National Journal Poll Tied at 47 with a D +7 Sample; Obama Carries Only 38% of White Vote

The just released United Technologies/National Journal poll shows the race dead-locked at 47 a piece. Lots of juicy details underneath this tied race.  Romney leads among Independents by a whopping 8-points but is only tied overall. In 2008 Obama won Independents by 8-points and won the overall race by 7-points.  Bottom line this a good poll for Romney and a very curious poll for Obama.  Considering it is incredibly unlikely Obama repeats his 7% turnout advantage with self-identified Democrats on election day, he almost certainly has a far tougher road ahead than the media and recent polling would lead readers to believe. Consistent with my essay from yesterday, earlier national polls were overstating Obama’s support among White voters as is this National Journal poll.  Interestingly in this survey Obama noses above the 80% support threshold among non-White voters at 81% but dips below the magic 40% threshold for White support at 38%.  In a D +7 sample, White Democrats were likely the culprits for much of the over-sampling pushing Obama’s true support among Whites even lower.

President Obama and Mitt Romney are deadlocked among likely voters as they prepare to square off in their first presidential debate, according to the latest United Technologies/National Journal Congressional Connection Poll. The survey showed that voters remain resistant to either Obama or Romney holding full control of the federal government. Obama and Romney each pulled in 47 percent support in the poll among likely voters. It is among the narrowest margins of several presidential surveys published ahead of the debate this week. Other polls have shown the president with a slim lead. In this survey, while the race is tied among likely voters, Obama has a 5-point lead, 49 percent to 44 percent, among registered voters. The survey was conducted Sept. 27-30 and has a margin of error of plus or minus 3.7 percentage points.

Sub-groups:

  • Romney led in the poll among independents, 49 percent to 41 percent
  • Both candidates winning more than 90 percent support from their respective parties

Demographic breakdown

  • In estimating the turnout on Nov. 6, the poll projects an electorate that is 74 percent white, 11 percent African-American, and 8 percent Latino.
  • The survey had Obama winning 81 percent of the nonwhite vote and Romney carrying 55 percent of white voters.
  • Obama carries only 38% of the white vote

Party ID is D +7

  • The likely-voter party splits are 36 percent Democratic, 29 percent Republican, and 30 percent independent
  • This is the same 7% advantage Obama had in 2008 which is not expected to repeat itself in 2012

23 Comments

  1. johnfisher
    Posted October 2, 2012 at 9:14 pm | Permalink | Reply

    I think I’m giving up on the state polls. There are essentially useless.

    • shane
      Posted October 2, 2012 at 9:17 pm | Permalink | Reply

      considering the RCP average for north Carolina has dropped to a tie…yet Obama hasn’t even really ever campaigned there

    • JeffR
      Posted October 2, 2012 at 11:30 pm | Permalink | Reply

      The samples tend to be pretty small state by state and can’t represent them demographically very well.

  2. margaret
    Posted October 2, 2012 at 9:53 pm | Permalink | Reply

    Doesn’t conventional wisdom say state polls lag behind national polls? So if the national polls are “tightening up” (or pollsters are using less atrocious turnout assumptions), then state polls should be changing as well.

    • Posted October 2, 2012 at 11:53 pm | Permalink | Reply

      Except only the campaigns really do large enough state samples to actually have any idea what’s going on state by state.

    • David Weed
      Posted October 3, 2012 at 1:48 am | Permalink | Reply

      Actually, wouldn’t you think a state poll of say, 1000LV would be much more accurate than a national poll of 1500LV? I mean, we’re talking about an average of on 30 likely voters PER STATE! I’ve never understood how a national poll of such a small sampling can be accurate.

  3. Posted October 2, 2012 at 10:00 pm | Permalink | Reply

    If Obama’s sub 40% white vote he most certainly loses. If my math was right I think he’ll lose Ohio if he’s sub 43% in the white vote. The white vote is crucial in this election, for Romney he needs to assume Minorities turnout the same for Obama this time as last, if they don’t that’s gravy. From there he just needs to chip away at the white Obama vote. Hannity’s piece on Carlson’s obama video tonight was a good start at chipping away at some white obama voters who may not have seen the racist side of him. I think you’ll see more of this strategy to turn them.

