Daily Breakdown: Bad Polls and Romney's Cash

by Nate Cohn | August 20, 2012

 

The most interesting poll comes from a joint effort by Monmouth, SurveyUSA, and Braun Research. The poll was conducted with both automated and live-interviews, and provided a breakdown of both likely and registered voters. Monmouth’s results were roughly the same as they were a few months ago, and still show Obama losing 3 points when transitioning to a likely voter screen. Now, the likely voter screen only includes voters who say they will definitely vote, which is arguably too tight, but the lesson is still important. There is one note of good news for team Obama: Democrats are now just as likely to say they will “definitely vote” as Republicans, which wasn’t the case a few months ago.

The two polls from JZ Analytics are interesting but hardly authoritative, and it’s worth flagging that the Florida poll was conducted online (not so sure about North Carolina). Given the history with Mr. Zogby, I’m taking a grain of salt. The Oklahoma poll shouldn’t be a surprise to anyone.

Odds and Ends

--The Washington Post teamed with Kaiser Family to produce a huge survey of divisions within the political parties. Although many findings are utterly predictable, they’ve provided an interesting look at cross-pressured and conflicted subsets of the two parties. Buried in the poll is the finding that Obama was up 7 among registered voters, 50-43. While the survey was conducted in late July and the first week of August, it would have joined a host of polls showing Obama with a 6-9 point edge among registered voters. It’s hard to say whether that was an aberration, or an indicator of an underlying and large Obama lead among registered voters obscured by the trackers and likely voter surveys.

--As a reminder of the difficulties of polling, two firms decided to release implausible polls today. A McKeon poll of Cook County found Obama up by just 12 points in Cook County, Illinois, home to Obama’s base of Chicago. Cook County has voted for Democratic presidential candidates by at least 40 points in each of the last four elections, and voted for Obama 76-23 in 2008. Similarly, a Florida poll gave Romney a 14 point lead, but just 1 percent of respondents were ages 18 to 30, and a whopping 64 percent were over age 64. Ninety percent of respondents were over age 50. Florida's  population is older than most states, but it’s not actually a giant retirement community.

--Team Romney has a $186 to $124 million cash edge at the end of July. The big question is how Romney’s going to deploy these staggering resources. How much goes into building up the ground game? Will Romney take to the offensive in currently uncontested states like Wisconsin, Michigan, Minnesota, or Pennsylvania? Would Romney double down on huge buys in Florida, Ohio, or Virginia? There is a question of diminishing returns here, but the slope of the downward curve isn’t as clear as you might expect.  No matter their choice, how Boston decides to deploy their cash edge will shape the fall campaign in a big way.

Source URL: http://www.newrepublic.com//blog/electionate/106328/daily-breakdown-bad-polls-and-romneys-cash