Making Sense of the Pre-Election Chaos

As we head into the final hours of the 2010 elections, there have been conflicting indications as to where we are headed on election night.  Here are some things to consider:

- Although tomorrow night will undoubtedly be a good night for us, the election is still very volatile and there is a wide disparity in terms of possible outcomes.  The election results could be similar to 1994 in which we picked up 8 Senate and 52 House seats.  On the other hand, it could be a super wave that could lead to a Senate majority, as well as up to 100 more seats in the House.  We therefore cannot rest until the polls close.  Call all your friends and family and make sure they vote Republican.

- Part of the reason why there is still so much uncertainty is because of the dichotomy between the macro polling and the individual state polling.  Let's put it bluntly.  Both the 15 point Gallup generic ballot lead and the state Senate polls cannot be true.  If the Republicans really lead by such a margin on the generic ballot polling, then there is no way that they are in such tight races in California, West Virginia, Washington, and Colorado.  If the generic macro polling data that gives the GOP a 30 point lead among independents and a 6 point lead among women is really accurate, then all those races are going in our direction.  In fact, the battle lines would move as far as to place Delaware and Connecticut in play.  Conversely, if the state polling is correct, then there is no way that the macro polling is accurate.

-  Turnout Models: This is connected to the previous point.  Most of the state polling that shows the GOP wave stalling in some Senate and Gov. races predict a turnout model similar to that of 2006.  Even Rasmussen has been doing this.  Their most recent Washington poll which has Rossi up 1, is pegged exactly to the 2006 party ID numbers.  The most recent Survey USA poll which had Boxer up by 8, assumed a more favorable Democrat turnout than in 2006.  This could provide us with a resolution to the contradiction between the state and macro polling data.  If we are to assume that the macro polling numbers (generic ballot, demographics, enthusiasm, etc.) are correct, then we have to add at least 5 points to every Senate race poll to account for the more favorable turnout models.  Nate Silver argues that this is one of the reasons why the wave can be even bigger than expected.

- House Polls: The House wave keeps growing even as we speak.  If the Republicans would just win the R rated districts, they would pick up 70 seats.  Yet, there are dozens more incumbents in deep D territory that are seriously vulnerable, and some are certain to go down.  All of the polling data confirms this.  Just today, there was a poll out of CT-5 in which Republican Sam Caliguiri was up 9 points in a D+2 district!  If this is true, there is no way that we aren't doing better in the Senate races.  Again, there is something flawed in the turnout models of these polls.