Another day, and another addition of the poll-o-meter. The day started out with some very encouraging news, that Obama was up by 3 on the ABC/WaPo national poll. The rest of the day was decent, pretty much the extension of the status quo. Overall, a 3 point lead in a highly reputable national poll, despite otherwise mediocre numbers, which put Obama into a near tie in the RCP average, bumps up the day's rating to 7.
The tracking polls today were mixed. Rasmussen showed Romney up 1, a 1 point improvement for Obama from yesterday. Gallup's Likely Voter tally showed Mittens up 49-47, with Obama up 48-46 among registered voters. These were identical numbers from yesterday. The IBD/TIPP tracking poll was essentially unchanged, showing Obama up by a fraction (don't you hate when pollsters use fractions?!?!?). The Reuters-Ipsos online poll showed Obama with a 2-point lead, which is an improvement for him. Similarly, the RAND poll, which measures support among a fixed group of voters, showed Obama with a 5 point lead. Basically, the data indicate very little national movement.
One other national poll, from GWU/Battleground, showed Obama up 1, the same as his lead last week.
The state polls were not particularly newsworthy today. PPP showed Obama up 7 points in Pennsylvania. That's a five point improvement for Romney from their last poll, but indicates that Obama can probably still rely on the Keystone State and it's 20 electoral votes. Two GOP-leaning robofirms also did some polls today. ARG showed Mitt up 1 in Virginia, and showed the candidates tied in Iowa (a state that has been seriously underpolled recently). Gravis Marketing showed Romney up 1 in Florida, pretty much where he's been in most recent polls there.
Overall, we can feel pretty decent about today's polls. Romney's bounce has subsided a bit, perhaps aided by a strong Joe Biden performance in the Vice Presidential debate. Tomorrow may not be so encouraging. This very site, along with PPP, will publish a poll, according to Markos, that will be the "worst news of the week for President Obama."
(Note: I'm ignoring the USA Today/Gallup poll showing Romney with a 5 point lead among likely voters in "swing states." First, Nate Silver and others have convinced me that taking a swing-state subsample isn't very reliable, since each firm has a different definition of swing states, and since we get several actual swing state polls every week! Second, the same poll showed Obama with a 2 point lead in those same swing states among REGISTERED voters. I'm just not buying a 7 point enthusiasm gap. And I won't even try to understand how the poll could possibly show the candidates tied among women.)