternoon for their opinion. The consensus was: “Yeah, that’s the BS system they’re using to steal elections in Maine. The Mafia couldn’t have thought of a better way to do it.”

    You just pack the ballot with unattractive second and third tier candidates and let the money pour in, hoping to keep subdividing the vote and force the election into the byzantine counting process. At the end, the winner of the election could be someone who got 10% of the vote.

    I’ve read it all. The Tufts analysis (which for all the negatives it identifies, including the Constitutionality, it still manages to not take a position on it (what a surprise), the 9 pages of the Act, what you have so graciously provided above, and quite a lot more and my answer is:

    “No. And Hell No.”

    Trust. Switching to ranked-choice voting would add a level of abstraction to the voting process. When voters leave the polling place they won’t actually
    know which candidate will get their vote: their top choice or their #7.

    This raises a risk that voters will feel unsure whether their ballots were counted properly — or even uncertain about the legitimacy of results as a
    whole. Here again, the research on voter confidence is limited, but one survey of Maine voters found a substantial drop-off in confidence after a
    2018 congressional election that was decided by multiple rounds of vote reallocation.7

    Logistical hurdles.
    Under our current system, individual precincts can tally their own votes and pass final numbers to the secretary of state’s office. This should still work in elections with just two candidates or those where the front-runner gets a clear majority after the first count.

    But anytime the front-runner has less than 50% of the first-round vote — and the multi-round ranked choice counting procedures take effect — all of the ballots would need to be moved to a central location so that rankings from each individual precinct could be combined and tallied. Alternatively, electronic records containing the ranking information from all ballots could be gathered in a central place.

    This would add a layer of logistical complexity, in managing the transfer of physical ballots or ballot records. It would also require new spending for
    secure transport, appropriate counting machines, and central counting facilities. Question 2 does allow for the possibility of electronic records, but the secretary of state’s office notes that over 50 municipalities still use
    hand-counted paper ballots. And even if we did capture digital versions of all ballots, the electronic records might still have to be physically transported to avoid cybersecurity risk.

    And a great deal more. This is why so many places that have tried it have repealed it.

    https://tischcollege.tufts.edu/sites/default/files/cSPA_ranked_choice_voting.pdf

    Read the full text of the act at the Ballotpedia page. It’s 9 pages long.

    https://ballotpedia.org/Massachusetts_Question_2,_Ranked-Choice_Voting_Initiative_(2020)

    The system is flawed but it’s something people can understand. The candidate who gets the most votes wins, and if there’s no majority you have a runoff. I don’t want my 7th choice pick being declared the winner. It’s regrettable that some third party and independent candidates can’t gather enough support for their positions to win or make a dent, but this is not voting. This is machine tabulation. It’s nonsense upon stilts.

    I’ll tell you what: the ballot initiative is written at the 14th grade level and has an FRE of 42, which puts it on par with papers written by economists and chemists ( https://en.wikipedia.org/wiki/Flesch%E2%80%93Kincaid_readability_tests ).

    When the teacher’s unions decide to go back to school and the average reading ability of a MA voter rises to Grade 14 so they can understand what they’re voting for, I’ll consider it.

    I’ll have more later, I’m pretty busy this afternoon and whenever I post too much in a thread, people get upset, even though the cost of having my stuff on this screen is $0.

    Speaking of expense, that’s also what this is going to mean.

    https://mainepolicy.org/press-release-new-analysis-shows-ranked-choice-voting-fails-to-live-up-to-its-promises/

    • In terms of what Mainers think of the reform, they already told us on multiple occasions. I 2016, it passed with 52% of the vote. Two years later, after using it once, they reaffirmed their support at the ballot with 54% voting in favor. Then, in November of 2018, exit polls had it up to 61% of people prefer it to plurality voting.

      That’s a 9% increase after one election cycle. The more Mainers are exposed to it, the more they like it.

      The links you are providing are to very conservative think tanks that have a partisan opposition to it, not because of a disagreement on the merits.

