The most plausible of these concoctions is Romney winning the popular vote but flunking out of the Electoral College. Almost everyone who writes about campaigns operates on the assumption that there is something historically aberrant about one candidate winning the most votes while not being awarded the keys to the Oval Office. This sense of alarm is understandable, given that the last victory by a minority-vote president was in the hanging-chad, constitutional-crisis election of 2000.

But an intriguing footnote in Partisan Balance, a 2011 book by Yale University political scientist David Mayhew, demonstrates how easy it is to mis-remember history. For in truth, any reasonable interpretation suggests that Richard Nixon—not John Kennedy—won the national popular vote in 1960. This is not a conspiracy theory about how Chicago Mayor Richard Daley stole the election for JFK. Instead, Mayhew highlights an error in how the correctly tabulated returns from Alabama have been recorded for posterity.

In 1960, Kennedy’s name was not on the Alabama ballot. Instead, 11 electors (some segregationist and some loyal to the national Democratic Party) were elected statewide on the Democratic line. The six segregationist electors (who went on to vote for Virginia Senator Harry Byrd) rolled up more votes than the five Kennedy loyalists.

All the electoral charts on the 1960 election award the entire Democratic vote in Alabama to Kennedy. But if you follow logic and reduce JFK’s proportion of the Alabama Democratic ballots to five-elevenths, he loses 175,000 votes. That single recalculation is large enough to wipe out Kennedy’s 118,000-vote national margin and very belatedly hand Nixon the popular vote lead.

Why is this historical recalibration relevant to contemporary campaign reporters?

Because we are all prone to grand pronouncements about the natural shape of American politics. But, as the 1960 election illustrates, some of the raw material that supports this glib theorizing should come with asterisks and caveats. Since there are so few presidential elections, even just one error in interpreting the results can alter perceptions.

(By the way, I wonder how many political and economic model builders include Kennedy’s purported popular vote victory in their database. Nate Silver gets at least partial credit for noting in a recent New York Times column that the 1960 election was so close that its “results could have been altered by essentially random factors.”)

If Romney does indeed win the popular vote but lose the White House, I hope that the election-night commentary notes this is the third time that we have had such an upside-down result in just 52 years. When something occurs that often, it may be weird, it may be undemocratic, but it is hard to call it aberrant.

If you'd like to get email from CJR writers and editors, add your email address to our newsletter roll and we'll be in touch.

Walter Shapiro just chronicled his ninth presidential campaign. He writes the “Character Sketch” political column for Yahoo News. Follow him on Twitter @WalterShapiroPD.