ncase patreon

[Patreon Exclusive] New Prototype!

🕑 Added 2016-11-05 16:50:52 +0000 UTC

Comments

Tisza Gergő

All voting systems have theoretical flaws (cf. Arrow's theorem, or more specifically, approval voting failing the majority criterion or Pareto optimality). The more important question is, how much do they matter in practice? Cloning vulnerability means there is a high bar of entry for new candidates and so the system prefers non-competitive "ideological areas", which are then going to underperform due to said lack of competitiveness. It also enables some foul play (e.g. the right-wing party funding far-left candidates to run just so they can split the left vote). It is not clear how non-monotonicity would cause similar problems.

Tom Lieber

Haha, that's great to know! Can't wait to learn more from your later prototypes. By the way, in your simulation of the spoiler effect, will you address the counter that the people who voted for the third candidate might not have voted at all? In the prototype above, the third candidate literally steals votes, but that's not always how it works.

Nicky Case

Oh nice, thanks for bringing that to my attention! I know Justin Trudeau, PM of Canada, is also planning to switch the nation to Instant Runoff soon as well. ...however. One thing my interactive will be explaining is why instant runoff/ranked voting is actually just as bad as first-past-the-post, if not worse. Y'see, instant runoff has this glitch, where if a candidate is *already* winning the polls, and they become *more* popular, then that can CAUSE THEM TO LOSE. This is called "non-monotonicity", because mathematicians suck at naming things. (But there is hope! Other *non-ranking* voting systems, like approval voting & range voting, are immune to both the spoiler effect *and* non-monotonicity! Maybe this interactive will have the power to convince Justin Trudeau of Canada to go for one of those, NOT ranked voting)

Tom Lieber

Sweet, ranked choice voting has come to Maine! <a href="http://www.boston.com/news/politics/2016/11/10/maine-became-the-first-state-in-the-country-to-pass-ranked-choice-voting" rel="nofollow noopener" target="_blank">http://www.boston.com/news/politics/2016/11/10/maine-became-the-first-state-in-the-country-to-pass-ranked-choice-voting</a>

Nicky Case

Thanks Punya! Ohhh I could go on for ages about gerrymandering, the electoral college, lack of proportional representation, and a zillion other structural problems with the way we (currently) do democracy. But for this project, I'll just focus on voting systems. I'll certainly mention the other (just as important, if not more important) problems – but one thing at a time!

Nicky Case

That's a great question, thanks Cai! Yeah right now my model is assuming a "normal" bell curve distribution of political positions, but who knows – maybe given how polarized politics is nowadays, we may have a bimodal ("two-humped") distribution instead. I'm not yet sure how to simulate/model that, but I'll keep it in mind as I flesh out the rest of this explorable explanation!

Nicky Case

Thank you too! I backed that IndieGoGo a while back, and I'm excited to see what the empirical results are. (I'll definitely link to your research in my interactive explainer, whenever it's out)

Punya Biswal

I love this so far! Would it take the discussion too far afield to consider the impact of aggregation at multiple levels (vs a popular vote), and long-term effects of gerrymandering?

Cai

Looks really great so far! Seems like a perfect topic to explore in this medium. A thought (which shouldn't be taken as a criticism!): I am left wondering about the distribution of views of the electorate versus the candidates. In the examples, the electorate are on-average moderate or centrist, with a big clump of voters who take approximately the same view. This seems as if there is one "true" view, and actual voter's views deviate randomly from it, with each of the candidates positioning themselves as more extreme in some direction compared to to the "true" view of the people. Of course, this is a simplifying assumption, but I wonder if the dynamics change in interesting ways if the distribution of voter preferences changes shape? Maybe they don't! Maybe this is to come in the full version!

Neal McBurnett

Wow - thank you - something like this could be really useful for grounding discussions of election systems, since it is such a pain to describe and test useful scenarios in a way that is easy to share with non-experts now. And there are of courselots of election systems to plug in to it. We'll even have some good multi-method data soon based on the current election via <a href="https://www.indiegogo.com/projects/polling-with-smarter-voting-methods-science" rel="nofollow noopener" target="_blank">https://www.indiegogo.com/projects/polling-with-smarter-voting-methods-science</a>#/

Feiya Wang

Oh yeah, that's very interesting, I would have never come up with such nuanced axes on my own, sometimes it's nice just to be fed ideas than coming up with them on your own you know? Those are great.

Nicky Case

Aw no, Red Hex is just doing what's reasonable! It's the *voting system* that's wonky, and that's what I hope to show with the rest of the explorable explanation! I'm not yet quite sure if I should avoid the yellow triangle / blue square thing, because while I *do* like a throwback, yeah, they're two totally different topics. (Maybe *blue* triangle and *yellow* square?)

Nicky Case

mmmmm multiple axes

Nicky Case

Thanks for this specific feedback! I was wondering if just leaving it as abstractly "two political axes" was enough or not. For a concrete case, I think "left vs right" is now rivaled by a new political axis this year, "institutions vs insurrections". So, for example, institution-left = Clinton, insurrection-left = Sanders, institution-right = Kasich, insurrection-right = Trump. I think this second axis also explains Brexit's "people have had enough of experts", and why some early Sanders supporters switched over to Trump (they're on opposite ends of the left-right axis, but they're both very anti-establishment)

Feiya Wang

I'm having a hard time feeling grounded with the blank axes, maybe some suggestions of what they could be?

Nick Schrag

Solid so far, demonstrates the system clearly. The poor red hex seems like such an evil third party, stealing all those votes and ruining the election! Not sure if the art is placeholder, but I'm finding myself thinking about racism because of the same art from Parable of the Polygons.