[stupendousman]

According to the link below, pollsters are using internal weighting that assumes that the Democrats will get up to 16 percent more voters to the polls than Republicans.

http://campaignspot.nationalreview.c...RmYmE1MzFlMjY=

What's wrong with that? Maybe the fact that even during an election year where Republicans knew they would lose and had a weak candidate (1996) and an analysis over the past 30 years, AT BEST, Democrats have gotten around a 3 point boost as far as party ID has gone. The last presidential election showed even turnout. Again the absolute record turnout when the Dems have done their absolute best is 1/4 of what Gallup is suggesting will be their current turn-out based on weighting. That makes absolutely NO SENSE.

Is it really reasonable to believe that the Democrats will quadruple their advantage this year when they've NEVER CAME EVEN CLOSE? Every recent election shows huge voter ID preference for Democrats that NEVER pans out. If you assume that Democrats will do a great job this year (in a close Presidential race) and get to 4 points, then you've got to assume that the poll outliers need maybe 6 points added to McCain and 6 subtracted from Obama. Again, unless you really think that with this year every trend the last 30 years simply no longer applies (and we hear that the Democrats are going have humungous turn-out every year, even when they don't). Pollsters even overestimated turnout in 1996 so much that some had the ending Presidential gap double what it ended up being. The only way I can see a quadruple change happening is with vote fraud. That doesn't take into account the large numbers of undecided or a possible "Bradley Effect" (as was shown during the primaries) - which no poll accounts for, either.

So...why the obvious phony numbers using weighting? If during the same time period one poll shows a 3 point Obama advantage and another an 11 point advantage as has been the case of late (all using over +3 points for Democrats) you have to assume that some of the polls are getting it MASSIVELY wrong. Margin of error normally is only 3 points.

The question is, why? Do the pollster really think that something that has never happen will happen, even when you can show where when they assumed the same thing in the past it never panned out? Or, are they doing it for another reason?

[Doofy]

Why? Propaganda and tactics.

Most people are so dumb that they'll do vote the way they think everyone else is voting. It's the "sheeple" effect.

[Paco500]

If your figures are correct and you are looking for an actual answer (both big assumptions, I know), a major component of this is the massive disparity in voter registration this cycle between the dems and repubs. The number of new dems being added to the rolls blows the repub effort out of the water.

I assume there will be a rants about voter registration fraud and about how historically, newly registered voters are less likely to actually vote, especially those that are being registered by the dems- the young, minorities, etc. However, if you are looking for a reason, this is likely it.

[Big Mac]

Originally Posted by Doofy 

Why? Propaganda and tactics.

Most people are so dumb that they'll do vote the way they think everyone else is voting. It's the "sheeple" effect.

Yeah, that seems to be the intended effect of these dubiously weighted polls.

[stupendousman]

Originally Posted by Paco500 

If your figures are correct and you are looking for an actual answer (both big assumptions, I know), a major component of this is the massive disparity in voter registration this cycle between the dems and repubs. The number of new dems being added to the rolls blows the repub effort out of the water.

I assume there will be a rants about voter registration fraud and about how historically, newly registered voters are less likely to actually vote, especially those that are being registered by the dems- the young, minorities, etc. However, if you are looking for a reason, this is likely it.

I understand why they claim they are doing it, but can't understand why they would actually do it given the facts. Unless the claim is that Democrats have registered something like 4 times the number of voters they normally do, then the weighing is likely going to be hugely inaccurate in favor of Obama.

My belief? I give the Democrats 4 points. That takes into account that Democrats have registered more voters (as they almost always do) and their base is excited. What takes away from their positives is that the Republicans are also excited and have a reason to go to the polls. Add in a possible "Bradley Effect" and they aren't likely going to do better than their all-time record which is well above what they did the last Presidential election.

What does that mean? On the lower end of the polls, you'd need to knock a point or 2 off Obama's numbers and add them to McCain. The lowest I can find where someone actually figures out the internals to the polls and figures how much they give the Democrats over the Republicans is 5.5. McCain is pulling in +10 independents and both are about the same in regards to their base. That puts the race at the statistical tie it had been at for some time. There really is no rational for taking polls like Gallup (which now uses 2 different formulas, one which boosts Democrats even more than their other) seriously they are such "outliers".

Again, the question remains why they would do something that makes so little sense, especially given past precedent. Someone might suggest (as they have in this thread) it's intentional due to bias. When was the last time a poll ended up finding that it overestimated the amount a Republican presidential candidate was going to get over a Democrat? I can't remember one, but they seem to do the opposite on a continual basis.

[ort888]

I'm sure if you righteous conservatives put your heads together you can figure out this nefarious democratic plot to pull one over on an unsuspecting American public.

[Chongo]

It's an attempt to demoralize McCain voters and hope they won't show up at the poll. (why vote, BO is going to win anyway) One guy I saw on TV said they will tighten up because, in the end, they do not like to be wrong.

[Paco500]

Originally Posted by Chongo 

It's an attempt to demoralize McCain voters and hope they won't show up at the poll. (why vote, BO is going to win anyway) One guy I saw on TV said they will tighten up because, in the end, they do not like to be wrong.

This guy was on TV? Well then it must be true.

[Chongo]

Originally Posted by Paco500 

This guy was on TV? Well then it must be true.

as true as the polls

[zerostar]

We use the new voter registration numbers (which is well below 16% more Dems in FL) as our baseline, we only poll registered voters, right now (Oct 11-13) we polled Obama +12 in Florida

We used 54 counties in our polling, there are I believe, 67 counties in FL.

Right now most polls show Obama +4, Obama +7.2, Obama +5, in Florida

When we base our #'s off of 2004 party turnout we are just above +9

So in out testing, in Florida, other polls seem low for many reasons....


In reality these polls don't matter much, if he is +1 or +2 nationwide if he clinches states that Kerry could not... that is the 50 state plan, that is the hard work Of Howard Dean, you see McCain ready to leave PA, you see him defending ($$$ spent) states that should have been a lock up! That is telling of how great a strategy.

Having worked with the Obama campaign, I will say if this is ANY indication of how things will run if he is President, I am very impressed. The foresight, the planning, the preparedness is above anything I have seen in a campaign, they know this is an uphill battle. A black man, who is being painted as a Muslim, against a White, older, war HERO and POW is going to be tight no matter what and the have shown they are ready for it.

[stupendousman]

Originally Posted by zerostar 

We use the new voter registration numbers (which is above 16% more Dems in FL) as our baseline, we only poll registered voters, right now (Oct 11-13) we polled Obama +12 in Florida

Who are "we"?

Are you saying "we" gives Democrats a 16+ advantage over Republicans?

What exactly was the turnout in Florida last election when Bush won?

[zerostar]

Originally Posted by stupendousman 

Who are "we"?
Are you saying "we" gives Democrats a 16+ advantage over Republicans?

Some people here know, but I would rather not say who I volunteer for... but we are privately run. Take that as you may, I have also volunteer for the Obama campaign, and in the past Clinton and HW Bush a bit.

No, I'm sorry I meant NOT +16%, we are around 4% more Dems which lies within the new voter registrations, as I said when its closer to 2004 turnout we are at +9, +8, depending on pool.

We see a clear lean in undecideds towards Obama, McCain still does lead in many areas as you could guess. (I fixed the word in my post)

What exactly was the turnout in Florida last election when Bush won?

I can find out for you... I don't aggregate the numbers but I do see the sheets and the results... I think it may have been 52 dem - 44 rep - 4 other? I think I remember correctly?

[finboy]

Originally Posted by stupendousman 

So...why the obvious phony numbers using weighting?

Wishful thinking -- pollsters are biased and hope the bias rubs off on undecided sheeps er voters.

Again, pretty sad when folks have to lie and cheat to get votes. What does that say about their legitimate efforts at reform? How flawed must one's ideology be to require a lie to sell it? Rough.

As Cosby would say, somebody's momma has to be pretty low down...

[stupendousman]

Originally Posted by zerostar 

No, I'm sorry I meant NOT +16%, we are around 4% more Dems which lies within the new voter registrations, as I said when its closer to 2004 turnout we are at +9, +8, depending on pool.

So whatever internal polling you do gives Dems an extra 4 points due as part of weighing the sample? That's about what I figured a national sample should be at.

I can find out for you... I don't aggregate the numbers but I do see the sheets and the results... I think it may have been 52 dem - 44 rep - 4 other? I think I remember correctly?

