Presidential polling in June: Flip a coin instead?

Facebooktwitterredditpinterestmail

“I know nothing except the fact of my ignorance.” — Socrates

With all of the polls, models and history at their disposal, political analysts should be able to figure out who is going to win a November presidential election by June, right?

Well, not quite. While we would modestly suggest to Socrates and our readers that we know more than nothing about the election, declaring the winner with certainty at this point is a fool’s errand, particularly when the current data argue only that the contest will be a close one. In the RealClearPolitics average of national horse race polls as of Wednesday, President Obama was narrowly ahead of Mitt Romney by 2.0 percentage points. Meanwhile, in last week’s Crystal Ball, Alan Abramowitz showed how his respected presidential election model forecasts a very tight race at this point, with Obama as a slight favorite.

But surely, this year is an outlier, many would assert. Because of the unique circumstances surrounding this election, including the great economic dislocation caused by the 2008 crash and the restless mood of Americans even after three straight wave elections, it’s understandable that this contest would remain hazy late into the spring.

That’s true. But uncertainty in June is not unique, at least not in modern history.

If anyone doubts that a reassessment — maybe several of them — will come as 2012 wears on, consider this: Over the past eight elections, Gallup — the most recognizable of polling organizations — has only identified the eventual popular vote winner twice in its early June horse race polling:

  • In June 1980, President Jimmy Carter led Ronald Reagan 39% to 32%, with independent John Anderson at 21%. In November, Reagan defeated Carter, 51% to 41%, with Anderson getting less than 7%. Remember that this race appeared close until the very end, with some polling even indicating that Carter might actually win just a few days before the election. But Reagan proved his mettle in a late debate, and Carter’s attempt to negotiate freedom for the American hostages in Iran failed. Those late developments helped turn a close election into a blowout. Note, also, Anderson’s strong early performance in polls: Third party candidates sometimes appear formidable in early surveys and then fade away as the election gets closer, victims of the voters’ desire not to “waste” their ballots.
  • The polling was fairly stable in 1984. In June, Reagan already led Walter Mondale by 53% to 44%. The incumbent won 59% in the fall. Such early polling, and Reagan’s strength, prompted Mondale to throw a Hail Mary by selecting Geraldine Ferraro as his running mate. Like most Hail Marys, the pass was incomplete.
  • By 1988, the June polling was far more misleading: Michael Dukakis was ahead of George H.W. Bush by a landslide, 52% to 38%. Bush ended up winning more than 53% in November.
  • The June 1992 polling projected the nation’s first independent president, Ross Perot. At 39%, Perot easily topped Bush (31%) and Bill Clinton at 25%. Less than five months later, the order was reversed: Clinton won with 43%, Bush (37%) was ousted and Perot finished last with 19%, failing to win a single electoral vote. However, Perot maintained his support to a greater degree than most independent candidates do down the stretch.
  • Gallup’s June 1996 survey got Clinton’s reelection percentage right on the nose (49%), but Bob Dole, at 33%, was well below his eventual 41% and Perot had 17% in June but finished with about 8% in November. Like 1984, Clinton’s reelection bid lacked drama.
  • The squeaker of 2000 was close even in June, but Gallup had George W. Bush up over Al Gore, 46% to 41%. Come November, Gore won the popular vote by half a percentage point, though of course he lost the Electoral College vote.
  • Gallup had John Kerry well on his way to avenging Gore’s loss in June 2004. Kerry led Bush outside the margin of error at 49% to 43%. Instead, Bush grabbed his second term with 51% in November.
  • It’s rarely recalled, but John McCain actually led Barack Obama by a whisker in Gallup’s daily tracking at the beginning of June 2008, 46% to 45%. It wasn’t close in the fall, with Obama winning 53%.

And the uncertainty goes back further. Jimmy Carter looked as though he would roll Gerald Ford in 1976; instead, the election ended up incredibly tight. So did the 1960 and 1968 contests. As we never tire of repeating, Harry Truman shocked the world in 1948 by defeating “President-elect” Thomas E. Dewey.

This is not meant to cast aspersions on Gallup; rather, it’s to say that presidential races are not static, and that polling conducted five months before the election is only a snapshot in time, as opposed to a reliable prediction as to how the race will eventually shake out. As of Wednesday, Obama and Romney were tied, 46%-46%, in the Gallup poll. Obviously, this is a matchup that could go either way.

Almost everything can change, and frequently does, during the course of the summer and fall in a presidential race. The economy can get decidedly better or worse. International crises can pop up — or peace can break out. Unexpected scandals can engulf one or both major party candidates. One or more independents or third-party candidates may prove influential in the presidential tally.

Politics, as we’ve insisted for years, is a good thing. And a fun thing, too, for people who do not treat American elections as a life or death affair. There will be many spectacles between now and Nov. 6, and plenty of unexpected developments in this semi-scripted human drama. But while we know the road to the finish line will be fascinating, let’s also grant that it will be somewhat unpredictable.

For those of you who can’t wait, just join the partisans on both sides who absolutely, positively know their side will win — in a landslide! One side will be right, more or less, and after the election, the winners will lord their perceptiveness over friends, family and the opposition.

And if your partisanship isn’t intense enough for this route, there’s always that coin in your pocket. With the prospect of a tight presidential race, a good flip may tell you as much as June polls.