Is Mitt Romney Already Winning the “Toss-Up” States?

Politico ran a Battleground State poll earlier this week that showed a tight race between the two candidates with Obama leading Romney by 2-points. But digging deeper into the internals you see they broke down the states even further into “Safe Republican”, “Safe Democrat” and “Toss-up” states. Mike Flynn at picked up on this and points out that Obama’s lead is based on big leads in the “safe Dem” states (like New York and California — how are they even included in the poll?) but Romney is leading in the true toss-ups:

Each candidate leads in states considered “safe” for their party. In safe GOP states, Romney leads by 8. In safe Democrat states, Obama leads by a massive 22 points. But, in the more numerous and more important “toss up” states, Romney leads by 4, hitting the critical 50% threshold.  In the slightly different category of “battleground” states identified by Politico, Romney leads by 2, 49-47. Romney’s lead over Obama is powered primarily by his edge with independents. Romney leads Obama by 4 among the important swing voters. By 11 points, these voters think Romney would do better on the economy than Obama, 51-40.  Romney also has a big edge with middle class families, who prefer him over Obama by 15 points, 56-41.

This is huge and massively under-reported (no surprise).  I went back and looked at the internals and the toss-up states are horrible for Obama. I’m trying to find out which states encompass each category according to Politico but here are the internals according to the Monday poll.

Among only the toss-up states:

  • Right track/wrong track is 34 Right track/61 Wrong track
  • Barack Obama Favorable/Unfavorable is 46/51
  • Obama strongly favorable/strongly unfavorable is 36/44 (this gets to voter intensity and turnout)
  • Mitt Romney Favorable/Unfavorable is 50/45 — his best this cycle
  • Obama job approval/disapproval is 45/53.  This nearly always correlates to an incumbents vote percentage and 45% is horrible
  • Presidential ballot: Obama 46 and Romney 50 (as Flynn pointed out)
  • Within the Presidential ballot both candidates are nearly certain to get the same 43% of the vote (we’re feeling more and more like a 50/50 country more similar to 2000 and 2004)
  • Obama job approval on the economy is 42 approve versus 57 disapprove
  • Better handle the economy, Romney leads 53 to 44
  • Better handle foreign policy Obama leads 49 to 41
  • Better handle medicare Obama leads 47 to 45 — essentially a draw which is a win for the GOP on this issue
  • Better handle jobs Romney leads 53 to 45
  • Better handle healthcare? Dead even at 47 a piece
  • Shares your values?  Romney leads 47 to 46 (this one surprised me)
  • Toss-up states respondents were the most likely voters to actually vote
  • Generic congressional ballot was democrat 45 and Republican 44 (another surprising result)

Toss-up state demos: This toss-up group was nearly 40% of the entire survey which contrasts greatly from the inconceivably small subset from the Washington Post yesterday. 76% of respondents were from the middle class, 19% lower class and 4% were upper class. Age breakdown was distributed evenly. Party ID was perfectly even at Dem 41, Rep 41, Ind 18 … very interesting. Demographics were hard to judge since we usually compare to national figures and these are state-specific breakdowns which won’t match the nation percentages. Relative to the national averages, the toss-ups were low on African-Americans and Hispanics but that is also true in states like Iowa, New Hampshire, Colorado and Nevada. Male/Female split was 49/51.


  1. margaret
    Posted October 3, 2012 at 3:28 pm | Permalink | Reply

    Wow. Rasmussen quietly released his September summary of party affiliation and it’s R+2.6 (down from R+4.3 from August). For the last quarter it was R+2.6. Has anyone seen any poll anywhere using a R+2.6 turnout model? Not even Ras uses that for his models, which people say are D+3 to D+5. But it’s a big clue.

  2. bman77
    Posted October 3, 2012 at 4:16 pm | Permalink | Reply

    That is curious. Does anyone know why Rasmussen is not following his own model? I wonder if pollsters like Rasmussen will begin to reduce the D party ID advantage in their polling even further.

  3. AussieMarcus
    Posted October 3, 2012 at 4:42 pm | Permalink | Reply

    Obama winning his states by 22 but Romney only by 8? Which states are they polling??

  4. Mike DeVault
    Posted October 3, 2012 at 4:56 pm | Permalink | Reply

    The article says California and New York. This is why I am always curious about national polls because you can “skew” the poll just by pollling more people from places like NY and California rather than folks from North Dakota or Texas.

  5. Posted October 3, 2012 at 5:03 pm | Permalink | Reply

    so clue me in…I thought Rasmussen’s way was to do the separate party id poll and then use that as the basis of his own polling? Is that not true does he just take the Party Affiliation temperature but not use it?

    • Posted October 3, 2012 at 5:36 pm | Permalink | Reply

      IIRC Party ID and Party Registration aren’t the same thing. Eg there are a lot of old school Democrats in the South or Midwest who still register D but ID as a R voter.

      Ras uses a D+3 (?) Registration model, not a D+3 Party ID

  6. Mike
    Posted October 3, 2012 at 11:56 pm | Permalink | Reply

    Like I said before, if you get your family, friends, and yourself to the voting place and cast the vote for Romney/Ryan, there is not way that he can’t win.

    Romney / Ryan 2012 to save our country!!

Leave a Reply

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

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

Google+ photo

You are commenting using your Google+ 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 )


Connecting to %s

%d bloggers like this: