Obama Over 50% in New Polls

by Bob Grundfest

A few new polls over the past couple of days continue to indicate a close race nationally and an increasing lead for president Obama in many of the key swing states. Mitt Romney has had a rocky week stemming from his overseas trip, where gaffes, not policy, became the main story.

The main headline poll from today comes courtesy of Quinnipiac/The New York Times/CBS News and shows Obama with leads in Florida (51-45%), Ohio (50-44%), and Pennsylvania (53-42%). The poll was of likely voters and was conducted during the last week of July.

On balance, the poll results are good news for the Obama campaign, which has faced rather dismal economic numbers and the president’s negative approval ratings. Most Americans polled also say that the country is on the wrong track.

So why is Obama still ahead? Much has been made of the likeability factor and the political ads that the Obama campaign has been running in the swing states that are savaging Romney’s business experience and his decision not to release more than two year of tax returns. These are not necessarily the kinds of issues that win or lose elections, but they are successful tools in defining an opponent who has not yet defined himself. Once the conventions are over and the real campaigning starts, the side issues will probably fade and Romney’s disparaging of the London Olympics will go the way of his wife’s Cadillacs and the size of trees in Michigan. Obama will have to defend an economy that is barely growing and a growing sense that perhaps the United States needs a new direction.

This poll’s party breakdown is solid in Florida, slightly favors Democrats in Ohio and undercounts Democrats in Pennsylvania, so the Ohio numbers might be overstated, but the other state figures are probably as reliable as a 95% confidence rate can establish.

For example, Q/NYT/CBS has party affiliation in Florida as D 42/R 36/I 20 while the latest figures from Florida show a D 41/R 36/I 23. Pretty good.

In Ohio, the poll has D 42/R 35/I 20 while the actual breakdown is D 36/R 37/I 27, so there’s a bit of a Democratic overstatement.

In Pennsylvania, the poll has D 46/R 40/I 11 and the actual voter registration shows D 51/R 37/I 12. This undercounts Democrats and overcounts Republicans, meaning that if Romney wants to win here, he’ll need a more concerted effort. It’s not out of the question, but time is of the essence.

In all three states, Barack Obama’s favorable ratings are in the low 50’s, while Romney’s are in the high 30s and low 40s. Florida and Pennsylvania voters say that their state’s economy is getting worse, while Ohio voters say theirs is getting better (which it is, statistically).

The complete results are here.

Another poll, this time out of Michigan (EPIC-MRA for the Detroit Free Press), shows the president ahead of Romney by 48-42%, a result that is in line with other pollsters, notably Rasmussen, and seems to suggest that Michigan will stay blue this year. As is the case with most state polls, the fact that Obama is running at under 50% in a reelection year has to concern him, especially since he’s been running hard on his saving Detroit’s automobile industry. The issue with this poll is that it doesn’t have any breakdowns associated with it, so it’s difficult to assess how accurate the methods and results are. That it mirrors other recent Michigan polls is helpful, but it would be nice to see some internal numbers.

The best news for Romney comes from Arizona, where a new PPP Poll has him ahead 52-41%, and a Rasmussen Poll of Missouri showing Mitt with a 50-44% lead.

We are fast approaching events which will have more profound effects on the race, with Romney’s Vice Presidential pick, the GOP Convention later this month and the Democratic Convention in September. It will be interesting to see how these affect the race. I suspect that Romney will gain enough of a bounce to make the swing states more competitive.

1 Comment

  1. The poll is a distorted mess – in Florida alone, the sample was skewed with a +9% Democrat sampling whereas the 2008 election results had Democrats showing up to the voting booth at a +3% … and in the 2010 election cycle at a minus1%. With this real world perspective, it is easy to see that the results from this “likely voter” poll is beyond belief in its results and what it really says about the intent of the folks who took it, and further, the folks who are touting it!

Post a Comment

Your email address will not be published.