Uh, oh. NPR? Looks like cracks are appearing in the cocoon. With 7 days to go it is time for vast turn arounds in polls for those who want to salvage credibility. With old people already threatening to Cock-Punch Romney the next week is going to be interesting.
The most interesting part to me from that Pew Poll hasn't been reported on much. Those who are R or Lean R have a likely to vote of 76%. Meanwhile D or lean D have a likely to vote of 62%. This enthusiasm gap is huge.
So, we have an electorate of: -- +1 to +3 R -- R's are significantly more likely to vote -- Romney is up 15-20 points with independents
Maybe these guys weren't just being crazily optimistic after all.
Here is an awesome Romney poll watching training booklet full misleading and false information for poll watchers. Part of that coveted Republican "ground game".
I am willing to deliver pizza to any-one but this election is OVER.
At Politico, we learn that Obama is ahead. At Nate Silver-land, we learn that Obama is ahead. At Huff-Post, Daily-Beast, TPM, we learn that Obama is ahead. Etc. Etc.
What part of the news are you selecting? The point is that this is not the needle in a haystack.
Every-one who is any-one in the media saying that the election is over, then why is Romney smiling?
EJ Dionne, the distinguished professor at Georgetown and a columnist among others. Judy Woodruff the distinguished reporter on PBS and a board member of Newseum among other things, all have said this many times.
The simple, known, confirmable fact is, no incumbent president polling under 50% has been reelected.
That's not determinative, of course, but this president has been polling under 50% for nineteen months now.
So, unless a critical mass of stupid votes prove utterly indifferent and immune to Obama's manifest failings, especially in Ohio, he will not be reelected.
I am willing to deliver pizza to any-one but this election is OVER.
At Politico, we learn that Obama is ahead. At Nate Silver-land, we learn that Obama is ahead. At Huff-Post, Daily-Beast, TPM, we learn that Obama is ahead. Etc. Etc.
What part of the news are you selecting? The point is that this is not the needle in a haystack.
Every-one who is any-one in the media saying that the election is over, then why is Romney smiling?
EJ Dionne, the distinguished professor at Georgetown and a columnist among others. Judy Woodruff the distinguished reporter on PBS and a board member of Newseum among other things, all have said this many times.
The media has been reporting bullshit polls suggesting that the race is close. Now the cynic in me thinks that the reporting is primarily designed to support the meme that the race is close. The pollsters will face their own armageddon on November 7 wherein they lose their contracts for marketing research.
Fortunately for the pollsters the American public is about as innumerate as is the president. And so the bullshit works.
Obama is going down hard. And the only thing of interest will be to read the "explanations" from the punditry post election.
With old people already threatening to Cock-Punch Romney the next week is going to be interesting.
If Obama wins, do we get to cock-punch some liberals? Do we get to burn some "mother-f***ing" cars with Obama stickers on them? Just curious what Move.On?Michael Moore are advocating here. Inquiring minds want to know.
I think the Mainstream press will see this as an opportunity to keep Benghazi and Obama's slipping poll numbers among Indy's and Women a secret.
Look for a week's worth of breathless NYT headlines about how a worker hired with Stimulus bill funding was able to save an Egyptian sculpture at the Met by placing a bucket under a roof leak.
ali karim bey (aka American Politico)--the most important question is, IMO, did you manage to get in Stephanie Cutters pants--she wont be having a mob soon, so the force may be with you :)
Alan Markus--I am sure that will be their excuse, but their track record is, in fact, a matter of record. Buyers of market research want results to further their bottom line, and many pollsters will be out in the cold (or under water).
Roger: No, but if Obama wins, I may get a 2nd chance.
My gut filling is that Sandy will amazingly help. Most voters have a short-term attention problem. They see a POTUS in situation room, and that is all that matters. The fact that the POTUS came back from FL after just a few hours there was a stroke of genius.
The election is over. Nate is correct. Politico is correct. Go here and you can see yourself.
If Romney wins it win will the inverse of the 2008 election. If he does win, will he have coattails? How far down will independents vote for republicans if the break for Romney? Will disaffected democrats vote for republicans for congress and state and local seats?
As for Obama, the worst thing that could happen to him is to win and have the republicans pick up the senate. Endless hearings over a multitude of scandals.
Obama and Biden may well turn out to be the democrats Nixon and Agnew. Anyway a week from today we shall know the outcome.
When President Barack Obama urged Americans under siege from Hurricane Sandy to stay inside and keep watch on ready.gov for the latest, he left out something pretty important — where to turn if the electricity goes out."
Even Christie said so. The Air-Force-One trip from FL to DC was a stroke of GENIUS. Axelrod has redeemed himself. What a genius. Now, we all know the POTUS Obama is Presidential. Even Christie said so.
Cuban Bob (and by the way my dad was born in Havana)--I do think Romney will have coat tails--we will find out on November 7th. In the interim I plan cooking up a batch of ropa vieja to celebrate November 6th. And I do ropa vieja very well.
The media is doing a wonderful job of keeping Obama in this race. If the average American had any clue what happened in the White House during Benghazi, this race would be over. Which, of course, is why the media is not reporting it.
But word is getting out, slowly but surely.
The bottom line of the current situation is that the national and state polls are in opposition. It is very hard to conceive of how Obama wins the majority of the biggest states by an overwhelming margin, Romney wins the popular vote, but somehow Obama wins enough swing states to win the electoral vote.
The math just doesn't work. There just aren't enough votes available to explain the contrast between Romney +4 nationally, but losing all the swing states.
Obama won in 2008 due to an historic combination of a nation tired of GW Bush and independents orgasmic over making an historic vote.
Neither of those are in play this year. Poll models based on 2008 election results are based on an outlier election, not a "normal" election.
Right now, in Ohio, Republicans are roughly 200,000 votes ahead of where they were in 2008. In the general election independents went for Obama by roughly 16 points. They are breaking for Romney by about that amount this year.
If that holds, there is no way Obama wins Ohio. In fact he is at risk of losing Minnesota and potentially Michigan and Pennsylvania as well.
Here is why the numbers are so low in Mass where they know well his malarkey: Barrett chalked Romney's veto of the Peabody project up to a lack of familiarity with infrastructure in the state.
"This was not unusual for him. He didn’t understand infrastructure improvements. It was just the bottom line. He never visited communities. He never understood the issues. He never sat down with mayors or city managers. He never understood why those things were in the budget," Barrett said. "That money was requested by locals. It was a major league problem.”
Here is an awesome Romney poll watching training booklet full misleading and false information for poll watchers. Part of that coveted Republican "ground game".
"While the Obama campaign would like to wish it is 2008, the reality is that they are now forced to “play defense” in least six states (Pennsylvania, Minnesota, Colorado, New Hampshire, Iowa and Wisconsin) that they once believed were “safe” Obama wins."
Its looking more and more ominous for the O, his Obamaness, the one we had we been waiting for.
Roesch/Voltaire, trying to use rational analysis to explain voting in Massachusetts is a joke. This is the state that voted an alcoholic murdering incompetent to the Senate for decades based on fond memories of a failed President. It is also the state that is about to vote for a faux "native american" who has demonstrably committed academic and legal fraud.
Thanks for the link to another Loony-Left site, garage. Leftist paranoids provide my favorite form of light reading. I've always enjoyed The Nation and Alternet but, Jesus, they get fewer comments in a month than Althouse gets in a day.
I have to admit that as much as I am hoping for a Romney victory for the good of the nation and the good of international geopolitical stability, I am also very much hoping to see people like garage and inga have to eat their smug words.
Roesch, in other states some of those municipal projects are what is called 'graft' and given to crony donors.
Now I'm not saying Romney was in the right if he did indeed ignore requests for local infrastrucutre improvements (which you haven't proven). But, if he was going to break the pork cycle, then he needed to be much more discriminant about approving these projects.
If you were familiar at all with the economics of the Big Dig in Boston, then you'd understand how it works in Massachusetts. It works well for those ensconced in power and those connected to them, but not so much for the larger electorate.
"This was not unusual for him. He didn’t understand infrastructure improvements. It was just the bottom line. He never visited communities. He never understood the issues. He never sat down with mayors or city managers. He never understood why those things were in the budget," Barrett said. "That money was requested by locals. It was a major league problem.”
No doubt, in South side Madison, this is the single most critical issue of the 2012 Presidential Election.
In the rest of America that is not committed to doubling down on failure, it is Obama's manifest, perpetual failures.
Failures he has no chance of fixing; failures he will only perpetuate.
Speaking of which, since you're canvassing for Obama in the wide open spaces of South side of Madison, tell us, please, the accomplishments they have you touting for Obama.
Cosmic Conservative: "Cosmic Conservative said... Roesch/Voltaire, trying to use rational analysis to explain voting in Massachusetts is a joke. This is the state that voted an alcoholic murdering incompetent to the Senate for decades based on fond memories of a failed President."
You know, I can remember all the way back (before John Kennedy Jr), when the Kennedy men used to kill their women just 1 at at time.
There is another reason I am hoping for a Romney win. And a convincing win would be even better.
Barack Obama ran the most cynical, scorched-earth campaign in my living memory. His campaign literally accused his opponent of being guilty of murder, or at best, negligent homicide. His campaign spent an estimated $300 million running nothing but ads of pure personal destruction. His entire strategy was wholly based on the concept that he could paint his opponent as such a monster that nobody would want to vote for him.
If Obama wins this election, that sort of campaign will become the norm. If he loses, we can at least hope that elections give at least some attention to actual issues, policy, agenda and record.
In case it wasn't clear, the Wed morning time off will be due to recovering from the day/night before, not from actually being at any polls....(of course you never know with the dems and their friendly judges keeping heavily dem poll sites open longer....)
Are the people reporting these polls really serious when they write that Romney is up 12% among independents, but is somehow going to lose?
Or do they know how silly that sounds, but somehow feel they *have* to write that?
"Ali Karim Bey said...
I am willing to deliver pizza to any-one but this election is OVER.
At Politico, we learn that Obama is ahead. At Nate Silver-land, we learn that Obama is ahead. At Huff-Post, Daily-Beast, TPM, we learn that Obama is ahead. Etc. Etc.
What part of the news are you selecting? The point is that this is not the needle in a haystack.
Every-one who is any-one in the media saying that the election is over, then why is Romney smiling?
EJ Dionne, the distinguished professor at Georgetown and a columnist among others. Judy Woodruff the distinguished reporter on PBS and a board member of Newseum among other things, all have said this many times.
Obama is here to stay till Jan. 2017.
What part of this are you all not following?"
I think there are a whole lot of people in this country who think exactly this. Next Tuesday is going to be a huge shock.
The interesting thing to see will be how many reach the correct conclusion: that the sources they look to to keep them informed have failed them.
This could be a day of reckoning for the MSM. But I doubt it.
I'm just really interested to see how the MSM will react -- what story will they come up with to explain how badly they got this wrong?
Seeing how Garage and RV seek to distract, how about this:
Adm. James A. Lyons, USN (Ret.) calls out Obama: "The Obama national security team, including CIA, DNI and the Pentagon, apparently watched and listened to the assault on the U.S. consulate and cries for help but did nothing."
So, Garage and RV, we all know your president left Americans, crying for help, to die. Tell us, please, what do you think?
The day of reckoning for the MSM has already come and gone. They are running on fumes and they know it. That, in fact, is the main reason they have deliberately shed any pretense of journalism in their naked attempts to drag their ideological favorite over the finish line.
