Polls show gap between preference and prediction

  • Sun Sep 16th, 2012 9:13pm
  • News

By Chris Cillizza The Washington Post

National polling suggests that the American public is deeply divided about who it wants to be the next president of the United States.

A Washington Post-ABC News survey released last week showed President Barack Obama at 49 percent to 48 percent for former Massachusetts governor Mitt Romney among likely voters. A New York Times-CBS News poll released around the same time put Obama at 49 percent to Romney’s 46 percent, among those most likely to vote.

But, if the pool of likely voters is almost evenly split on who should occupy 1600 Pennsylvania Avenue in January, they are far less indecisive on who they think will be the next president.

Nearly six in 10 (59 percent) registered voters in a late August Post-ABC poll said they believed Obama would win, while just 34 percent chose Romney even as the head-to-head vote in that same poll stood at 47 percent for Romney and 46 percent for Obama among registered voters.

The story was the same in an early July when a Post-ABC poll found the two candidates tied at 47 percent among registered voters in the horse race, but Obama held a 58 percent to 34 percent edge on the “who will win” question.

What explains the wide gap between who voters want to win and who they think will win? Theories abound.

“The ballot question is driven by who voters want to win,” Republican pollster Whit Ayres said. “The ‘who do you think will win’ question is driven by pundits and commentators who want Obama to win.”

Added Ayres: “⅛Comedian€ Bill Maher and his ilk affect the thinking of a lot of people on the predictive question, but fortunately do not affect the thinking of that many people on the preference question.”

Under Ayres’s line of thinking, the general tone of the coverage of the race leads people who want Romney to win to believe that he ultimately won’t – although that sentiment doesn’t change how they plan to vote.

Republicans point to the last week of media coverage of Libya and other anti-American protests in the region as evidence of the media swaying the predictive question. The coverage, they insist, was heavily focused on whether Romney jumped the gun with his statement on Libya, not on whether the Obama administration’s actions (or lack thereof) may have led to a lack of American leadership in the region.

There’s also a less partisan explanation available. That is that people tend to struggle to imagine someone other than the current occupant of the White House as the president until he, well, isn’t anymore – even if they don’t like him or plan to vote for him. (Obviously this is less of an issue in an open-seat race.) So, it’s possible that until Nov. 6 – or maybe until the general election debates start early next month – the numbers on the “who will win” question won’t change.

Just as there are competing theories as to why there is such a gap between the ballot test and the prediction question, there are a number of operating assumptions about what it could mean for the actual vote.

Human nature is such that people like to be with the person they think looks like a winner – are there really that many people born fans of the L.A. Lakers or the New York Yankees? – and that reality suggests that if the general sense in the electorate is that Obama is going to win, it may tip some undecideds his way just so they can say they voted for the victor.

Not so, according to Ed Goeas, a longtime Republican pollster. “Large numbers of people saying that your candidate is going to win is not a good thing,” he said. “It has a dampening effect on driving your candidate’s vote to turn out. It allows other things in a person’s busy day to become more important because their candidate is going to win anyway.”

There is some evidence that Goeas is on to something. As we mentioned above, Obama and Romney were running neck and neck among likely voters in an Post-ABC poll conducted earlier this month. But, among registered voters Obama had a 50 percent to 44 percent lead. That trend has been apparent in a slew of national polls for months; the more the sample is trimmed to those absolutely planning to vote on Nov. 6, the better Romney performs.

There’s not enough evidence on either side of the argument to conclusively say that Obama’s lead on the “who will win” question is good, bad or indifferent to his chances in November. But, it’s a fascinating window into the difference between “want” and “will” in politics.