Thursday, November 6, 2008

Congressional races?

Prior to the election, polls and groups like Congressional Quarterly and the Cook Report estimated that Democrats would pick up between 25-35 seats in the House. Right now, they have picked up 19, with 8 still too close to call. Even if all 8 go to the Democrats (which they won't), that is still at the low end of the spread. Senate races that had favorable polling for Democrats ended up being much closer than expected (see Merkley, Jeff and Stevens, Ted). There seem to be two questions that come out of this. The first is why were the Congressional results different than the polling indicated? The second is why did Democrats not do as well as expected?

Initially, it seems to me that support for Obama did not translate as clearly down ballot. Republicans who were willing to vote for Obama may not have been as willing to also support Democrats in Congress. Perhaps the argument in favor of divided government that began to be introduced by Republican candidates towards the end of the race was effective. I haven't seen much analysis yet of the Congressional numbers, in part because so much is still up in the air, but I am curious what others think.


Larry Becker said...

I think most of the Senate races turned out as expected with the exception of Stevens. Merkley was a little closer than expected but it was not that far off from the polling I had seen (a few points). Franken was a total tossup and Chambliss was probably a slight favorite to win and they'll go to a runoff. Stevens is a whole different story (see Russ's post above). In the House, it does seems there was a slight shift back to the Republicans at the last moment. No idea why but one key indicator of it was the demise of Elwyn Tinklenberg. He had a small lead in the late polls I saw and ended up losing. Seems like there was a shift in most places of a few points at the last minute to GOP House candidates.

Chris Shortell said...

Merkley was at a six point lead in the last polling and the result was razor thin. The polls had Chambliss at 47, but he ended up just a hair under 50. McConnell was polling at 48 and ended up at 53. The shifts weren't huge, but they were real. I think the same thing that happened in the Senate happened in the House. The question is why.

Larry Becker said...

The margins in the final polling is the best measure as there were undecided voters in all these polls though. Merkley was ahead by 3 in the final Rasmussen poll and ahead by 6 or 7 in some others. He won by 3 from what we can tell now. The final polling had Chambliss winning by between 2 and 5 points. He appears to have won by 3. Polls in KY had McConnell winning by 3 to 8 points and he won by 6. I think the polling was right on the mark in the Senate ... again, with the exception of Alaska.

There may have been some slight narrowing at the end but it was not much.