The Democrats’ New Normal (Nate Silver's take on the Gallup poll)

[FBaggins]

The poll is probably an outlier of sorts, by which I mean that if you were to take the exact same survey and put it into the field again — but interview 1,450 different registered voters, instead of the ones Gallup surveyed — you would most likely not find the G.O.P. with a 10-point advantage. This week’s generic ballot survey by Rasmussen Reports actually bounced back toward the Democrats somewhat (although still showing them with a 6-point deficit); polling averages have them trailing by around 5 points instead; and there was no specific news event last week that would have warranted such a large shift in voter preferences.

...snip...

Still, even if the poll is an outlier, that doesn’t mean it should simply be dismissed. Instead, the question is: an outlier relative to what? If the Democrats’ true deficit on the generic ballot were 5 points, it would not be all that unusual to have a poll now and then that showed them trailing by 10 points instead, nor would it be so strange for a couple of polls to show the race about tied. Indeed, that seems to be about where the generic ballot sits now. No non-Internet survey has shown the Democrats with a lead larger than 1 point on the generic ballot for over a month now, whereas their worst results of late seem to put them in the range of 10 points to 11 points behind.

...snip...

The “good news” for the Democrats is that the generic ballot almost certainly isn’t the only metric you should look at when forecasting midterm elections, and the other salient statistical indicators, while poor for Democrats, are not quite this poor. More on that when we release our House model, which is coming soon.


http://fivethirtyeight.blogs.nytimes.com/2010/08/31/the-democrats-new-normal/

[Liberal_Stalwart71]

1. My take on polls is that until they start asking more people of color, they don't mean squat.

I'm a life long, very active Democrat, as are my friends and family. None of us have ever received calls.

[Stuart G]

2. On the 2008 election..Nate Silver got it correct withing

a percentage pt...Someone to at least listen to, not necessarily agree with..

[sadbear]

3. Republicans aren't running generic candidates

When a republican name is put on the ballot, people will vote Democratic.

[FBaggins]

4. To many voters they are.

Especially in a midterm election, a significant number of voters just look at the party affiliation.

Regardless, the generic polling does have predictive value for congressional elections. Mr Silver certainly takes them seriously.

[KingFlorez]

5. And at the same time, Republican leaning Rasmussen shows a narrower gap

For months, they've had Republicans up 9 or 10 and now that gap is only 6. Rasmussen doesn't have Republicans with narrow leads much, so I'd say there is something to that. Gallup generic ballot is all over the place, weeks ago Democrats had a lead.