This is being reported as horrible news for Democrats:
... Republicans have captured a lead in the areas home to the year’s most competitive races, according to a new POLITICO poll.While this isn't good news for Democrats, I question how dire it is. Go to the fine print of the poll data and you learn that 68 House districts were surveyed. Of those, a majority -- 36 -- are already represented by Republicans.
In the congressional districts and states where the 2014 elections will actually be decided, likely voters said they would prefer to vote for a Republican over a Democrat by 7 points, 41 percent to 34 percent. A quarter of voters said they were unsure of their preference....
So in a sample of districts that's 6% more Republican than Democratic -- 53% Republican, 47% Democrat -- Republicans have, um, a 7-point lead. Forgive me if I'm not horrified.
The Senate battleground states are a lot more problematic for Democrats -- in 14 of the 16 races, the current senator is a Democrat.
But it's long been known that this would be a challenging year in the Senate, because Democrats are defending seats in red states (Alaska, Arkansas, Louisiana, Montana, South Dakota, West Virginia) and a couple of purple states (North Carolina, Virginia). You'd have to be really naive to think that Democrats won't lose seats. They just have to avoid losing six or more to hold the Senate.
But why ask a generic ballot question about a Senate race anyway? Despite party sentiment, a lot of Democrats are surprisingly competitive in red states -- Mark Begich in Alaska, Mark Pryor in Arkansas, Michelle Nunn in Georgia, Alison Grimes in Kentucky, Mary Landrieu in Louisiana, Kay Hagan in North Carolina. Don't voters already know the candidates in these races? If so, why ask about unnamed Democrat vs. unnamed Republican?
So the poll is bad. But it's not as awful as it appears on the surface.