National Tracking Poll Thread

[Likely Voter]

Past exit polling shows that only 22%-25% haven't made up their minds by the end of September. This year I bet the number could be as low as 20%. So for the other 80%, watching lat night is like watching sports. You root for your team, but you arent going to become a fan of the opponent just because your team lost.

That being said, I still suspect that within the 20% of persadables, Romney has probably changed some minds and that will probably show up as a bump in the polls. Losing a debate has a ripple effect in social media, among friends and family and via the media. There is a week or so of Obama as a loser and that has to hurt.

[Mynheer Peeperkorn von Thurn und Taxis-Hohenlohe]

What a joke:

The poll found that Obama's seven-point advantage over Romney had narrowed to a five-point lead, 48 percent to 43 percent, but a truly accurate reading on the debate's impact will not be possible for several days, he said.

That is because the poll drew on a smaller pool of voters, taken over a shorter period of time, than the Reuters/Ipsos daily tracking poll that measures the prospects of each candidate.

The online poll surveyed 536 registered voters on Wednesday and Thursday after the debate. It has a credibility interval of 4.8 percentage points.

[Reds4]

Wow, that "poll" is a joke.. I'll wait for some real numbers.

[J. J.]

It's not a joke, but unless they polled everything from 11:00 PM yesterday to 11:30 AM today, it won't get any post poll numbers.

[J. J.]

Also unchanged on Gallup, with Obama 49, Romney 44.

We probably won't the full debate effects, if any, until Monday.

[Bandit3 the Worker]

In other words, it's DebateFail for Romney.

[sobo]

The online poll surveyed 536 registered voters on Wednesday and Thursday after the debate. It has a credibility interval of 4.8 percentage points.

They say that the interviews were conducted after the debate though I'm not sure how they could have done many Wednesday interviews if they were waiting until after the debate to do them. Still, there's no reason to publish this smaller sample poll rather than their tracking poll if they're including old interviews.

[asexual trans victimologist]

Interesting if true.

[True Federalist]

In other words, it's DebateFail for Romney.

Hardly. I expect the tracing polls released today will have little if any post-debate sampling, and none of what little there might be will be from the Eastern or Central Time Zones.

[Skill and Chance]

So the first data point to come in is a 2 point Romney bump with Obama retaining a lead (likely narrower among LV).  That lines up almost exactly with Bush 2004 and Nate Silver's regression.  If it happened across the board, it would also bring the race back in line with the economic models currently showing Obama by <1%.     

[J. J.]

In other words, it's DebateFail for Romney.

In other words, we don't any post debate numbers.

[Reds4]

Just read through the topline results on their page.. they compared pre-debate poll to post-debate... likely voters went from Obama +9 pre-debate to Obama +5 post-debate.. for what it's worth.. only time will tell.

[The Mikado]

only 538 people

This is a perfectly adequate sample size, if the people polled are members of the Electoral College.

[sobo]

Here are the topline results.
http://www.ipsos-na.com/download/pr.aspx?id=12047

[ajb]

Just read through the topline results on their page.. they compared pre-debate poll to post-debate... likely voters went from Obama +9 pre-debate to Obama +5 post-debate.. for what it's worth.. only time will tell.

If those numbers are borne out by other polls (and it's too early to know if they will be) that would be both a relatively large shift for a debate to cause, and also not nearly enough on its own to change the election result.

[Umengus]

the bounce is +4 with a pollster overestimating democrats by great mesure.

[Pacific Councillor LiberalJunkie]

The one bright side of last night is, that if turns out Romney is still losing to Obama, then we can be pretty sure this election is basically over.

Yep. People are overreacting way too much.

the bounce is +4 with a pollster overestimating democrats by great mesure.

Oh stop complaining about the Poll ID.

[Californian Tony]

only 538 people

This is a perfectly adequate sample size, if the people polled are members of the Electoral College.

Well, that would give Obama 258 electors, Romney 231, and 49 electors who wouldn't vote or vote for someone else. Which means the election is thrown to the House where Republicans have a majority.

Bad news for Obama.

[Umengus]

The one bright side of last night is, that if turns out Romney is still losing to Obama, then we can be pretty sure this election is basically over.

Yep. People are overreacting way too much.

the bounce is +4 with a pollster overestimating democrats by great mesure.

Oh stop complaining about the Poll ID.

certainly not...

Facts are stubborn things.

[ajb]

Rand American Life Panel, Oct 2-3-4:

Obama   49.76   49.17   49.87
Romney  44.22   44.83   43.98

[afleitch]

Rand American Life Panel, Oct 2-3-4:

Obama   49.76   49.17   49.87
Romney  44.22   44.83   43.98

Curious.

[State Comptroller Atkins]

Rand American Life Panel, Oct 2-3-4:

Obama   49.76   49.17   49.87
Romney  44.22   44.83   43.98

If those numbers are accurate, then (barring a major Obama gaffe or scandal) Romney is almost certainly done.

[J. J.]

Rasmussen:  49/47 Obama, unchanged.

[Lief]

Obama also leads 50 (-1) to 45 (nc) in the battlegrounds according to Scott.

lol mitt sucks

[J. J.]

Obama also leads 50 (-1) to 45 (nc) in the battlegrounds according to Scott.

lol mitt sucks

I wouldn't, he's released two state polls that are closer.  Obama could be surging in PA or MI, and falling apart every place else.