Every major liberal rag is bleeding through the pores and cannot possibly continue. You will soon see desperate pleas for government buyouts and subsidies (which is in part why they are so desperate for an Obama win). They will argue that they perform a "necessary service" for the country.
The sad reality is that they had a responsibility to do exactly that. And they failed.
"I have always been a believer in data telling me the full story. Truth is, nobody knows what will happen on Election Day. But here is what we do know: 220,000 fewer Democrats have voted early in Ohio compared with 2008. And 30,000 more Republicans have cast their ballots compared with four years ago. That is a 250,000-vote net increase for a state Obama won by 260,000 votes in 2008."
Here's the line immediately preceding the excerpt R-V is quoting:
"Every time it rained, it wiped out their downtown," Barrett told HuffPost.
Wow, everything's getting "wiped out" now.
Peabody isn't some newly settled exurb; it was first settled in 1626 and incorporated as a city in 1916. If this has been such a problem for so long, why haven't the taxpayers of Peabody fixed it? It's home to a large regional mall, so it's not exactly lacking a tax base.
Could it be that the mayor wanted state funding for some ridiculously over-priced boondoggle?
Naaah. Obviously Romney just wanted Democrats to drown in those recurring, downtown-destroying floods.
clint: "I'm just really interested to see how the MSM will react -- what story will they come up with to explain how badly they got this wrong?"
Easy peasy.
First, Peter Jennings circa election night 1994 has already provided part of the answer when the dinosaur media could no longer pretend that "incumbents" were losing but instead it was just dem incumbents that were losing: "The voters threw a tantrum".
Second, as Bill Clinton has already laid out: "The voters are impatient" and "nobody, including me, could have fixed this economy in just 4 years".
That comment by Clinton lays the foundation for obama-ites to claim that the coming economic recovery under the Romney Admin is really all due to obambi's "hard work". LOL, but the left will believe it.
Third: "Racist!!"....yes yes I know logical thinkers will say "that doesn't make sense, since we already elected the guy once". Silly silly logical thinkers! These are lefties we are talking about! Of course we will see a "reset" to "Amerikkka the racist"!!
After all, it couldn't be obama's fault.
He was magical! He was a lightworker! He was "sort of a God"! He makes my legs tingle! (more likely incontinence)
Above all else, it won't be "because of obamas leftist policies".
It has long been axiomatic that when liberal/progressive politicians "fail" it is because the people or the system failed them, or that the opposition stymied them.
That's what will happen here too, IF Romney wins.
When conservatives fail, the blame is usually placed on the politician directly (GHW Bush, Nixon) or it is blamed on the conservative base (usually the horrid "social conservatives").
This is actually not all bad. In practice it means that conservatives tend to learn from their mistakes and liberals/progressives keep making the same mistakes over and over.
That's why Obama's administration will, in a few years, be remembered as a Carter rerun.
Honestly, the folks that put out these polls, the so-called 'experts', make me want to pull my hair out. They declare the race tied nationally, but Obama leads by 4 in the swing states! Did anyone performing this poll stop and question this? Either 1) the swing states aren't really swing states, or 2) their polling methodology is in error.
Meanwhile Rasmussen and Gallup both have Romney leading nationally and in swing states, not coincidentally by about the same margins.
If for nothing more than consistency, which outfits do you think are right?
Garage: "It has a copy of the booklet and examples of wrong information. What do you think?"
I think you're OT - again.
The post is about Obama losing more and more ground in the polls. How many points do you believe Mitt Romney will lose in the polls as a result of this booklet? Is there any relevance? You're like the lawyer without the law on his side and without the facts on his side who decides he's got nothing left except to pound on the table.
The post is about Obama losing more and more ground in the polls
From the link, which you apparently did not read:
But Democratic pollster Stanley Greenberg and Republican pollster Whit Ayres found that Obama leads by 4 points in the 12 battleground states that appear ready to pick the winner for the rest of the country next Tuesday. And they suggest that Romney's post-debate surge has "stalled."
That's probably why Romney is throwing a hail Jeep pass in Ohio, and sending booklets out to poll watchers to try and prevent as many votes being cast as possible.
The "Mitt-mentum has stalled" meme has been around for a couple weeks now Garage. A few weeks that have shown Mitt gaining about four points in national polls, moving into a tie in Ohio and threatening to take Wisconsin.
That's some stall there.
You might well believe that, but based on where Obama is sending his biggest assets a week from the election, it doesn't appear that Obama believes it.
A vegan pizza delivered any-where if Obama loses either VA or OH. He won't. So, I WILL NOT deliver.
The election is over. Heard at the morning WH-Chicago call. The move from FL to DC on AFOne was the STROKE of Genius - even Christie said so.
That stopped the bleeding. The voters have short-term. They will automatically go to he polls and vote for Obama. IT IS AUTOMATIC. They will not even think.
If I were an Obama supporter, the one number that would worry me the most is: "47".
As in both the Gallup and Rasmussen polls show Obama with 47% of the nation saying they will vote for him.
Both of those polls have Romney higher, but that's not really the problem Obama has. The problem Obama has is that he is a known commodity coming down to the wire in an election where people are not likely to radically change their opinion about him at the last minute.
Romney is not as well known, and the last month has been nothing but a national re-assessment of Romney's character and competence. It is likely that a LOT of people will think that it might be better to "take a chance" on the Romney they don't know than hold their nose and vote for an Obama they do know and do not like.
This is one of the most important polling measurements in politics. Incumbents who can't get to 50% in the final polls have a low rate of re-election. Incumbents who can't get to 48% have a dismal re-election rate.
If that 47% figure holds solid for another week, Barack Obama will almost certainly be kicked out of office.
With all the buzz about Benghazi, Obama's childish, simpering campaign and the ongoing economic woes that people are suffering, it will take something dramatic to get people to rethink their opinion of Obama at this time.
Which is why I keep waiting for the inevitable "October Surprise."
That's pretty much all Obama has left to hope for at this point.
One week left garage. I've said all along this would be a nail-biter of an election. That's still how I feel about it.
The fact that Obama is not running away with the election with the media in his pocket, Hollywood in his pocket and having used trillions of taxpayer dollars to pay off his constituents... well, I think that more or less speaks for itself.
Parts of the Big Dig in Boston were so poorly constructed that the roof fell in and killed some drivers—possibly because of graft. I've been there and driving in it is scary.
"And the 10% non-lefty is Car Talk and Whad'Ya Know."
But the 10% doesn't include "Wait, Wait, Don't Tell Me." I went to a taping of that once and it was so crudely smug and partisan that I was honestly surprised. I never heard that particular show on the air, so I don't know what made the final cut, but I thought the tone was quite rude for a general audience that paid good money to be there.
"If for nothing more than consistency, which outfits do you think are right?"
Question wasn't directed at me, but I don't think any of the outfits are right. They are all statistical samples. While some are better than others (ie those that call cellphones use live callers and call people back) most of them are valid estimates, subject to unavoidable statistical error. Based on the law of large numbers, statistical error decreases the more observations you have. That's why taking averages of all the polls is the best approach.
Cosmic: "You might well believe that, but based on where Obama is sending his biggest assets a week from the election, it doesn't appear that Obama believes it."
garage does not think about what people are actually doing.
garage likes to bask in the warm, warm glow of what he is told to think.
All you have to do is watch where the dems are sending their assets and it tells you what their internal polls are telling them.
I'd hate to be a Dem. They're going to be so unhappy the next four years regardless of who wins. It's either Romney as Prez or Obama the universally despised lame duck. So so sad.
BTW, is the left still running around talking about the vaunted obama early voting "advantage" that has turned out to be not an advantage?
In the swing states that matter, where the campaigns are active, Obama is holding a big lead. Which would seem to indicate a superior ground game. No wonder Republicans want to eliminate early voting.
AF said... Alrighty then! Great discussion, thanks for that.
A good rule of thumb, if you want great discussion don't start off insulting everyone first. Oh, you mean this is just something you say when people give you the respect you've earned? What a preening schmoe.
AF said... A good rule of thumb, if you want great discussion don't start off insulting everyone first.
Check out my first comment Marshal. Nothing insulting about it.
There's a famous psychiatric case about a guy who believed every new stream of thought was his first moment of sentience. He had a problem with his long term memory, so every time his short term memory was cleared he could remember nothing from before.
AF - If polling was nothing more than the raw collection of information, you'd have a point as to averaging the polls together. But part of the job of polling is also to forecast participation on election day. Raz and Gallup appear to use one (roughly even b/t R and D) while (for the most part) all others seem to assume more D than R. Gallup and Ras have their own polling to back up their predictions - I haven't seem why others think it'll be closer to 2008. Even the NPR poll of this thread is D +6. I can't believe anyone thinks it'll be even close to that on Nov 6.
My point is averaging polls with different assumptions together won't give a superior (or even meaningful) result.
I can't believe anyone thinks it'll be even close to that on Nov 6.
I can't either, but that is what worries me. I know there is some polling by Ras & Gallup about turnout, but how do the others decide? I assume they do their own polling on the issue. Why do they hear something so different?
If I were a Democrat, I'd be plenty nervous about Nov. 6. No matter what polls you read, Romney has zoomed ahead several points since his campaign was declared dead a month ago.
Whether Romney has overtaken Obama is arguable, but there is no question that Romney is attacking and Obama is defending.
Furthermore, the Romney campaign looks organized, confident, on-message and expanding into Obama territory. Meanwhile the Obama campaign is flailing about with silly attacks (Big Bird and binders), shifts in messaging ("New Economic Patriotism" brochure), frantic appeals to his base (talk shows, Lena Dunham video), and avoidance of serious questions.
Plus there is no good news in the pipeline for Obama, while Benghazi might rear up and bite his legs off.
I expect Romney to win by a solid five or six points with the possibility of a blow-out.
Patrick: "Why do they hear something so different?"
They are utilizing different built-in assumptions about the electorate (the "population").
This leads to different results when they sample that population.
The most glaring and easily understood is that many of the non-Rasmussen/Gallup org's were tweaking their samples to reflect higher dem participation than was actually realized from 2008.
That, my friends, is a very unrealistic assumption.
Which is what led to so many conservative complaints that many polling outfits were making themselves nothing more than an arm of the dems by using polls as a reinforcing tool for the dems strategic plan to paint the election as "over" last spring/summer.
This attempt to convince voters that there was simply no way any republican could ever beat obama was combined with the "kill Romney" campaign theme which attempted to paint Romney as so far beyond the pale that he (Romney) would be electorally destroyed long before the Fall campaign.
The obama-ites knew way back when just how weak obama's "support" was and easily it could be moved away from him if he didn't look invincible.
It might have worked to, if it hadn't been for those meddling kids.....
Oh, and I'm sure the Federal Gov'ts investigation into Gallup which began last summer had nothing whatsoever to do with axelrod's displeasure with Gallup's polling methodology which was at odds with the obama strategy.........
The most glaring and easily understood is that many of the non-Rasmussen/Gallup org's were tweaking their samples to reflect higher dem participation than was actually realized from 2008.
They are still doing that? Do they have anything to base it upon beyond "that's what happened 4 years ago?"
You're about 25% correct. Strictly speaking, the differing methodologies of different polls mean that an aggregate of all polls is not the same as a single poll with a gigantic sample size. Practically speaking, it's a reasonable approximation. That's why the Nate Silvers and RCPs of the world do better than any given poll.
Patrick: "They are still doing that? Do they have anything to base it upon beyond "that's what happened 4 years ago?"
Yes.
Of course, it all depends on what you mean by "the electorate looks the same as 2008".
