RCP is okay. I am SO WORRIED about the Nate Silver pulling a trick - he weighs the Obama higher, this helps the fund-raising, this in turn gets reporters to use his work, etc. It is a massive fraud at NYT.
Romney cannot himself battle three ways: WH, Press, and International. The GOP must support him. But, how? There is a MO senator and an IN senator (both candidates) who have backward views. SO, poor Romney.
I hope he wins. I voted for him. BUT, I AM SCARED. My instincts are always right. Obama may win due to five reasons:
1. NYT (total support, nothing but super-positive) 2. Benghazi (lack of focus by NYT, NPR, and PBS) 3. Sandy (which gave WH Photo-OPs). 4. GOP backward senate candidates (MO and IN) 5. GOP ugly primary with Newt and Rick damaging Romney.
I admit, I'm baffled by the polls. The MU law school poll shows a big swing in Obama's favor, and yet the Democrat mayor of Denver, campaigning for Obama in Milwaukee, said this:
"Hancock even broke news on that Wisconsin trip, telling voters if the election were to be held right now the president would lose Wisconsin and its coveted 10 electoral votes.
“We have not turned out the vote early,” Hancock told the newspaper. “The suburbs and rural parts of Wisconsin – the Republican base – are voting. President Obama’s base has yet to go vote. We’ve got to get our people to go vote.”"
Can somebody tell me why the mayor of Denver is stumping in Wisconsin? Wouldn't it be a bit smarter for him to campaign for Obama in, oh I don't know - Pueblo? Or Colorado Springs? Would it make sense for Tom Barrett to campaign in Denver?
But for every poll I see that looks good for Romney, I can find an opposite. Obama is pulling ahead in Wisconsin at the same time he's losing ground in Minnesota and Pennsylvania? The early voting estimates I've seen in Ohio are all over the place.
Really, I give up trying to make sense of it all. The only thing I know for sure is that some polling organizations are going to end up as discredited next Tuesday as Zogby was.
This is just cruel. We know that Dems handle losing really bad; we've seen it over and over, so why do this to them. It's gonna be like waking up with a transvestite in your bed, assuming that was not your intention.
The only significant Obama lead (+5 in the National Journal poll) uses a voter model where Democrats actually *increase* their turnout advantage by one point from 2008.
This is just cruel. We know that Dems handle losing really bad; we've seen it over and over, so why do this to them. It's gonna be like waking up with a transvestite in your bed, assuming that was not your intention.
10/31/12 7:24 PM
Hey, now. The Vice President has told us that bigotry against the transgendered is THE great civil rights struggle of our time.
The real tell here isn't that the RCP average has the race tied -- they've generally been within one point since the first debate -- it's that Althouse is citing the RCP average. Why not cherry-pick Rasmussen or Gallup that still have Romney winning?
garage mahal said... The Gallup poll was giving Romney some real buoyancy and evidence of 'mo. Then came Sandy at the worst time for Romney.
Who would you prefer to handle the cleanup of this mess? A lot of people are thinking about that now. Obama has never handled a disaster, or anything else.
Who would you prefer to handle the cleanup of this mess? A lot of people are thinking about that now. Obama has never handled a disaster, or anything else.
I think it will help him, if anything.
A WaPo poll had a 78% positive rating on how Obama handled Sandy, to only 4% negatively. Not to mention the lovefest from Christie.
Some of us know, Inga--the democratic mayor and democratic governor failed so miserably that by the time FEMA showed up, despite being third responders, they had to scramble to do first and second responder work.
garage mahal said...The Gallup poll was giving Romney some real buoyancy and evidence of 'mo. Then came Sandy at the worst time for Romney.
Will Sandy cause anyone to vote for Obama that wouldn't have anyway? I doubt it. Especially when clean up starts and people start finding out what Obama isn't doing and Romney is.
Martin Bashir and his insane freak out come to mind. Reasonable people see his kind of partisan foaming at the mouth and hate it. Obama's media is starting to hurt him.
Oh good grief Michael, we all know what happened in Katrina."
Tell me.
I know. Bush kept all those people from getting on the buses that Mayor Nagin and Governor Blanco had arranged.
I flew into NO the year before for the American College of Surgeons meeting. Until the last day, we didn't know if it would be cancelled. I arrived the day after hurricane Ivan missed the city by 100 miles and blasted Pensacola.
I saw NO EVIDENCE of any precautions. Windows were not boarded up. There had been no evacuation. The hurricane swerved away a day before.
The money that had been voted by Congress (Republicans) for levee improvement had been spent on casino parking lots.
I had friends in Mississippi which was hit as badly as NO. They got no publicity and pretty much cleaned up the state with little help. People who drove surplus army vehicles to NO to help out left after they were shot at.
The TV cameras were looking for bodies and blasting Bush.
We'll see how Obama does but we have to use alternate media to find out.
The RCP average is just that, an average of all the polls, of whatever quality, mishmashed together. Some of them may be right, many of them may be wrong, but they're all mixed into the stew.
We'll find out what's slouching toward Babylon to be born next Tuesday.
"Some of us know, Inga--the democratic mayor and democratic governor failed so miserably that by the time FEMA showed up, despite being third responders, they had to scramble to do first and second responder work. "
This ^
Remember the parking lot full of flooded buses that could have made a big dent in the evacuation..the evacuation that never happened?
Remember the kid that hot wired one before the flooding and drove himself and a bunch of people to safety?
As horrible as the current disaster is, unlike N.O., the effected area is above ground..the water is removing itself for the most part.
A lot easier to strut around for photo ops when you don't need a divers suit.
So what exactly did Obama do with regard to Sandy that we're all supposed to be so impressed with?
He "monitored" the storm from Washington, he spoke to Christie on the phone at midnight, and he "expedited the designation of NJ as a major disaster area."
I lived in Louisiana for many years. Most of my family and my wife's family lived in Louisiana. All of them took in refugees during Katrina.
They blamed Mayor Nagin and Governor Blanco. You might notice neither of them are still in office.
The media, as it always does, looked for a Republican to blame for the mess and left Bush holding the bag. The same media that is ignoring the worst foreign policy scandal of my lifetime.
I am glad to see the dead tree legacy media dying. They deserve to die. They squandered their responsibility to be fair-minded journalists and instead became partisan agents of the Democrat party. And they will pay for it with the loss of their entire industry in the next few years.
I won't shed a tear. The new media will be better, faster, more accurate and more ethical. Because they will have to be.
President Ladyparts is their last hurrah, and they know it. Boy do they know it.
The reality is there is very little movement left so the range of these polls reflect inaccurate/imprecise polling. So there is just as much reason to gain insight from other sources.
In this case enthusiasm and turnout will be decisive.
The poling is actually all fairly consistent once you normaliIze it for partisan weighting.
And Romney is leading in every poll when reweighted to a partisan participation rate similar to any recent cycle where the GOP had an enthusiasm advantage over the Dems - basically GOP 36%, Dem 36% and Indis 28%.
The undecided in these polls massively dispprove of Obama and will almost certainly break to an already leading Romney or stay home.
The campaigns are moving into MI, MN and PA for a reason. Romney is moving from a 2004 level win to a 1988 level win.
I'll say it again, 'cause I just love repeating myself. Nationally, the indep voters are breaking by about 8-10 points for Romney, and that's been steady for some time now. Historically, that translates to a 3-4 lead for a Republican candidate.
Romney 52-48 next week. Not a landslide, but a solid win.
You'll notice that despite overwhelming evidence that has been building for weeks, that not one of our lefties has developed enough character and seriousness to ask that the guy they intend to vote for tell them what they need to know to be sure they don't vote for a man who willfully abandoned four brave selfless Americans heros, and then lied about it all for his own selfish political fortune.
It's bad enough to vote for someone with a record of failure, but this level of lazy partisanship will never be lived down. And it won't matter if he somehow squeaks out of it. The fact will remain that you didn't even want to know before you voted. One thing is certain. You didn't even care if the awful thing was true.
Romney is leading in every poll when reweighted to a partisan participation rate . . . [of] GOP 36%, Dem 36% and Indis 28%. .
Dan danoso said...
I'll say it again, 'cause I just love repeating myself. Nationally, the indep voters are breaking by about 8-10 points for Romney, and that's been steady for some time now.
Both of these points are correct, but if you think about it for a moment they cancel each other out. The reason that polls are finding more Ds than Rs is that a significant number of conservatives are identifying themselves as independents. And that is also the reason that independents favor Romney: they are conservative.
Yeah -- the trend in the average is caused by two polls that had Romney up falling off the back end of the average.
Note also that the RCP average is done without weighting by sample size -- the Gallup (R+5) and National Journal (O+5) cancel out to zero.
That's just bad math.
Taking everything *completely* at face value, even ignoring the National Journal's claim that Democrats are *more* excited to vote this time around than they were in 2008, you're still left with:
Gallup: sampling 2700 LVs concludes that the final result is Romney by 5% +/- 2%. That is anywhere from Romney +3 to Romney +7.
National Journal: sampling 716 LVs concludes that the final result is Obama by 5% +/- 4.4%. That is anywhere from Obama +0.6 to Obama +9.4.
Clearly, these two ranges don't overlap. But equally clearly, the Gallup result is a *much* stronger statement -- it's claiming that an Obama victory in the popular vote is a 2000:1 long shot. (2.5x the 95% confidence interval...) In contrast, the National Journal result claims that a Romney victory in the popular vote is a 30:1 long shot.
Obviously, one of the pollsters has made a fundamental error. (Like believing that Democrats will have a +8% party ID advantage at the polls this year, perhaps.) But even if you don't try to judge which is the better crafted poll, just averaging the results makes no mathematical sense.
The Gallup pollsters sampled nearly as many likely voters as the next two largest polls combined. (And, interestingly, those two polls found Romney +2 and Romney +0.)
To just average their result with that of the National Journal's much smaller poll is just silly.
STORY SAID TO INVOLVE POWERFUL SENATOR, SOURCES TELL DRUDGE. DAILY CALLER PLANS LATE NIGHT RELEASE.
---
Ace is claiming to have heard rumors -- it's a Democratic Senator, running for reelection, who was in the Caribbean at a party thrown by a wealthy donor where there were prostitutes. The prostitutes apparently weren't paid and are telling all in hopes of getting paid for their stories.
Credibility level of foreign prostitutes looking for a payday? You be the judge.
"To just average their result with that of the National Journal's much smaller poll is just silly."
Yes, taken to the extreme you could have a poll with 1 respondent who likes Obama and call it equal to 550 people for Romney in a 1000n poll. The sample size is most important. They should weigh them by sample size, it would be very easy. You could do it by hand in 2 minutes.
Out of 12,317 respondents across 10 polls, Romney got 114 more than Obama. A statistical tie.
Much more important I think is methodology, especially party weighting. With that you can make any result you want when it's this close. How else do you explain one poll Romney +5, and another Obama +5 in the same week. The Romney +5 has a sample size of almost 4 times the Obama +5, but that should not make that much difference.
One clue is discernable from following the links to the poll sources. The +5 Romney poll is Gallup and when you go there it's pretty neutral and only really concerns polling issues. The +5 Obama poll is from National Journal at who's site you see a pretty one-sided content including very little mention of anything Benghazi. That says strong bias to me.
AF: Both of these points are correct, but if you think about it for a moment they cancel each other out. The reason that polls are finding more Ds than Rs is that a significant number of conservatives are identifying themselves as independents. And that is also the reason that independents favor Romney: they are conservative.
You might be right, but independents tend to swell from the party out of favor, which is the Dems this time around (certainly more so since 2008). As for finding more D than R out there - well, I could find more reformed druids than either party if that was my goal. If your goal is to accurately model the voting public in 2012, that's simply not the case this time around.
Re: The Bob Menendez scandal, the prostitutes claim they were promised $500 but were only paid $100. This is the same sort of thing that got the Secret Service johns in trouble. They stiffed the girls (in more ways than one), but payback is a bitch!
