Some polls now have Romney ahead.

[Rainsford]

Don't you think the sample should be looked at when trying to decide what the poll is really saying? If they are just recording what they get then maybe their polling methodology introduces some sort of bias unintentionally.

If the national poll that had an 8% advantage for Democrats is correct then Romney is going to lose. Considering that 2008 had a 7 point lead I find that possibility to be absurdly low. Obama will get less turnout this time around and Republicans more for a whole host of reasons that I outlined earlier.

The problem is how do you adjust things unless you have a more accurate baseline to work from? Or to put it another way, adjusting for party identification requires you to know what percentage of the population identifies with which party from some source outside of (and more accurate than) your poll. Looking at previous years can provide somewhat of a gut check, but unless it highlights and obvious flaw in their methodology, it's hard to come up with something that can actually be changed. And to be perfectly honest, asking a lot of randomly chosen people what their party identification is intuitively seems like it would produce fairly accurate results unless people are lying.

And while the polling organizations should consider that their polls might be non-representative for some reason, you should also consider that party identification might not mean what you think it means. The "over-representation" theory says that the respective percent of Democrats and Republicans in the polls should closely track with how many votes each candidate is going to get. So if self-identified Democrats outnumber Republicans in a race expected to be close, there's something wrong with the sample. Except that's ignoring that a huge number of people identify as "independent" in the polls while the vast majority of the vote goes to either the Democrat or Republican candidate. There's nothing saying Romney supporters HAVE to identify as Republicans, and I wouldn't be surprised if the rise of the Tea Party leads many people who wouldn't vote for Obama no matter what to identify as independent.

 

[cybrsage]

buckshot: You keep focusing on splits and "weighting" (of course, only for the polls whose results you don't like). You do understand that they don't decide ahead of time how many Ds, Rs and Is to poll? They decide whom to poll based on demographics and ask people for party ID as part of the poll.

So if a poll has 100 Ds in it, 1 R and 1 I and it shows Obama winning by a landslide, you are saying the number of Ds in it is irrelevant?

By that measure, an online poll where you answer questions on the FoxNews website is just as valid...even though it would be heavily skewed by the vast number of Rs who would take part in the poll...

 

[Farang]

Few state polls came out today

Iowa +1 Romney PPP poll (yesterday NBC has Obama up 8)
Virginia +3 Romney Rasmussen
Florida +1 Romney CNN
Florida +5 Romney Rasmussen
Florida +3 Romney Fox
New Hampshire +1 Romney PPP
Wisconsin +2 Obama Rasmussen
Ohio +3 Obama Fox (+8 Dem sample which was the gap in 2008)

Overall a very good day for Romney. That Iowa poll is especially interesting given that NBC had Obama up 8 just yesterday. The Ohio poll seems to have too many Democrats in it. If it is plus 8, as it was in 2008, then Ohio will simply be out of reach for Romney. I doubt it will be plus 8 however.

Where are you culling this list? Seems a bit selective (i.e. where is PPP's Virginia Obama +3)

You also have a tendency to qualify only polls that you don't like the result of.

Also, short term trends only matter if they're coming from the same place. Given two polls, one yesterday and one today, where Obama is +8 and Romney is +1, I'd say that is better news for Obama. More likely the real result is somewhere in the middle, rather than there being some massive 9 point swing within 24 hours.

So far in what I've seen today--no real winner, but good day for Obama in that any day there is no evidence of a Romney breakthrough in Ohio is going to be a good day for Obama.

 

[buckshot24]

The relevance of exit polling from 2004 escapes me, sorry.

More hand waving.

You're the one making the assertion, but you haven't provided any rational basis for it.

If you actually look at the polls, a good chunk of people oppose Obamacare because they don't think it's liberal enough. Those people are not going to vote for Mitt Romney.

Obamacare is a known quantity and its impact is already "priced into" all the polls.

I don't know what polls you are looking at. But if you look at the internals of a CNN poll from June/July you see that your assertion isn't valid.

Democrats oppose repeal 80-16
Liberals oppose repeal 76-23
Conservatives favor repeal 73-25
Republicans favor repeal 86-13

http://www.realclearpolitics.com/docs/2012/CNN_healthcare_0702.pdf

Clearly this is a partisan/ideological split that couldn't be clearer given this poll.

I think it should be pointed out that each point I make should be taken as a set instead of individually. I'm giving reasons why I think that the splits in a lot of these polls are way off. Not that each point, on there own, is what is going to make the difference. It's easier to make a case that Obamacare unpopularity will hurt the president rather than help him.

 

[buckshot24]

The problem is how do you adjust things unless you have a more accurate baseline to work from? Or to put it another way, adjusting for party identification requires you to know what percentage of the population identifies with which party from some source outside of (and more accurate than) your poll. Looking at previous years can provide somewhat of a gut check, but unless it highlights and obvious flaw in their methodology, it's hard to come up with something that can actually be changed. And to be perfectly honest, asking a lot of randomly chosen people what their party identification is intuitively seems like it would produce fairly accurate results unless people are lying.

