• The numbers were amazingly static for quite a long time, until late summer or early fall, and even now are not really all that different than they were almost 14 months ago. You could swap results from a September 2005 poll into a current one and no one would be immediately the wiser -- it wouldn't look out of place.



• If I were in charge of getting Democrats elected at the DCCC, I would be just a wee bit worried that in what is supposedly going to be a "wave" election, the Democrats' numbers aren't really taking off all that much. Yes, they're going up, judging by the trendline, but pretty gently so. I'd think I'd want to see some hurrying up going on, but it doesn't seem to be there.



• On the other hand, what's good is that all the gains coming from the Undecideds seem to be going to the Democrats, which foretells the possibility that they will be breaking our way this election. Combine that with the Republicans' numbers being amazing flat, and that looks like a good combination. (Of course, it would be even better if the Dems were drawing from both the Undecideds and the Republicans, causing the Republicans to go down, but that doesn't seem to be happening.)

That's about all I can come up with until I get a little sleep. I'll add anything that occurs to me in my fever dreams later on today.

Update (5pm): How nice to finally be able to edit posts after being locked out by Blogger problems for that last few hours. I've cleaned up some of the messiest of my morning writing.

Update: Incidentally, if you'd prefer real analysis from professionals who actually understand what they're talking about, instead of the simplistic amateur bullshit you get from me, check out any of these links:

B-E-W study / Mark Blumenthal / Jay Cost / Charles Franklin / Matt Shugart

Or just hang out at Pollster.com for a while and absorb what they have to offer.

Update: More simplistic stuff from me. A look at how the Democrats' lead on the generic ballot has increased: