Morning Jay: Is Obama's Ad Blitz Moving the Polls?

Over the last six weeks, President Obama has launched a sustained advertising blitz focused primarily in nine swing states – Colorado, Florida, Iowa, Nevada, New Hampshire, North Carolina, Ohio, Pennsylvania, and Virginia. Everybody is wondering: has it moved the needle in his direction?

Opinions are mixed.

Writing in yesterday’s Wall Street Journal, Karl Rove argued:

These ads have not moved him up in the polls. The race is tied in the July 30 Gallup poll at 46%. Neither have the ads strengthened public approval of Mr. Obama's handling of the economy, which is stuck at 44% in the July 22 NBC/WSJ poll, nor have they erased Mr. Romney's seven-point lead in that poll regarding who has "good ideas for how to improve the economy.

Empirically, this is true. The RealClearPolitics national polling average between Obama and Romney has not shown any net movement in the last month. The story is the same with the president’s net approval rating, which remains mired in slightly negative territory as of this writing.

But it is possible that the ad blitz is having an effect in the targeted states. Bill Kristol raised this possibility yesterday when he wrote:

A savvy friend, a Romney supporter who has an excellent track record of reading election trends, emails:

“The national numbers aren't changing much because Romney is actually gaining in the states that are not being bombarded with media. Yesterday’s Connecticut poll has Obama by only 8 for example. And red states seem to be getting even redder. This is happening because the daily news is about the economy, Washington problems, etc. and that is the main message getting through. So, polls in these states reflect how voters who only see national news and national advertising (to the degree there is any) respond.

“But.

“In the swing states they are being assailed with ads and campaigning, as well as the news. And here Obama seems to be building a bit of a margin. He now is ahead by solid margins in the most recent surveys in Ohio, Florida, Pennsylvania, Michigan, Nevada, Virginia, Iowa, Wisconsin and New Hampshire. It's very close in Colorado, and Romney has a slight lead in North Carolina."

This is a very real possibility, and entirely consistent with the data that Rove cites.

Writing yesterday at RealClearPolitics, Sean Trende argued that the swing state polls are not tilted to Obama. His weighted average of the 50 states plus the District of Columbia points to the same, modest nationwide lead that Obama sports in the nationwide polling.

Yet I think there is evidence the ad blitz is having a modest effect.

It’s a tough case to make because ultimately we are dealing with different types of polls from many different pollsters – on both the state and national level. The possibility for “house effects” is very high. One way around that is to look at one pollster who is polling on the state and national level. Fortunately, there is one pollster who does regular polling like this – Rasmussen Reports.

Over the last month – during the heyday of Obama’s ad blitz – Rasmussen has conducted a daily national tracking poll and also polled in five of the nine targeted states – Florida, Nevada, North Carolina, Ohio, and Virginia. This offers us an opportunity to see if Obama is doing better than we should “expect” him to.

But what to base those expectations on? We can look at 2008 – and, specifically, compare how Obama did in those states relative to the national average. We then take that to the 2012 polls to see if the president has enjoyed a boost. For instance, Obama did two points worse in Ohio than he did nationwide in 2008 (51 percent in the Buckeye State compared to 53 percent nationwide). So, we might “expect” Obama to be doing two points worse in the Rasmussen poll of Ohio than he did in the nationwide poll conducted at the same time.