[ALUmnus]

Polling has become a tool to shape public opinion rather than gauge it.

And Newt was right about having a Lincoln-Douglas debate, because that actually would make a difference in public perception as it leaves no place to hide behind moderators, commercial brakes, and applause.

[Purple Haize]

anyone else feel like the obama tv ads make it seem like he hasn't already been president for 4 years?

I agree 100%. It's almost like he is saying that ROMNEY has been leading us these past for years. Very Rules For Radicals of Him!

[adam42381]

rassmeusen is generally deemed the best polling website year after year after year. they have it as a dead heat.

They're currently predicting the Electoral College as follows:

Obama 237
Romney 196
Toss-up 105

Not exactly a dead heat.

[jbock13]

But look at the toss-up states. There's plenty that are "in-play" that Romney will likely win. Like North Carolina. And I don't think Iowa and Wisconsin are out of reach for Romney. But who knows. It's not worth predicting this far out, but I doubt much really changes until we all vote in November.

[Purple Haize]

But look at the toss-up states. There's plenty that are "in-play" that Romney will likely win. Like North Carolina. And I don't think Iowa and Wisconsin are out of reach for Romney. But who knows. It's not worth predicting this far out, but I doubt much really changes until we all vote in November.

I would add Michigan as well

[jbock13]

I'm not one of those who believes that polls are rigged, but the polls in Wisconsin saying Obama is up by 10 make absolutely no sense. In fact it seems all these polls make no sense. But who knows.

[RubberMallet]

rassmeusen is generally deemed the best polling website year after year after year. they have it as a dead heat.

They're currently predicting the Electoral College as follows:

Obama 237
Romney 196
Toss-up 105

Not exactly a dead heat.

thats pretty close.

[flamehunter]

The polls in at least one, if not both Bush elections said he was losing pretty substantially until the election was over. I don't pay attention much to polls any more. You don't know if you can trust them or not.

[adam42381]

The polls in at least one, if not both Bush elections said he was losing pretty substantially until the election was over.

[tinfoilhatresponse]Maybe there's a reason for that.[/tinfoilhatresponse]

[flamehunter]

I'm sure that there was. Pick a side.

[ATrain]

I really don't see North Carolina as a battleground state. That's going to Romney. States that I see that are "in play,":
New Mexico
Colorado
New Hampshire
Wiscosin
Iowa
Virginia
Florida
Ohio

[LUminary]

The polls in at least one, if not both Bush elections said he was losing pretty substantially until the election was over. I don't pay attention much to polls any more. You don't know if you can trust them or not.

That's right. I was trying to remember which, but either the 2000 or '04 polls were way off. I don't care much for polls. As interesting as they might be, they can skew things for sure and possibly alter some elements of an election such as turnout. At least the networks are now supposed to keep the lid on all those exit polls until after actual results are in.

[belcherboy]

I just don't see how Romney wins this election. He has to win just about every battleground state to win the election, from what I've seen and heard. Unfortunately, I just don't think he has enough support to win. The only chance the GOP have of remaining viable is if they can keep control of the House, because (again from what I've seen from polls) the GOP are looking to drop a couple of seats in the Senate.


Outside of an October surprise, I think Obama will get a second term. Even then, they estimate that 40%+ of the ballots will already have been completed in the next month or so. So an October surprise will likely not affect anything.

[jbock13]

I'm not sure where you're getting any of that information from.

[belcherboy]

I'm not sure where you're getting any of that information from.

I read this today:

http://www.cnn.com/2012/09/23/politics/ ... index.html

To win the White House, Romney needs to win all the states that Sen. John McCain carried in 2008, plus grab back about half a dozen that Obama turned from red to blue four years ago.

Romney campaigned Thursday in Florida, where two nonpartisan live operator polls conducted over the past two weeks both indicate Obama has a five-point advantage, which is within the surveys' sampling errors. Both the NBC News/Wall Street Journal/Marist and Fox News polls have the race at 49%-44%. Other, partisan surveys released since the end of the Democratic convention suggest a closer contest. Florida's 29 electoral votes are the biggest catch of the nine or so battleground states that both campaigns are heavily contesting.

