The first polls after the debate show Obama widening his lead overnight by three points in Gallup, unchanged in Rasmussen. Gallup now gives Obama an eight point lead compared with a six point lead for Rasmussen.
Again, Gallup reacts faster than Rasmussen. This suggests that Obama had a GREAT response to the debates and we'll see that response get baked in even more over the coming days. Undecided voters obviously felt Obama was plausible as a Commander In Chief and felt comfortable making up their minds in his favor. This may turn into an Obama blowout.
(Here's a great analysis by James Fallows of why this debate worked SO WELL for Obama and his overall campaign, and how McCain continues to focus on tactics, even in this debate, losing the bigger picture of the strategy needed to sway undecided voters at this point. I've been thinking the same things about the debate but couldn't have said it better than Fallows does.)
My sense is that Palin has little chance to change this dynamic in her debate on Thursday. If she wins the debate over Biden (which could happen: she beats expectations, gets a zinger, Biden makes a gaffe) it would slow Obama's momentum but not erode his overall lead. The VP's have become increasingly irrelevant since the financial meltdown.
Debate two is a town forum. McCain is much better at these types of things and will likely get some points back on that, but my sense is it will only be a fraction of the three-point bump Obama got from Friday.
Debate three is an economic debate. Judging from McCain's performance Friday and Obama's dominance on this issue, my sense is this would erase any McCain gains from the VP debates and debate 2.
So my prediction: Unless there is another major game-changing event or big October surprise (terrorist attack, Iran nukes Israel, McCain has heart-attack, Michelle Obama comes out as a lesbian, etc.), we're likely looking at an 8 point Obama lead in polls before election day. For those of you who believe there's a Bradley effect masking Obama's real votes (and the only state where I think this might be happening is Pennsylvania), this should cover that margin. At this point, Obama could lose Pennysylvania and still plausibly win Virginia, Nevada, Colorado, and New Hampshire.