But are they totally random? I don't have a replacement model for the pollsters but I'm pretty sure that the split won't be bigger than 2008. 2008 was like a perfect storm for Democrats and they got smoked badly in 2010.

And while the polling organizations should consider that their polls might be non-representative for some reason, you should also consider that party identification might not mean what you think it means. The "over-representation" theory says that the respective percent of Democrats and Republicans in the polls should closely track with how many votes each candidate is going to get.
So if self-identified Democrats outnumber Republicans in a race expected to be close, there's something wrong with the sample. Except that's ignoring that a huge number of people identify as "independent" in the polls while the vast majority of the vote goes to either the Democrat or Republican candidate. There's nothing saying Romney supporters HAVE to identify as Republicans, and I wouldn't be surprised if the rise of the Tea Party leads many people who wouldn't vote for Obama no matter what to identify as independent.

This very well could all be true except this hasn't generally been seen in polls from this cycle. For instance the poll I looked at yesterday had only 15% independents. There were around 30% in 2008 according to the CNN exit poll.

 

[buckshot24]

Where are you culling this list? Seems a bit selective (i.e. where is PPP's Virginia Obama +3)

Those are from yesterday. The PPP was from another day.

http://www.realclearpolitics.com/epolls/latest_polls/president/

You also have a tendency to qualify only polls that you don't like the result of.

Probably guilty as charged.

Also, short term trends only matter if they're coming from the same place. Given two polls, one yesterday and one today, where Obama is +8 and Romney is +1, I'd say that is better news for Obama. More likely the real result is somewhere in the middle, rather than there being some massive 9 point swing within 24 hours.

Oh yeah, I agree that there probably wasn't a swing that large. Somebody is completely off or they are both off a little.

 

[Charles Kozierok]

More hand waving.

You clearly don't even understand what "hand waving" means.

It wasn't "hand waving", it was a simple question. You brought up the exit polls from an election two cycles ago featuring two people who aren't running this time around. It's up to you to make something approaching a rational argument for why it is relevant.

It's easier to make a case that Obamacare unpopularity will hurt the president rather than help him.

Not based on the data you've provided.

What you don't seem to understand is that of the 51% who support repeal, some percentage support repeal because they want something more liberal. They want better coverage, or tighter rules on insurance companies, or they want a single-payer system.

Those people are not going to vote for Romney.

I haven't seen a single poll showing that a majority of Americans want to vote against Obama over the health care law.

 

[buckshot24]

You clearly don't even understand what "hand waving" means.

It wasn't "hand waving", it was a simple question. You brought up the exit polls from an election two cycles ago featuring two people who aren't running this time around. It's up to you to make something approaching a rational argument for why it is relevant.

It wasn't a question so it appeared to be hand waving.

It isn't relevant who were the candidates nor is it relevant who won. What is relevant is that there was no gap between Democrats and Republicans (according to those exit polls) in 2004.

Not based on the data you've provided.

What you don't seem to understand is that of the 51% who support repeal, some percentage support repeal because they want something more liberal. They want better coverage, or tighter rules on insurance companies, or they want a single-payer system.

Which is absolutely irrelevant to my point. I am not assuming Romney will win because of the majority support repeal of Obamacare. I am saying more Republicans will show up to vote this year because of their desire to get rid of Obamacare. This is all about turnout. Republicans stayed home in 2008 because they weren't enthusiastic and Obama was so far ahead in the polls. Obama is not far ahead in the polls this year.

Take your liberals who want to repeal Obamacare because it isn't liberal enough. Are they more likely to show up to the polls or less likely? Seems that if they just want to scrap it that they would be less enthusiastic to vote at all this time. Romney doesn't need to capture these voters be benefited by them.

 

[OneOfTheseDays]

Obama actually is far ahead in the polls that matter.

Romney has no realistic chance to win this election if he loses Ohio. Obama has been consistently up in this state since the election started earlier this year. Romney cannot crack him in Ohio, and there simply isn't enough time left to do so.

 

[buckshot24]

Obama actually is far ahead in the polls that matter.

Romney has no realistic chance to win this election if he loses Ohio. Obama has been consistently up in this state since the election started earlier this year. Romney cannot crack him in Ohio, and there simply isn't enough time left to do so.

Cool story.

 

[Balt]

Obama actually is far ahead in the polls that matter.

Romney has no realistic chance to win this election if he loses Ohio. Obama has been consistently up in this state since the election started earlier this year. Romney cannot crack him in Ohio, and there simply isn't enough time left to do so.

I think this is probably true, but I wouldn't say it's 100%. Romney has made some, but not enough, traction in Ohio following the first debate. Ohio is basically Obama's "firewall" at this point. If he loses Ohio, he's screwed. If he keeps Ohio, he can win through so many different combinations that a Romney victory is very unlikely.

I think that in the end, Romney's connection to Bain (an outsourcer) and his "let Detroit go bankrupt" stance will cost him the election. If I had to guess, I'd say that's what keeping him from making bigger gains in Ohio (a rust belt state).

Of course, I've been wrong before. I thought that choosing Paul "Medicare voucher" Ryan as a running mate would alienate seniors and lead to Obama winning Florida, but that clearly is not the case. It looks like Romney is certainly going to win Florida, despite Clinton doing some heavy campaigning in the state on behalf of Obama.

 

[Charles Kozierok]

It isn't relevant who were the candidates nor is it relevant who won. What is relevant is that there was no gap between Democrats and Republicans (according to those exit polls) in 2004.

That doesn't make it relevant at all to this current race. Enthusiasm and party ID fluctuate over time -- sometimes as little as a few weeks, much less years.

I am saying more Republicans will show up to vote this year because of their desire to get rid of Obamacare.

And as I said, that is all already priced into the polls and the LV models.

 

[First]

That doesn't make it relevant at all to this current race. Enthusiasm and party ID fluctuate over time -- sometimes as little as a few weeks, much less years.



And as I said, that is all already priced into the polls and the LV models.

He doesn't get it; he doesn't know how polling methodology works, he doesn't understand contingent valuation and he knows nothing about statistics. It's pointless.

 

[buckshot24]

That doesn't make it relevant at all to this current race. Enthusiasm and party ID fluctuate over time -- sometimes as little as a few weeks, much less years.

Of course it does. 2004 is merely a point of reference.

And as I said, that is all already priced into the polls and the LV models.

And yet most of the polls have a gap larger than 2008. If we get what we got in 08 then it will be a short night on November 6th.

Do you really think there will be up to 10% more Democrats vote this year than Republicans when in a "perfect storm" year there were only 7%?

I suppose there could have been a fundamental change in party affiliation over the last eight years but considering that the Democrats got smoked in 2010 I doubt it.

 

[Charles Kozierok]

Of course it does. 2004 is merely a point of reference.

Yes, it's a point of reference, but you have to explain how it translates into 2012, or it's meaningless. That is my point.

Do you really think there will be up to 10% more Democrats vote this year than Republicans when in a "perfect storm" year there were only 7%?

I suppose there could have been a fundamental change in party affiliation over the last eight years but considering that the Democrats got smoked in 2010 I doubt it.

Ironically, you've answered your own question. There has been a significant shift of conservatives away from GOP party affiliation specifically because of the "tea party" movement associated with the 2010 election.

 

[buckshot24]

Yes, it's a point of reference, but you have to explain how it translates into 2012, or it's meaningless. That is my point.



Ironically, you've answered your own question. There has been a significant shift of conservatives away from GOP party affiliation specifically because of the "tea party" movement associated with the 2010 election.

Then independent sampling should be higher when that isn't the case. Some polls have half as many, on a percentage basis, independents as the exit polls from 2008. Two of the polls that have Obama up nationally have a 15% share for independents. Politico and Uconn

 

[OneOfTheseDays]

It's all about Ohio, Ohio, Ohio. Romney's stance on the auto bailout has ruined his chances of ever winning this state. He cannot win without Ohio, and he cannot crack Obama's lead here with so little time left.

He couldn't crack it after the 1st debate, which was his best opportunity to make up ground. He didn't win the 2nd debate, and with the final debate being about foreign policy which is a non issue to most people he won't have another moment left in the campaign.

 

[buckshot24]

I think we are close enough to the election that I can wait to see if I am right. I'm done arguing about how the polls aren't accurately sampling voters.

 

[Doppel]

I think we are close enough to the election that I can wait to see if I am right. I'm done arguing about how the polls aren't accurately sampling voters.

Well in less than three weeks Romney will lose and you'll know for sure, like you said. I really hope you are prepared for that eventuality.

 

[buckshot24]

Well in less than three weeks Romney will lose and you'll know for sure, like you said. I really hope you are prepared for that eventuality.

I have been practicing meditation...

 

[Doppel]

I have been practicing meditation...

If it's any consolation I don't think it matters much which of these guys wins anyway ^_^

 

[Charles Kozierok]

Well in less than three weeks Romney will lose and you'll know for sure, like you said. I really hope you are prepared for that eventuality.

I hope Obama supporters are prepared for the possibility of "President Romney".

I hate the guy more than any other politician I've ever encountered (well, other than Sarah Palin). He's a fraud, a used-car salesman, a professional liar with no principles and no honor -- but he very well could win.

Just saw a PPP poll showing Obama's lead down to 1 in Ohio. Obama has lost pretty much his entire lead even in the EC at this point.

 

[buckshot24]

Just saw a PPP poll showing Obama's lead down to 1 in Ohio. Obama has lost pretty much his entire lead even in the EC at this point.

Just for informational purposes. They have D +8 in that poll as well.:whiste:

 

[Charles Kozierok]

Just for informational purposes. They have D +8 in that poll as well.:whiste:

That's not "informational", it just demonstrates your ongoing ignorance on that subject.

 

[First]

That's not "informational", it just demonstrates your ongoing ignorance on that subject.

lol, bump