    • JeffR
      Posted October 2, 2012 at 11:30 pm | Permalink | Reply

      I also think the national security issues will diminish his support among whites. The Fast & Furious reporting by Univision could hurt him with hispanics, at least turnout wise. Continued bad economic news is likely to embolden his opponents while making his supporters less enthusiastic.

      I think sub 40% of the white vote in 2008 would have lost it for him then, or made it much more interesting. Declined black and hispanic turnout coupled with an increase in white male turnout could turn this into a laugher.

      Much will depend on Romney tomorrow night. Expectations with undecideds are low, so he just has to do enough to convince them to either not show up for Obama or show up for him.

  4. Posted October 2, 2012 at 11:04 pm | Permalink | Reply

    Offtopic but has anyone seen this: http://www.nytimes.com/2012/10/03/us/politics/obama-outspending-romney-on-tv-ads.html

    It talks how Obama and friends are putting out 7 ads to 5 for Romney and friends.

    Thoughts? That seems to differ greatly from the “Money where your mouth is” posts.

    -ZR

    • WillBest
      Posted October 2, 2012 at 11:34 pm | Permalink | Reply

      Its possible that Romney is going after certain target demographics with his buys now, and then blanketing the negative ad buys the last 2-3 weeks. I saw an article not too long ago saying that negative ads only leave an impression in people’s minds for 2-3 weeks. Which is part of the reason gaffes don’t matter.

    • JeffR
      Posted October 2, 2012 at 11:52 pm | Permalink | Reply

      Seems to me that Obama and his people are taking a “kitchen sink” approach. I see his ads on Fox News here in Florida. HotAir.com has his ads sometimes in the sidebar. That’s poor targeting.

      I don’t see Romney’s ads on Fox News here in Florida. That’s preaching to the choir. I do see them on local network affiliates, like tonight during Dancing With The Stars. Same for the SuperPAC ads. I see them during football games too.

      But I’ve seen Obama ads on MSNBC and Fox News. Those are two worthless ad buys for him – one if preaching to the choir, the other is a pointless attempt at conversion. I’m not sure his people really understand demographic targeting. Seen it on the ground too here. The GOP GOTV effort is built on personal contact and very specific demographic targets. We don’t spend time with people guaranteed to show up and vote GOP nor do we spend any time with people who have no history of voting or who are highly unlikely to vote GOP. Collectively, we’ve knocked on over a million doors in Florida and the people I’ve spoken to personally have only ever been robocalled by Obama’s camp. We’re getting strong commits from over 80% of all personally contacted swing voters.

      • WillBest
        Posted October 3, 2012 at 1:18 am | Permalink

        Fox news has about 30% liberal viewership last I heard. But most of Obama’s ads on fox would be to suppress conservative turnout

        it wouldn’t surprise me that Obama wastes money in his own campaign. Its not like he has ever had to manage a competitive campaign ever.

      • WillBest
        Posted October 3, 2012 at 1:33 am | Permalink

        Incidentally, I like it when I happen across Obama ads online. the effort of clicking on the link is worth the satisfaction of charging him 50 cents or whatever

    • No Tribe
      Posted October 3, 2012 at 1:08 am | Permalink | Reply

      That was last week: On ads.

      Obama spent $19.2M in ads ending Oct 1st, Romney spent Romney spent $14M. DNC spent $0 and RNC spent 2.2 So 19.2 to 16.2, which is as close as Romney has ever been in a week of ad spending so far.

      $3.9 was spent by Obama’s outside groups, and just $2.8 in ads by outside groups for Romney, which is probably a first– where Obama’s outside groups outspend Romney’s in a week on ads.

      Not this week. From here out, Romney and his side will be outspending Obama and his.

  5. AussieMarcus
    Posted October 2, 2012 at 11:57 pm | Permalink | Reply

    What’s interesting now is that even with exaggerated 2008-like samples, it’s basically a tie. Not even D+7-8 can get Obama above margin of error anymore….

    The number of ridiculous “Obama up 10” polls is decreasing rapidly, and the ones that remain are now more obvious outliers.

