Oliver Roeder in Nautilus:
It’s not hard to find close elections.
In 2015, a Mississippi state house race ended in a tie, after which the winner was decided by drawing straws. A 2013 mayoral race in the Philippines was deadlocked and resolved with a coin toss. A 2013 legislative election in Austria was decided by a single vote, after wrangling over the validity of a ballot featuring a vulgar cartoon. Heck, I didn’t have to look far to find examples: In 1990, my own uncle lost his bid for Congress by less than 1 percent of the vote.
But there is one election that is so consistently close, and so important, that it deserves special consideration—the United States presidential election.
I plotted the top two popular vote-getters in every U.S. presidential election since 1824, using data from The American Presidency Project. The top two contenders, typically a Democratic and a Republican, but occasionally a Whig, have danced closely around the 50-50 mark for nearly 100 years. Only four times since 1824 has the winner received more than 60 percent of the popular vote. Since 2000, the candidates have been separated by an average of 3.5 points. The median and average separations have been 8.2 and 9.5 points since 1824—a figure skewed upward due to a few outlying and not particularly close races. (The electoral tally doesn’t usually appear so close because the Electoral College tends to magnify differences in the popular vote.)
This is a feature of U.S. politics that many of us have become accustomed to. So is it unsurprising? Not really. “Considering all of the factors that go into what would make an election close or not close—incumbency, the brand of the parties—my perspective is that there’s a surprising rate of close elections,” Eitan Hersh, a Yale political scientist and the author of Hacking the Electorate: How Campaigns Perceive Voters, told me.
The question is, why?
More here.