Hey, Bob? Come se dice "cheap bastard" in Espanol?
The interesting thing to me is that Princeton Survey Research Associates conducted two of the polls over roughly the same time frame and came up with Very different results:
I have finally just decided to accept that the polls are at best a very rough approximation of the state of the election.
I think they are directionally useful, meaning that I think the Romney boost from the first debate was real.
Furthermore I think the Obama decline in approval the last two weeks as Benghazi has been leaking around the media firewall is also real.
At this point I think it all boils down to enthusiasm, strategy and ground game.
At this point I couldn't really say who I thought had the edge. But since a month ago I thought we were looking at another McCain style rout, I can at least say that I have some hope going into the last weekend.
That's something. Romney has done much better than I expected.
I am a participant in the Amazon Services LLC Associates Program, an affiliate advertising program designed to provide a means for me to earn fees by linking to Amazon.com and affiliated sites.
Encourage Althouse by making a donation:
Make a 1-time donation or set up a monthly donation of any amount you choose:
७० टिप्पण्या:
It's a tie because National Journal's garbage poll had Obama +5. It's a garbage poll because it assumes a Dem+8 turnout (2008 had a Dem+7 turnout).
Must be the coastal cities theory!
The fact that NPR, of all entities, has Romney up by 5 is significant.
I agree with Dick Morris -- Romney in a landslide.
(trolls responding in 3 ... 2 ...1)
RCP is okay. I am SO WORRIED about the Nate Silver pulling a trick - he weighs the Obama higher, this helps the fund-raising, this in turn gets reporters to use his work, etc. It is a massive fraud at NYT.
Romney cannot himself battle three ways: WH, Press, and International. The GOP must support him. But, how? There is a MO senator and an IN senator (both candidates) who have backward views. SO, poor Romney.
I hope he wins. I voted for him. BUT, I AM SCARED. My instincts are always right. Obama may win due to five reasons:
1. NYT (total support, nothing but super-positive)
2. Benghazi (lack of focus by NYT, NPR, and PBS)
3. Sandy (which gave WH Photo-OPs).
4. GOP backward senate candidates (MO and IN)
5. GOP ugly primary with Newt and Rick damaging Romney.
What do you think?
Hide yo kids, hide yo wife.
I admit, I'm baffled by the polls. The MU law school poll shows a big swing in Obama's favor, and yet the Democrat mayor of Denver, campaigning for Obama in Milwaukee, said this:
"Hancock even broke news on that Wisconsin trip, telling voters if the election were to be held right now the president would lose Wisconsin and its coveted 10 electoral votes.
“We have not turned out the vote early,” Hancock told the newspaper. “The suburbs and rural parts of Wisconsin – the Republican base – are voting. President Obama’s base has yet to go vote. We’ve got to get our people to go vote.”"
http://blogs.denverpost.com/thespot/2012/10/31/denver-mayor-michael-hancock-stumps-obama-wisconsin-early-voting-favor-president/85190/
Can somebody tell me why the mayor of Denver is stumping in Wisconsin? Wouldn't it be a bit smarter for him to campaign for Obama in, oh I don't know - Pueblo? Or Colorado Springs? Would it make sense for Tom Barrett to campaign in Denver?
But for every poll I see that looks good for Romney, I can find an opposite. Obama is pulling ahead in Wisconsin at the same time he's losing ground in Minnesota and Pennsylvania? The early voting estimates I've seen in Ohio are all over the place.
Really, I give up trying to make sense of it all. The only thing I know for sure is that some polling organizations are going to end up as discredited next Tuesday as Zogby was.
This is just cruel. We know that Dems handle losing really bad; we've seen it over and over, so why do this to them. It's gonna be like waking up with a transvestite in your bed, assuming that was not your intention.
The only significant Obama lead (+5 in the National Journal poll) uses a voter model where Democrats actually *increase* their turnout advantage by one point from 2008.
I see I'm not the only one to notice this...
bagoh20 said...
This is just cruel. We know that Dems handle losing really bad; we've seen it over and over, so why do this to them. It's gonna be like waking up with a transvestite in your bed, assuming that was not your intention.