  6. Posted October 3, 2012 at 12:07 am | Permalink | Reply

    first if you take the RCP average and throw out the highest O+9 and lowest National Journal Tie you have an O+2.5 average which is more reasonable.

    Jeff I think you MAY have a point. Obama’s appeals are to the under 30 crowd so year the fact they data mine facebook and pretty much stalk their prey may be cool to them. But to most people having someone knock on your door, talk to you and ASK for your vote still means something to some people. I think Obama’s super double secret data mining operation is awesome for raising money, getting volunteers, etc but actually getting people out to vote…..remains to be seen. Demos have always been great with the homeless, unemployed, etc they can pay to go vote. 2008 contrary to public opinion he didn’t get the youth out in any major increased numbers he just took an overwhelming majority of those who did vote. 4 years is a whole other world. Those bright eyed 18 year olds who cried at his election are now 22, graduated college and probably unemployed and living with mom and dad making a 1/3 of what they thought they would with their degree. They won’t come and vote for Romney but I have my doubts they vote.

    It all comes down to turn out if it stays close. If it breaks open after the debates. we shall see. The columns I am reading tonight speculating that “insiders within the GOP” are planning to pull out of the POTUS race and focus on the House and Senate are hilarious…NO ONE is pulling out of a POTUS race with a month to go and the race tied. Those are the BS trying to suppress republican vote campaigns.

    • WillBest
      Posted October 3, 2012 at 1:31 am | Permalink | Reply

      The tie clearly isn’t the outlier. For starters we know Gallup is an outlier with a +17 or more data point. Second its MoE is 4.5 meaning that every poll except Gallup and WashTimes is inside its 95% confidence band.

    • Posted October 3, 2012 at 9:18 am | Permalink | Reply

      I think the problem with trying to suppress the GOP vote is that it won’t work this time. Republicans often do fall in line and work far less off emotions. But I think 2008 was a huge wave and a lack of enthusiasm due to many of us being tired of Bush and not completely happy with him. McCain, with his complete lack of economic care or knowledge, didn’t help.

      The problem with trying to suppress GOP turnout is that you also are telling Dems that the election is over. There are a lot of people in their coalition who will not show up if that’s the message. Democrats are overconfident and that may cause their turnout to go down. The media may be helping that, but they’re in a bind. Show it as a tight race and the GOP won’t be suppressed, show it as a wide open lead for Obama and his voters may not show as much. This isn’t 2008 where emotions were riding high. Liberals run off emotions and unless he can get people angry at Romney, the emotion isn’t there. The amount of passionate Obama supporters has gone down dramatically in the last four years.

  7. Kevin
    Posted October 3, 2012 at 1:37 am | Permalink | Reply

    The DNC and the Alinsky Media are trying to run out the clock on Fast, and Furious, and now the Libya attacks since they know they make Watergate look like petty theft. If we had an honest media that would report accurately about Fast and Furious, and the Libya attacks. If we had a media that did investigative journalism into those two scandals, Obama would be down 8 to 12 points in every poll.

    I told my girlfriend back on July 4th, after Labor Day, you won’t see Obama trailing in one poll from the bias polling groups, and that if anything bad would come up, the media would suppress the stories. Sure enough, I was right.

  8. Ron
    Posted October 3, 2012 at 7:07 am | Permalink | Reply

    Anybody seen Peter???….(crickets)..

  9. No Tribe
    Posted October 3, 2012 at 10:21 am | Permalink | Reply

    I got an idea, let’s do a poll with an RDI of +7 Democratic. Let’s have a Democratic firm do it for us. Let’s do it on the taxpayer’s dime. Obama up by 7% yea:
    General Election: Romney vs. Obama NPR Obama 51, Romney 44 Obama +7

  10. Greg
    Posted October 3, 2012 at 10:42 am | Permalink | Reply

    It would be interesting to do a chart of Obama lead in each individual poll over time and the +D skew. Saw NBC poll tighten in Vir and Fl. just because their skew narrowed in the past month.

    • Travis
      Posted October 3, 2012 at 11:21 am | Permalink | Reply

      Go to unskewedpolls.com

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: