Some polls now have Romney ahead.

[OneOfTheseDays]

Any polling that doesn't take cellphones into account is likely to oversample older adults and married families who traditionally do have land lines.

 

[buckshot24]

Oh, IBD seems to have updated their poll. It's now Obama up 4 instead of 6.

 

[TraumaRN]

Any polling that doesn't take cellphones into account is likely to oversample older adults and married families who traditionally do have land lines.

Exactly. Demographically speaking cell phone only users tend to be younger(typically under 35), and lean more Democrat than Republican, educated and all the other attributes that go with younger votes.

But I am looking at this more from the fact of getting accurate research and not from a particular political viewpoint.

 

[randomrogue]

I'm not sure how they poll but I question the accuracy. You would exclude a certain type of voter at least (technologically speaking).

When they poll do they have demographic information from phone records? Can they tell that a user at 555-5555 is 25 years old, white, and a college graduate or are they having to call and call until they get enough answers from a certain demographic to make the poll sound?

I'm so curious how the final results are going to end up and how it will affect future polling.

 

[cybrsage]

? Cell phone users tend to be more educated? What planet do you live on, it is not Earth!

 

[buckshot24]

I'm not sure how they poll but I question the accuracy. You would exclude a certain type of voter at least (technologically speaking).

When they poll do they have demographic information from phone records? Can they tell that a user at 555-5555 is 25 years old, white, and a college graduate or are they having to call and call until they get enough answers from a certain demographic to make the poll sound?

I'm so curious how the final results are going to end up and how it will affect future polling.

I think they weigh under presented demographic participants in their poll result. So if they only got in touch with 4 black people and the demographics suggest there should be 12 they multiply each black respondent by 3.

 

[randomrogue]

I think they weigh under presented demographic participants in their poll result. So if they only got in touch with 4 black people and the demographics suggest there should be 12 they multiply each black respondent by 3.

And how does that affect the error of the poll? It is statistics so does that kind of polling have huge margins of error?

 

[buckshot24]

And how does that affect the error of the poll? It is statistics so does that kind of polling have huge margins of error?

I think it makes the poll less certain the more you have to do this.

 

[TraumaRN]

Nate Silver in his update today said that polls with cell phones tend on average to lean towards Obama by 2 points compared to polls without them. So I'd say given how close this race is, that is a statistically important amount. And in swing states it could provide the margin of victory.

 

[buckshot24]

That poll that had Obama up 6 the other day is now down to Obama up 2.

Two polls from Ohio yesterday have Obama up 3 at 50-47 (CBS) and a tied race (suffolk).

Rasmussen Romney up 4 in Colorado.

Rasmussen tied in Iowa

At this point in 2004 Bush had a 2.7 point edge in RCP Avg at 48.5 to 45.8.

Romney is up 47.9 to 47.2 today.

 

[OneOfTheseDays]

Ohio, Ohio, Ohio.

Romney is toast. He's not going to make up 3-5 percentage points at this point.

 

[buckshot24]

Ohio, Ohio, Ohio.

Romney is toast. He's not going to make up 3-5 percentage points at this point.

He's not down that much.

If Romney is up 3 in popular vote then Ohio won't matter.

ABC just posted their poll

Obama was up 3 on 10/10 to 10/13 but now Romney is up 1.

 

[Charles Kozierok]

All that matters are the state polls. The national polls, especially with their dubious LV screens, are mostly noise.

I still see no path for Romney to 269, and I don't see what exactly he's going to do over the next two weeks to move the chains in the states. Sure, he could still win. But he's behind.

 

[TraumaRN]

All that matters are the state polls. The national polls, especially with their dubious LV screens, are mostly noise.

I still see no path for Romney to 269, and I don't see what exactly he's going to do over the next two weeks to move the chains in the states. Sure, he could still win. But he's behind.

I figure there are two things left unseen. What if any bounce for Obama from the debate last night and the October jobs numbers coming out. If job creation is like 150-200,000 in October I definitely think you'll see a second Obama term.

 

[Farang]

I figure there are two things left unseen. What if any bounce for Obama from the debate last night and the October jobs numbers coming out. If job creation is like 150-200,000 in October I definitely think you'll see a second Obama term.

