« How to compare a two-person and a six-person election? | Main | Statistical evidence is both powerful and limited: fraud in baseball and lotteries »


Feed You can follow this conversation by subscribing to the comment feed for this post.

Tom Dietterich

If you group the candidates into left-wing vs. centrist, the centrist vote was larger than the left-wing. So I don't think you can say that this is equivalent to Bernie winning 60%. What is the underlying stochastic model?


TD: I'm framing the problem as looking for equivalence in terms of strength of the winner in a N-person versus 2-person contest. I want something that works generally, regardless of who the candidates are. The goal is to put some rigor behind the intuitive notion that the vote shares can't be directly compared.

I thought about a stochastic model but couldn't get past the need to model migration behavior. In the counterfactual scenario, what would the #4 candidate's voters do if there were only 2 contestants? That requires arbitrarily setting a large number of parameters, with no data to support them.

Separately, any analysis that adds together left-wing and centrist groupings makes several assumptions: (a) that all voters make decisions primarily based on their identification with those factions, and (b) that voters consider the candidates within each grouping as exchangeable.

Also, Bernie winning 60% is not the blowout that the media portrayed it to be. An even match is 50/50; 60/40 is a good win but not huge. Clinton won most states on Super Tuesday by a bigger margin than 60/40!

Noah Motion

A new instance from Megan McCardle, in a slightly different context:

“The most important theory these primaries should have killed is that of our nascent socialist revolution. That theory was plausible in 2016, when Bernie Sanders (I-Vt.) won New Hampshire with 60 percent of the vote. It is less so now that he’s leaving the same state with only a quarter of its votes. Even if every one of Andrew Yang’s and Warren’s voters had picked him, he’d still only have gotten a bit over a third, and his polls suggest that’s about where he’d be nationally, too. So what looked, four years ago, like a sharp leftward shift in the electorate now seems more like a mass protest against the party’s slavish fealty to the Clintons.”



The comments to this entry are closed.

Get new posts by email:
Kaiser Fung. Business analytics and data visualization expert. Author and Speaker.
Visit my website. Follow my Twitter. See my articles at Daily Beast, 538, HBR, Wired.

See my Youtube and Flickr.


  • only in Big Data
Numbers Rule Your World:
Amazon - Barnes&Noble

Amazon - Barnes&Noble

Junk Charts Blog

Link to junkcharts

Graphics design by Amanda Lee

Next Events

Jan: 10 NYPL Data Science Careers Talk, New York, NY

Past Events

Aug: 15 NYPL Analytics Resume Review Workshop, New York, NY

Apr: 2 Data Visualization Seminar, Pasadena, CA

Mar: 30 ASA DataFest, New York, NY

See more here

Principal Analytics Prep

Link to Principal Analytics Prep