The problem with electoral history predicting Hillary’s political demise

“If you torture the data enough, nature will always confess.” – Ronald Coase.

Writing in The New Republic on Sunday, John Judis argued that Hillary Clinton is likely to lose in 2016.

The reason is relatively simple: Political parties generally have a difficult time holding on to the presidency for three consecutive terms. Exiting presidents typically have low popularity, and the electorate blames the incumbent party for eight years of pent-up frustrations.

He isn’t the first to make this argument, and he won’t be the last.

Before I go into why this is a weak argument, I’ll add another pillar to it: The more presidential elections a party loses, the more moderate their nominees become, which theoretically raises their chances of ending their losing streak.

Hillary Clinton

The reports of her political death have been greatly exaggerated.

While Judis isn’t wrong on its face — parties seeking a third consecutive term have only won twice (three times if you count Al Gore) in the modern era, with Harry Truman and George H. W. Bush maintaining control of the White House — he’s basing his case on a dubiously small dataset.

After all, there have only been eight such elections (1948, 1952, 1960, 1968, 1976, 1988, 2000 and 2008).

To put that in perspective, Judis’ analysis is the statistical equivalent of saying that a baseball player won’t get a hit because his batting average through the first two games of the season is .250.

There are a number of data-driven ways to poke holes in Judis’ analysis. For starters, if you walk your timeline back to include the full duration of the two-party era (1856 onward) the incumbent party has won four of eleven times (five counting Gore) when going for a third consecutive term. One could also point out that Hillary Clinton is an unprecedentedly strong candidate as measured by pre-campaign polling.

This speaks to a point that anyone with a cursory knowledge in statistics could make: It is difficult to draw significant conclusions from a small number of observations. Given how few federal elections we have had — let alone presidential elections — every one is likely to defy history in some way or another. We can identify trends, but we have nowhere near enough observations to derive political Truth.

For example, here’s the Washington Post‘s breakdown of fifteen “firsts” that happened in the 2012 elections. The big one? First president in the modern era to be re-elected with unemployment above 7.2 percent: Barack Obama. Journalists started throwing that stat around about this time four years ago, producing nifty clickbait to keep the 2012 cycle interesting. The stat, of course, ignored the fact that a president had recently been re-elected with unemployment at 7.2 percent. And that president was Ronald Reagan, who won by 15 points. At least Nate Silver was there to remind us how silly the argument was.

All this is to say that 70+ years of modern presidential history is useful qualitative guide for handicapping presidential races, but you want to make quantitative claims, you need a lot more to go on. Judis himself makes such concessions in his last paragraph, where he basically rejects his whole premise and points out that Republicans need a lot more than eight years of Barack Obama to take back the White House:

The Democrats could benefit if the Republicans nominate a relatively inexperienced right-winger or someone who possesses the temperament of a high school football coach rather than a president. But in the last elections, they opted for the more centrist contenders who had some credibility as presidential candidates. If they opt for an experienced centrist in 2016—Florida’s Jeb Bush is the obvious example—and if the party’s right wing doesn’t demand he toe the line, they could stand a good chance of reclaiming the White House and of confirming Americans’ reluctance to keeping the same party in the White House three terms in a row.

After spending a whole article (mistakenly) asserting that historical fundamentals are likely to make for a red night on Election Eve 2016, Judis accidentally sums up the 2016 cycle perfectly. History aside, if the GOP stops acting like the GOP, moves to the left and finds a way to keep from saying flamingly derogatory things about large segments of the American electorate, they have a reasonable shot at winning the presidency — even against a strong candidate like Clinton. But that’s an argument based in a qualitative analysis of current political trends, not a quantitative analysis based on decades of data.

Of course, that move to the left is highly unlikely to happen, and even if it does it won’t be Jeb Bush making it. Mother Jones has 23 reasons why.

Jon Green graduated from Kenyon College with a B.A. in Political Science and high honors in Political Cognition. He worked as a field organizer for Congressman Tom Perriello in 2010 and a Regional Field Director for President Obama's re-election campaign in 2012. Jon writes on a number of topics, but pays especially close attention to elections, religion and political cognition. Follow him on Twitter at @_Jon_Green, and on Google+. .

Share This Post

© 2018 AMERICAblog Media, LLC. All rights reserved. · Entries RSS