In Ohio, an NBC/WSJ/Marist poll and a Fox News survey each have the president holding a seven-point lead, while an American Research Group survey shows Obama with a two-point edge, well within that poll's sampling error.

And in Virginia, a Washington Post poll indicates Obama leading by eight points, while a Fox News survey shows the president up by seven. According to a Quinnipiac University/CBS News/New York Times poll, Obama holds a four-point advantage, which is within that survey's sampling error.

That is just the latest. I've read many other things saying the same the past few weeks. Bottom line is that Romney has to basically win all the battleground states. So even if he manages to turn a few, he would still be short according to most predictions. I know many polls seem to be biased, so maybe I'm just buying into the hype, but I really don't see how Romney wins this election.

Also according to this : http://www.cnn.com/2012/09/21/politics/ ... index.html

Come the first Tuesday in November, when millions are streaming into polling stations across the country, as much as 40% of Americans will have already voted.

In 2004, 22% of Americans voted early and that rate rose to 34% in 2008, according to Paul Gronke, Professor of Political Science who founded and runs the Early Voting Information Center at Reed College in Oregon.

Not only is early voting changing the way Americans cast their ballot but it's also changing the way candidates run their campaigns.

[Purple Haize]

Polling has become the big scandal in this election. Because I'm a nerd I have heard several interviews about what the polling numbers say. To not get too in depth, basically what is happening is that most polling agencies are using a model designed on the 2008 election. The problem is that the 2008 election was truly historic, in terms of turn out and demographics. It was like no other election since polling started. In polls that use a hybrid of the 2004 and 2008 elections, the race is a tie. When you factor in 'undecideds' in the same hybrid model Romney is ahead. Also, Obama has scored higher on his exit polls then he actually scores. So people actually go, vote for someone else but when asked as they leave the polling place say they voted for Obama.
Ok, hopefully that didn't put anyone to sleep.

[jbock13]

What Haize said.

CNN is part of the Democratic narrative to convince us that the election is over. They want us to be discouraged and think our vote doesn't matter. They're doing this, because they know we're ahead.

Perfect example, Wisconsin Governor Recall. All the polls showed it was a dead heat in Wisconsin. Even Fox. On election night, Governor Walker won by 8 points. Had you paid attention to the media, you'd had never known it. CNN had all their pundits ready for this "long night" after the exit polls were released. After it became clear that Democrat challenger Tom Barrett would be blown out, CNN switched to Piers Morgan coverage of the queen about 13 minutes into the contest.

[belcherboy]

I don't know about the CNN polls, but this site is usually the best for talking about all polls. It is BRUTAL for Romney in most polls:

http://realclearpolitics.com/epolls/lat ... president/

Obama leads in North Carolina, Nevada, Iowa, Florida, Michigan, Wisconsin, Pennsylvania, and Minnesota by 5-12 points.

It just isn't even close right now. I just find it hard to believe that these polls are skewed that much. I think it is just wishful thinking to believe that Romney isn't in serious trouble here, and likely has little to no chance of winning in November.

[jbock13]

I'm not one who believes in a polling conspiracy. However, I do believe they widely over sample Democrats. This has been shown to be the case recently in Wisconsin, as well as occurred in 2004 when it seemed W was going to lose in a landside (and won by 5 points).

Obama has no chance in NC. Polls surveying likely voters have shown that. However, the media uses "registered voters", and even worse, "adults", because when you use that pool, the Democrat performs better. Thus, it creates the news that Obama has somehow already won the election.

As for the other states, who knows. I don't know if Mittens wins Wisconsin, but the fact that a Senator and Governor won by 5-8 doesn't give credence that Obama is up by 10. That's not logical.

Here's a fun little site for laughs though.

http://unskewedpolls.com/

[LUminary]

I don't know about the CNN polls, but this site is usually the best for talking about all polls. It is BRUTAL for Romney in most polls:

http://realclearpolitics.com/epolls/lat ... president/

Obama leads in North Carolina, Nevada, Iowa, Florida, Michigan, Wisconsin, Pennsylvania, and Minnesota by 5-12 points.

It just isn't even close right now. I just find it hard to believe that these polls are skewed that much. I think it is just wishful thinking to believe that Romney isn't in serious trouble here, and likely has little to no chance of winning in November.