Remember, when Gallup announced yesterday (day before?) that their expected electorate "looked" like 2008, they actually "buried the lede" so to speak.
Even if the demographics break out "close" to 2008, small differences in those demographics make a huge difference.
For instance, will whites make up 72% of the electorate, or up to 75%?
Big difference.
Further, even if the demographics for 2012 are similar in terms of age brackets, there has been a marked shift in party self-identification towards the republicans.
There are so many ways to look at this stuff it can make your head spin.
The questions you must always ask yourself are these: 1) What is it that my client actually needs? 2) What is it that my client is actually willing to pay for?
.....what? Those questions don't make sense to you?
AF: "That's why the Nate Silvers and RCPs of the world do better than any given poll."
Nate Silver doesn't just take an average of lots of polls, he WEIGHTS those polls differently, giving select polls MORE influence in his "average of polls".
What does Nate base his different "weightings" on?
Who knows.
Probably whatever keeps him employed at that bastion of objectivity and non-bias, the NYT.
garage - are you going to be your life savings on it?
No. But I have no idea what data Romney supporters declaring a blowout are looking at either.
Gallup LV: Romney +5 (+/- 1) over most of the last two weeks.
Rasmussen National: Romney +2-4 for weeks.
Rasmussen Swing States: Romney +4-6 for over a week.
Significant signs of Republican voter enthusiasm matching 2010. No signs of Democrat voter enthusiasm matching 2008.
Just about every national poll finds Romney leading among independents, usually by double digits.
In short, everything is pointing to a blowout of major proportions. The national electorate is going to be somewhere between D+2 and R+3. Ohio is going to be R+0 to R+5.
Obama has nothing to campaign on. His "successes" (ObamaCare and the Stimulus) are highly unpopular with everyone other than his base. So he's trying to make it a base-turnout election, but that's a losing strategy when the other side's base is fired up, and you're losing independents by 8+ points.
Bush could do it in 2004 because IIRC he only lost independents by 1%, more Democrats cross over than Republicans, and he got Republican voter turnout higher than it had been in a long time.
"Is there a reason the left doesn't consider Gallup to be a decent poll?
Yes, there is: it's not giving the answers they want to hear."
Gallup is a reputable pollster. This year, it seems likely that it is overestimating the percentage of likely voters who prefer Romney over Obama for the simple reason that there have been dozens of other polls, also by reputable pollsters, and none of them show Romney leading by as much Gallup does.
By the same token, it's unlikely that Obama is leading by 4 in Virginia, as a recent Washington Post poll found, or that he is leading in Ohio by 4 or 5, as recent polls by PPP and Time found. The laws of statistics state that if you take many random samples of a population, you will get a range of results and the true population value is most likely in the middle of the range.
"What does Nate base his different "weightings" on?
Who knows."
Actually, he explains it on his site.
But I actually agree with you here: I'm more comfortable just taking a simple average than applying Silver's special sauce. He has a good track record, but I find the averaging method to be more transparent. That's why I like RCP (which, however, injects its own bias by more or less arbitrarily excluding certain polls from its averages.)
I think part of the problem here is our imprecise use of certain terms.
For instance: Blowout.
What does "blowout" actually mean and what would it look like?
I mentioned in an earlier thread that in 1980, Reagan's "blowout" of Carter was only Reagan 50.7%, Carter 41%, Anderson 6.6%.
That was considered a "blowout" since Reagan beat Carter ty 9%, but Reagan didn't clear 50% by much. (EV's were alot higher naturally, since it's winner take all).
So, I would probably call Reagans 1980 win a solid win.
If Reagan had received 53% or more, I would call it a blowout.
I would call Reagans 1984 win a tsunami.
I would call Bush's 2000 win over Gore a "delightful exercise in schadenfreude".
In the swing states that matter, where the campaigns are active, Obama is holding a big lead. Which would seem to indicate a superior ground game. No wonder Republicans want to eliminate early voting.
Do you have any actual evidence for that? It's a serious question, not snark. Do you have any sources of real numbers for how many absentee ballots have been returned / how many people have voted early in person in the various swing states? Even better would be having that information, paired with the equivalent numbers from 2008 and 2010.
If you've got that, I'd love to see it.
If all you have is polls where the pollsters do not validate every claim of early voting, OTOH, IMHO you've got nothing.
You seem to want to make the 'truth is in the middle' argument, and it is likely we'll be somewhere b/t R+3 and D+7 on election day. Using a prediction model based on the last election isn't much of a justification, imho, particularly when there's strong evidence to indicate R+3 will be closer in Nov.
I don't take the folks at unskewed too seriously, but they do something that I'm surprised we don't see more often, that is, normalizing the data for participation biases. What's interesting is just how far off the cw is when this bias is removed.
"What does Nate base his different "weightings" on?
Who knows."
Actually, he explains it on his site.
He does? He gives you a formula that you can use, that would let you come up with the exact same weights that he does?
No? Then he doesn't "explain" it on his site.
But I actually agree with you here: I'm more comfortable just taking a simple average than applying Silver's special sauce. He has a good track record
Does he? How good was he in 2010, when the results weren't to his liking?
10 days out in 2010, he had the Republican chance of getting over 60 House seats to flip as less than 30%. They flipped 63. (I read all his posts about the House races after that one. He never again gave odds for the Republicans getting over 60 seats, so there's nothing else to compare it to. His final prediction was R+ 54 - 55. Off by 15% = not that special, IMHO.)
For whatever it is worth, looking at the polls and the current mood of the country, I think if the election were held today, Romney would scrape out a small victory.
Romney is still riding a wave from the debates. I frankly thought it would subside before now, but I think the Benghazi situation is still keeping him afloat a bit.
What happens between now and Tuesday will be the deciding factor. In fact Hurricane Sandy may be all that Obama needs to regain the advantage. I hope not, but that possibility is there.
Still, either way I think the race is going to be much closer than the vast majority of people think, and the reason I think it's going to be close is because both sides have impressive ground games.
In 2008 McCain had no ground game at all. Team Romney has built up a very effective ground game that is nearly the equal of the Obama machine.
There may be some surprises, especially in Wisconsin where Romney was able to exploit the highly effective and well-organized Walker recall election machine.
If Romney supporters maintain their current enthusiasm, that might be just enough to overcome Obama's paid machine.
We will see. The strategies have mostly all played out. Now it's just pure execution of the final tactics.
Drago, there's no perfect method, but neither is it rocket science. As with horshoes and hand grenades, close enough counts. Given the volume of polling this year, I'm comfortable with RCP's methodology -- simple average of all the polls from the last week, excluding multiple polls of the same jurisdiction by the same pollster. The only thing I would change is include all the polls.
I wouldn't be adverse to giving more weight to polls with stronger methodologies -- Gallup should get more weight than the other tracking polls for example because it doesn't use robocalls and has larger samples -- but that puts you on a slippery slope to Nate Silver's approach.
AF: "I'm comfortable with RCP's methodology -- simple average of all the polls from the last week, excluding multiple polls of the same jurisdiction by the same pollster. The only thing I would change is include all the polls."
Your comfort level is irrelevant.
And when you speak of "all the polls", you are giving more of the game away as it is patently obvious many of the "obama is winning!!!eleventy" polls are D+8/9/10/11/12/15/17!
AF: "Exactly right. When it comes to statistical samples, the "truth is in the middle" approach will get you the most accurate answer more often than not."
Incidentally, all of you who are so confident that Romney is more likely than not to win should not be wasting time on this site. Go invest in the betting markets! There are some serious arbitrage opportunities for you out there.
AF: "AF: "Exactly right. When it comes to statistical samples, the "truth is in the middle" approach will get you the most accurate answer more often than not."
So, if I take 4 parts meatloaf, and mix in 1 part dog crap and another part elephant dung, it's still "mostly" meatloaf, right?
AF: "Incidentally, all of you who are so confident that Romney is more likely than not to win should not be wasting time on this site. Go invest in the betting markets! There are some serious arbitrage opportunities for you out there."
Yes, it would be easier for you if no one from the other side were here.
FWIW, my take on Romney's "momentum" and the polls is this:
Suppose most people decided a long time ago to base their vote on the economy, somewhat in terms of past performance but mostly in terms of likely future performance. These people can all agree on their basic criteria, but differ in terms of when they think they've got enough data points to be confident enough to make up their minds about how to vote.
In this case, much of Romney's momentum would simply be b/c nobody thinks there's going to be any good economic news b/w now and the election. So a large share of these late deciders become pro-Romney.
This causes "public opinion" as tracked by the polls to appear to be changing, when what's actually happening is that it's converging to what it always would have been if we'd only used vote-forecasting models that were based on economic fundamentals rather than "public opinion" surveys.
BTW, wasn't Lucas running around last year whining that he was out of the Stars Wars "bidness" since he was getting raked over the coals for the last 3 films?
I think that AF and Drago are talking about two different statistical concepts: efficiency and bias.
AF is right that as sample size increases the distribution converge on the actual sample mean. And Drago's correct that this is only unambiguously valuable if the sample mean is the true population mean. If a sample is biased, then increasing its size simply induces convergence on an incorrect number.
In the swing states that matter, where the campaigns are active, Obama is holding a big lead. Which would seem to indicate a superior ground game. No wonder Republicans want to eliminate early voting.
AF: Exactly right. When it comes to statistical samples, the "truth is in the middle" approach will get you the most accurate answer more often than not.
Not because of anyone's track record -- because of the laws of statistics.
Argh. And you accused me of being 25% right. ;)
You're making unsupportable assumptions about your data if this is your argument. I'll leave it at that.
I'll leave you with this - Nate Silver saw the lay of the land in 2010 and choose not to believe it. As pointed out by others, he wasn't particularly accurate that year (+53R when it was +63R). It's not difficult to get a high percentage of house races correct - if you do nothing more than pick 'incumbent' you'll hit around 90% on average.
From 2010, the generic ballot question was polling around +1R, which was pretty much unheard of, so much so that Silver et al discounted it. A simple regression fit of final generic ballot preferences versus final results though the years indicated repubs would pick up 65 seats in 2010 (based on being +1R IIRC). Ten minutes of a back-of-the-envelope analysis did better than Silver.
I suspect 2012 is a repeat. Silver sees the same things the rest of us do, but can't bring himself to properly weigh it in his analysis. You can speculate as to why, I really don't care myself. Like in 2010, he'll move somewhat from his 2008 model, but not enough to capture reality. There are simply too many generalized, large scale data points working against his analysis. Quick example - independents are breaking for Romney by about 12 points. Again, a simple comparison to past elections indicates the repub candidate wins by about 5.5 points on average when leading the indy vote by this amount.
We'll know in a week, but there appear to be far too many indicators pointing to a Romney victory for the RCP and Silver to be correct this year.
Chip S.: "AF is right that as sample size increases the distribution converge on the actual sample mean. And Drago's correct that this is only unambiguously valuable if the sample mean is the true population mean. If a sample is biased, then increasing its size simply induces convergence on an incorrect number."
Correct.
My problem with the polls have everything to do with the initial and stubborn "assumptions" made by the major polling outfits.
Further, I postulated that the use of these unrealistic assumptions was something more than just "chance" and in fact were in service to the obama campaign strategy.
An interesting question would be: When Silver is wrong, how often is he wrong by finding a stronger Republican vote? Does he over estimate their chances?
Patrick: "An interesting question would be: When Silver is wrong, how often is he wrong by finding a stronger Republican vote? Does he over estimate their chances?"
Given that Pinch signs his checks, something tells me that Silver is always looking for the Dems silver lining.
