Looks like the white vote is coming through for McCain.
PA has gone from O+13 to O+4.
If McCain takes PA, as I think is likely, Obama has lost the election.
It’s going to be the greatest comeback ever. I’m confident.
It’s very scary to read these Daily Kos types. There must be some kind of Internet virus that they’ve caught …
And if “the One” loses, will Wednesday be “Dawn of the Dead?”
Small world department: I actually have met the Daily Kos diarist in question. She’s a lovely woman, and a UCLA alumna, not that the two have much to do with each other. Personally, given my heritage, I would find numbers tattooed on the arm deeply troubling, but I seem to recall her having these cute bracelets that afternoon, so whatever. I think the motivational ink is a little quirky, but she’s not a nut-case.
That said, I’m not sure what polls “Anonymous” is reading. PA is more like a O+8 or 9 state, with Obama already over 50% in many polls (CNN, Quinnipiac, and Survey USA all have him at 53). You can argue over polling and cherry pick the ones you like the best, but honestly, Anonymous, you should be on to bargaining at the very least by now.
Rand, please warn us when you’re linking to sites like Kos. I feel icky just being there.
Jane, see Rasmussen today. They have the best weighting methodology as even Nate Silver agrees. PA has come down from O+13 to O+4 in a few weeks. meanwhile 25% of Democrats in PA are now voting McCain.
You will be very surprised when the PA results come out. Same story with OH and FL. When McCain had these 3 states, it is impossible for Obambi to win – he has to sweep all the toss-ups.
Obama may win the popular vote but he will lose the electoral college. A lot of whites in some parts are not going to be voting for him.
Anonymous, my best guess is that Obama is more likely to win the electoral college by over 100 EV than lose it. I read the Rasmussen poll, but I also read Nate Silver’s site and I find his methodology extremely persuasive. Rasmussen has a noticeable GOP lean of a couple of points. So, citing Silver back at you, take a look at his site, fivethirtyeight.com today, and scroll down to the post “Pennsylvania Sanity Check.” SIlver’s simulations are pretty much in line with the predictive market results, too, for what it’s worth, even though the methodology is utterly different.
Nate Silver is a Democratic Obamabot. I’m not taking anything he says seriously.
So he’s an Obamabot, fine.
But could you point out where his numbers are wrong.
By the way, +4 on Ras with 72 hours to go, is still a win…
Actually, I think Obama CAN lose PA, FL and OH and still win.
I think Daveon’s right on this one. If he takes Colorado, Virginia, and Nevada, he wins, even if Florida, Pennsylvania, and Ohio go the other way.
Hmmmmm.
Frankly I think anybody arguing polls is wasting their time.
The real determining factor in this race is just how much ballot-box stuffing the Democrats can get away with.
The real determining factor in this race is just how much ballot-box stuffing the Democrats can get away with.
And how many votes that Republicans can prevent being cast 🙂
Your electoral system is a delight to watch.
Rationally speaking the polls generally do call things within their MOEs. While Obama is way outside that MOE at the moment I’m still not convinced he’s going to win, but the odds certainly look favourable to him.
In honour of the fact I now have taxation without representation I’m going to throw a tea bag into Elliott Bay on Tuesday. It seems fitting.
Then I’m going to the pub to watch the results come in.
And how many votes that Republicans can prevent being cast
Dude, do you just make stuff up out of your ass? Have you ever heard of an actual Republican attempt to supress a legitimate vote, cast in a reasonable way, following the agreed-upon rules?
Disgusting but strangely appropriate considering how the Kossite proto-/pseudo-“national socialists” are at the front leading the US into the slaughterhouse –and the possibility of such a slaughter is this European’s #1 fear.
Forget civil war and give a thought to the consequences of Obama’s geopolitical polar shift. Change the world indeed. Friends and enemies alike can live with a steady gentle giant sure of its ways but that is not on the table with Obama in charge.
Well, that is not a real tattoo. Looks like either transparent tape or a digital patch.
J.B. – “Personally, given my heritage, I would find numbers tattooed on the arm deeply troubling, but I seem to recall her having these cute bracelets that afternoon, so whatever. I think the motivational ink is a little quirky, but she’s not a nut-case.”
Likely not a nut-case and likely not with a tattoo. That will be for others.
Dating tip: a girl with tattoos has already shown a willingness to live with her mistakes for the rest of her life.
I don’t trust any of the polls this year. Most years, for that matter, but this year is even worse. Consider this bit of logic, and it’s not tortured logic by any means:
1. If polls have about a 3.5% MOE, then there is a 7% spread between their high and low end results. That’s not opinion, just statistics.
2. Polls typically report a 95% confidence interval, which means that asking the same questions of equivalent groups would result in 19 of 20 polls falling within the error range. Again, just statistics.
3. This means that if the MOE is 3.5% and the confidence interval is 95%, then 19 of 20 polls should fall in a 7% range. Still, just statistics.