Wow. The only poll that matters is the one taken election day. These polls now can very much be skewed. It's happened before. A lot of these polls over-sample democrats and, like the Purple guy aptly said, much depends on the model. I'm not saying that Obama won't win, because I don't know. No one does. It will come down to a few key swing states. But I do know it's not over. It might actually be a good sign for Romney that so many people are intently trying to convince everyone that it's over. Things can still happen between now and election day, and I sure wouldn't trust these polls.

[Purple Haize]

I don't know about the CNN polls, but this site is usually the best for talking about all polls. It is BRUTAL for Romney in most polls:

http://realclearpolitics.com/epolls/lat ... president/

Obama leads in North Carolina, Nevada, Iowa, Florida, Michigan, Wisconsin, Pennsylvania, and Minnesota by 5-12 points.

It just isn't even close right now. I just find it hard to believe that these polls are skewed that much. I think it is just wishful thinking to believe that Romney isn't in serious trouble here, and likely has little to no chance of winning in November.

RCP is a metanalysis. And if the input is faulty so will be the output.

[belcherboy]

I don't know about the CNN polls, but this site is usually the best for talking about all polls. It is BRUTAL for Romney in most polls:

http://realclearpolitics.com/epolls/lat ... president/

Obama leads in North Carolina, Nevada, Iowa, Florida, Michigan, Wisconsin, Pennsylvania, and Minnesota by 5-12 points.

It just isn't even close right now. I just find it hard to believe that these polls are skewed that much. I think it is just wishful thinking to believe that Romney isn't in serious trouble here, and likely has little to no chance of winning in November.

RCP is a metanalysis. And if the input is faulty so will be the output.

I agree, but used it so you could look at all the individual polls. It's nice to have one place to see them all, unfortunately just about every one has a big lead for Obama.

[jbock13]

Basically I would tell you don't panic yet.

[Purple Haize]

Belcher- that's sort of the point. The models used by the polls mentioned at RCP are flawed. Just because you say something over and over again doesn't make it true. That's basically what these polls are saying.

[belcherboy]

Belcher- that's sort of the point. The models used by the polls mentioned at RCP are flawed. Just because you say something over and over again doesn't make it true. That's basically what these polls are saying.

I need to find some better news! Anyway, this is another article that makes me worry that we may be seeing November's results right now. The bolded part is what makes really question your logic on RCP polling.

http://www.examiner.com/article/why-rom ... ls-suggest

Why Romney really is losing as badly as the polls suggest

Over the past two weeks President Obama has opened up a wide lead on Mitt Romney in both the national polls and, more importantly, the key state polls across the country. Romney now trails Obama by 3.5 points in the Real Clear Politics average of polls, and trails in nine of the ten swing states. These numbers have caused Democrats to express a great deal of cautious optimism, and some conservatives to bemoan the state of the Romney candidacy.

But alas, many conservatives claim that the polls are all wrong, and that Romney is ahead of Romney despite all the numbers to contrary. Commonly, these critics point to party identification numbers within the polls that show a higher sample of Democrats than Republicans. Some also claim that a high percentage of the undecided vote will break towards Romney, or that some people who currently say they will vote for Obama will change their mind by Election Day. Finally, some simply throw out all the polls which show Obama ahead and keep all the polls that show Romney tied or with small lead.

As discussed in earlier article, there is good reason to be skeptical of polls in general, and especially upon the reliance of one poll from one organization. All polls are ultimately conducted by humans, and humans are certainly capable of error.

Having conceded this truth to the pollster critics, there is still good reason to believe that Obama is in just as good of a position, if not better, than the current polls suggest.

Party Identification

First let us address the party identification “problem” brought up by many critics of the current set of polls. The critics contend that many of the polls sample too many Democrats and too few Republicans. If the two parties were weighted more equally, then the critics contend that Romney would be polling much better in the key states. After a pollster reports there numbers on Twitter these critics will usually quickly ask, “What is the party ID numbers?” If the pollster reports any positive number for Democrats then the poll is usually quickly dismissed as biased and/or innacurate.

The problem with the party identification argument is that it is old as time, and hardly ever correct. Republicans made the same argument about state polls in 2008 that showed President Obama in the lead, and Obama ended up winning in those states just as the pollsters predicted. Democrats claimed that polls unfairly favored Republicans before the 2010 election, but just as the pollsters predicted the Democrats lost a significant number of House and Senate seats in 2010. Party identification changes from election to election, and those changes typically favor the winning party. If pollsters are showing samples with more Democrats than Republicans, it may simply be because Democrats are winning this election.

It is also worth noting that many states are not balanced when it comes to the party identification of the voters. Party identification numbers have trended toward the Democrats, especially in presidential elections. If a pollster reports that their survey showed eight percent more Democrats in Wisconsin than Republicans, it may very well be that there really are more Democrats in Wisconsin than republicans.

All the polls, all the time

One poll might be wrong some of the time, but it is hard to believe that all of the pollsters are wrong all of the time. Polling is both an art and a science, and the pollsters have all found ways to improve their craft over time by using tools such as sampling adjustments. The critics may be right in saying that “this poll” or “that poll” could be wrong, but it is implausible to suggest that all the polls are wrong.

Consider the polls taken before the 2008 election. When averaging the results of all the pollsters, using the Real Clear Politics average, the pollsters correctly predicted the winner in 49 out of 50 states. That is 95 percent success rate. The one state where pollsters got it wrong was North Carolina, where the RCP average gave McCain a 0.4 percentage point lead, but Obama won by 0.4 percentage points instead. Still, the pollsters predicted a very tight race in North Carolina, and that is exactly what happened. In the key swing states, like Florida, the RCP average often came within one or two points of the actual percentages on Election Day.

The RCP averages currently show Obama ahead in nine out of the ten key swing states. In many of the states all of the pollsters have Obama ahead, which means that every pollster would have to wrong in order for the critics to be right.

Even Rasmussen Reports, the pollster commonly favored by conservative critics, has Obama ahead in Florida, Ohio, Virginia, Wisconsin, and Nevada. In order accept the critics’ contention, namely that the pollsters all are wrong and manipulating their numbers to help Obama, one must accept that Scott Rasmussen (a man whose polling organization is infamously known to favor Republican candidates) is purposefully or negligently cooking his numbers to put Obama up in these states.

The Undecided Vote

Finally, poll critics also claim that there is still a massive amount of undecided voters who will turn towards Romney in the final six weeks of the campaign. But there is a problem with this theory as well.

First, in many states President Obama is polling at or above the 50 percent mark. If President is just vote over 50 percent in these states it does not matter if Romney wins 100 percent of the undecided vote – Obama will still win.

Secondly, the critics commonly contend that the undecided will break towards Romney in overwhelming numbers. There is some support for this claim in a study that showed majority of undecided voters breaking towards the challenger instead of the incumbent. However, as documented by Nate Silver, there is much less support for the "incumbent rule" in presidential elections. More specifically, there is no real support for the undecided voters in this election breaking towards Romney. The pollsters that have asked undecided voters to pick between Obama and Romney have generally found the undecided voters split, or favoring Obama slightly.

Finally, polls have generally found that a very small percentage of the population (approximately seven to ten percent) remains undecided in the election. The number of undecided voters expands slightly when including "soft" supporters for each candidate, but including that number assumes Romney "steal" soft Obama supporters while keeping all of his own soft support.

Given the how tiny the undecided voter block is, Romney would have to win an implausible percentage of the undecided vote just to make up a three or four percentage-point deficit on Obama.

For example, if a poll shows President Obama leading 48 percent to 44 percent, this means that Romney would need to win 75 percent of the undecided voters to overcome the current deficit he has among decided voters. This would be especially difficult for Romney since candidates like Gary Bauer will often take one to two percent of the vote in many of these states.

Conclusion

Hope springs eternal, but as President Obama found in his first term hope also must often meet up with the harsh realities of the world we live in. Conservatives certainly have reason to hope that Romney can turn things around with debate or with the hundreds millions left unspent by conservative Super PAC’s. Still, at some point unrealistic hope meets up with the reality. Conservatives can rightly argue that the polls may change between now and the election, but they are dreaming an unrealistic dream if they believe that the polls showing Romney losing today are all wrong.