In a move sure to not register on garage at all, now, like Susquehana, Mason-Dixon is pulling out of Florida because their polling shows Romney pulling away there:
Silver's "secret sauce", in 2008 at least, included having the Obama campaign's internals. Not sure that was ever widely known when he was congratulated for his forecasts in 2008 being so close to the actual outcome.
Most of these state polls that assume turnout for the Dems this year will be equal to or greater than 2008 are smoking something well beyond Maui Wowie.
Silver's "secret sauce", in 2008 at least, included having the Obama campaign's internals.
avwh: Yes. Furhtermore Silver did not disclose this nugget of information about his 2008 work because of a confidentiality agreement with the Obama campaign.
In 2008 Silver had a diary on DKos. He voted for Obama. He now blogs for and is paid by the NY Times, a very liberal organization in the tank for Obama.
These facts do not mean that Silver is intentionally juicing his numbers for Obama, but he has had a secret relationship with the Obama campaign before and he is a Kossack who favors Obama. Who knows.
In his current column Silver has done more than any other single writer to shore up liberal hopes for the Obama campaign after Obama's wipeout in the first debate.
For all these reasons Silver has conscious and unconscious reasons to nudge his numbers in Obama's direction. And at some level his secret sauce has got to be subjective.
I haven't trust his numbers this year. With a Romney win, Silver becomes another tarnished star.
I think that AF and Drago are talking about two different statistical concepts: efficiency and bias.
AF is right that as sample size increases the distribution converge on the actual sample mean. And Drago's correct that this is only unambiguously valuable if the sample mean is the true population mean. If a sample is biased, then increasing its size simply induces convergence on an incorrect number.
the problems interacting with each other. No power means gasoline can't be pumped at the stations, as in much of New Jersey, and it means humongous lines at the few stations that do have power. I know that power is coming back on, much too slowly for people's liking, of course, but it's going to be bad for a while. Tax Preparation NJ
Support the Althouse blog by doing your Amazon shopping going in through the Althouse Amazon link.
Amazon
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.
Support this blog with PayPal
Make a 1-time donation or set up a monthly donation of any amount you choose:
১৯৫টি মন্তব্য:
Just think how much that number will change in practice now that the hurricane has destroyed all the coastal cities in the nation!
Hooray for hurricanes!
the hurricane has destroyed all the coastal cities in the nation!
I had no idea that Miami, New Orleans, and SF had been hit so hard.
This is terrible. Perhaps the election should be postponed for a couple of years.
Uh, oh. NPR? Looks like cracks are appearing in the cocoon. With 7 days to go it is time for vast turn arounds in polls for those who want to salvage credibility. With old people already threatening to Cock-Punch Romney the next week is going to be interesting.
The most interesting part to me from that Pew Poll hasn't been reported on much. Those who are R or Lean R have a likely to vote of 76%. Meanwhile D or lean D have a likely to vote of 62%. This enthusiasm gap is huge.
So, we have an electorate of:
-- +1 to +3 R
-- R's are significantly more likely to vote
-- Romney is up 15-20 points with independents
Maybe these guys weren't just being crazily optimistic after all.
Romney still wins by a landslide, I don't see how the storm helps Obama.
NPR is a known right-wing propaganda site. Barry is leading by 20 points easy . . .
It must have killed them to say that, but, yes, we'll now see that miraculous swing all the media types have been ignoring for 4 months.
Here is an awesome Romney poll watching training booklet full misleading and false information for poll watchers. Part of that coveted Republican "ground game".
Is this a great country or what?
I am willing to deliver pizza to any-one but this election is OVER.
At Politico, we learn that Obama is ahead.
At Nate Silver-land, we learn that Obama is ahead.
At Huff-Post, Daily-Beast, TPM, we learn that Obama is ahead.
Etc. Etc.
What part of the news are you selecting? The point is that this is not the needle in a haystack.
Every-one who is any-one in the media saying that the election is over, then why is Romney smiling?
EJ Dionne, the distinguished professor at Georgetown and a columnist among others. Judy Woodruff the distinguished reporter on PBS and a board member of Newseum among other things, all have said this many times.
Obama is here to stay till Jan. 2017.
What part of this are you all not following?
The simple, known, confirmable fact is, no incumbent president polling under 50% has been reelected.
That's not determinative, of course, but this president has been polling under 50% for nineteen months now.
So, unless a critical mass of stupid votes prove utterly indifferent and immune to Obama's manifest failings, especially in Ohio, he will not be reelected.
"Is this a great country or what?"
It is. And with a little luck it will be even greater once Barry is sent packing.
Garage, I feel like your avatar can see into the depths of my soul. So, question: is it more marshmallowy or would you say gooey-creamy like in there?
I am willing to deliver pizza to any-one but this election is OVER.
At Politico, we learn that Obama is ahead.
At Nate Silver-land, we learn that Obama is ahead.
At Huff-Post, Daily-Beast, TPM, we learn that Obama is ahead.
Etc. Etc.
What part of the news are you selecting? The point is that this is not the needle in a haystack.
Every-one who is any-one in the media saying that the election is over, then why is Romney smiling?
EJ Dionne, the distinguished professor at Georgetown and a columnist among others. Judy Woodruff the distinguished reporter on PBS and a board member of Newseum among other things, all have said this many times.
Obama is here to stay till Jan. 2017.
What part of this are you all not following?
The media has been reporting bullshit polls suggesting that the race is close. Now the cynic in me thinks that the reporting is primarily designed to support the meme that the race is close. The pollsters will face their own armageddon on November 7 wherein they lose their contracts for marketing research.
Fortunately for the pollsters the American public is about as innumerate as is the president. And so the bullshit works.
Obama is going down hard. And the only thing of interest will be to read the "explanations" from the punditry post election.
With old people already threatening to Cock-Punch Romney the next week is going to be interesting.
If Obama wins, do we get to cock-punch some liberals? Do we get to burn some "mother-f***ing" cars with Obama stickers on them? Just curious what Move.On?Michael Moore are advocating here. Inquiring minds want to know.
"For myself, I expect Romney to win by just over 52 to 46 percent, with two minor candidates gathering about 2 percent between them."
Michael Novak in National Review, works for me.
http://www.nationalreview.com/articles/332002/why-romney-will-win-michael-novak
"Is this a great country or what?"
Oh-Bah-Muh, Oh-Bah-Muh, Oh-Bah-Muh, Oh-For-Four, Oh-For-Four, Oh-For-Four.
Obama trolls are funny.
But in a pathetic sort of way.
Oh-Bah-Muh, Oh-Bah-Muh, Oh-Bah-Muh, Oh-For-Four, Oh-For-Four, Oh-For-Four.
I have tried the making my own scenario many times.
http://elections.nytimes.com/2012/electoral-map
The result is the same:
243 (status quo) + VA + OH gives Obama the victory.
243 (status quo) + CO + OH gives Obama the victory.
Etc.
As I have said: The election is over.
No pollster's reputation will suffer more than Nate Silver's. Rassmussen's reputation will be enhanced.
This storm is a big help for Obama.
I think the Mainstream press will see this as an opportunity to keep Benghazi and Obama's slipping poll numbers among Indy's and Women a secret.
Look for a week's worth of breathless NYT headlines about how a worker hired with Stimulus bill funding was able to save an Egyptian sculpture at the Met by placing a bucket under a roof leak.
The pollsters will face their own armageddon on November 7 wherein they lose their contracts for marketing research.
Don't they already have the impact of Sandy lined up as an excuse? I thought I read that polling was being suspended.
ali karim bey (aka American Politico)--the most important question is, IMO, did you manage to get in Stephanie Cutters pants--she wont be having a mob soon, so the force may be with you :)
Alan Markus--I am sure that will be their excuse, but their track record is, in fact, a matter of record. Buyers of market research want results to further their bottom line, and many pollsters will be out in the cold (or under water).
Roger: No, but if Obama wins, I may get a 2nd chance.
My gut filling is that Sandy will amazingly help. Most voters have a short-term attention problem. They see a POTUS in situation room, and that is all that matters. The fact that the POTUS came back from FL after just a few hours there was a stroke of genius.
The election is over. Nate is correct. Politico is correct. Go here and you can see yourself.
http://elections.nytimes.com/2012/electoral-map
Bottom line:
Romney cannot win without VA and OH.
I told you that he will get neither.
Ali KB--I have enjoyed your commentary no end--I do hope you will continue even if you dont have your window office in the west wing. Best
Obama supporters should be required to install pollution control devices on their excuse machines.
Ali--keep your eyes on the prize: Stephanie Cutter's lady parts :)
Typically Garage refutes bad polls with good polls, now he resorts to a Romney training book.
If Romney wins it win will the inverse of the 2008 election. If he does win, will he have coattails? How far down will independents vote for republicans if the break for Romney? Will disaffected democrats vote for republicans for congress and state and local seats?
As for Obama, the worst thing that could happen to him is to win and have the republicans pick up the senate. Endless hearings over a multitude of scandals.
Obama and Biden may well turn out to be the democrats Nixon and Agnew. Anyway a week from today we shall know the outcome.
Ann, you have got to do a post on this:
From the very tech savvy president,
"FEMA, W.H. send storm victims to Internet
When President Barack Obama urged Americans under siege from Hurricane Sandy to stay inside and keep watch on ready.gov for the latest, he left out something pretty important — where to turn if the electricity goes out."
"Typically Garage refutes bad polls with good polls, now he resorts to a Romney training book."
There are no good polls for Barry. Even the skewed ones are turning out badly for him now.
Why Obama will win.
Even Christie said so. The Air-Force-One trip from FL to DC was a stroke of GENIUS. Axelrod has redeemed himself. What a genius. Now, we all know the POTUS Obama is Presidential. Even Christie said so.
http://www.politico.com/blogs/burns-haberman/2012/10/christie-obamas-been-wonderful-through-sandy-147683.html
Cuban Bob (and by the way my dad was born in Havana)--I do think Romney will have coat tails--we will find out on November 7th. In the interim I plan cooking up a batch of ropa vieja to celebrate November 6th. And I do ropa vieja very well.
"did you manage to get in Stephanie Cutters pants"
I've noticed Stephanie Cutter has a big mouth, better to tell big old lies with, I guess.
Dreams--perhaps for Ali's perspective Stephanie Cutter is a better choice than Debbie Wasserman--although it is admittedly a bad selection pool.
"Typically Garage refutes bad polls with good polls, now he resorts to a Romney training book."
So what did you think of the booklet full of errors to inform voters?
HAHAHAHA
Stupid question, just kidding bro!
20% of newspapers that endorsed Obama in 2008 are now endorsing Romney.
Sweet momentum.
USA Today: Everyone feels it's now OK to criticize the lightbringer.
The media is doing a wonderful job of keeping Obama in this race. If the average American had any clue what happened in the White House during Benghazi, this race would be over. Which, of course, is why the media is not reporting it.
But word is getting out, slowly but surely.
The bottom line of the current situation is that the national and state polls are in opposition. It is very hard to conceive of how Obama wins the majority of the biggest states by an overwhelming margin, Romney wins the popular vote, but somehow Obama wins enough swing states to win the electoral vote.
The math just doesn't work. There just aren't enough votes available to explain the contrast between Romney +4 nationally, but losing all the swing states.
Obama won in 2008 due to an historic combination of a nation tired of GW Bush and independents orgasmic over making an historic vote.
Neither of those are in play this year. Poll models based on 2008 election results are based on an outlier election, not a "normal" election.
Right now, in Ohio, Republicans are roughly 200,000 votes ahead of where they were in 2008. In the general election independents went for Obama by roughly 16 points. They are breaking for Romney by about that amount this year.
If that holds, there is no way Obama wins Ohio. In fact he is at risk of losing Minnesota and potentially Michigan and Pennsylvania as well.
Here is why the numbers are so low in Mass where they know well his malarkey: Barrett chalked Romney's veto of the Peabody project up to a lack of familiarity with infrastructure in the state.
"This was not unusual for him. He didn’t understand infrastructure improvements. It was just the bottom line. He never visited communities. He never understood the issues. He never sat down with mayors or city managers. He never understood why those things were in the budget," Barrett said. "That money was requested by locals. It was a major league problem.”
Here is an awesome Romney poll watching training booklet full misleading and false information for poll watchers. Part of that coveted Republican "ground game".
TP stands for Think Progress?
garage mahal said...
"Typically Garage refutes bad polls with good polls, now he resorts to a Romney training book."
So what did you think of the booklet full of errors to inform voters?
I bet they fixed it by now.
222,000
"While the Obama campaign would like to wish it is 2008, the reality is that they are now forced to “play defense” in least six states (Pennsylvania, Minnesota, Colorado, New Hampshire, Iowa and Wisconsin) that they once believed were “safe” Obama wins."
Its looking more and more ominous for the O, his Obamaness, the one we had we been waiting for.
http://www.nationalreview.com/corner
So what did you think of the booklet full of errors to inform voters?
You are OT dude, that's the point
HAHAHAHAHAHA
Stupid comment-not kidding
Who is ineligible (in Wisc.)?
He/she is a convicted felon and has not completed probation or parole.
A little different than on the Think Leftist site, don't you think?
...he left out something pretty important — where to turn if the electricity goes out.
Cut the man some slack, he still uses a Blackberry.
Roesch/Voltaire, trying to use rational analysis to explain voting in Massachusetts is a joke. This is the state that voted an alcoholic murdering incompetent to the Senate for decades based on fond memories of a failed President. It is also the state that is about to vote for a faux "native american" who has demonstrably committed academic and legal fraud.
Massachusetts. LOL.
Isn't a training guide for poll atchers to inform poll watchers, not voters?
Thanks for the link to another Loony-Left site, garage. Leftist paranoids provide my favorite form of light reading. I've always enjoyed The Nation and Alternet but, Jesus, they get fewer comments in a month than Althouse gets in a day.
I have to admit that as much as I am hoping for a Romney victory for the good of the nation and the good of international geopolitical stability, I am also very much hoping to see people like garage and inga have to eat their smug words.
"I am also very much hoping to see people like garage and inga have to eat their smug words."
A heaping helping of crow topped off with an equally ample helping of humble pie can be good for body and soul.
Roesch, in other states some of those municipal projects are what is called 'graft' and given to crony donors.
Now I'm not saying Romney was in the right if he did indeed ignore requests for local infrastrucutre improvements (which you haven't proven). But, if he was going to break the pork cycle, then he needed to be much more discriminant about approving these projects.
If you were familiar at all with the economics of the Big Dig in Boston, then you'd understand how it works in Massachusetts. It works well for those ensconced in power and those connected to them, but not so much for the larger electorate.
Thanks for the link to another Loony-Left site, garage.
It has a copy of the booklet and examples of wrong information. What do you think?
America's Politico, maybe you should go after some of the women in the GOP. Gail Gitcho, a Romney team member, is quite comely.
"This was not unusual for him. He didn’t understand infrastructure improvements. It was just the bottom line. He never visited communities. He never understood the issues. He never sat down with mayors or city managers. He never understood why those things were in the budget," Barrett said. "That money was requested by locals. It was a major league problem.”
No doubt, in South side Madison, this is the single most critical issue of the 2012 Presidential Election.
In the rest of America that is not committed to doubling down on failure, it is Obama's manifest, perpetual failures.
Failures he has no chance of fixing; failures he will only perpetuate.
Speaking of which, since you're canvassing for Obama in the wide open spaces of South side of Madison, tell us, please, the accomplishments they have you touting for Obama.
You can do that, right?
The entire Obama campaign strategy at this point can be described in one word:
"Squirrel!"
"It has a copy of the booklet and examples of wrong information. What do you think?"
It is nothing.
Anyone who has ever poll-watched below knows it is nothing.
Care to hazard a guess as to why?
Dig deep...
It's worse than that. Look at the numbers among independents: Obama's getting hammered.
I'm going to be having a really nice celebration Tuesday night.
Heh, I scheduled next Wed as a day off of work a long time ago, mostly because I didn't want to deal with being depressed.
Now I'm hoping I'm dealing with a hangover...
But, as the master would say. "Don't get cocky."
Ali Karim Bey said...
Bottom line:
Romney cannot win without VA and OH.
I told you that he will get neither.
And you're wrong, because VA is in the bag for Romney, and Ohio is leaning for him.
Romney is a winner and Obama a loser. Even the loons at NPR can see it. An incumbent president polling under 49.5% --- is a loser every time.
Cosmic Conservative: "Cosmic Conservative said...
Roesch/Voltaire, trying to use rational analysis to explain voting in Massachusetts is a joke. This is the state that voted an alcoholic murdering incompetent to the Senate for decades based on fond memories of a failed President."
You know, I can remember all the way back (before John Kennedy Jr), when the Kennedy men used to kill their women just 1 at at time.
Good times, good times.
There is another reason I am hoping for a Romney win. And a convincing win would be even better.
Barack Obama ran the most cynical, scorched-earth campaign in my living memory. His campaign literally accused his opponent of being guilty of murder, or at best, negligent homicide. His campaign spent an estimated $300 million running nothing but ads of pure personal destruction. His entire strategy was wholly based on the concept that he could paint his opponent as such a monster that nobody would want to vote for him.
If Obama wins this election, that sort of campaign will become the norm. If he loses, we can at least hope that elections give at least some attention to actual issues, policy, agenda and record.
Cosmic: "Heh, I scheduled next Wed as a day off of work a long time ago, mostly because I didn't want to deal with being depressed."
I've scheduled time off from work as a Poll Checker for Romney on Tuesday and Wed morning.
I'll be back in the office on Wed afternoon 'cuz I got's to keep working for the man.
In case it wasn't clear, the Wed morning time off will be due to recovering from the day/night before, not from actually being at any polls....(of course you never know with the dems and their friendly judges keeping heavily dem poll sites open longer....)
Oh, and by the way, NPR is now clearly racist.
For anyone who thinks the election will be over at the scheduled poll closing time, I have two words for you.
McCaskill
Franken
If Dems CAN win by cheating, they WILL win by cheating. As Hugh Hewitt would say, to win, we have to win by enough to make cheating impossible.
Anyone who has ever poll-watched below knows it is nothing.
The Romney campaign went through the trouble of producing a booklet for his "team". Full of errors. Probably figured they're too dumb to know.
Are the people reporting these polls really serious when they write that Romney is up 12% among independents, but is somehow going to lose?
Or do they know how silly that sounds, but somehow feel they *have* to write that?
"Ali Karim Bey said...
I am willing to deliver pizza to any-one but this election is OVER.
At Politico, we learn that Obama is ahead.
At Nate Silver-land, we learn that Obama is ahead.
At Huff-Post, Daily-Beast, TPM, we learn that Obama is ahead.
Etc. Etc.
What part of the news are you selecting? The point is that this is not the needle in a haystack.
Every-one who is any-one in the media saying that the election is over, then why is Romney smiling?
EJ Dionne, the distinguished professor at Georgetown and a columnist among others. Judy Woodruff the distinguished reporter on PBS and a board member of Newseum among other things, all have said this many times.
Obama is here to stay till Jan. 2017.
What part of this are you all not following?"
I think there are a whole lot of people in this country who think exactly this. Next Tuesday is going to be a huge shock.
The interesting thing to see will be how many reach the correct conclusion: that the sources they look to to keep them informed have failed them.
This could be a day of reckoning for the MSM. But I doubt it.
I'm just really interested to see how the MSM will react -- what story will they come up with to explain how badly they got this wrong?
Seeing how Garage and RV seek to distract, how about this:
Adm. James A. Lyons, USN (Ret.) calls out Obama: "The Obama national security team, including CIA, DNI and the Pentagon, apparently watched and listened to the assault on the U.S. consulate and cries for help but did nothing."
So, Garage and RV, we all know your president left Americans, crying for help, to die. Tell us, please, what do you think?
Re: poll training document
I can't believe you fell for such a simple hoax, garage.
Remember the website where the Obama campaign used the Romney symbol?
This is the exact same thing, obviously.
You never have encountered a false flag operation too obvious to penetrate your gullibiity, have you?
The day of reckoning for the MSM has already come and gone. They are running on fumes and they know it. That, in fact, is the main reason they have deliberately shed any pretense of journalism in their naked attempts to drag their ideological favorite over the finish line.
Every major liberal rag is bleeding through the pores and cannot possibly continue. You will soon see desperate pleas for government buyouts and subsidies (which is in part why they are so desperate for an Obama win). They will argue that they perform a "necessary service" for the country.
The sad reality is that they had a responsibility to do exactly that. And they failed.
Tell us, please, what do you think?
Fire don't melt steel, yo!
The Mayans predicted a Romney victory. So did that lobster that is never wrong and the parrot, too.
If Big Bird isn't on suicide watch, we have no common decency.
"I have always been a believer in data telling me the full story. Truth is, nobody knows what will happen on Election Day. But here is what we do know: 220,000 fewer Democrats have voted early in Ohio compared with 2008. And 30,000 more Republicans have cast their ballots compared with four years ago. That is a 250,000-vote net increase for a state Obama won by 260,000 votes in 2008."
Deal with it liberal Dems.
http://www.commentarymagazine.com/2012/10/30/obamas-early-voting-strategy-flops/#more-809496
NPR will probably use these foreboding numbers to fundraise.
Obama's firewall is on fire!
Bill Clinton is in MN, Joe Biden in PA, instead of OH or VA? With a week to go?
Here's the line immediately preceding the excerpt R-V is quoting:
"Every time it rained, it wiped out their downtown," Barrett told HuffPost.
Wow, everything's getting "wiped out" now.
Peabody isn't some newly settled exurb; it was first settled in 1626 and incorporated as a city in 1916. If this has been such a problem for so long, why haven't the taxpayers of Peabody fixed it? It's home to a large regional mall, so it's not exactly lacking a tax base.
Could it be that the mayor wanted state funding for some ridiculously over-priced boondoggle?
Naaah. Obviously Romney just wanted Democrats to drown in those recurring, downtown-destroying floods.
Never let a crisis go to waste.
clint: "I'm just really interested to see how the MSM will react -- what story will they come up with to explain how badly they got this wrong?"
Easy peasy.
First, Peter Jennings circa election night 1994 has already provided part of the answer when the dinosaur media could no longer pretend that "incumbents" were losing but instead it was just dem incumbents that were losing: "The voters threw a tantrum".
Second, as Bill Clinton has already laid out: "The voters are impatient" and "nobody, including me, could have fixed this economy in just 4 years".
That comment by Clinton lays the foundation for obama-ites to claim that the coming economic recovery under the Romney Admin is really all due to obambi's "hard work". LOL, but the left will believe it.
Third: "Racist!!"....yes yes I know logical thinkers will say "that doesn't make sense, since we already elected the guy once". Silly silly logical thinkers! These are lefties we are talking about! Of course we will see a "reset" to "Amerikkka the racist"!!