4. This has not happened, with a wide variety of national polls showing results all along a 15% spread, twice what the MOE says should happen. As a result, the polls have failed a very basic test of statistical validity. (They actually fail several tests of validity and polling standards, but any failure is sufficient to render the result invalid.)
5. Given that the polls are not statistically valid, it is not possible to say that the results can be relied on in any way. In other words, if you have twenty invalid polls, all showing candidate A leading by between 1 and 30 points, then you cannot validly say that candidate A is leading. Once a poll has been shown to be statistically invalid, even the trends that the poll shows over time are suspect, and cannot be relied on.
Worse yet, not only is every national poll invalid in one or more ways this year, but many people are relying on poll aggregates, like fivethirtyeight.com, mentioned above, or Real Clear Politics. Here’s the thing: poll aggregations are less, rather than more, reliable than individual polls, because poll aggregations are not statistically valid even if every poll in the aggregate is statistically valid.
My personal feeling is that Obama will likely win — the Republicans have been in the White House for 8 years, and are unpopular; the Democrats have made a strong case (whether or not you believe them, a lot of people will) that the Republicans are the primary motive force behind the current financial crisis and that their plans to fix it will fail; and Obama is a compelling speaker with the advantage of no record of significant accomplishments to misrepresent, meaning that he has a built-in rhetorical advantage. But I could not reach the conclusion that Obama will win using the polls, because they are a crock this time around, even more than usual.
Bradley effect Jane, Bradley effect.
Not to mention the remark Obama made about Bankrupting the Coal industry will be all over eastern Pennsylvania and Southwestern Virginia by tomorrow evening.
Early voting in Nevada is showing McCain performing beyond where Bush was in 2004.
McCain takes those three and it is all over.
Last 2 presidential elections, as I recall, polls were forecasting a democrat win leading up to the election. G. W. Bush won both elections. Beyond that, the more we see of Obama, the worse he looks. McCain on the other hand, is a know quantity. His negatives were as high as they were going to get months ago.
If Obama wins it will be by fraud, whether by lying about who he is, or the more decidedly illegal variety.
PeterH: To build on what you were saying, the polls reported on election day itself in 2004 were saying “Kerry Landslide!” until about mid-afternoon, based on “exit poll extrapolations” from that morning made by woefully innumerate reporters.
I look forward to the news updates – Does Intrade have a line on if the internet itself will be noticeably slowed by so many people hitting “refresh” in every spare moment at work on Tuesday?
Don’t worry.
Looks like the white vote is coming through for McCain.
PA has gone from O+13 to O+4.
If McCain takes PA, as I think is likely, Obama has lost the election.
It’s going to be the greatest comeback ever. I’m confident.
It’s very scary to read these Daily Kos types. There must be some kind of Internet virus that they’ve caught …
And if “the One” loses, will Wednesday be “Dawn of the Dead?”
Small world department: I actually have met the Daily Kos diarist in question. She’s a lovely woman, and a UCLA alumna, not that the two have much to do with each other. Personally, given my heritage, I would find numbers tattooed on the arm deeply troubling, but I seem to recall her having these cute bracelets that afternoon, so whatever. I think the motivational ink is a little quirky, but she’s not a nut-case.
That said, I’m not sure what polls “Anonymous” is reading. PA is more like a O+8 or 9 state, with Obama already over 50% in many polls (CNN, Quinnipiac, and Survey USA all have him at 53). You can argue over polling and cherry pick the ones you like the best, but honestly, Anonymous, you should be on to bargaining at the very least by now.
Rand, please warn us when you’re linking to sites like Kos. I feel icky just being there.
Jane, see Rasmussen today. They have the best weighting methodology as even Nate Silver agrees. PA has come down from O+13 to O+4 in a few weeks. meanwhile 25% of Democrats in PA are now voting McCain.
You will be very surprised when the PA results come out. Same story with OH and FL. When McCain had these 3 states, it is impossible for Obambi to win – he has to sweep all the toss-ups.
Obama may win the popular vote but he will lose the electoral college. A lot of whites in some parts are not going to be voting for him.
Anonymous, my best guess is that Obama is more likely to win the electoral college by over 100 EV than lose it. I read the Rasmussen poll, but I also read Nate Silver’s site and I find his methodology extremely persuasive. Rasmussen has a noticeable GOP lean of a couple of points. So, citing Silver back at you, take a look at his site, fivethirtyeight.com today, and scroll down to the post “Pennsylvania Sanity Check.” SIlver’s simulations are pretty much in line with the predictive market results, too, for what it’s worth, even though the methodology is utterly different.
Nate Silver is a Democratic Obamabot. I’m not taking anything he says seriously.
So he’s an Obamabot, fine.
But could you point out where his numbers are wrong.
By the way, +4 on Ras with 72 hours to go, is still a win…
Actually, I think Obama CAN lose PA, FL and OH and still win.
