Obama Widens Lead (10-6-08)

[MannyIsGod]

Poll interviews were conducted with 524 adult Americans who watched the debate and were conducted by telephone on September 26. All interviews were done after the end of the debate. The margin of error for the survey is plus or minus 4.5 percentage points.


The results may be favoring Obama simply because more Democrats than Republicans tuned in to the debate <more precisely, because more dems than reps were questioned as to who won the debate>. Of the debate-watchers questioned in this poll, 41 percent of the respondents identified themselves as Democrats, 27 percent as Republicans and 30 percent as independents.


http://www.cnn.com/2008/POLITICS/09/27/debate.poll/

Here's a question for you, are there moe Dems nationide than Republicans?

[Wild Cobra]

Here's a question for you, are there moe Dems nationide than Republicans?

In my experience, thay are both somewhere between 40% to 45%. I cannot speak for everywhere, but among the people I know, republicans refuse to participate in polling more than democrats. I am a conservative liberatarian that refuses to do polls when asked. This does skew the results, and there is no accurate way to adjust for it. If we simply normalize to an approximate 45% democrat, 45% republican, 10% other, the rebublican vote often exceeds the democrat vote. I showed how a poll in a pst thread recently, that went for Obama, actually favored McCain if you did such adjustments.

[dg7md]

Don't forget Al Gore. Wasn't he up by 11 points on this day 8 years ago?

no clue what you're talking about.

Plus, that wasn't exactly a case where Bush was ELECTED over Gore.

[Nbadan]

I see what you were referring too though. OH for instance went from 49-43% Kerry on the 24th, to 47-42% Bush on the 25th...

No, it didn't

Was the 2004 Election Stolen?

Republicans prevented more than 350,000 voters in Ohio from casting ballots or having their votes counted -- enough to have put John Kerry in the White House.

Rolling Stone

What's ironic is wing-nut pundits cry about voter fraud in this years general election thanks to a new Democratic Sec. of State in Ohio, but what I wanna know is where were these 'preservers of democracy' in 04 when the GOP was purging Democratic voter rolls and e-voting machines were counting negative votes for Kerry? Not so fun when the shoe is on the other foot is it?

[jochhejaam]

Here's a question for you, are there moe Dems nationide than Republicans?

Back to my point, if you poll 41% Dems and 27% Reps, the outcome should favor the Democrats.


Now, for your question; 41% Dems and 27% Reps comes out to 61 percent of the 2 major Party's voters being registered as Dems.

Question for you; Do you believe that to be an accurate percentage?

[Findog]

Don't forget Al Gore. Wasn't he up by 11 points on this day 8 years ago?

He was up by 8 points in a single Gallup poll at this time. The other polls at the time showed a much tighter race. Gore's lead in the polls evaporated after the debates. Obama's polling numbers have 2 debates baked in them.

First Gore Debate: Sigh, Roll Eyes, Condescend to Bush
Second Gore Debate: Overcompensate by saying "I agree with the Governor" 80 times
Third Gore Debate: Weirdly invade Bush's personal space

[Buddy Holly]

He was up by 8 points in a single Gallup poll at this time. The other polls at the time showed a much tighter race. Gore's lead in the polls evaporated after the debates. Obama's polling numbers have 2 debates baked in them.

Gore was down 8 points in the Gallup poll.

[Findog]

Gore was down 8 points in the Gallup poll.

There was a single poll where he was ahead by 8 at the beginning of October, so thanks for your correction. I do know that the polling reflected a very close race in the final month. Right now the polls are suggesting a solid Obama victory in the popular vote and an electoral college landslide. I think the point stands that no candidate has ever really made that up at this late a date.

Truman came from behind to beat Dewey, but polling methods were much less sophisticated back then, and Dewey's victory seemed like such a foregone conclusion that most polling stopped well before the election. Truman started his surge well before October. Humphrey quickly made up ground on Nixon once he broke with LBJ on Vietnam. Ford at one point was down 33 points to Carter in the summer but he steadily cut into that lead from that point forward.

If McCain is going to win this election, he needs something to dramatically alter the landscape.

[TomBrady]

A single by itself is USELESS. Look at the aggregate of polling done over the past month and you can clearly see a positive trend towards Obama. He has leads outside the MOE in nearly every major national poll and in just about every key battleground state. He is challenging in IN which has not gone Democratics since 1964.