After all, it couldn't be obama's fault.
He was magical!
He was a lightworker!
He was "sort of a God"!
He makes my legs tingle! (more likely incontinence)
Above all else, it won't be "because of obamas leftist policies".
No sireee. It's never about leftist policy.
It can't be.
Why?
Because shut up.
It has long been axiomatic that when liberal/progressive politicians "fail" it is because the people or the system failed them, or that the opposition stymied them.
That's what will happen here too, IF Romney wins.
When conservatives fail, the blame is usually placed on the politician directly (GHW Bush, Nixon) or it is blamed on the conservative base (usually the horrid "social conservatives").
This is actually not all bad. In practice it means that conservatives tend to learn from their mistakes and liberals/progressives keep making the same mistakes over and over.
That's why Obama's administration will, in a few years, be remembered as a Carter rerun.
Because it was.
Honestly, the folks that put out these polls, the so-called 'experts', make me want to pull my hair out. They declare the race tied nationally, but Obama leads by 4 in the swing states! Did anyone performing this poll stop and question this? Either 1) the swing states aren't really swing states, or 2) their polling methodology is in error.
Meanwhile Rasmussen and Gallup both have Romney leading nationally and in swing states, not coincidentally by about the same margins.
If for nothing more than consistency, which outfits do you think are right?
garage mahal said...
"The Romney campaign went through the trouble of producing a booklet for his "team". Full of errors. Probably figured they're too dumb to know."
Wow, garage, are you really so stupid you don't understand the difference between idiocy and ignorance?
People might be "too dumb to UNDERSTAND" or "too IGNORANT to know". But no one is "too dumb to know".
I do appreciate your willingness to make yourself look bad.
LOL
It's a good thing "no one" is politicizing this storm.......
Just the civility BS.
We see once again that the left really doesn't believe a single thing that they lecture the rest of us about.
Fools, you don't know the outcome. I Do.
Have you seen the movie, Fields of Dreams?
People will just come to vote for Obama.
They will do it automatically.
The bottom line: Obama HAS WON VA and OH (already!)
The election is OVER.
Dan: "If for nothing more than consistency, which outfits do you think are right?"
LOL
Leftists don't "think".
They are told what to say/do/feel.
They are voice-actuated automatons.
Nothing more.
As an earlier commenter mentioned:
"o-ba-muh....o-ba-muh....o-ba-muh...."
Hey Ali Karim Bey,
You're a Republican troll, pretending to be a Democrat lunatic, right?
Garage: "It has a copy of the booklet and examples of wrong information. What do you
think?"
I think you're OT - again.
The post is about Obama losing more and more ground in the polls. How many points do you believe Mitt Romney will lose in the polls as a result of this booklet? Is there any relevance? You're like the lawyer without the law on his side and without the facts on his side who decides he's got nothing left except to pound on the table.
garage's yelping suggests that the felon vote is crucial to the Democrats.
I don't doubt that.
The post is about Obama losing more and more ground in the polls
From the link, which you apparently did not read:
But Democratic pollster Stanley Greenberg and Republican pollster Whit Ayres found that Obama leads by 4 points in the 12 battleground states that appear ready to pick the winner for the rest of the country next Tuesday. And they suggest that Romney's post-debate surge has "stalled."
That's probably why Romney is throwing a hail Jeep pass in Ohio, and sending booklets out to poll watchers to try and prevent as many votes being cast as possible.
garage mahal said...
Tell us, please, what do you think?
Fire don't melt steel, yo!
LOL
NPR is about 60/40 liberal in coverage of news stories. But their 60% is brutal to hear.
But the day NPR makes fun of Obama like they do GOP guys and gals will be the day hell freezes over.
The "Mitt-mentum has stalled" meme has been around for a couple weeks now Garage. A few weeks that have shown Mitt gaining about four points in national polls, moving into a tie in Ohio and threatening to take Wisconsin.
That's some stall there.
You might well believe that, but based on where Obama is sending his biggest assets a week from the election, it doesn't appear that Obama believes it.
A vegan pizza delivered any-where if Obama loses either VA or OH. He won't. So, I WILL NOT deliver.
The election is over. Heard at the morning WH-Chicago call. The move from FL to DC on AFOne was the STROKE of Genius - even Christie said so.
That stopped the bleeding. The voters have short-term. They will automatically go to he polls and vote for Obama. IT IS AUTOMATIC. They will not even think.
The election is OVER. GO home every-body.
If I were an Obama supporter, the one number that would worry me the most is: "47".
As in both the Gallup and Rasmussen polls show Obama with 47% of the nation saying they will vote for him.
Both of those polls have Romney higher, but that's not really the problem Obama has. The problem Obama has is that he is a known commodity coming down to the wire in an election where people are not likely to radically change their opinion about him at the last minute.
Romney is not as well known, and the last month has been nothing but a national re-assessment of Romney's character and competence. It is likely that a LOT of people will think that it might be better to "take a chance" on the Romney they don't know than hold their nose and vote for an Obama they do know and do not like.
This is one of the most important polling measurements in politics. Incumbents who can't get to 50% in the final polls have a low rate of re-election. Incumbents who can't get to 48% have a dismal re-election rate.
If that 47% figure holds solid for another week, Barack Obama will almost certainly be kicked out of office.
With all the buzz about Benghazi, Obama's childish, simpering campaign and the ongoing economic woes that people are suffering, it will take something dramatic to get people to rethink their opinion of Obama at this time.
Which is why I keep waiting for the inevitable "October Surprise."
That's pretty much all Obama has left to hope for at this point.
A few weeks that have shown Mitt gaining about four points in national polls, moving into a tie in Ohio and threatening to take Wisconsin.
yes, if you only read Rasmussen and Gallup polls, Romney is doing well.
One week left garage. I've said all along this would be a nail-biter of an election. That's still how I feel about it.
The fact that Obama is not running away with the election with the media in his pocket, Hollywood in his pocket and having used trillions of taxpayer dollars to pay off his constituents... well, I think that more or less speaks for itself.
@Cosmic
I'm sure it will be close. Should be exciting! And good luck to you.
"NPR is about 60/40 liberal in coverage of news stories."
I think 60 means something different on your planet.
Here that number is 90.
And the 10% non-lefty is Car Talk and Whad'Ya Know.
Parts of the Big Dig in Boston were so poorly constructed that the roof fell in and killed some drivers—possibly because of graft. I've been there and driving in it is scary.
"And the 10% non-lefty is Car Talk and Whad'Ya Know."
But the 10% doesn't include "Wait, Wait, Don't Tell Me." I went to a taping of that once and it was so crudely smug and partisan that I was honestly surprised. I never heard that particular show on the air, so I don't know what made the final cut, but I thought the tone was quite rude for a general audience that paid good money to be there.
yes, if you only read Rasmussen and Gallup polls, Romney is doing well.
Well, they have the polls that most closely resemble polls based on national party affiliation, don't they?
Heard at the morning WH-Chicago call
This was funny. Once.
Have you seen the movie, Fields of Dreams?
People will just come to vote for Obama.
They will do it automatically.
He's talking about dead Democrat voters in Iowa!
"If for nothing more than consistency, which outfits do you think are right?"
Question wasn't directed at me, but I don't think any of the outfits are right. They are all statistical samples. While some are better than others (ie those that call cellphones use live callers and call people back) most of them are valid estimates, subject to unavoidable statistical error. Based on the law of large numbers, statistical error decreases the more observations you have. That's why taking averages of all the polls is the best approach.
@kcom "Wait, Wait, Don't Tell Me."
Yeah, NPR never has bake sales for fundraising because everything tastes of bitter and bile.
Cosmic: "You might well believe that, but based on where Obama is sending his biggest assets a week from the election, it doesn't appear that Obama believes it."
garage does not think about what people are actually doing.
garage likes to bask in the warm, warm glow of what he is told to think.
All you have to do is watch where the dems are sending their assets and it tells you what their internal polls are telling them.
AF: "That's why taking averages of all the polls is the best approach."
LOL
I can see you've never actually had to use statistics to earn a living.
But that's ok.
In a week we'll know everything.
BTW, is the left still running around talking about the vaunted obama early voting "advantage" that has turned out to be not an advantage?
I'd hate to be a Dem. They're going to be so unhappy the next four years regardless of who wins. It's either Romney as Prez or Obama the universally despised lame duck. So so sad.
BTW, here's an article that captures in a much more articulate way what I've been thinking about this election thus far:
http://www.realclearpolitics.com/articles/2012/10/30/are_we_looking_at_an_undertow_election.html
Drago: "I can see you've never actually had to use statistics to earn a living."
I can see that you've either never taken stat 101 -- or you pretend not to.
BTW, is the left still running around talking about the vaunted obama early voting "advantage" that has turned out to be not an advantage?
In the swing states that matter, where the campaigns are active, Obama is holding a big lead. Which would seem to indicate a superior ground game. No wonder Republicans want to eliminate early voting.
Maybe next time.
Bwahahahaha.
I give Obama New York, New England, New Jersey,Maryland, DC, and Delaware. Also California, Nevada, New Mexico, Hawaii, and Washington.
I'm still not sure about Minnesota and Oregon. They might stay blue.
The rest are all Romney.
Romney by 100 electoral votes, give or take.
AF: "I can see that you've either never taken stat 101 -- or you pretend not to."
LOL
Come out into the real world laddie.
It's not as scary as you think....unless you're on the left.
Drago:"LOL. Come out into the real world laddie. It's not as scary as you think....unless you're on the left."
Alrighty then! Great discussion, thanks for that.
AF: "Alrighty then! Great discussion, thanks for that."
You're more than welcome.
One does what one can.
Is there a reason the left doesn't consider Gallup to be a decent poll?
AF said...
Alrighty then! Great discussion, thanks for that.
A good rule of thumb, if you want great discussion don't start off insulting everyone first. Oh, you mean this is just something you say when people give you the respect you've earned? What a preening schmoe.
A good rule of thumb, if you want great discussion don't start off insulting everyone first.
Check out my first comment Marshal. Nothing insulting about it.
And by the way, I wasn't complaining about lack of respect, I was simply noting the lack of content.
AF said...
A good rule of thumb, if you want great discussion don't start off insulting everyone first.
Check out my first comment Marshal. Nothing insulting about it.
There's a famous psychiatric case about a guy who believed every new stream of thought was his first moment of sentience. He had a problem with his long term memory, so every time his short term memory was cleared he could remember nothing from before.
AF - If polling was nothing more than the raw collection of information, you'd have a point as to averaging the polls together. But part of the job of polling is also to forecast participation on election day. Raz and Gallup appear to use one (roughly even b/t R and D) while (for the most part) all others seem to assume more D than R. Gallup and Ras have their own polling to back up their predictions - I haven't seem why others think it'll be closer to 2008. Even the NPR poll of this thread is D +6. I can't believe anyone thinks it'll be even close to that on Nov 6.
My point is averaging polls with different assumptions together won't give a superior (or even meaningful) result.
Dan: "My point is averaging polls with different assumptions together won't give a superior (or even meaningful) result."
This.
In AF's defense, I didn't take his comments to be snarky or off-putting.
I can't believe anyone thinks it'll be even close to that on Nov 6.
I can't either, but that is what worries me. I know there is some polling by Ras & Gallup about turnout, but how do the others decide? I assume they do their own polling on the issue. Why do they hear something so different?
garage - are you going to be your life savings on it?
If I were a Democrat, I'd be plenty nervous about Nov. 6. No matter what polls you read, Romney has zoomed ahead several points since his campaign was declared dead a month ago.