I think Daveon’s right on this one. If he takes Colorado, Virginia, and Nevada, he wins, even if Florida, Pennsylvania, and Ohio go the other way.
Hmmmmm.
Frankly I think anybody arguing polls is wasting their time.
The real determining factor in this race is just how much ballot-box stuffing the Democrats can get away with.
The real determining factor in this race is just how much ballot-box stuffing the Democrats can get away with.
And how many votes that Republicans can prevent being cast 🙂
Your electoral system is a delight to watch.
Rationally speaking the polls generally do call things within their MOEs. While Obama is way outside that MOE at the moment I’m still not convinced he’s going to win, but the odds certainly look favourable to him.
In honour of the fact I now have taxation without representation I’m going to throw a tea bag into Elliott Bay on Tuesday. It seems fitting.
Then I’m going to the pub to watch the results come in.
And how many votes that Republicans can prevent being cast
Dude, do you just make stuff up out of your ass? Have you ever heard of an actual Republican attempt to supress a legitimate vote, cast in a reasonable way, following the agreed-upon rules?
Disgusting but strangely appropriate considering how the Kossite proto-/pseudo-“national socialists” are at the front leading the US into the slaughterhouse –and the possibility of such a slaughter is this European’s #1 fear.
Forget civil war and give a thought to the consequences of Obama’s geopolitical polar shift. Change the world indeed. Friends and enemies alike can live with a steady gentle giant sure of its ways but that is not on the table with Obama in charge.
Well, that is not a real tattoo. Looks like either transparent tape or a digital patch.
J.B. – “Personally, given my heritage, I would find numbers tattooed on the arm deeply troubling, but I seem to recall her having these cute bracelets that afternoon, so whatever. I think the motivational ink is a little quirky, but she’s not a nut-case.”
Likely not a nut-case and likely not with a tattoo. That will be for others.
Dating tip: a girl with tattoos has already shown a willingness to live with her mistakes for the rest of her life.
I don’t trust any of the polls this year. Most years, for that matter, but this year is even worse. Consider this bit of logic, and it’s not tortured logic by any means:
1. If polls have about a 3.5% MOE, then there is a 7% spread between their high and low end results. That’s not opinion, just statistics.
2. Polls typically report a 95% confidence interval, which means that asking the same questions of equivalent groups would result in 19 of 20 polls falling within the error range. Again, just statistics.
3. This means that if the MOE is 3.5% and the confidence interval is 95%, then 19 of 20 polls should fall in a 7% range. Still, just statistics.
4. This has not happened, with a wide variety of national polls showing results all along a 15% spread, twice what the MOE says should happen. As a result, the polls have failed a very basic test of statistical validity. (They actually fail several tests of validity and polling standards, but any failure is sufficient to render the result invalid.)
5. Given that the polls are not statistically valid, it is not possible to say that the results can be relied on in any way. In other words, if you have twenty invalid polls, all showing candidate A leading by between 1 and 30 points, then you cannot validly say that candidate A is leading. Once a poll has been shown to be statistically invalid, even the trends that the poll shows over time are suspect, and cannot be relied on.
Worse yet, not only is every national poll invalid in one or more ways this year, but many people are relying on poll aggregates, like fivethirtyeight.com, mentioned above, or Real Clear Politics. Here’s the thing: poll aggregations are less, rather than more, reliable than individual polls, because poll aggregations are not statistically valid even if every poll in the aggregate is statistically valid.
My personal feeling is that Obama will likely win — the Republicans have been in the White House for 8 years, and are unpopular; the Democrats have made a strong case (whether or not you believe them, a lot of people will) that the Republicans are the primary motive force behind the current financial crisis and that their plans to fix it will fail; and Obama is a compelling speaker with the advantage of no record of significant accomplishments to misrepresent, meaning that he has a built-in rhetorical advantage. But I could not reach the conclusion that Obama will win using the polls, because they are a crock this time around, even more than usual.
Bradley effect Jane, Bradley effect.
Not to mention the remark Obama made about Bankrupting the Coal industry will be all over eastern Pennsylvania and Southwestern Virginia by tomorrow evening.
Early voting in Nevada is showing McCain performing beyond where Bush was in 2004.
McCain takes those three and it is all over.
Last 2 presidential elections, as I recall, polls were forecasting a democrat win leading up to the election. G. W. Bush won both elections. Beyond that, the more we see of Obama, the worse he looks. McCain on the other hand, is a know quantity. His negatives were as high as they were going to get months ago.
If Obama wins it will be by fraud, whether by lying about who he is, or the more decidedly illegal variety.
PeterH: To build on what you were saying, the polls reported on election day itself in 2004 were saying “Kerry Landslide!” until about mid-afternoon, based on “exit poll extrapolations” from that morning made by woefully innumerate reporters.
I look forward to the news updates – Does Intrade have a line on if the internet itself will be noticeably slowed by so many people hitting “refresh” in every spare moment at work on Tuesday?