« Palin Praises The "Energy" Of The Crowds At Her Rallies | Home | Webb: Obama Has Roots In Appalachia »

TPM Track Composite: Obama's Lead Holds Steady In First Day Of Post-Debate Trackers

Here's our daily composite of the five major national tracking polls. With one day of post-debate sampling now incorporated into the three-day tracking poll, there hasn't been much of a change so far in Barack Obama's big lead:

Gallup*: Obama 51%, McCain 45%, with a ±2% margin of error, unchanged from yesterday.

Rasmussen: Obama 50%, McCain 46%, with a ±2% margin of error, unchanged from yesterday.

Hotline/Diageo: Obama 50%, McCain 40%, with a ±3.4% margin of error, compared to a 49%-41% Obama lead from yesterday.

Research 2000: Obama 52%, McCain 42%, with a ±3% margin of error, compared to a 52%-41% Obama lead from yesterday.

Zogby: Obama 49%, McCain 44%, with a ±2.9% margin of error, compared to a 49%-43% Obama lead yesterday.

Adding these polls together and weighting them by the square roots of their sample sizes, Obama is ahead 50.4%-43.9%, a lead of 6.5 points, pretty much the same as the 50.3%-43.7% Obama lead from yesterday.

*ed. note: Gallup has begun offering two different sub-samples of likely voters, one using a traditional likely-voter model and the other using a modified likely-voter model for an expected higher turnout. For all our composites and Poll Tracker entries, we will be using that second model under the expectation that newly-registered voters and other factors will contribute to a higher turnout this cycle.


22 Comments

| Leave a comment
user-pic

There should be one day after the debate which happened on Wednesday.

user-pic

I see you removed the paragraph. Well played.

user-pic

I see you removed the paragraph. Well played.

user-pic

Still pretty #@!*&%##!!@?!!* close . . .

user-pic

O come the hell on man - these are the biggest margins I've seen in 3 elections.


WTF?


user-pic

It's Biden's fault; he sent a personal email to me (yeah, right, uh huh)
Anyone who tells you this election is already decided is dead wrong. Let's not forget the 2000 election, when Al Gore was up by double digits in October.

The surest way to lose a race is to slow down with the finish line in sight [...]

I didn't remember Gore being up by double digits but I don't think he made it up. Excuse me but I have to get back to tearing my hair out.

user-pic

Yes, well, Obama's saying that Democrats know how to snatch defeat from the jaws of victory didn't do anything but scare me, either.

And I think that's the point - ;)

user-pic

Yeah, no kidding. Obama is counting on high turnout and the ground operation. If people get complacent and don't show up, we'll lose.

So people damn sure better get out, vote, and volunteer.

user-pic

(quote goes up to ellipsis)

user-pic

So, can we say "Lead Possibly Widening" in today's headline? After all, 0.1% is nearly as statistically significant as 0.3%.

user-pic

my thoughts exactly...

user-pic

Meanwhile, Joe Scarborough is running around telling anyone who will listen to ignore all these fancy numbers and that this is going to be a nail biter. Loser.

http://pufferfish.typepad.com/

user-pic

The numbers will tighten up. It always happens in every Presidential election as the time to vote draws nearer.

The critical poll numbers are 1) those in the battleground states and 2) the ones that come out in November.

user-pic

I understand the CW seems to be that things will tighten, but saying it always gets closer simply isn't true. That didn't happen in '96, '92, '88, '84, or '80. In none of those contests was the result within six points. You're speculating because the last two elections that have been close. Right now there's just as good a chance of this thing tightening as there is McCain getting his doors blown off by an even bigger margin.

user-pic

Good summary, thanks.

What's the deal with "averaged together, weighted by the square root of the sample sizes"? That makes no mathematical sense for lots of reasons. Mainly because there are reasonable apples-to-oranges problems among many of these polls that argues against attempting to aggregate at all, but also ... square root of the sample size? What is the rationale for that?

user-pic

The problem with weighting by raw sample size is that accuracy does not double as sample size doubles. It only increases by a smaller amount.

I got the advice to go with square roots from Charles Franklin over at Pollster, who is also my former professor.

That said, on the scale we're working here it doesn't make too much of a difference. The methods of weighting between one poll that has Obama at 50 and another that has him at 49 don't produce much of a change.

user-pic

Obamanos everywhere should work like this is gonna be a nail-biter.

Don't let up.

I will vote on Monday, yee-haw!

user-pic

Everyone I run into says: Don't forget to vote.

:)

user-pic

As far as Wisconsin goes (well at least as far as I know) Obama leads with 100% of the vote. The current tally for Dane County (again as far as I know) is Obama 2, McCain 0. Everyone in my household voted early (except the cat... some confusion as to his real name and visa status...)

Take _that_ John McCain!

user-pic

Yay on the Missouri +6 poll. That is the best news this week so far.

OT but this is funny. Check out Ron Pauls wiki page really quick.

http://en.wikipedia.org/wiki/Ron_Paul

user-pic

Look above his picture on the right specifically.

user-pic

Dang, they fixed it already if you missed it. I'll upload a picture of what it said at some point.

Leave a comment

Advertise Liberally
Share
Close Social Web Email

"To" Email Address

Your Name

Your Email Address