Whether Romney has overtaken Obama is arguable, but there is no question that Romney is attacking and Obama is defending.
Furthermore, the Romney campaign looks organized, confident, on-message and expanding into Obama territory. Meanwhile the Obama campaign is flailing about with silly attacks (Big Bird and binders), shifts in messaging ("New Economic Patriotism" brochure), frantic appeals to his base (talk shows, Lena Dunham video), and avoidance of serious questions.
Plus there is no good news in the pipeline for Obama, while Benghazi might rear up and bite his legs off.
I expect Romney to win by a solid five or six points with the possibility of a blow-out.
Patrick: "Why do they hear something so different?"
They are utilizing different built-in assumptions about the electorate (the "population").
This leads to different results when they sample that population.
The most glaring and easily understood is that many of the non-Rasmussen/Gallup org's were tweaking their samples to reflect higher dem participation than was actually realized from 2008.
That, my friends, is a very unrealistic assumption.
Which is what led to so many conservative complaints that many polling outfits were making themselves nothing more than an arm of the dems by using polls as a reinforcing tool for the dems strategic plan to paint the election as "over" last spring/summer.
This attempt to convince voters that there was simply no way any republican could ever beat obama was combined with the "kill Romney" campaign theme which attempted to paint Romney as so far beyond the pale that he (Romney) would be electorally destroyed long before the Fall campaign.
The obama-ites knew way back when just how weak obama's "support" was and easily it could be moved away from him if he didn't look invincible.
It might have worked to, if it hadn't been for those meddling kids.....
Oh, and I'm sure the Federal Gov'ts investigation into Gallup which began last summer had nothing whatsoever to do with axelrod's displeasure with Gallup's polling methodology which was at odds with the obama strategy.........
garage - are you going to be your life savings on it?
No. But I have no idea what data Romney supporters declaring a blowout are looking at either.
The most glaring and easily understood is that many of the non-Rasmussen/Gallup org's were tweaking their samples to reflect higher dem participation than was actually realized from 2008.
They are still doing that? Do they have anything to base it upon beyond "that's what happened 4 years ago?"
Is there a reason the left doesn't consider Gallup to be a decent poll?
Besides the recent polls the left does not like and the fact that one of their guys appeared on FNC a few times, I can't think of any reason.
Blogger Patrick said...
Is there a reason the left doesn't consider Gallup to be a decent poll?
Yes, there is: it's not giving the answers they want to hear.
garage: "No. But I have no idea what data Romney supporters declaring a blowout are looking at either"
What we are looking at is a RANGE of possibilities.
That range has shifted in Romney's favor significantly since August.
Does that mean that Romney can't lose?
No.
It just means that most likely scenario is a Romney 3 to 4 point win with decreasing possibilities for a Romney blowout OR Obama close win.
What no one is talking about anymore is an obama blowout, something was routinely discussed 4 and 5 months ago.
Dan danoso:
You're about 25% correct. Strictly speaking, the differing methodologies of different polls mean that an aggregate of all polls is not the same as a single poll with a gigantic sample size. Practically speaking, it's a reasonable approximation. That's why the Nate Silvers and RCPs of the world do better than any given poll.
Patrick: "They are still doing that? Do they have anything to base it upon beyond "that's what happened 4 years ago?"
Yes.
Of course, it all depends on what you mean by "the electorate looks the same as 2008".
Remember, when Gallup announced yesterday (day before?) that their expected electorate "looked" like 2008, they actually "buried the lede" so to speak.
Even if the demographics break out "close" to 2008, small differences in those demographics make a huge difference.
For instance, will whites make up 72% of the electorate, or up to 75%?
Big difference.
Further, even if the demographics for 2012 are similar in terms of age brackets, there has been a marked shift in party self-identification towards the republicans.
There are so many ways to look at this stuff it can make your head spin.
The questions you must always ask yourself are these:
1) What is it that my client actually needs?
2) What is it that my client is actually willing to pay for?
.....what? Those questions don't make sense to you?
Hey man, this isn't about you....
AF: "That's why the Nate Silvers and RCPs of the world do better than any given poll."
Nate Silver doesn't just take an average of lots of polls, he WEIGHTS those polls differently, giving select polls MORE influence in his "average of polls".
What does Nate base his different "weightings" on?
Who knows.
Probably whatever keeps him employed at that bastion of objectivity and non-bias, the NYT.
LOL
garage mahal said...
garage - are you going to be your life savings on it?
No. But I have no idea what data Romney supporters declaring a blowout are looking at either.
Gallup LV: Romney +5 (+/- 1) over most of the last two weeks.
Rasmussen National: Romney +2-4 for weeks.
Rasmussen Swing States: Romney +4-6 for over a week.
Significant signs of Republican voter enthusiasm matching 2010. No signs of Democrat voter enthusiasm matching 2008.
Just about every national poll finds Romney leading among independents, usually by double digits.
In short, everything is pointing to a blowout of major proportions. The national electorate is going to be somewhere between D+2 and R+3. Ohio is going to be R+0 to R+5.
Obama has nothing to campaign on. His "successes" (ObamaCare and the Stimulus) are highly unpopular with everyone other than his base. So he's trying to make it a base-turnout election, but that's a losing strategy when the other side's base is fired up, and you're losing independents by 8+ points.
Bush could do it in 2004 because IIRC he only lost independents by 1%, more Democrats cross over than Republicans, and he got Republican voter turnout higher than it had been in a long time.
None of these are going to be true for Obama.
"Is there a reason the left doesn't consider Gallup to be a decent poll?
Yes, there is: it's not giving the answers they want to hear."
Gallup is a reputable pollster. This year, it seems likely that it is overestimating the percentage of likely voters who prefer Romney over Obama for the simple reason that there have been dozens of other polls, also by reputable pollsters, and none of them show Romney leading by as much Gallup does.
By the same token, it's unlikely that Obama is leading by 4 in Virginia, as a recent Washington Post poll found, or that he is leading in Ohio by 4 or 5, as recent polls by PPP and Time found. The laws of statistics state that if you take many random samples of a population, you will get a range of results and the true population value is most likely in the middle of the range.
Not necessarily, but most likely.
"What does Nate base his different "weightings" on?
Who knows."
Actually, he explains it on his site.
But I actually agree with you here: I'm more comfortable just taking a simple average than applying Silver's special sauce. He has a good track record, but I find the averaging method to be more transparent. That's why I like RCP (which, however, injects its own bias by more or less arbitrarily excluding certain polls from its averages.)
I think part of the problem here is our imprecise use of certain terms.
For instance: Blowout.
What does "blowout" actually mean and what would it look like?
I mentioned in an earlier thread that in 1980, Reagan's "blowout" of Carter was only Reagan 50.7%, Carter 41%, Anderson 6.6%.
That was considered a "blowout" since Reagan beat Carter ty 9%, but Reagan didn't clear 50% by much. (EV's were alot higher naturally, since it's winner take all).
So, I would probably call Reagans 1980 win a solid win.
If Reagan had received 53% or more, I would call it a blowout.
I would call Reagans 1984 win a tsunami.
I would call Bush's 2000 win over Gore a "delightful exercise in schadenfreude".
But that's just me.
Nate to get the Pulitzer next year. He will get the GENIUS award. He is highly sought from NYT for fast-track on Op-Ed pages to replace others.
Obama will DEFEAT Romney.
Why?
Both VA and OH are going blue for Obama.
AF: "Actually, he explains it on his site."
LOL
I know he does.
It's still a blackbox.
Ali: "Both VA and OH are going blue for Obama."
Maybe they're just holding their breath........
garage mahal said...
In the swing states that matter, where the campaigns are active, Obama is holding a big lead. Which would seem to indicate a superior ground game. No wonder Republicans want to eliminate early voting.
Do you have any actual evidence for that? It's a serious question, not snark. Do you have any sources of real numbers for how many absentee ballots have been returned / how many people have voted early in person in the various swing states? Even better would be having that information, paired with the equivalent numbers from 2008 and 2010.
If you've got that, I'd love to see it.
If all you have is polls where the pollsters do not validate every claim of early voting, OTOH, IMHO you've got nothing.
In short, everything is pointing to a blowout of major proportions
The betting markets aren't looking at Rasmussen/Gallup only.
Betfair give Obama a 68% chance
Intrade give Obama a 63% chance
Iowa Electronic Markets give Obama a 61% chance.
RCP, shows Romney up by 0.8 percent nationally, but shows Obama up in Ohio, New Hampshire, Iowa, Nevada, Wisconsin, Pennsylvania and Michigan.
AF: "That's why I like RCP (which, however, injects its own bias by more or less arbitrarily excluding certain polls from its averages.)"
Which is it's own "problem".
I mean, how long should you keep a poll around?
Should you "mix" polls from 3 weeks ago with a poll from today?
Does it matter if someone tries to inject a RV poll into the last month?
What if they only poll 600 people? Compared to other polls that are up around 1000+?
etc.
Hey, if we could truly model human behavior as well as we thought, we'd make a killing in the market.
AF - Gee, 25% - thanks.
You seem to want to make the 'truth is in the middle' argument, and it is likely we'll be somewhere b/t R+3 and D+7 on election day. Using a prediction model based on the last election isn't much of a justification, imho, particularly when there's strong evidence to indicate R+3 will be closer in Nov.
I don't take the folks at unskewed too seriously, but they do something that I'm surprised we don't see more often, that is, normalizing the data for participation biases. What's interesting is just how far off the cw is when this bias is removed.
Blogger AF said...
"What does Nate base his different "weightings" on?
Who knows."
Actually, he explains it on his site.
He does? He gives you a formula that you can use, that would let you come up with the exact same weights that he does?
No? Then he doesn't "explain" it on his site.
But I actually agree with you here: I'm more comfortable just taking a simple average than applying Silver's special sauce. He has a good track record
Does he? How good was he in 2010, when the results weren't to his liking?
10 days out in 2010, he had the Republican chance of getting over 60 House seats to flip as less than 30%. They flipped 63. (I read all his posts about the House races after that one. He never again gave odds for the Republicans getting over 60 seats, so there's nothing else to compare it to. His final prediction was R+ 54 - 55. Off by 15% = not that special, IMHO.)
For whatever it is worth, looking at the polls and the current mood of the country, I think if the election were held today, Romney would scrape out a small victory.
Romney is still riding a wave from the debates. I frankly thought it would subside before now, but I think the Benghazi situation is still keeping him afloat a bit.
What happens between now and Tuesday will be the deciding factor. In fact Hurricane Sandy may be all that Obama needs to regain the advantage. I hope not, but that possibility is there.
Still, either way I think the race is going to be much closer than the vast majority of people think, and the reason I think it's going to be close is because both sides have impressive ground games.
In 2008 McCain had no ground game at all. Team Romney has built up a very effective ground game that is nearly the equal of the Obama machine.
There may be some surprises, especially in Wisconsin where Romney was able to exploit the highly effective and well-organized Walker recall election machine.
If Romney supporters maintain their current enthusiasm, that might be just enough to overcome Obama's paid machine.
We will see. The strategies have mostly all played out. Now it's just pure execution of the final tactics.
The storm has created a national political time out.
I don't think anyone can predict how that is going to affect the election.
Governor Rendell used a basketball analogy and said it hurts the team with the most momentum--Romney.
Could be.
In any event elections ride on emotion in the final days. We will see an emotional shift of some sort, but what's the impact.
