Democratic pollster Mark Mellman insisted that recent history doesn’t back up that theory. He pointed to the 2008 and 2010 Senate races, when 11 incumbents went into Election Day with less than 50 percent of the vote in the polls. Five won; six lost.
While the 50 percent rule may have applied in the past — in Senate races between 1994 and 2004, on average, 70 percent of the undecideds broke to the challenger — it no longer holds true, Mellman said. In the 1980s and 1990s, most incumbents were much better known than their challengers, but undecided voters learned enough by Election Day to vote against the status quo. More recently, as campaign spending has skyrocketed, fewer challengers are unknown before Election Day.
“People really know the candidates in races where millions are spent, and they know Mitt Romney,” Mellman said. “They may be negatively disposed toward the incumbent, but they are not ignorant about either one.”
Data compiled by Public Opinion Strategies, a Republican firm, and Hart Research Associates, a Democratic firm, for NBC News and The Wall Street Journal suggest that undecided voters lean toward Obama. In a profile of the “up for grabs” voter, defined as a voter leaning toward Obama or Romney, favoring another candidate, or favoring neither candidate, Obama was viewed favorably by 43 percent and negatively by 33 percent. By contrast, only 26 percent had a favorable image of Romney and 40 percent had a negative view of Romney.
Lynn Vavreck is a UCLA associate professor of political science and a principal investigator at the Cooperative Campaign Analysis Project, which has been interviewing 1,000 people per week since January. She said the data show that undecided voters have been breaking for Obama since the summer, and there’s no reason to think that trend, while not as strong in the last three weeks, would change in the final days of the campaign.
“My hunch is that a lot of undecideds stay home, maybe one-third of them, and the remaining ones break for Obama,” she said. “People know what life is like under Obama. There’s much more uncertainty about a Romney presidency, and generally uncertainty is bad.”
John Sides, a George Washington University associate professor of political science who is also working on the campaign project, comes down in the middle. He expects the undecided voters to break fairly evenly between Romney and Obama. “You don’t know who exactly is going show up at the polls and who is not. You don’t know whether the impact of independent voters will be outmatched by the mobilization of base partisans. You need a microscope to see this very small sliver of undecided voters, but you can’t get it focused enough and they keep moving.”
At least until Tuesday.