    • @Greg: And it’s not a surprise that the links to supporters in Maine are Democrats further into the doppler shift left. We know that RCV is designed to upend plurality. It means that one vote is many votes. It’s a fraud. I can’t help it if more people like fraud.

  6. @Alex The argument that it’s too much of a burden for localities to send copies of their ballots to a central location to be merged is just idiotic. Why is it any different at all for localities to send their single-choice vote summaries, or to send the ranked ballot data? Sending all of that data takes about 1 second using this new fangled thing the kids are calling “internet”, don’t know if you have tried it yet. Security courtesy of https and public key encryption that you probably trust for your online banking and healthcare records already.

    More transparency is better, the public *should* be able to see what the raw vote data is from each polling place. It’s kind of shady that it’s so hard to get that data now.

    . You seem hung up on the terrifying prospect that the definition of a ‘vote’ is changing. Of course it is being changed. That is the whole purpose of this exercise. What kind of circular reasoning argument is that?

    • It’s not a circular argument, it’s a fundamental change in the way votes are cast and counted, introducing a completely new tabulation method and eliminating plurality. Even the Tufts paper talks all about that.

  7. Ranked choice voting didn’t work so well in Oakland, CA ten years ago. Jean Quan won the election amidst two stronger front-runners of opposing parties, simply because she was most everybody’s second choice. She proved to be at best a mediocre mayor, and was voted out in the next election.

    • Jean Quan was the strongest of the challengers in that race, and the correct representation of what voters wanted at the time. The incumbent, Don Perata, was not popular, which is why he had so many challengers to his incumbency, and she had the most first choices of any of the challengers. The final round showed that more voters preferred her to Perata than the other way around. See: https://www.acgov.org/rov/rcv/results2010-11-02/rcvresults_2984.htm

      And then, yeah, in 2014 voters decided they weren’t happy with her performance and voted her out. That was also the correct representation of what voters wanted at the time. I think it’s a good example of how elected officials don’t seem to have the same lock on incumbency under ranked choice voting as they do under plurality today.

      Voters are perfect at predicting which candidate will be the best at holding office. And there are certainly plenty of people elected under the current system who we could point to as mediocre! 🙂 But following the will of the majority of the voters is the best we can do — it’s a better predictor than the current plurality system, which allows someone to win with less than a quarter of the vote.

    • Well Greg, it’s unconstitutional also and I hope if it passes in Massachusetts thanks to people who can’t read well enough to understand what it really means, that a vigorous challenge is brought and it’s overturned. We know what it’s about: it’s about overturning plurality by packing the ballot with candidates that can’t win, hoping to get the 2nd place candidate over the top. The progressives are all behind it for precisely that reason, and the rest of this stuff is the academic argument they’re using to “add bits of information” to the ballot and turn the voting process into a tabulation scheme. And I suspect you know that’s true, which is why you want it so badly.

      It’s the best way anyone ever thought of to say:

      Candidate A would have won with 40%, but when we added up Candidate B with 35%, candidate C with 10%, candidate D with 7%, Candidate E with 5% and candidate F with 3%, it looks like Candidate B won.

      That’s all it is, folks. Vote for it if that’s what you want, Philip.

  8. At my local Rod and Gun club with about 400 members, each year we vote on a slate of new officers and Board members and last year, the idea for ranked-choice voting was proposed during a meeting, so we had a vote on it. The results were approx. 95% no, 5% yes. We only have runoffs if there’s a tie, and otherwise everyone accepts that the person with the most votes won the office. If someone wins the Presidency of the Club with just a plurality and not a majority, they know that they have to govern the Club comprised of 399 other gun owners while not making them too angry, so we don’t worry much about it. 🙂 The differences in the “platforms” usually amount to things like:

    “Should we add 3 new stations to the pistol range or extend the rifle range out another 100 yards this year?”

    Armed citizens are polite citizens.

Comments are closed.