I think the jobs numbers are overstated.. I've yet to see evidence they effect polling at all. If I recall, bad jobs numbers preceded Obama's slow rise in September

The question is are we going to continue to see these polls in OH of tie to 3+ point Obama lead? As of now I give him the advantage based on ties in VA and CO, tiny leads in OH and IA. The premise being the Democrats have a better ground game (more experience + more time to have prepared + track record) and also have more votes to mine for (i.e. if there wer 100% turnout, Obama would win easily).

Question mark on the effect of early voting.. think it gives a big advantage to Obama in that he has more time to increase turnout, but how many were his base and how many were undecideds? If they got people in the middle to come out early, that is a big plus

 

[KlokWyze]

Polls schmolls. It comes down to how people vote on election day.

Anyone with half a brain can see through Mitt's complete facade that he gives a shit about anyone making less then 2 mil a year. Come the fuck on....

 

[Blackjack200]

I figure there are two things left unseen. What if any bounce for Obama from the debate last night and the October jobs numbers coming out. If job creation is like 150-200,000 in October I definitely think you'll see a second Obama term.

I think that would be weak. 250k+ with wage growth and unemployment going down to 7.5 or less is what I'm hoping for. Does that number come out on the 5th or before?

 

[Blackjack200]

Polls schmolls. It comes down to how people vote on election day.

Anyone with half a brain can see through Mitt's complete facade that he gives a shit about anyone making less then 2 mil a year. Come the fuck on....

And just like that, you've stumbled on the Obama campaign's biggest problem. :awe:

 

[buckshot24]

All that matters are the state polls. The national polls, especially with their dubious LV screens, are mostly noise.

National polls are made from these things called people which live in these things called states. /obama last night

I still see no path for Romney to 269, and I don't see what exactly he's going to do over the next two weeks to move the chains in the states. Sure, he could still win. But he's behind.

So you see no path but Romney could win? OK

 

[OneOfTheseDays]

Unless the job creation numbers are really bad or really good I don't see that moving the needle much. People more or less know where the economy is at today and you either believe it's going to get better slowly over the next 4 years with Obama or not.

 

[buckshot24]

Question mark on the effect of early voting.. think it gives a big advantage to Obama in that he has more time to increase turnout, but how many were his base and how many were undecideds? If they got people in the middle to come out early, that is a big plus

I think Obama won Ohio by 3 last time and had a huge gap in early voting but Romney has closed that gap a bit this year.

 

[uclaLabrat]

I think that would be weak. 250k+ with wage growth and unemployment going down to 7.5 or less is what I'm hoping for. Does that number come out on the 5th or before?

I think that's setting the bar a bit high...the unemployment rate just went down to 7.8%, it might get to 7.5 with seasonal employment, but I bet it stays the same or hits 7.7% after adjustments.

Also, 250K jobs is a pipe dream, 160 would be better than what we're expecting, 200K if we're wishing upon a star...again, after seasonal employment adjustments.

 

[Farang]

I think Obama won Ohio by 3 last time and had a huge gap in early voting but Romney has closed that gap a bit this year.

Basically the way I see it, higher turnout in general favors Obama, and for that reason any ground operation is going to have more potential on the Democratic side than on the Republican. So if a strong ground operation is given more time to work (early voting), that will favor Obama. They're literally going to have days to go to Democratic neighborhoods and carry warm bodies to the polls.

So as it stands I don't think 0-3+ Obama poll results in Ohio are going to cut it for Romney. If the election were held today I think Obama wins. However, a 0-3 point advantage is far from a firewall.

 

[TraumaRN]

I only brought jobs up because people do pay attention to it and not to sell the voting public short but many people are not quite intelligent and a good/great could get a few peoples sentiments up for Obama. Stranger things have happened.

 

[Charles Kozierok]

National polls are made from these things called people which live in these things called states. /obama last night

Yes, most of whom live in states where the outcome is already settled. Only the swing states matter.

Romney appears to have gotten a 2 point bounce from the first debate. Most of these national polls have him a head because of LV screens. And I think a lot of it is also right-wingers in red states coming around to voting for the stooge, which is meaningless.

So you see no path but Romney could win? OK

By "no path" I mean I do not see any current set of polls where Romney is ahead. That could change -- there's two weeks left -- but right now, Romney is behind, no matter how many national polls are conducted.

Here is the RCP map showing who's leading in each state. Obama 281. For Romney to win, he needs to either flip Ohio, or some combination of NV, IA, WI and NH adding to at least 12. And he's behind in all of them.

As for the jobs report, I think at this point it will only matter if it's dramatically high or low, and even then, probably not much.