Skip to main content

View Diary: Daily Kos Elections Polling Wrap: Attack of the Republican Snap-Pollsters (306 comments)

Comment Preferences

  •  Where does (6+ / 0-)
    Recommended by:
    Supavash, pollbuster, itskevin, askew, Linda1961, IM

    the national polling show any overall change in the race? Gallup.O+1, Ras. No change. Reuters? R +3. Not much movement at all and certainly not enough to lead one to believe that somehow Obama will be down a couple to few points across the board a week from now.

    •  Gallup is ONE DAY out of seven (2+ / 0-)
      Recommended by:
      auapplemac, Losty

      That's post-debate!  Reuters is 2 of 5, and the lead has gone from 9 to 5!

      How is this 'no change'?  

      And now we use Raz to prove a point?

      Why such aversion to reality here?

      •  Dude, you're a little unhinged (15+ / 0-)

        You put forward a "reality" which doesn't yet and may not ever exist (your national polling prediction) and are criticizing everyone who isn't as pessimistic as you.  It's frankly a little silly, since you do it on a constant basis.

        The pleasure of hating...eats into the heart of religion...[and] makes patriotism an excuse for carrying fire, pestilence, and famine into other lands. - W. Hazlitt

        by rfahey22 on Fri Oct 05, 2012 at 06:47:51 PM PDT

        [ Parent ]

      •  Reuters is four daily samples, not five (4+ / 0-)
        Recommended by:
        itskevin, EcosseNJ, LordMike, askew

        It shows a five date range, but the last "day" is only the morning of the release day. The first day of the sample is from the mid-morning or mid-day cutoff on, so it's four 24 hours samples.

        That changes the math somewhat on how much of an influence one day can have.

      •  Yeah, but you'd expect (4+ / 0-)

        a small move in Gallup. It was clear in Gallup that Obama was getting a bump during the convention, even though the poll never moved by more than 2 net points in a single day. That Obama actually gained yesterday suggests the day was nothing special for Romney.

        Rasmussen is a 3 day track and is worth something, if only for trends.

        Even the Reuters poll isn't that bad Obama was at 46-41 on both Monday and Tuesday in that poll - and his RV score was actually worse on Monday.

        This said, I do think Romney will gain from his debate performance, and I think the totality of polling shows this. He has been performing under - almost certainly - his floor, which I'd say is 47% (no pun intended), so there was room for him to grow if he could give those inclined to support him a reason to vote for him. But your prediction that he will be down in all national polls next week is  much too extreme and is not born out by data.

      •  I understand very well (3+ / 0-)
        Recommended by:
        Micheline, Stroszek, LordMike

        That Gallup is on a 7 day rolling. And however terrible Ras is overall as a pollster, it doesn't mean that it's totally worthless for comparison. Also, Obama's top line number has been unchanged in Reuters, showing no loss in support.

        So, I'd say its relatively insane to predict a national trend based on 2 days of just 1 poll.

        I still might be wrong, and you might be right. But that doesn't mean that your prediction is based in sound evaluation of data.

        The theory that somehow clutching our pearls, standing by the fainting couch, and setting our hair on fire will help us win elections has been proven false over the last 30-50 years.

        I'm more the type who invests in popcorn futures.

Subscribe or Donate to support Daily Kos.

Click here for the mobile view of the site