10/31/12 7:24 PM
Hey, now. The Vice President has told us that bigotry against the transgendered is THE great civil rights struggle of our time.
The Gallup poll was giving Romney some real buoyancy and evidence of 'mo. Then came Sandy at the worst time for Romney.
"This is just cruel. We know that Dems handle losing really bad; we've seen it over and over, so why do this to them."
Well ya...
http://www.sfgate.com/politics/joegarofoli/articleComments/Dems-nervous-GOP-upbeat-as-vote-nears-3989238.php?gta=commentlistpos
I have a hard time mustering up any pity though.
:o)
Relax. They're gaming RealClearPolitics to try to discourage us.
Somewhere, shilol is screaming "First Ned Silver, now this! Say it ain't so! Say it ain't so!".
Wonder what this is about (from Drudge Report - no link yet)?
SEX SCANDAL TO HIT CAMPAIGN... DEVELOPING...
Gloria Allred back in action?
Or Corsi finally found somebody who'd speak up.
If it's the latter, the Gospel According to Saint Bette of Davis shall be broadcast across the land.
The real tell here isn't that the RCP average has the race tied -- they've generally been within one point since the first debate -- it's that Althouse is citing the RCP average. Why not cherry-pick Rasmussen or Gallup that still have Romney winning?
garage mahal said...
The Gallup poll was giving Romney some real buoyancy and evidence of 'mo. Then came Sandy at the worst time for Romney.
Who would you prefer to handle the cleanup of this mess? A lot of people are thinking about that now. Obama has never handled a disaster, or anything else.
I think it will help him, if anything.
Who would you prefer to handle the cleanup of this mess? A lot of people are thinking about that now. Obama has never handled a disaster, or anything else.
I think it will help him, if anything.
A WaPo poll had a 78% positive rating on how Obama handled Sandy, to only 4% negatively. Not to mention the lovefest from Christie.
Chris Christie seems to think Obama is doing a outstanding job. No Katrina repeat.
Sandy will definitely help President Ladyparts. The question is, will it help him enough.
That remains to be seen.
I told you all yesterday that Sandy will get POTUS the love from NYT.
It did. I told you, I am the best politico.
" No Katrina repeat."
Do you actually know what happened at Katrina or are you just repeating what you saw on TV?
Oh good grief Michael, we all know what happened in Katrina.
"A WaPo poll had a 78% positive rating on how Obama handled Sandy, to only 4% negatively. Not to mention the lovefest from Christie."
No argument but nothing has been "handled "yet except photo ops and we know who is the expert on them.
I still think people are thinking about it.
Inga said...
Oh good grief Michael, we all know what happened in Katrina.
You mean how the media lied their asses off?
Some of us know, Inga--the democratic mayor and democratic governor failed so miserably that by the time FEMA showed up, despite being third responders, they had to scramble to do first and second responder work.
garage mahal said...The Gallup poll was giving Romney some real buoyancy and evidence of 'mo. Then came Sandy at the worst time for Romney.
Will Sandy cause anyone to vote for Obama that wouldn't have anyway? I doubt it. Especially when clean up starts and people start finding out what Obama isn't doing and Romney is.
Martin Bashir and his insane freak out come to mind. Reasonable people see his kind of partisan foaming at the mouth and hate it. Obama's media is starting to hurt him.
"Blogger Inga said...
Oh good grief Michael, we all know what happened in Katrina."
Tell me.
I know. Bush kept all those people from getting on the buses that Mayor Nagin and Governor Blanco had arranged.
I flew into NO the year before for the American College of Surgeons meeting. Until the last day, we didn't know if it would be cancelled. I arrived the day after hurricane Ivan missed the city by 100 miles and blasted Pensacola.
I saw NO EVIDENCE of any precautions. Windows were not boarded up. There had been no evacuation. The hurricane swerved away a day before.
The money that had been voted by Congress (Republicans) for levee improvement had been spent on casino parking lots.
I had friends in Mississippi which was hit as badly as NO. They got no publicity and pretty much cleaned up the state with little help. People who drove surplus army vehicles to NO to help out left after they were shot at.
The TV cameras were looking for bodies and blasting Bush.
We'll see how Obama does but we have to use alternate media to find out.
The RCP average is just that, an average of all the polls, of whatever quality, mishmashed together. Some of them may be right, many of them may be wrong, but they're all mixed into the stew.
We'll find out what's slouching toward Babylon to be born next Tuesday.
So what exactly did Obama do with regard to Sandy that we're all supposed to be so impressed with?
I've been without power for two days, and I distinctly recall Garage claiming that there was no evidence for Romney momentum.
Now that he thinks the momentum has been squashed, he's happy to crow about Romney's momentum being squashed.
Tool.
"Some of us know, Inga--the democratic mayor and democratic governor failed so miserably that by the time FEMA showed up, despite being third responders, they had to scramble to do first and second responder work. "
This ^
Remember the parking lot full of flooded buses that could have made a big dent in the evacuation..the evacuation that never happened?
Remember the kid that hot wired one before the flooding and drove himself
and a bunch of people to safety?
As horrible as the current disaster is, unlike N.O., the effected area is above ground..the water is removing itself for the most part.
A lot easier to strut around for photo ops when you don't need a divers suit.
Yes, Bush was so bad in Katrina that Louisiana is now one of the most solidly Republican states in the nation.
So what exactly did Obama do with regard to Sandy that we're all supposed to be so impressed with?
He "monitored" the storm from Washington, he spoke to Christie on the phone at midnight, and he "expedited the designation of NJ as a major disaster area."
Wow!!!!!!!
I lived in Louisiana for many years. Most of my family and my wife's family lived in Louisiana. All of them took in refugees during Katrina.
They blamed Mayor Nagin and Governor Blanco. You might notice neither of them are still in office.
The media, as it always does, looked for a Republican to blame for the mess and left Bush holding the bag. The same media that is ignoring the worst foreign policy scandal of my lifetime.
I am glad to see the dead tree legacy media dying. They deserve to die. They squandered their responsibility to be fair-minded journalists and instead became partisan agents of the Democrat party. And they will pay for it with the loss of their entire industry in the next few years.
I won't shed a tear. The new media will be better, faster, more accurate and more ethical. Because they will have to be.
President Ladyparts is their last hurrah, and they know it. Boy do they know it.
AF: "Why not cherry-pick Rasmussen or Gallup that still have Romney winning?"
LOL
Yeah, Rasmussen and Gallup polls are real "johnny come lately's" without much history or reputation!
What we need are more PPP polls with D+8 (over Repub) and 45% of all respondents self-identified as D's to really get an accurate picture of the race!
LOL
Sandy will be irrelevant once the jobs report (including CA) comes out.
I'll concede that Sandy may marginally help "The Bringer of Light by,
1)Being plastered all over the LSM in a positive light.
2)Helping to push Bengazi below the surface.
The percentage of the I.JUST.CAN"T.FIGURE.OUT.HOW.TO.VOTE crowd that has a propensity to eat their car keys could be swayed...nothing else.
"SEX SCANDAL TO HIT CAMPAIGN... DEVELOPING..."
Mitt/Ann sex tape found on Obama's laptop in a folder named "In case of Emergency"
The reality is there is very little movement left so the range of these polls reflect inaccurate/imprecise polling. So there is just as much reason to gain insight from other sources.
In this case enthusiasm and turnout will be decisive.
Folks,
The poling is actually all fairly consistent once you normaliIze it for partisan weighting.
And Romney is leading in every poll when reweighted to a partisan participation rate similar to any recent cycle where the GOP had an enthusiasm advantage over the Dems - basically GOP 36%, Dem 36% and Indis 28%.
The undecided in these polls massively dispprove of Obama and will almost certainly break to an already leading Romney or stay home.
The campaigns are moving into MI, MN and PA for a reason. Romney is moving from a 2004 level win to a 1988 level win.
I'll say it again, 'cause I just love repeating myself. Nationally, the indep voters are breaking by about 8-10 points for Romney, and that's been steady for some time now. Historically, that translates to a 3-4 lead for a Republican candidate.
Romney 52-48 next week. Not a landslide, but a solid win.
Look out!
But for what?
Look at the graph.
If the trend is your friend Obama has the edge.
You'll notice that despite overwhelming evidence that has been building for weeks, that not one of our lefties has developed enough character and seriousness to ask that the guy they intend to vote for tell them what they need to know to be sure they don't vote for a man who willfully abandoned four brave selfless Americans heros, and then lied about it all for his own selfish political fortune.
It's bad enough to vote for someone with a record of failure, but this level of lazy partisanship will never be lived down. And it won't matter if he somehow squeaks out of it. The fact will remain that you didn't even want to know before you voted. One thing is certain. You didn't even care if the awful thing was true.
A little more off Drudge:
SEX SCANDAL TO HIT CAMPAIGN...
STORY SAID TO INVOLVE POWERFUL SENATOR, SOURCES TELL DRUDGE. DAILY CALLER PLANS LATE NIGHT RELEASE.
Harry?
Chuckie?
Ma'am? Naked?
Bart DePalma said:
Romney is leading in every poll when reweighted to a partisan participation rate . . . [of] GOP 36%, Dem 36% and Indis 28%.
.
Dan danoso said...
I'll say it again, 'cause I just love repeating myself. Nationally, the indep voters are breaking by about 8-10 points for Romney, and that's been steady for some time now.
Both of these points are correct, but if you think about it for a moment they cancel each other out. The reason that polls are finding more Ds than Rs is that a significant number of conservatives are identifying themselves as independents. And that is also the reason that independents favor Romney: they are conservative.
David said...
Look at the graph.
If the trend is your friend Obama has the edge.
10/31/12 9:21 PM
Heh.
Yeah -- the trend in the average is caused by two polls that had Romney up falling off the back end of the average.
Note also that the RCP average is done without weighting by sample size -- the Gallup (R+5) and National Journal (O+5) cancel out to zero.
That's just bad math.
Taking everything *completely* at face value, even ignoring the National Journal's claim that Democrats are *more* excited to vote this time around than they were in 2008, you're still left with:
Gallup: sampling 2700 LVs concludes that the final result is Romney by 5% +/- 2%. That is anywhere from Romney +3 to Romney +7.
National Journal: sampling 716 LVs concludes that the final result is Obama by 5% +/- 4.4%. That is anywhere from Obama +0.6 to Obama +9.4.
Clearly, these two ranges don't overlap. But equally clearly, the Gallup result is a *much* stronger statement -- it's claiming that an Obama victory in the popular vote is a 2000:1 long shot. (2.5x the 95% confidence interval...) In contrast, the National Journal result claims that a Romney victory in the popular vote is a 30:1 long shot.
Obviously, one of the pollsters has made a fundamental error. (Like believing that Democrats will have a +8% party ID advantage at the polls this year, perhaps.) But even if you don't try to judge which is the better crafted poll, just averaging the results makes no mathematical sense.
The Gallup pollsters sampled nearly as many likely voters as the next two largest polls combined. (And, interestingly, those two polls found Romney +2 and Romney +0.)
To just average their result with that of the National Journal's much smaller poll is just silly.
/nerdrant
edutcher said...
A little more off Drudge:
SEX SCANDAL TO HIT CAMPAIGN...
STORY SAID TO INVOLVE POWERFUL SENATOR, SOURCES TELL DRUDGE. DAILY CALLER PLANS LATE NIGHT RELEASE.
---
Ace is claiming to have heard rumors -- it's a Democratic Senator, running for reelection, who was in the Caribbean at a party thrown by a wealthy donor where there were prostitutes. The prostitutes apparently weren't paid and are telling all in hopes of getting paid for their stories.
Credibility level of foreign prostitutes looking for a payday? You be the judge.
Perhaps there are pictures?
Oh please, please let McCaskill be one of the prostitutes.
"To just average their result with that of the National Journal's much smaller poll is just silly."
