Eve Of The Storm

Gallup's final generic ballot numbers for tomorrow's election are gruesome for the Dems, with a couple of big assumptions.

The results are from Gallup's Oct. 28-31 survey of 1,539 likely voters. It finds 52% to 55% of likely voters preferring the Republican candidate and 40% to 42% for the Democratic candidate on the national generic ballot -- depending on turnout assumptions. Gallup's analysis of several indicators of voter turnout from the weekend poll suggests turnout will be slightly higher than in recent years, at 45%. This would give the Republicans a 55% to 40% lead on the generic ballot, with 5% undecided.

Republicans' 15-point lead among likely voters contrasts with their 4-point lead, 48% to 44%, among registered voters, highlighting the importance of higher GOP turnout to the election outcome. This wide difference between the GOP's margin among registered voters and its margin among likely voters is similar to the 2002 midterms, in which Democrats led by 5 points among all registered voters in Gallup's final pre-election poll, while Republicans led by 6 points among likely voters -- an 11-point gain.

Gallup's traditionally been very accurate with these numbers, enough so that Nate Silver's famous chart from April makes its appearance here:

As you can see, a 15-point GOP lead equals 90 seats for the Republicans, a complete and total bloodbath.  I don't know if the Republicans will do this well, we're right on the edge of Nate's model and all and it's a linear one, out into "Here be dragons" territory on the map.  But the Republicans have to be feeling really, really good.

Gallup is basically confident in saying that Republicans will have a presidential election year level of turnout, while the Democrats will have the worst midterm turnout in modern history, a combination that will result in an electoral abattoir.

We'll see what happens in about 36 hours.

[UPDATE]  Nate himself is greatly hedging his bets because there are so many House seats in play, it makes a model difficult at best.

Our model is a lot more sophisticated than that. It does look at the generic ballot, but it doesn’t necessarily assume that it is right. It also looks at local polls in each congressional district, expert ratings, fundraising data — the whole kit and caboodle. Unlike the political science models, it formulates an estimate of the result in each individual congressional district, and not just the overall seat count.

But it tells you basically the same thing. Tonight, our forecast shows Republicans gaining 53 seats — the same as in recent days, and exactly the same answer you get if you plug the generic ballot average into the simple formula. Our model also thinks the spread of potential outcomes is exceptionally wide: its 95 percent confidence interval runs from a 23-seat Republican gain to an 81-seat one.

Now, this is actually something of a coincidence; our model doesn’t think the confidence interval is wide because there is disagreement in the generic ballot polls, but rather for other reasons.

So it could be anything in there, and Nate's 52 is right dead in the middle of that range.  Individual races still count, and that means voter turnout counts too.