This race is not tied. If it was McCain would not have to resort to the Ayers attack. He is clearly behind and if you can't accept that you're in for a big disappointment Nov. 4th.

[CubanMustGo]

A single by itself is USELESS. Look at the aggregate of polling done over the past month and you can clearly see a positive trend towards Obama. He has leads outside the MOE in nearly every major national poll and in just about every key battleground state. He is challenging in IN which has not gone Democratics since 1964.

This race is not tied. If it was McCain would not have to resort to the Ayers attack. He is clearly behind and if you can't accept that you're in for a big disappointment Nov. 4th.

Neocon definition of tie:

[ElNono]

Neocon definition of tie:

You did not tag NY as red...

[MannyIsGod]

In my experience, thay are both somewhere between 40% to 45%. I cannot speak for everywhere, but among the people I know, republicans refuse to participate in polling more than democrats. I am a conservative liberatarian that refuses to do polls when asked. This does skew the results, and there is no accurate way to adjust for it. If we simply normalize to an approximate 45% democrat, 45% republican, 10% other, the rebublican vote often exceeds the democrat vote. I showed how a poll in a pst thread recently, that went for Obama, actually favored McCain if you did such adjustments.

In your experience.

This is something that is easily quantified. Here's a hint, its no where near the 40% mark for Republicans.

[MannyIsGod]

Back to my point, if you poll 41% Dems and 27% Reps, the outcome should favor the Democrats.


Now, for your question; 41% Dems and 27% Reps comes out to 61 percent of the 2 major Party's voters being registered as Dems.

Question for you; Do you believe that to be an accurate percentage?

Why can't you simply answer the question? Its obvious that if 41% of the polled are Democrats they will likely come out ahead. Thats why its important to ask whether or not 41% of the people registered in this country are actually Democrats or not.

Its not about a figure gives a party an advantadge, its about whether or not that figure is correct.

So I ask you once again, What is the party breakdown in the United States? If you are assuming what is shown is incorrect, what is correct?

[Yonivore]

A blog I steal from regularly had this to say about the polls:

Polling in the Presidential race is getting stranger. The latest Reuters/CSpan/Zogby poll has Obama up by only two points. Rasmussen shows the race tightening from eight points to six. At the same time, the Gallup Poll has Obama jumping out to an unprecedented eleven-point margin.

I'm pretty sure the Gallup result is an outlier, but beyond that it's hard to say what is happening. Either there is a tremendous amount of volatility in the electorate, or the pollsters are using pretty drastically different assumptions. At the moment, it is hard to be sure which way the tide is flowing.

the polls, I think they're useless...and have been for some time.

[MannyIsGod]

In 2004 this was the breakdown

29% Republicans
33% Democrats

Back to my point, if you poll 41% Dems and 27% Reps, the outcome should favor the Democrats.


Now, for your question; 41% Dems and 27% Reps comes out to 61 percent of the 2 major Party's voters being registered as Dems.

Question for you; Do you believe that to be an accurate percentage?

Now, using your own methods, in 2004 the percentage was 53 percent. You think its somehow inconceivable that changed by 8% in those four years? Its not nearly as outlandish as you try to make it sound.

[MannyIsGod]

And more recent info



http://pewresearch.org/pubs/773/fewe...as-republicans

[TomBrady]

When you look at individual polls and see them all over the map that doesn't mean that polling is inaccurate.

Let me give you an example. Say there are 10 people in a group. I want to conduct a poll on their opinion of abortion. I don't have the resources to poll all 10 people so I will only poll 5 of them. Now in this group, 7 people are PRO abortion and 3 people are ANTI abortion.

In my first poll, in which I poll 5 people, let's say I end up calling the 3 who are ANTI abortion and only 2 that are PRO abortion. From this sample it would appear that 60% of the people are ANTI abortion and 40% are PRO abortion. Doesn't make sense with what we know right?

Now, let's say I repeat this poll many many times over the course of months and years. Probability will eventually win out and the polls will reflect that indeed more people are PRO abortion than ANTI abortion. That's why I highly recommend you go to www.fivethirtyeight.com. They have statistical models that help make sense of all this polling data.

Polling isn't an exact science by any means, but it CAN give you a good idea of where the race stands if there is enough data.

[MannyIsGod]

When you look at individual polls and see them all over the map that doesn't mean that polling is inaccurate.

Let me give you an example. Say there are 10 people in a group. I want to conduct a poll on their opinion of abortion. I don't have the resources to poll all 10 people so I will only poll 5 of them. Now in this group, 7 people are PRO abortion and 3 people are ANTI abortion.

