We Ask America Polls Ohio, Florida and Virginia — Maybe Democrats Were Too Upset to Respond?

I’ll give PPP one thing, they are at least consistently biased.  Even though I don’t blog their polls when the results fly through a news crawl or twitter feed I simply adjust the margin 5-points in favor of the Republican and I generally get what is likely a fair result.  We Ask America, on the other hand, is all over the map. When they first showed up on the scene they were accused of being Republican shills.  Then they went completely in the tank for Obama with survey outputs impossible to achieve this election cycle.  Now their post debate polls show large swings of support towards Romney in the tightest of Battleground States. You have to wonder if they are smoking some wacky tobacky in their data crunching centers.  Anyway here are their three post-debate polls:

Few disagree that Mitt Romney won Wednesday night’s presidential debate. Even fewer would argue against the notion that Romney sorely needed that performance to put his campaign back on track. With the national polling averages hovering around three percent in Barack Obama’s favor, Romney’s inability to gain traction was resulting in a pundit-driven drum beat of inevitability. Not so fast there, muchachos. Polling in key states has shown a widening gap, but in many states, the spreading numbers have been more meandering that directional. True, if it had continued that way the election could have gently slid into the point of no return for Romney, but they didn’t. There’s still time for this thing to get real interesting. Could Romney’s debate performance be the key that unlocks a surge?

To find out, we cranked up the blowers last night in Virginia, Ohio and Florida to ask the electorate who they would vote for if the election were held today. Here are those topline results along with the splits among Republican/Democrat/and Independent voters (self-described by participants): Each of these automated polls had 1,200+ responses from likely voters. MoE 3%. Each was weighted to correct for over-/under-sampling among select demographics.

Ohio

Mitt Romney leads by 1%, 47 to 46 with 1% going to Gary Johnson and 6% Undecided.  The party ID is reasonable with a 4 point drop from Obama’s 2008 turnout advantage.
Party ID was D +4 (Dem 38, Rep 34, Ind 28). This compare to 2008 of D +8 (Dem 39, Rep 31, Ind 30) and 2004 of R +5 (Dem 35, Rep 40, Ind 25)

Florida

Mitt Romney leads by 3%, 49 to 46 with 1% going to Gary Johnson and 3% Undecided.  The party ID is aggressive in favor of Romney outpacing Bush’s 2004 turnout advantage. With the steep change in registrations and purging of illegal voters this party ID may eb achieved on election day but for polling purposes, it is too aggressive in my opinion.
Party ID was R +5 (Dem 33, Rep 38, Ind 29). This compare to 2008 of D +3 (Dem 37, Rep 34, Ind 29) and 2004 of R +4 (Dem 37, Rep 41, Ind 23)

Virginia

Mitt Romney leads by 3%, 48 to 45 with 2% going to Gary Johnson and 5% Undecided.  The party ID is again a little aggressive in favor of Romney, only 2-points off Bush’s 2004 turnout advantage. This state feels like a 50/50 state to me for party ID so R +2 isn’t way off but it does skew slightly Republcian for my tastes.
The party ID was R +2 (Dem 32, Rep 34, Ind 34). This compare to 2008 of D +6 (Dem 39, Rep 33, Ind 27) and 2004 of R +4 (Dem 35, Rep 39, Ind 26)

17 Comments

  1. Pete
    Posted October 5, 2012 at 11:27 am | Permalink | Reply

    Re-post from earlier thread:

    Further on Rasmussen poll from Washington Examiner : “But among the stunning 92 percent of all voters in the state who say that they are certain to go to the polls on Election Day, Romney leads 51 percent to 48 percent. And among the 83 percent who have already made up their minds how they will vote, Romney is ahead 52 percent to 48 percent.”

    • Pete
      Posted October 5, 2012 at 11:28 am | Permalink | Reply

      Regarding Ohio

  2. Eric
    Posted October 5, 2012 at 11:29 am | Permalink | Reply

    I’m in the minority here I know, but I believe that party IDs for 2012 will not be between 2004 and 2008’s numbers. I think they will be more Republican than 2004. Nationally 2004 was even, while 2008 was Dem+7. I see 2012 somewhere around Rep+2-3.

    About 25% of the electorate makes up their minds in the month preceding the election. Romney will probably win these late deciders overwhelmingly. Late deciders went to McGovern in 1972, to Reagan in 1980, to Mondale in 1984, to Dole in 1996, and to Kerry in 2004. The 1992 numbers were skewed by Perot. The last time late deciders went to an incumbent president was in 1956 to Eisenhower.

    These late deciders will identify as Republicans more often on Election Day when they talk to exit pollsters, so the final exit polls will have more Republicans than Democrats. An early indicator that things are headed this direction is Romney’s lead among independents already.

    I see Romney winning by 3-5 points on Election Day, and it could push up to 6 or 7 points should Obama do horribly in the next 2 debates.

    Remember 2010. Most pundits saw it as a tossup or slight Republican advantage for winning the House. 2 years ago I predicted that Republicans would win between 60-80 seats. I saw 70-75 as the most likely scenario. Most pundits had it between 35-45 seats (something around there). The result was 63 seats gained. They were shocked by how many Republicans seats were picked up. I was a bit disappointed on election day because I wanted to see about 10 more.

    The pundits are always wrong. Romney is the favorite for this election.

    • Posted October 5, 2012 at 11:38 am | Permalink | Reply

      Eric, Team Romney may achieve that on election day but for polling purposes I think the correct and conservative measure is a split between 2004 and 2008. If a campaign wants to run internal polls at D +70 or R +70, they are more than welcome to do that. But I think fairness from “independent” organizations calls for something far more balanced than we have been seeing. Thanks for the comment. I hope you’re right!

      • Eric
        Posted October 5, 2012 at 12:07 pm | Permalink

        You’re right. Today’s polls are measuring voter’s opinions today. That is a static analysis though, and it’s the best that we can objectively do. I’m just saying that we should expect changes in the polls from now until election day. Normally, polls move around 6-8 points in the final month before an election. So, trends are very important.

        We saw this happen in the primaries. I studied the primary battles between Romney and Santorum as they were happening. It was tough to tell who would win a given state because oftentimes the polls were not necessary wrong, but they were out-of-date. I picked the correct winner in every single election in the primary with the exceptions of Iowa and Colorado, where I missed Santorum winning those 2. Understanding that a lot of people don’t decide until they really have to decide is key. It’s human nature to procrastinate. If a pollster calls someone who hasn’t really thought much about the election, and the pollster asks them who they are going to vote for, they’re likely to tell them one of three things. Either they are undecided, or they’ll say Obama because they voted for Obama last time around, or they’ll just give out a random answer.

        When this late-deciding voter has time to finally sit down and figure out who they’re going to vote for, that’s when it really matters, and usually the late-deciding voters break in one direction. So, someone who called themselves an independent on Sept. 30 (because they hadn’t really thought about it yet) and decides to vote Republican on Nov. 6th is more likely to call themselves a Republican on Nov. 6th. Someone who considered themselves a Democrat on Sept. 30th when they were undecided (but have voted Democrat in the past) and then they finally decide to vote for Romney on Nov. 6th. That person will probably describe themselves as an independent to an exit pollster if asked.

        So, it’s hard for pollsters to really give an accurate assessment because they are trying to measure a moving target (especially in the last couple of weeks before an election). Trends are really important, and the first trend we see in October is one moving towards Romney.

  3. margaret
    Posted October 5, 2012 at 11:34 am | Permalink | Reply

    There was a telephone polling outfit like this, I think in 2000, that consistently showed Bush in the tlead, and it had been over-polling Republicans. I would like to think the results are accurate but I’m wary. I’ve read too many reports from our side saying they pretended to be something they were not for a telephone poll e.g low-income minority voter for Romney or something, just to mess up the poll.

  4. housebroken dad
    Posted October 5, 2012 at 11:45 am | Permalink | Reply

    I’m really liking that OH number since I think it will be right around a break even D/R split given the early voting numbers and reduction of D registrations. Ras has O +1 and it appears he oversampled D’s since Indys like Romney 58-40.

    • damien
      Posted October 5, 2012 at 2:02 pm | Permalink | Reply

      gotta be a oversample somewhere…58-40 among indies and still lose?

  5. William Jefferson
    Posted October 5, 2012 at 12:04 pm | Permalink | Reply

    It depends. Pollsters may, indeed, be letting their partisan ID “float” based upon whoever makes it through their likely voter screens. Most likely voter screens are affected by enthusiasm; thus, we might have seen huge D oversamples after their convention because they were particularly enthusiastic. Now, after the debate, the pendulum has swung. Although I would like pollsters to make a realistic guess at what turnout will be on election day–and weight accordingly–there is logic in letting the sample float according to the different tides of partisan enthusiasm at the moment.

  6. Posted October 5, 2012 at 1:21 pm | Permalink | Reply

    Even Ras with a D+3 (?) has Romney up in FL.

  7. Kevin
    Posted October 5, 2012 at 1:27 pm | Permalink | Reply

    What none of these polls are showing is the lack of enthusiasm for Obama. The young people that voted for Obama four years ago aren’t going to be as nearly jazzed up for him this time around. Four years ago, he was “cool” and new. He was the young guy with no gray hair that had a blackberry, the most popular phone at the time. Now, Obama is Mr. “eh, yeah, whatever dude”. This is a voting block that wants to trade in their cell phones every 9 to 12 months because their phones are “outdated”. This is a voting block that “goes with the flow”. Obama isn’t “cool” anymore, and he’s “outdated” in the world they live in.

    Obama will still get a lot of the young voters, but not near the same number as he did four years ago. The college students turned out to be the deciding factor in North Carolina, and Florida four years ago. They won’t be the deciding factors this time around since he’s not drawing the massive crowds on college campuses he did four years ago.

    • jeff
      Posted October 5, 2012 at 1:36 pm | Permalink | Reply

      Sometimes I think Scott Rasmussen is a paid hack. With the empirical evidence Showing that Republican enthusiasm is way up and that they are exrtremely energized why is he stikk sticking to D+3? Why cant he just split the difference just to be on the safe side and use the 2004 model with Ds and Rs evenly split? Most likely wete goibg to see a 2010 model

      • jeff
        Posted October 5, 2012 at 1:39 pm | Permalink

        I have the feeling that Scott is weary of showing numbers with Romney ahead for fear of beong blasted by yhe msm and BO Chicago hit squad.Maybe hes waiting till the very end to start using more realistic models.

  8. damien
    Posted October 5, 2012 at 2:00 pm | Permalink | Reply

    I think Rasmussen is holding it close…so afraid to go to deep for either guy…

    Ohio is the key…I just dont see how romney gets 270 without Ohio..Unless he pretty much runs a sweep thru all the other swing states and i just dont trust iowa or nevada to come thru..

    • No Tribe
      Posted October 5, 2012 at 5:31 pm | Permalink | Reply

      Romney should win Iowa easily. I think that the voters are holding back there. Obama has showered Iowa and Ohio with federal largesse the past four years. Romney bashing the green jobs, well, alot of that pork has gone into making wind energy in Iowa, subsidizing it of course.

      Ohio has one out of seven that are getting money from the subsidized auto industry.

      So, those, out of all the battlegrounds, are going to be the toughest states for Romney to win. But he can, the underlying demos are in his favor.

  9. Posted October 5, 2012 at 4:44 pm | Permalink | Reply

    Honestly….I would much rather continue to see Romney dig, claw and scrape his way to a lead with these polls showing D+2-1000 THEN i would see him go up with polls expecting a huge Republican turnout.
    Because that way if and when he wins and the liberals cry…you can point to the fact the polls were off on how the electorate turned out. If the polls turn to a R+2 breakdown and he wins by .5% they will all cry that it was republican leaning polls, etc.
    It is a better story he is clawing his way to the lead and it will just further demoralize the democrats.

    • No Tribe
      Posted October 5, 2012 at 5:32 pm | Permalink | Reply

      It doesn’t matter. When Romney wins, no matter what, they will all cry out. Just enjoy it.

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: