Skip to main content

View Diary: There Was Probably No Debate Bounce - And No Convention Bounce Either (27 comments)

Comment Preferences

  •  Here is how you know it was always going to be (0+ / 0-)

    close: Look at Obama's job approval rating.

    I don't agree that there wasn't a bounce after the Democratic Convention, and I would note the best single day in state polling for Obama was September 30th.

    I DO think you can make the argument that the first debate accelerated waking GOP voters who had slept.  I could see that visually in Florida.  In this sense you can argue the bounce Obama got from the convention was never going to last, and the debate actually just served to restore the race to a proxy for Obama's job approval.

    I don't really buy that, though.

    I don't think your subsamples of geogrphical shifts are really valid.  You can look at state polling shift:

    Here are the red, blue and close state margins:

                  July          Aug          Sept        Oct(post debate)
    Red         2.24    1.53    4.99    0.72
    Close       3.14    1.40    3.93    0.70
    Blue         1.21    1.81    3.82    0.87

    The notion that there is a difference in state polling between close states, red states and blue states does not survive this analysys.  When you look at real state polls and not small samples - I see no evidence that Romney or Obama are outperforming in their respective bases. It does show, though, that the first debate bounce was bigger in red states than elsewhere.

    I don't actually expect people to care about this fact: people are looking for an explanation for the divergence between national and state polling - and the simple answer is that they cannot be reconciled.  

    The bitter truth of deep inequality has been disguised by an era of cheap imported goods and the anyone-can-make-it celebrity myth - Polly Toynbee

    by fladem on Thu Oct 25, 2012 at 11:28:31 AM PDT

    •  State polling is better (1+ / 0-)
      Recommended by:
      MBishop1

      When the states are polled! The problem is that the states where the national data shows the biggest swings, and also the states where a large chunk of the national change comes from, are rarely if ever polled: how many polls of TX do you have each month in your database?

      But you are definitely right about looking at the approval rating...

      •  In September (0+ / 0-)

        there were 24 polls in Red states and 44 in Blue states. Since the debate there have been 19 polls in red states and 38 in blue states.

        Ironically, Obama projects to a national lead of 3.13 in southern states. He underperforms in the Mountain West (by about 1 point) and the Northeast (by about .75).  His numbers in the Pacific West are identical to the projected state national number.  

        There were three polls in Texas in September.

        The bitter truth of deep inequality has been disguised by an era of cheap imported goods and the anyone-can-make-it celebrity myth - Polly Toynbee

        by fladem on Thu Oct 25, 2012 at 12:32:33 PM PDT

        [ Parent ]

        •  The last Texas polls showed a shift of -7 (0+ / 0-)

          to Romney from 2008, which is consistent with a tied national race.  None of the Texas polling supports a massive shift from '08 beyond that seen in other regions or states.  For example, Michigan and Nevada have seen larger shifts.

          The bitter truth of deep inequality has been disguised by an era of cheap imported goods and the anyone-can-make-it celebrity myth - Polly Toynbee

          by fladem on Thu Oct 25, 2012 at 12:35:56 PM PDT

          [ Parent ]

Subscribe or Donate to support Daily Kos.

Click here for the mobile view of the site