With all these polls coming out - I thought it'd be useful to have one diary where we could update with new polls.
Political Wire has broken three big polls this morning:
The Republican Strategic Vision has Obama up big in Florida, and in front in Ohio as well.
Florida: Obama 52%, McCain 44%
Ohio: Obama 48%, McCain 46%
A new Civitas poll in North Carolina, meanwhile, has Obama up by 5 with leaners, and by seven among those who are committed.
According to the live telephone survey of 600 likely General Election voters, Obama leads McCain 46-39 among those who initially voiced support for a candidate. However, when undecided voters are asked which way they lean, Obama’s lead shrinks to 48-43. Libertarian candidate Bob Barr receives two percent of the vote. Seven percent of voters remained undecided.
Money quote for this poll:
"One of the theories we’ve advanced is that for Obama to win North Carolina he needed to obtain 37% of the white voters. In this poll he is now receiving that amount," added De Luca. "McCain has a lot of ground to make up and not much time to do so."
Let me know what else you're seeing and I'll update.
Update I meant to add this story from Bloomberg on a nearly-always-right forecasting model that predicts Obama will win nationally with 52% of the vote:
"The model has predicted all along that the Democrat will get the majority of the two-party vote, and it's still saying that," [Yale University Professor Ray] Fair, who has been forecasting a Democratic victory since November 2006, said in a telephone interview from New Haven, Connecticut...
The only economic data that Fair's vote equation uses that are still unknown are third-quarter gross domestic product growth and inflation. Assuming a 0.2 percent decline in GDP this quarter, the median of a recent Bloomberg survey of 52 economists, and 3 percent inflation, the model forecasts Republican John McCain will receive 48 percent of the vote.
Accurate? Usually.
Fair's model would have correctly predicted the winner of the two-party vote in all but three elections since 1916, according to a 2002 paper published on his Web site. Before the three most recent elections, he accurately forecast who would get more votes without always getting the margin right.