If that's the case, it would be insane to only give the Democrats +4, when the last election cycle actually produced +8 in turnout. Based on what you appare to be saying, it would appear that voter turn-out this year would be worse for Democrats this election than last?

[zerostar]

Originally Posted by stupendousman 

If that's the case, it would be insane to only give the Democrats +4, when the last election cycle actually produced +8 in turnout. Based on what you appare to be saying, it would appear that voter turn-out this year would be worse for Democrats this election than last?

No its still plus 4 for Dem, Rep + Ind (in this state you can register as either of the 3) is 48% so only 4 lean dem. I think the polls are a little low right now actually. How they break is a different matter....

I don't know is that is 100% but its correct AFAIRemember

[stupendousman]

Case in point.

IBD poll has it at a statistical tie.

NYT/CBS has Obama up 14 points.

Either one or both polls are so far off the mark that it can't be a matter of simply not getting the math right. Someone has to be trying to mess up if you're getting a poll wrong by more than double the normal margin of error.

Something is fishy in Pole-mark.

[zerostar]

That poll is +2, and that poll is also the oldest, starting 10/6 to 10/13
Looking here:

http://www.realclearpolitics.com/polls/

You are comparing the LARGEST leading poll to the SMALLEST leading poll.... something is fishy, don't think its the polls.

If what I have seen is correct, it takes 5-6 days to get our polled #'s, we showed +6 about 6 days ago, FOX reported +5 yesterday. Now we are showing +12 hopefully early next week FL will see a few +8 to +9 polls

[stupendousman]

Originally Posted by zerostar 

That poll is +2, and that poll is also the oldest, starting 10/6 to 10/13
Looking here:

http://www.realclearpolitics.com/polls/

You are comparing the LARGEST leading poll to the SMALLEST leading poll.... something is fishy, don't think its the polls.

That's my point. If you look at the spread between the smallest and largest, it is way too large to be a difference that can be chalked up to "margin of error".

You can even take polls where the exact same days where used for polling and see the same crazy gulf between poll numbers. Gallup itself has FOUR DIFFERENT POLLS with 6 points between the smallest and largest.

If what I have seen is correct, it takes 5-6 days to get our polled #'s, we showed +6 about 6 days ago, FOX reported +5 yesterday. Now we are showing +12 hopefully early next week FL will see a few +8 to +9 polls

Again, you can find gulfs 10 points wide between 2 pollsters who polled on the very same days. You can't chalk this up to when the people where polled. It doesn't work that way.

[stupendousman]

Right now, the only clear "outlier" in the bunch is CBS/NYT. I have no idea who does their polling, but it's clear they aren't doing their job - or based on CBS/NYT's pattern of bias - maybe they are. I don't say this just because their results give Obama a lead of about double what everyone else shows, but CBS/NYT has a history of WAY over estimating the Democrats. I believe that they underestimated the 96 election by at least 7 points in Clinton's favor - way too much to be a "margin of error" problem. They were using bogus math.

Most of the mainstream polls have it at between 3-6 points. Some as low as 2, some as high as 8 of late. If you add up all the mainstream polls ending with the sample taken on the 14th, you've got an average of about 5 points. Yet, you still hear members of the media repeating that Obama has a 14 point lead or some kind of insurmountable advantage. I'm pretty sure this isn't an accident or due to ignorance.

The only place I've heard knowledgeable people openly mock the NYT poll was on The O'Reilly Factor last night (and I hate O'Reilly) - and none of them took it seriously. Granted, that's probably a biased source, but you'd think that someone at the NYT would want to do everything it could to ensure it wasn't further screwing up it's reputation. Gallup for whatever it's worth, is involved with 2 different polls but at least when they change their weighing formula to reflect an edge for the Democrats that goes above and beyond any past trends, they also give their traditional analysis.

This with the fact that there's 3-4 points to be had in the undecideds (which have traditionally helped Republicans), and experts have claimed that based on the primary polls Obama needs to be up at least 6 points to get over the already observed "Bradley Effect" - Obama still has a lot of work to do. But that's not what most of the media is reporting. They've already measured the curtains for Obama just as they were doing before the conventions.

Of course, you've got people like James Carville who knows the score already hedging about what will happen if Obama is up more than 5 and still loses (given that it's a highly probably scenario). He saw what happened with Obama scoring big on pre-primary polling only to lose in many cases to Hillary.