Drago, there's no perfect method, but neither is it rocket science. As with horshoes and hand grenades, close enough counts. Given the volume of polling this year, I'm comfortable with RCP's methodology -- simple average of all the polls from the last week, excluding multiple polls of the same jurisdiction by the same pollster. The only thing I would change is include all the polls.
I wouldn't be adverse to giving more weight to polls with stronger methodologies -- Gallup should get more weight than the other tracking polls for example because it doesn't use robocalls and has larger samples -- but that puts you on a slippery slope to Nate Silver's approach.
TOMORROW the press will cover POTUS 24/7 on Sandy.
Every-one will love us.
On Thurs/Fri/Sat/Sun: Photos of POTUS!
YES! We check-mated Romney.
The move from FL to DC on Sun. by Stepahnie Cutter was excellent. She will be the next Chief of Staff in 2nd term.
We are doing to bury Romney thanks to Sandy.
Benghazi: Who cares? No one remembers in America.
We win. GOP: Suckers, once and always!
David: "Governor Rendell used a basketball analogy and said it hurts the team with the most momentum--Romney."
What if the team with the most momentum is the team that is also leading?
Then wouldn't the timeout keep the losing side from sparking?
What if the timeout lasted 5 days?
garage mahal said...
Me: In short, everything is pointing to a blowout of major proportions
You: The betting markets aren't looking at Rasmussen/Gallup only.
I know they're not, and I know the betting markets are going the other way. I'm all in for Romney, so we'll see who's right.
Yesterday was the first day in the Rasmussen poll where less than 50% expected Obama to win. Those kind of expectations affect the markets.
shrug. We'll see. But I expect to be celebrating a week from tonight.
AF: "Drago, there's no perfect method, but neither is it rocket science."
LOL
You keep stepping into additional areas of my expertise.
"You seem to want to make the 'truth is in the middle' argument"
Exactly right. When it comes to statistical samples, the "truth is in the middle" approach will get you the most accurate answer more often than not.
Not because of anyone's track record -- because of the laws of statistics.
AF: "I'm comfortable with RCP's methodology -- simple average of all the polls from the last week, excluding multiple polls of the same jurisdiction by the same pollster. The only thing I would change is include all the polls."
Your comfort level is irrelevant.
And when you speak of "all the polls", you are giving more of the game away as it is patently obvious many of the "obama is winning!!!eleventy" polls are D+8/9/10/11/12/15/17!
LOL
It's all part of the axelrod strategy.
I'm not surprised you thinks it's kosher.
AF: "Exactly right. When it comes to statistical samples, the "truth is in the middle" approach will get you the most accurate answer more often than not."
Define: "most accurate".
Take your time.
This should be good.
Incidentally, all of you who are so confident that Romney is more likely than not to win should not be wasting time on this site. Go invest in the betting markets! There are some serious arbitrage opportunities for you out there.
AF: "AF: "Exactly right. When it comes to statistical samples, the "truth is in the middle" approach will get you the most accurate answer more often than not."
So, if I take 4 parts meatloaf, and mix in 1 part dog crap and another part elephant dung, it's still "mostly" meatloaf, right?
"Define: "most accurate"."
Closest to the true population value, which in this case means the voting preferences of the American electorate.
AF: "Incidentally, all of you who are so confident that Romney is more likely than not to win should not be wasting time on this site. Go invest in the betting markets! There are some serious arbitrage opportunities for you out there."
Yes, it would be easier for you if no one from the other side were here.
AF: "Closest to the true population value, which in this case means the voting preferences of the American electorate."
Closest compared to what?
Politico has announced that tomorrow is going to be all day live-streaming of Obama visits to Sandy's trail.
YES! Yes! yes!
We got Romney all boxed in. NYT loves. NPR loves. All voters in America will love us.
We will leave no one behind from Sandy.
We got it right. Thanks to Stephanie was suggesting that we leave FL on Sun for DC over AF-One.
What a genius move! Check-mate to Romney.
Forget you, Benghazi!
"Closest compared to what?"
Compared to other ways of estimating that value, for example, choosing your favorite single poll, or guessing.
FWIW, my take on Romney's "momentum" and the polls is this:
Suppose most people decided a long time ago to base their vote on the economy, somewhat in terms of past performance but mostly in terms of likely future performance. These people can all agree on their basic criteria, but differ in terms of when they think they've got enough data points to be confident enough to make up their minds about how to vote.
In this case, much of Romney's momentum would simply be b/c nobody thinks there's going to be any good economic news b/w now and the election. So a large share of these late deciders become pro-Romney.
This causes "public opinion" as tracked by the polls to appear to be changing, when what's actually happening is that it's converging to what it always would have been if we'd only used vote-forecasting models that were based on economic fundamentals rather than "public opinion" surveys.
In more important than election news, Disney will acquire Lucasfilm and will force a Star Wars 7 on us.
Ugh. With Disney involved, I'm betting we see a doubling down on the Jar Jar Binks character.
AF: "Compared to other ways of estimating that value, for example, choosing your favorite single poll, or guessing."
Examples please.
Michael Moore is correct in his new anti-Romney AD:
We will C***-P***** Romney every-hour. We will use Sandy to C***-P***** Benghazi.
We will C***-P***** all the way to Nov. 6 victory.
What to make of The Oregonian poll which shows that obama only garners 47% support at this time?
The Oregonian buries this number in paragraph 7.
Ali: "We will C***-P***** all the way to Nov. 6 victory."
How sill Cold Pastry help?
Not that I don't like Cold Pastry mind you.
BTW, wasn't Lucas running around last year whining that he was out of the Stars Wars "bidness" since he was getting raked over the coals for the last 3 films?
I think that AF and Drago are talking about two different statistical concepts: efficiency and bias.
AF is right that as sample size increases the distribution converge on the actual sample mean. And Drago's correct that this is only unambiguously valuable if the sample mean is the true population mean. If a sample is biased, then increasing its size simply induces convergence on an incorrect number.
FWIW, ABC/WAPO has Mitt at 49% to the President's 48%, with a D+5 sample.
It is unclear how the Star Wars/Disney deal may affect this. I think both campaigns are worried.
In the swing states that matter, where the campaigns are active, Obama is holding a big lead. Which would seem to indicate a superior ground game. No wonder Republicans want to eliminate early voting.
Do you have numbers and links?
Gallup's party ID poll as >9,000 respondents. That's larger than any of the other polls, isn't it?
AF:
Exactly right. When it comes to statistical samples, the "truth is in the middle" approach will get you the most accurate answer more often than not.
Not because of anyone's track record -- because of the laws of statistics.
Argh. And you accused me of being 25% right. ;)
You're making unsupportable assumptions about your data if this is your argument. I'll leave it at that.
I'll leave you with this - Nate Silver saw the lay of the land in 2010 and choose not to believe it. As pointed out by others, he wasn't particularly accurate that year (+53R when it was +63R). It's not difficult to get a high percentage of house races correct - if you do nothing more than pick 'incumbent' you'll hit around 90% on average.
From 2010, the generic ballot question was polling around +1R, which was pretty much unheard of, so much so that Silver et al discounted it. A simple regression fit of final generic ballot preferences versus final results though the years indicated repubs would pick up 65 seats in 2010 (based on being +1R IIRC). Ten minutes of a back-of-the-envelope analysis did better than Silver.
I suspect 2012 is a repeat. Silver sees the same things the rest of us do, but can't bring himself to properly weigh it in his analysis. You can speculate as to why, I really don't care myself. Like in 2010, he'll move somewhat from his 2008 model, but not enough to capture reality. There are simply too many generalized, large scale data points working against his analysis. Quick example - independents are breaking for Romney by about 12 points. Again, a simple comparison to past elections indicates the repub candidate wins by about 5.5 points on average when leading the indy vote by this amount.
We'll know in a week, but there appear to be far too many indicators pointing to a Romney victory for the RCP and Silver to be correct this year.
Patrick: "It is unclear how the Star Wars/Disney deal may affect this."
LOL
EMD: "Gallup's party ID poll as >9,000 respondents. That's larger than any of the other polls, isn't it?"
Yes.
Chip S.: "AF is right that as sample size increases the distribution converge on the actual sample mean. And Drago's correct that this is only unambiguously valuable if the sample mean is the true population mean. If a sample is biased, then increasing its size simply induces convergence on an incorrect number."
Correct.
My problem with the polls have everything to do with the initial and stubborn "assumptions" made by the major polling outfits.
Further, I postulated that the use of these unrealistic assumptions was something more than just "chance" and in fact were in service to the obama campaign strategy.
That's all.
Dan danoso:
Get thee to a betting market!
Heck, you can bet me. Tell me what you think the odds are that Romney will prevail, and give me those odds.
An interesting question would be: When
Silver is wrong, how often is he wrong by finding a stronger Republican vote? Does he over estimate their chances?
Patrick: "An interesting question would be: When
Silver is wrong, how often is he wrong by finding a stronger Republican vote? Does he over estimate their chances?"
Given that Pinch signs his checks, something tells me that Silver is always looking for the Dems silver lining.
In a move sure to not register on garage at all, now, like Susquehana, Mason-Dixon is pulling out of Florida because their polling shows Romney pulling away there:
http://www.wokv.com/news/news/local/mason-dixon-calling-fl-mitt-romney/nSrLj/
Of course, Talking Points Memo and Pravda aren't covering this so in garage's world it's not really happening.
My more hysterical family members are letting everyone know that a vote for Romney = a vote for George Wallace!! on Facebook right now.
Silver's "secret sauce", in 2008 at least, included having the Obama campaign's internals. Not sure that was ever widely known when he was congratulated for his forecasts in 2008 being so close to the actual outcome.
Most of these state polls that assume turnout for the Dems this year will be equal to or greater than 2008 are smoking something well beyond Maui Wowie.
Silver's "secret sauce", in 2008 at least, included having the Obama campaign's internals.
avwh: Yes. Furhtermore Silver did not disclose this nugget of information about his 2008 work because of a confidentiality agreement with the Obama campaign.
In 2008 Silver had a diary on DKos. He voted for Obama. He now blogs for and is paid by the NY Times, a very liberal organization in the tank for Obama.
These facts do not mean that Silver is intentionally juicing his numbers for Obama, but he has had a secret relationship with the Obama campaign before and he is a Kossack who favors Obama. Who knows.
In his current column Silver has done more than any other single writer to shore up liberal hopes for the Obama campaign after Obama's wipeout in the first debate.
For all these reasons Silver has conscious and unconscious reasons to nudge his numbers in Obama's direction. And at some level his secret sauce has got to be subjective.
I haven't trust his numbers this year. With a Romney win, Silver becomes another tarnished star.
Bed-wetting liberals in California. LOL funny.
When the election is over Nate Silver is gonna have to hide.
From St. Croix's link:
"Kay Edelman says she sometimes finds it hard to exercise as she worries about an Obama loss."
Classic.
a vote for Romney = a vote for George Wallace!!
Are you speaking of George Wallace, Democrat?
Chip S. said...
I think that AF and Drago are talking about two different statistical concepts: efficiency and bias.
AF is right that as sample size increases the distribution converge on the actual sample mean. And Drago's correct that this is only unambiguously valuable if the sample mean is the true population mean. If a sample is biased, then increasing its size simply induces convergence on an incorrect number.
Very nicely put, chip.
the problems interacting with each other. No power means gasoline can't be pumped at the stations, as in much of New Jersey, and it means humongous lines at the few stations that do have power. I know that power is coming back on, much too slowly for people's liking, of course, but it's going to be bad for a while. Tax Preparation NJ
একটি মন্তব্য পোস্ট করুন