Gallup: Democratic Edge on Generic Ballot at Perilous 2%

Interesting analysis you may find pretty uplifting. Gallup tries to avoid saying that a 2% Democratic lead means that Republicans are actually ahead and will win, at least in so many words. But they kinda say that, with a lot of caveats and hedging.

Let's go to 1994: When the Republicans trounced the Democrats and swept the party out of power in both Houses of Congress.

In 1994, leading up to a Republican landslide, the generic ballot suggested it could be a very promising year for Republicans, as the party actually enjoyed a slight lead among registered voters in several polls. When the likely voter models were applied in the final months of the campaign, Republicans held solid leads among this group of voters.

On the other hand, when Democrats do well in the elections, their numbers on the generic ballot poll show double-digit leads:

Gallup doesn't offer a prediction at this time, except to say that the Democrats' edge in the House is so large that it is "unlikely" Republicans can take that chamber.

But the numbers at this point are close to being 1994-ish.

The generic ballot question does determine, within a range or so of 10 seats either way, how many seats each party will win. If Republicans actually manage to grab 50% of the national vote in 2010 -- which seems quite possible -- they are not nearly guaranteed of taking the House, but it's fairly likely.

Thanks to AHFF Geoff.

But There are 435 Independent Races! People say this to denigrate the importance of the generic ballot. They're wrong.

Yes, there are 435 independent races. Many seats are not up for grabs, ever. But among the competitive ones, they share something in common: the partisan split is somewhere in the range of... well, not quite even, but not so huge that either party can feel safe.

That means competitive districts tend to be purple. They tend to look a lot like Missouri or, lately, Virginia.

And just as in presidential elections, purple states tend to vote the same way. Not always. Some are more red and some are more blue, so those sorts of states may break from the general trend.

But there is a general trend: If the Democrat gets, say, 53% of the national vote, we can expect all bluish-purple states to go his way easily, and most of the purples, and even one of the reddish-purples too.

House races are the same way. There are different politicians in each and different local issues, but they are determined, to a fair extent, by national issues and national mood. People in the middle think more or less the same way on health care in Kentucky District 5 as in New York District 23.

So these races aren't truly "independent" in the sense that there is no common external force present in all of them, as if they were truly independent flips of a coin. They're not independent in that sense; they'e interrelated. When we say that "a team that gives up 4 turnovers is 95% likely to lose a game," each of those particular games is "independent" in the sense they're different games. But they're strongly related by the fact that in each game, one team gave up four or more turnovers. And that uniting factor allows us to speak intelligently about the likeliest outcome in each of those games. Guaranteed? No. But 90%.

The Democrats have thrown two or three interceptions and seem intent on throwing a couple more.

The generic ballot is fairly predictive. Given a party's getting a vote-share of 55%, for example, it's a pretty damn solid lock that that party will get somewhere between 283 seats and 304 seats, with 294 the most likely outcome.

So when Republicans are polling, essentially, at parity, that's an indicator that 218 seats is not very unlikely. The Democrats have advantages, like incumbency, but incumbency becomes a negative rather than a positive when your party is supporting unpopular measures like PelsoiCare.

posted by Ace at 02:47 PM

| Access Comments