That is a very good point. I believe that Ras weights according to a 3-week moving average of party affiliation. That means that there will be automatic D oversampling when people are swinging toward saying “I am an R”.
According to another thread, 20 minutes into the debate, GOP online voter registrations started zooming upward, and that has not yet abated, according to last I heard.
Very cool, I hadn’t heard of that. If you can find that thread again, can you ping me to it?
According to another thread, Rasmussen was on O’Reilly on Thursday night expecting Romney to lead by at least 2 by Monday. He said that on Thursday night. Curious thing to say if Monday back then was more than 3 days away and you have a 3-day tracking poll.
I think he was seeing a big surge in R responses and predicting when the R weighting would start showing up, rather than going off by that day’s polling data.
I meant Independents, not undecided.