Taking advantage of a slow news week, Tom Edsall takes to the virtual pages of the New York Times‘ Opinionator blog today to assess Republican odds of regaining the White House in 2016, which is audacious since many news consumers groan about early discussions of the 2016 nominating process, much less the general election.

Edsall marshals his evidence for better-than-even odds of a GOP victory pretty effectively, with two exceptions. The first is a dubious data point about the recent history of elections after one party has held the White House for two consecutive terms:

Steve Ansolabehere, a political scientist at Harvard, is…bullish on Republican prospects:

“I think they have a better than 50-50 shot at the presidency. It is very hard for a party to win the presidency three times in a row — since WWII, only the run of Reagan-Bush. There are a lot of divisions in the G.O.P., but there is no shortage of respectable candidates.”

The post-World War II historical record strongly favors a Republican victory in 2016, according to David Mayhew, a political scientist at Yale. He pointed out that in five of the six “open” elections since World War II (open meaning without an incumbent candidate), the candidate representing the party holding the White House lost. In 2016, the Democratic nominee will fit that definition.

This is a pretty good example of brushing over conflicted empirical evidence from a very limited data set. There have been seven elections since WWII when a party was vying for a third straight presidential win. One, which Ansolabehere excludes for some reason, was in 1948 when Truman won. Another, which he notes, was in 1988 when George H.W. Bush won. A third, in 2000, was rather famously a fluke wherein Democrats won the popular vote and were denied a third straight win by an intervention of the U.S. Supreme Court. Three others, in 1960, 1968, and 1976, were among the closest presidential elections in history (at least between 1876 and 2000). Only 2008 actually fits the template of a party clearly running out of steam in a third-term bid, and you can make the case the GOP hasn’t entirely removed the cloud of events that helped make that happen. If you eliminate the arbitrary World War II limit on the data set, of course, you see Democrats winning five straight times from 1932 through 1948, and before that, Republicans winning three straight from 1920 through 1928. So it’s not exactly a clear case of any “rule” governing third-term elections.

The elections Mayhew’s talking about largely overlap with the ones just discussed, with the exception of 1952, when the “out” party won after losing five five straight presidential elections, benefitting from the candidacy of a national hero so popular that both parties vied for his allegiance (can you imagine that now?), who among other things was promising to end an unpopular war.

If the “history” data points for a Republican win in 2016 are weaker than they initially look, at least they are backed by significant evidence. You can’t quite say that about Edsall’s closing argument:

The future of the Republican Party will in large part be determined by the outcome of the internal struggle between the Tea Party faction and the pro-business establishment wing. Over the long haul, this is not a fight between equals. Prospects favor a revived pro-business, anti-regulatory Republican Party that purposefully narrowcasts — that is, carefully restricts to a select audience — its focus on divisive social and cultural issues, just as the Democratic Party, which had lost three presidential elections in a row, lowered its liberal profile in the 1990s. The compelling mandate for a national political party in the United States is not to serve as ideological advocate, but to win.

This is stated as dogma, which is supposed to override perceptions of the actual behavior of the Republican Party. Edsall doesn’t explicitly say this, but often those predicting final Establishment victory in the struggle for control of the GOP cite the party’s nomination of two relatively non-ideological figures, John McCain and Mitt Romney, in the last two presidential cycles. But they both lost, of course, which enormously undermines the Establishment’s case for claiming it knows how to win. Unless Republican “pragmatists” have the contemporary equivalent of Dwight David Eisenhower somehow lurking in the wings, it’s very unlikely that in 2016 the Tea Party is going to be overwhelmed by assertions that Father Knows Best.

Ed Kilgore

Ed Kilgore is a political columnist for New York and managing editor at the Democratic Strategist website. He was a contributing writer at the Washington Monthly from January 2012 until November 2015, and was the principal contributor to the Political Animal blog.