Tuesday, August 3, 2010

Burr ahead by a couple

The North Carolina Senate race continues to look very competitive, with Richard Burr's approval numbers hitting a record low and Elaine Marshall pulling within 2 points. Burr's at 39% to 37% for Marshall and 7% for Libertarian Michael Beitler.

The main thing that's changed since the last poll, when Burr led by 5, is that Marshall is shoring up her support from within the party. 65% of Democrats say they'll vote for her, up from 57% a month ago. Burr continues to lead because of a 44-25 advantage with independents and because with 73% Republican support his party is more unified around him than Marshall's is around her.

There's not much doubt the closeness of this race is more an indictment of Burr than a reflection of Marshall's popularity. The Democratic challenger continues to be largely an unknown to North Carolina voters with 58% saying they have no opinion of her. Among those who do 23% see her favorably and 19% unfavorably. But what's really changed since February when Burr held a ten point lead is perceptions of the incumbent. At that time voters split evenly on his job performance with 35% approving of him and 35% disapproving. Now just 32% approve with 44% disapproving.

Burr's numbers with Republicans are exactly what they were six months ago. But independents have gone from splitting evenly on him to disapproving by a 33/45 margin, and Democrats have hardened in their feelings with their level of disapproval rising from 50% to 62%.

The race is closer at this point than the Kay Hagan/Elizabeth Dole race was at the same point two years ago. PPP's late July poll of that contest found Dole leading Hagan 49-40. By late August Hagan took a lead she would never give back, but that was after a large amount of outside advertising attacking Dole over the course of that month. It does not seem likely that will be replicated this year, at least not at this early a stage.

Turnout in the Triangle may end up being the key to Marshall's chances. She trails in most of the state but has a 50-33 lead in the region where the greatest numbers of votes is likely to be cast and that's what's making the race competitive.

The number of voters who dislike Burr has been rising this year. The number who like him has not. The big question now is whether his millions will be able to get that turned around.

Full results here

21 comments:

Anonymous said...

sorry I can't laugh enough over this poll

Rasmussen has Burr up by a lot. Notice Shuler in trouble by SUSA poll and u know Dems are getting runs in their shorts

No way will NC vote for Dem and 7% for Libertarian candidate is beyond laughable

Anonymous said...

RCP average is close to 10. Seems to be an outlier

Moe Lane said...

Jeebus, Tom. This was precisely the mistake that everybody who wasn't you or the Raz made about NJ-GOV: overstating third-party candidate support. When's the last time a Libertarian Senate candidate hit 7% in NC?

Anonymous said...

Looking at the Obama/McCain question, you seem to imply the same likely voter in 2010 that existed in 2008. Perhaps this is a result of your weighting to 2008 demographics. You should consider addressing this issue at some point as it appears to be at odds with the current polling narrative.

Anonymous said...

The money didn't save Arlen Specter or Bob Inglis. It isn't making Jeff Green any more popular either.Money only goes so far.

Unknown said...

Will the Washington poll be released today as well?

Anonymous said...

Anon- If you're not going to take the most recent election as a baseline, then what would you suggest for the demographics, especially in a non-Presidential election year? Can you pin point exactly what turn out will be and where it will come from? Is it not possible that the GOP will be less energized than expected because of Burr's lackluster performance. In the same breath, is it not possible that Dems will not be energized because they are not excited about their candidate? Nobody knows the answers, hints why the 2008 is used as a demographic baseline. Unless you have a magic 8 ball that will predict turnout, I think using 2008 is about the best we can ask for from a pollster.

Moe Lane- If you think the Libertarian getting 7% is the reason for the closeness of the race, then I'm sure you'd agree that the race is probably even closer than you think because right now Marshall is only getting 65% of the Democratic vote. Probably not exact, but when Libertarian support goes down and Marshall's Dem support goes up, those are likely to balance out giving us a similar result, or even a Marshall lead.

It's clear all you folks in the GOP are scared that Burr could lose to a fairly unknown Democrat, so you attack PPP. Maybe your time would be better spent trying to help your candidate explain why he has been totally ineffective for the last 6 years. Then again, Burr was way too busy checking his ATM receipts to worry about his constituents.

NRH said...

Rasmussen is the only pollster to have Burr with an approval rating north of forty, and in fact north of fifty as well. Everyone else shows a substantial proportion of North Carolinians don't have much of an opinion of Burr at all. If Rasmussen's screen is picking up only the people who are engaged in the race in the middle of the summer, then that could very well explain why Ras' results are so far off from everyone else. Then again, Ras has a distinct history of trying to set narratives instead of uncovering reality, manipulating his 'likely voter screen' to Republican advantage - he had McCain winning Florida in 2008, running even in Ohio, winning North Carolina, winning Indiana, and running six points closer in New Hampshire. Ras pegged Rand Paul to a 25-point lead immediately after the Kentucky primaries and then sheepishly backed off of the blatant attempt to scare off Democratic support when multiple other polls showed a neck-and-neck race. Ras tried to prop up Angle in Nevada until her incompetence made her too obviously a loser. Even in New Hampshire's blowout gubernatorial race, Rasmussen consistently has had Democratic governor John Lynch a full five points weaker than any other pollster.
As PA-12 reminded everyone but Ras, highly-motivated votes end up counting exactly the same as apathetic ones. Rely on voter screens this early at your peril.

DBL said...

The last comment is funny. "Unless you have a magic 8 ball that will predict turnout, I think using 2008 is about the best we can ask for from a pollster."

Every single poll tells us that less Democrats will show up this year and more Republicans will show up. Last year New Jersey went from D+16 in 2008 to D+10. Virginia went from D+6 to R+4. The Democratic base, 18-29 and African-Americans, dropped significantly from 2004 to 2006 despite 2004 skewing Republican and 2006 heavily Democratic. These groups show up in much smaller numbers in off-year elections.

In 2004 North Carolina was R+1. In 2008 it was D+11. PPP has it D+12, meaning that Democrats are going to show up in even greater numbers than 2008. PPP is the only pollster that continually shows this.

Anonymous said...

It has more to do with PPP consistently being an outlier on this race. Rasmussen has Burr up by 15, Survey USA has Burr up by 10, and Civitas has Burr up by 7. When other pollsters start to verify PPP's numbers, I will take a closer look at this race.

Anonymous said...

This polls makes me laugh. PPP polls are so bias these days. To think the the 2010 is going to have the same voters in 2008 is just bias and stupid. But hey, if you want to make it look like Burr is going to lose then use your polls to make people think Democrats are not doing to bad then go ahead.

Anonymous said...

Every single poll tells you that Dems will show up less because the pollster is asking less Dems because that's what he/she believes, not because the pollster knows these individuals won't go vote.

Also, NJ and VA are much different than NC Senate, especially in terms of Democratic support. There is no weak Democratic candidate or incumbent. The GOP is the incumbent. We have seen time and time before that this is much more a anti-incumbent year than an anti-Democatic year. In both NJ and VA Democrats were the incumbents and voter anger was directed towards Dems. In NC that anger is going to be directed towards the incumbent GOP.

What's the sign of a nervous candidate/supporter? Attack a pollster.

Moe Lane said...

"Moe Lane- If you think the Libertarian getting 7% is the reason for the closeness of the race, then I'm sure you'd agree that the race is probably even closer than you think because right now Marshall is only getting 65% of the Democratic vote."

No, Sparky, I'm extending a semi-professional courtesy (blogger to pollster) to Tom by pointing out that his results are historically atypical for the state. Now hush: grownups talking.

Tom Jensen said...

Moe,

I don't think the Libertarian will really get 7% in the end but that's what the respondents to the poll said at this point in time. It's acceptable for pollsters to weight for things like gender, race, and age but when you start weighting for what percentage of the vote you think a particular candidate's going to get that's when you start getting on pretty shaky ground. I'm sure Beitler won't get more than 3% in the end but that's where he is for today.

Anonymous said...

Did you also test "Are you more or less likely to vote for a candidate that opposed Obamacare" or "Are you more or less likely to vote for a candidate that opposed the stimulus bill" in addition to TARP?

Why not "Are you more or less likely to vote against a candidate that opposes Arizona's immigration law"?

Moe Lane said...

Tom: Fair enough. I was misremembering the third-party polling for NJ-GOV anyway.

Timothy said...

The enthusiasm gap alone will push Burr over the finish line. However, Burr is SUCH a GOP squish, it's no wonder the GOP base isn't exactly gung-ho about him.

Anonymous said...

Moe Lane- leave the polling to the grown ups.

Dustin Ingalls said...

"Did you also test "Are you more or less likely to vote for a candidate that opposed Obamacare" or "Are you more or less likely to vote for a candidate that opposed the stimulus bill" in addition to TARP?

Why not "Are you more or less likely to vote against a candidate that opposes Arizona's immigration law"?"

No to all of those. The point in asking what we did is that, as pointed out in the linked press release, Burr voted for TARP, and Republicans overwhelmingly oppose it. A majority of all North Carolinians do as well.

Anonymous said...

PPP did fine in a Dem cycle. Watch what happens this time around. Tom will start to weight his polls correctly in early to mid October.

Anonymous said...

>Burr voted for TARP, and Republicans overwhelmingly oppose it. A majority of all North Carolinians do as well.

..... and this is why I think you're wrong when you say the Libertarian (Beitler) won't get more than 3% in November. A LOT of Conservatives are pissed at Burr for his support of TARP, The Patriot Act, nation building (Iraq and Afghanistan), and more. They will vote - but not for him. The GOP will eventually get the message that either they kick the neoconservatives out or they will continue losing to the Democrats.

 
Web Statistics