Yes, taken to the extreme you could have a poll with 1 respondent who likes Obama and call it equal to 550 people for Romney in a 1000n poll. The sample size is most important. They should weigh them by sample size, it would be very easy. You could do it by hand in 2 minutes.
Stacy McCain is saying it is Bob Menendez of NJ. Maybe another Dem seat is in play?
Rumor is also the hookers are underage.
Ted Kennedy's legacy lives!
OK, Here it is sample size weighted.
I think I did it right:
So same data, sample weighted, it's Romney ahead by 0.9%
Sample size Rom% Oba% Romney Obama
1128 0.46 0.46 519 519
1500 0.49 0.47 735 705
1288 0.49 0.49 631 631
563 0.47 0.48 265 270
713 0.45 0.5 321 357
1495 0.47 0.47 703 703
2700 0.51 0.46 1377 1242
1000 0.48 0.47 480 470
930 0.44 0.45 409 419
1000 0.48 0.49 480 490
12317 5919 5805
+114
+0.9%
So to clarify that gobbeldygook:
Out of 12,317 respondents across 10 polls, Romney got 114 more than Obama. A statistical tie.
Much more important I think is methodology, especially party weighting. With that you can make any result you want when it's this close. How else do you explain one poll Romney +5, and another Obama +5 in the same week. The Romney +5 has a sample size of almost 4 times the Obama +5, but that should not make that much difference.
One clue is discernable from following the links to the poll sources. The +5 Romney poll is Gallup and when you go there it's pretty neutral and only really concerns polling issues. The +5 Obama poll is from National Journal at who's site you see a pretty one-sided content including very little mention of anything Benghazi. That says strong bias to me.
AF: Both of these points are correct, but if you think about it for a moment they cancel each other out. The reason that polls are finding more Ds than Rs is that a significant number of conservatives are identifying themselves as independents. And that is also the reason that independents favor Romney: they are conservative.
You might be right, but independents tend to swell from the party out of favor, which is the Dems this time around (certainly more so since 2008). As for finding more D than R out there - well, I could find more reformed druids than either party if that was my goal. If your goal is to accurately model the voting public in 2012, that's simply not the case this time around.
"I agree with Dick Morris -- Romney in a landslide"
Me too.
"Credibility level of foreign prostitutes looking for a payday?"
-- People said the same thing regarding the Secret Service scandal.
Re: The Bob Menendez scandal, the prostitutes claim they were promised $500 but were only paid $100. This is the same sort of thing that got the Secret Service johns in trouble. They stiffed the girls (in more ways than one), but payback is a bitch!
Hey, Bob? Come se dice "cheap bastard" in Espanol?
Credibility level of foreign prostitutes looking for a payday? You be the judge.
Credibility level of Democrat running for re-election?
I'll take the prostitute's word.
The interesting thing to me is that Princeton Survey Research Associates conducted two of the polls over roughly the same time frame and came up with Very different results:
National Journal, 10/25 - 10/28, 713 LV, Obama +5
Pew Research, 10/24 - 10/28, 1495 LV, Tie
http://legalinsurrection.com/2012/10/the-death-of-polling-industry-credibility/
Its also interesting that the polls with the larger samples tend to show a Romney lead
http://www.volokh.com/2012/10/31/two-takes-on-the-state-national-poll-divergence/
http://pjmedia.com/instapundit/155416/
I have finally just decided to accept that the polls are at best a very rough approximation of the state of the election.
I think they are directionally useful, meaning that I think the Romney boost from the first debate was real.
Furthermore I think the Obama decline in approval the last two weeks as Benghazi has been leaking around the media firewall is also real.
At this point I think it all boils down to enthusiasm, strategy and ground game.
At this point I couldn't really say who I thought had the edge. But since a month ago I thought we were looking at another McCain style rout, I can at least say that I have some hope going into the last weekend.
That's something. Romney has done much better than I expected.
Broken record time:
Electoral college seats count.
Popular vote percentage doesn't.
270 seats are needed to win.
538 blog = Obama just over the 300 mark.
Electoral-vote.com = 299.
टिप्पणी पोस्ट करा