Yesterday I noted that, with two days of post-debate tracking poll results now in, it was starting to look as if that final presidential debate had killed the momentum Mitt Romney seemed to have going into it. Now, with three days of tracking poll results in, things look, if anything, a bit worse for Romney than they looked yesterday.
Yesterday the key number was 0.6. Eight tracking polls that had reported results for Wednesday -- two days after the debate -- showed an average movement of 0.6 points in Obama's direction compared to the final pre-debate polling numbers they had reported.
Today's numbers, which include polling done three days after the debate, showed an average movement of 1.0 points toward Obama relative to the final pre-debate polls.
Two asterisks, one that works in Romney's favor and one that works in Obama's favor:
First, on Romney's behalf, it should be said that the poll that shows the strongest net gain for Obama -- the RAND poll -- has the most eccentric methodology. It interviews the exact same people every day, week after week. Presumably the knowledge that they'll be interviewed each day makes them behave atypically--by, say, focusing more than the average voter on news about the candidates. If we got rid of this poll (which I originally included because it was one of the eight polls summarized by Nate Silver of The New York Times), Obama's net gain would drop from 1.0 to 0.6.
However, there's also a point that works in Obama's favor. In general, as I noted yesterday, tracking polls are somewhat opaque, because they don't release the results for individual days, but rather just the latest rolling multi-day average. So none of the numbers I reported yesterday reflected only post-debate polls. However, because two of the polls (PPP and Rasmussen) are three-day tracking polls, today, for the first time, we have numbers that reflect only post-debate polls--polls conducted on Tuesday, Wednesday, and Thursday, following the Monday night debate. And these two polls show an average post-debate gain for Obama of 1.5 points -- 2 points in the PPP poll and 1 point in the Rasmussen poll. So the only two polls that give us a precise pre-debate versus post-debate contrast reflect even more favorably on Obama than the eight tracking polls taken as a whole.
OK, here are the results broken down by poll. With two exceptions, which I'll explain shortly, these are from polling done on Thursday and published on Friday, and the number for each poll represents the change relative to polling done on Monday, the last pre-debate polling. Gallup: no change; IBD/TIPP: no change; Ipsos/Reuters: no change; PPP: Obama +2; RAND: Obama +4; Rasmussen: Obama +1; UPI/C-Voter: Obama +1; Washington Post/ABC: no change. [Notes: (1) The Ipsos/Reuters poll number, though reported today, actually includes some polling done today. (2) The UPI/C-Voter poll number is the same number I reported yesterday. I can't find an update for today, and, anyway, I now realize that the UPI number I reported yesterday was unusual in that it included polling done yesterday. So in that sense it is comparable to the other numbers I'm reporting today (except Ipsos/Reuters) in that the most recent polling it includes was done Thursday. All these polls can be found (in regularly updated form) here, except for RAND, which is here.]
This article available online at: