Franklin Countys First News

Politics & Other Mistakes: How not to fix voting

Al Diamon

If only there was a way to hold elections that didn’t produce crappy results.

Unfortunately, there isn’t. That doesn’t mean Maine is going to accept whatever septic-tank overflow the system spews out. Currently, the state is divided over whether to retain plurality voting (the awful candidate with the most votes wins) or expand ranked-choice voting (the awful candidates are ranked in order of awfulness with the least favored eliminated until somebody awful secures a majority of the remaining ballots). In general, Republicans favor the former, while Democrats support the latter. Those not enrolled in a party tend to think both methods suck, and neither guarantees the winners will be even remotely qualified.

Social scientists have been studying this problem for centuries. They’ve made lots of interesting discoveries (I’m lying about the interesting part), which indicate that partisan voters are always going to game the system. As Charles Dodgson – a mathematician better known as Lewis Carroll, author of “Alice’s Adventures in Wonderland” and “Through the Looking Glass” – noted back in the 19th century, most voting methods are vulnerable to strategic balloting, which “make[s] an election more of a game of skill than a real test of the wishes of the electors.”

Take, for example, a trendy new proposal for something called approval voting. Actually, it’s not trendy or new at all. According to a Lewiston Sun Journal story, the Vatican has used it since the 13th century whenever cardinals select a new pope, and the United Nations employs it for choosing its secretary general. Baseball writers utilize a modified version to decide who gets in the Hall of Fame.

Approval voting works like this: All candidates are listed on the ballot, and voters check off as many as they approve of. The one with the most checkmarks wins. Ideally, voters don’t concern themselves with strategy, but instead hand out as many checks as they feel are merited. In reality, only the least informed members of the electorate do that. Those with more knowledge of the candidates or a bigger stake in the outcome tend to bullet vote, that is to mark their ballots for the single candidate they most want to win.

In essence, approval voting devolves into plurality voting.

There are plenty of other voting methods – Borda (a first-place vote is worth 10 points, a second-place nine points and so on; the one with the most points wins), Condorcet (all candidates are matched one-on-one with every other candidate to see who prevails; some sort of tiebreaker is often needed), Score voting (points awarded for style and competence – actually that could be kinda fun to watch), exhaustive ballot (same as ranked-choice, except there’s a new election after every round), Kemeny, Fishburn, uncovered set, Black, majority judgment, top cycle and Lexmin (none of which I’ve bothered to research because I’ve got this uneasy feeling they aren’t going to be any less terrible).

And there’s always dictatorship, decided by who has the biggest militia and the snazziest mustache.

In short, every system can produce results that don’t make sense – even without interference from Russian hackers and bots. As Robert Wilcox put it in an article for Deadspin (upon which I’ve relied heavily in an effort to conceal my ignorance of the finer points of numerical and sociological complexities), “There are more than a dozen mathematical criteria to evaluate voting systems and it is impossible to satisfy every criteria. The relative importance of each criteria is a matter of opinion and heavily depends on what is being voted on.”

To put that more bluntly, no matter what method you use to vote, chances are you’re going to get a crappy outcome.

Emails sent to aldiamon@herniahill.net are tallied in an irrational manner.

Print Friendly, PDF & Email

7 Responses »

  1. It's really pretty simple....one vote, one winner....and that is that!!

  2. > In essence, approval voting devolves into plurality voting.

    This is simply false. With Plurality Voting, "strategic voting" means not voting for one's favorite candidate, e.g. a Green Party supporter who insincerely votes Democrat. With Approval Voting, that voter would vote Democrat and Green.

    > Those with more knowledge of the candidates or a bigger stake in the outcome tend to bullet vote

    Absolutely false. A voter's best strategy is to approve every candidate they prefer to the "expected value" of the winner. So if you rate the candidates in your head on a 0-5 scale, and figure you expect to get about a 3 from the election outcome, you want to approve every candidate you think is better than a 3. Case in point: that Green Voter wants to tactically vote Democrat and also sincerely vote Green.

  3. Clay Shentrup why does the Green Party always go with the Democratic Party in your comments? Very interesting statements, but let’s remember that is your opinion not facts. I bet you read those points from some poll that was taken, and polls are always so accurate of people’s opinions.

  4. Thankfully we don't have to be as pessimistic about voting science as this article. Approval Voting can only ever devolve into plurality voting if every single voter decides to only vote for one candidate. The moment one voter decides to endorse multiple candidates, that election is better off.

    Consider the 2016 Republican primary or the current Democratic 2020 primary. Is there only one candidate you would support?

    The solution to failed voting requires two elements: A good voting method, and a diverse set of compelling candidates. You can't have a functioning democracy without both.

    We can do a lot better than Plurality voting, and the research proves it. The Center for Election Science is supporting education and campaigns throughout the country to improve voting methods. Our current campaign is in St. Louis, MO where a recent poll showed an astounding 72% support of the ballot measure. Check out https://stlapproves.org and https://electionscience.org. It's not hopeless - we can make politics better, and enact real lasting change.

    Felix Sargent
    Chair of the Board
    Center for Election Science

  5. One vote each. The candidate with the most votes wins. It's as simple as that. Except in a presidential election which the Electoral College decides, which is only fair (Maine is a good example).

  6. > Clay Shentrup why does the Green Party always go with the Democratic Party in your comments? Very interesting statements, but let’s remember that is your opinion not facts

    My facts are in order. According to American National Election Studies (ANES), 90% of self-described Nader supporters in a 2000 exit poll claimed to have voted for someone other than Nader, most of them for Democratic nominee Al Gore.

    From the Wikipedia page on tactical voting: In voting methods, tactical voting (or strategic voting, sophisticated voting or insincere voting) occurs in elections with more than two candidates, when a voter supports another candidate more strongly than their sincere preference in order to prevent an undesirable outcome.

    So even if everyone using Approval Voting "bullet voted" for their sincere favorite, and no one else, that would obviously look dramatically different than the current system, where significant numbers of voters vote for someone other than their favorite.

    In large exit poll studies, in which voters are allowed to vote using both a standard choose-one ballot as well as an Approval Voting ballot, we often see dramatic differences in outcomes. See this example from the 2014 Maine gubernatorial election.
    http://scorevoting.net/Maine2014Exit

  7. "The solution to failed voting requires two elements: A good voting method, and a diverse set of compelling candidates. You can't have a functioning democracy without both."

    Good thing we're not a democracy. Tho I know one party that is trying hard to make it so.

    One person, ONE vote.

    Want a runoff? Fine, hold ANOTHER election - when faced with "A OR B ONLY", votes may (will) change, in ways not knowable on "RCV Day". Tho I see no reason to satisfy people who couldn't be bothered to show up for the first vote. "51%" of those who vote is really more like 30%, given how many DON'T....so it's a fraud from the get go, simply designed to catch the Green and Independent vote as a 'second chance'.

Categories

Archives