In my first poll, in which I poll 5 people, let's say I end up calling the 3 who are ANTI abortion and only 2 that are PRO abortion. From this sample it would appear that 60% of the people are ANTI abortion and 40% are PRO abortion. Doesn't make sense with what we know right?

Now, let's say I repeat this poll many many times over the course of months and years. Probability will eventually win out and the polls will reflect that indeed more people are PRO abortion than ANTI abortion. That's why I highly recommend you go to www.fivethirtyeight.com. They have statistical models that help make sense of all this polling data.

Polling isn't an exact science by any means, but it CAN give you a good idea of where the race stands if there is enough data.

I've tried explaining it to them time after time. They don't want to listen to anything about trendlines and averaging the polls. They want to focus on the individual numbers and find ways to discredit an Obama lead. Its not like its going to do them any good because all the discrediting in the world won't make McCain win.

[jochhejaam]

In 2004 this was the breakdown

29% Republicans
33% Democrats



Now, using your own methods, in 2004 the percentage was 53 percent. You think its somehow inconceivable that changed by 8% in those four years? Its not nearly as outlandish as you try to make it sound.

Merely provide us with the info that shows where 61% of the registered Major Party voters are Democrats and I'll agree that the poll wasn't contoured to favor the Dems. (I don't think that's an outlandish request)
You're giving us a lot of lip service, and a bunch of numbers, but without that 61% stat, you've submitted nothing that substantiates that the poll was fairly conducted (and don't cite just one source for the percentage, to be fair, cite several credible sources in order to provide us with the more balanced aggregate number).


And in regards to polls, I believe that everyone knows that an aggregate of credible polls is a more accurate barometer of results than a single poll.

With that in mind, why didn't you get all up in arms with the author of this thread who submitted an opinion (Obama widens lead) based on a single poll? <---rhetorical

[MannyIsGod]

The graphic I provided is up to 58% using your contorted stat. I'm not sure why you feel the need to manipulate the numbers into a different stat - oh wait yes I am - Intelltecual dishonesty.

I'm not going to run around the internet looking for statistics to show what every objective observer of the current political climate has already acknowledged. If you feel the numbers are incorrect you're free to do so. You've obviously done no research into it, you've just assumed it was incorrect. I've provided you with some statistics on the matter and if thats not enough then thats fine but I've no interest in finding anymore. I'm not the one assuming here.

Anyway, if you think the poll is off based on that difference, simply weight it to the party identification numbers that I posted above and tell me how much McCain won the debate by. Of course it will be a negative number but thats besides the point, isn't it?

[jochhejaam]

The graphic I provided is up to 58% using your contorted stat. I'm not sure why you feel the need to manipulate the numbers into a different stat - oh wait yes I am - Intelltecual dishonesty.

The 41% dems and 27% reps polled are from CNN, so if it's a distortion it's on CNN.

A charge of intellectual dishonesty is code for "I can't back up my premise with stats". That's being intellecutally dishonest.



I've spent more time in this thread than it deserves, I'm calling it a day here.

[MannyIsGod]

A charge of intellectual dishonesty is code for "I can't back up my premise with stats". That's being intellecutally dishonest.

This is exactly my point. So when will you be backing up your premise that CNN is incorrect with stats?

[jochhejaam]

This is exactly my point. CNN's polling of 41% D's and 27% R's comes out to 61% of the registered Major Party voters being D's, and that's a stat I am unable to back up.

Alright then, we'll consider that the last word.

[MannyIsGod]

LOL I did notice you ignored my point that even if you throw out those figures and use figures you're comfortable with and weight the poll Obama still wins by a large margin thereby rendering that statistic irrelevant.

You would never knowingly ignore a point though. Never.

[fyatuk]

And more recent info



http://pewresearch.org/pubs/773/fewe...as-republicans

That's a pretty interesting graph. Really bears out the historical claim of 33/33/33% roughly. There's a current adjustment going underway due to the backlash from popular opinion of the last 8 years of neocon dominance, but it'll likely go back to the historical split over the next few years once the Bush administration fades into the background. I mean, it's not likely those missing percentage of Republicans all changed their idealogies, just stopped identifying themselves out of disgust with party leadership.

Not soon enough to help McCain. The economic crises could not have happened at a worse time for McCain, especially after admitting he wasn't solid on the economy.