Do presidential debates sway voters? Or are they political entertainment that just affirm electoral choices Americans have already made?
Mitt Romney hopes they’re the former. He and his campaign are looking to Wednesday night’s debate in Denver as a way to overcome President Obama’s stubborn lead in the polls. Romney supporter Gov. Chris Christie (R) of New Jersey has gone so far as to predict that the outcome of the verbal tussle will turn the race upside down.
If so, that will be out of step with the historical trend, say some pollsters and political scientists.
Gallup, for instance, has gone back through a half century of its polling results and found only a few examples of presidential debates that made an impact on election outcomes.
There have been nine sets of presidential debates since 1960, points out Gallup. (Lyndon Johnson refused to debate in 1964, and Richard Nixon followed suit in 1968 and 1972.) In only two of these nine political cycles did the candidate who trailed prior to the debates come from behind to win.
And those two were perhaps the most famously close elections of the past 60 years. In 1960, then-Vice President Richard Nixon was up by one percentage point when he met Sen. John F. Kennedy in the first televised presidential debate on Sept. 26. By the time of the fourth debate, in late October, Mr. Nixon trailed Senator Kennedy by four points.
Ultimately, Kennedy won the popular vote in 1960 by 0.2 percentage points.
In 2000, then-VP Al Gore led George W. Bush by eight percentage points right before their first debate, in October, according to Gallup’s records. The first three days after the event, Gallup polls showed the race tied.
Mr. Gore came back a little bit prior to the next debate, only to fall back again. The same pattern held for the third debate. Ultimately Mr. Bush prevailed in an election so close it was, in essence, decided by a Supreme Court ruling.
“The debates were less likely to be catalyst events in years when one candidate was a strong front-runner, including 1984, 1988, and 1996,” writes Gallup analyst Lydia Saad.
The reasons for that are fairly obvious, according to George Washington University associate professor of political science John Sides. Presidential candidates tend to be fairly evenly matched in terms of capability, preparation, and political experience. Over a series of debates, neither is likely to be able to dominate the other so thoroughly that undecided or wavering voters will judge them a superior possible president.
Even if polls do move during a debate period, it can be hard to determine whether the debates themselves were the cause of the movement, or whether they were the result of other events, such as overseas crises or economic troubles.
“What history can tell us is that presidential debates, while part of how the game is played, are rarely what decide the game itself,” writes Mr. Sides in an article on debate effects in the Washington Monthly.
But “rarely” isn’t the same as “never." And it’s possible that 2012 could be an outlier in this historical data set.
First of all, the election is close. It’s close enough so that one stray gaffe might send just enough voters fleeing to the other guy. And the media environment surrounding the debate is different as well. The Twitter-fueled political news cycle is faster and more ferocious than ever, meaning that any perceived victor or loser could find their gain or loss exaggerated by the sheer volume of media hype.
Don’t think so? We’ve got one word for you: “oops." That’s what Texas Gov. Rick Perry said when he fumbled an answer in a GOP primary debate, if you recall. It’s true that wasn’t a general election debate, but if any single moment sealed a candidate’s fate this cycle, that was it. Governor Perry never recovered.