Some have suggested that the media, you clearly overwhelmingly support Obama, are doing this to project an air of "inevitability" to his election. The problem is, that's the same strategy Hillary went with and it's also a strategy that can cause your own base to stay home because they are sure you'll win. It doesn't really make any sense, if you ask me, but neither do polls which project the Democrats getting turnout of DOUBLE their best years over the past 30 years. It's not credible.

[ort888]

So your theory is that the entirety of the media is working together to put out misleading polling information in order to sway the election toward their candidate of choice?

That sounds plausible.

[stupendousman]

Originally Posted by ort888 

So your theory is that the entirety of the media is working together to put out misleading polling information in order to sway the election toward their candidate of choice?

That sounds plausible.

It doesn't really work that way with media bias. They don't have to "work together". If you are biased, you'll spin the news in the way that best suits your agenda. It shouldn't be a surprised if they all do it the same way.

If you are a member of the media that wants Obama to win, they aren't going to explain that it's a 4-5 point race right now. You're going to go with the poll that has the biggest lead for your candidate of choice. Otherwise, there'd be no reason to quote such a clear "outlier" poll when trends show a much closer race. Especially when even the polls which show the race the closest are giving the Democrats 2 points on top of it's previous all-time record of 4 points. Last Presidential election, the turn-out was even. Of course, they aren't going to explain that either, because people might be asking why the NYT and other pollsters have given the Democrats so much weight despite those numbers being highly unlikely.

Again, there's no way for the polls that have shown the race as a statistical tie, and those who show something like a 14 point lead can be accurate. One (or both) of the polls are engaging in bad bath. The thing is, this wouldn't be the first time that the NYT had polls which has "bad math" which helped their candidate of choice.

[vmarks]

When the media confirm suspicions repeatedly on both a local and national level that they are clearly biased in favor of all things leftist, why is it not plausible?

National level: having to remove Olbermann from the debates because he's too obviously biased.

Local level: http://gearino.com/?m=200809&paged=14 http://gearino.com/?m=200809&paged=12 http://gearino.com/?m=200809&paged=11 wherein a local editor of the state's paper of record, owned by the McClatchy group, reveals that the paper is partisan.

Two examples. More are abundant, but I figured I'd provide a few.

Note that even when editors and reporters make serious attempts at taming their biases, that it still is apparent through story choice. Unfortunately, serious attempts at taming biases in the newsroom are to be commended, and recorded - they are rarely sighted any longer.

Now, on to why the fantasy polling?

http://www.zombietime.com/lefts_big_blunder/

"Facts and events in and of themselves are no longer important; what's important is how everyone reacts to them. And how do we find out the public's mood concerning this or that incident? Why, the media tells us, that's how. "

[kobi]

You can't look at the national polls, you have to look at the state polls to get a true number as to what candidate is doing what, after it's the electoral college votes that count, the popular doesn't. Just look at the state polls in the link below.

I don't believe the 14 Point CBS poll either, but it's a double digit lead in 7 other national polls, media bias or not McCain is in trouble especially based on the early voting numbers.

I'm not to claim a Obama victory until Nov 5th, having said that both sides need to vote and voice their opinions.

From Fivethirtyeight.com

Perhaps the CBS poll that shows Barack Obama with a 14-point lead among likely voters (12 points when third-party candidates are included) is a modest outlier. But if so, John McCain has more and more outliers that he has to explain away these days. There are now no fewer than seven current national polls that show Obama with a double-digit advantage: Newsweek (+11), ABC/Post (+10), Democracy Corps (+10), Research 2000 (+10), Battleground (+13), Gallup (+10 using their Likely Voter II model) and now this CBS News poll.

Fivethirtyeight.com has the best list of state polls.

http://www.fivethirtyeight.com/searc...ay%27s%20polls

[ort888]

So the next question is how does an Obama presidency benefit the media? Why are they conspiring to get lefties elected? What is their nefarious plot?

Does the end of the world = better ratings and electing Obama will send the country into such a tragic spiral of doom that they will have more news to report and better ratings?

Don't stop with them just manipulating the data. Let's figure out why!

[stupendousman]

Originally Posted by kobi 

You can't look at the national polls, you have to look at the state polls to get a true number as to what candidate is doing what, after it's the electoral college votes that count, the popular doesn't. Just look at the state polls in the link below.

State polls usually follow national polls. When national polls start getting closer, state polls do the same about a week or so later.

Regardless, we are talking about national polls in this thread. I'm sure there are similar crazy things going on in the state polls as well.

I don't believe the 14 Point CBS poll either, but it's a double digit lead in 7 other national polls, media bias or not McCain is in trouble especially based on the early voting numbers.

Which other national polls give Obama a double digit lead? I looked on Realclearpolitics.com and I can't find any. If you ditch the outliers, it's about a 4-5 point race and there are about 4 percent who haven't decided yet. Polls that simply assume that the Democrats will at meet their previous record turnout have it at a dead heat.

Trust me...no one in Obamaville are really all that secure in what's going on with the polls. Otherwise, Carville wouldn't be threatening race wars if they lose.

[Chongo]

Uh oh, Gallup has it with the margin of error.

[ort888]

Originally Posted by Chongo 

Uh oh, Gallup has it with the margin of error.

Well, it shouldn't matter since apparently polls are nothing more then media manipulated propaganda.

You can't quote them when they are in your favor and mock them when they aren't. You can't have it both ways. Either polls are meaningless brainwashing tools or they aren't.

Which one is it?

[macintologist]

Polls are wrong when you don't like the results. I get it now.

[Chongo]

oops, I forgot the sarcasm smiley The only poll that matters is the one on election day.

[stupendousman]

The polls have been giving Obama between 6 and 12 points extra weighting of Democrats, despite the Democrats NEVER getting above 4 points in the last 30 years of measurement of turn-out. Last election saw equal turn-out. If you believe that the Democrats might double their best turn-out, then you have reason to believe the polls. Given that even when Republicans aren't energized and the Democrats do their absolute best, they can only get 4 points up, you can decide for yourself if this is reasonable.

Having said that, despite the unreasonable extra weighting the polls have been giving to benefit Obama, there are a bunch of polls that show the race at a statistical tie. That's REAL bad news for Obama, because that's pretty much the same as him being behind in the polls given what's really going on. It's not that now that it's a statistical tie we believe the polls, but rather that now that it's a statistical tie (despite the still faulty math) it's further evidence that the race isn't as the media and Obama's campaign had previously portrayed it. It's my opinion that if there are polls showing it a statistical tie at election time, barring vote fraud, Obama will lose.

[kobi]

Originally Posted by stupendousman 

Which other national polls give Obama a double digit lead? I looked on Realclearpolitics.com and I can't find any.

Here's a list of the seven polls that show a double digit lead for Obama that I quoted from Fivethirtyeight.com from my earlier post.

From Fivethirtyeight.com

Perhaps the CBS poll that shows Barack Obama with a 14-point lead among likely voters (12 points when third-party candidates are included) is a modest outlier. But if so, John McCain has more and more outliers that he has to explain away these days. There are now no fewer than seven current national polls that show Obama with a double-digit advantage: Newsweek (+11), ABC/Post (+10), Democracy Corps (+10), Research 2000 (+10), Battleground (+13), Gallup (+10 using their Likely Voter II model) and now this CBS News poll.

[stupendousman]

Originally Posted by kobi 

Here's a list of the seven polls that show a double digit lead for Obama that I quoted from Fivethirtyeight.com from my earlier post.

Newsweek (+11):
From about a week ago, a lifetime in an election = not current.

ABC/Post (+10):
From about a week ago, a lifetime in an election = not current.

Democracy Corps (+10):
A poll taken by Democrats. Not even RealClearPolitics uses them anymore.

Research 2000 (+10):
The "Daily Kos" poll. Again..a partisan poll, regardless of the claims of those who do the numbers. Again, other aggregate polling sites don't use Democrat or Republican partisan sources for polls.

Battleground (+13):
The current numbers have them at 6 points

Gallup (+10 using their Likely Voter II model) and now this CBS News poll:
Even with their model that uses the disproportionate number of democrats for turn-out, it's still only 6.

Based on your evidence, the only other polls showing anywhere near the NYT's (and they are still under the NYT's margin of error) are week old polls and ones taken for Liberal/Democrat sources.

So, the NYT's is still way past the "margin of error" on every single one of the major polls.