Please register to participate in our discussions with 2 million other members - it's free and quick! Some forums can only be seen by registered members. After you create your account, you'll be able to customize options and access all our 15,000 new posts/day with fewer ads.
If you read Nate Silvers latest posting, he explains this pretty well. All of the polls are simply a "snap shot" in time. And while they may all be showing Obama with a lead, if one or two is wrong, its likely they are all wrong.
Only about 10% or less of the population has been polled in states like Ohio, others don't answer the phone.
But, this level of polling participation have proven accurate in 2000, 2004, 2008, and 2010. Could they all be wrong this time? Yes, it is a possibility.
And that possibility is the only thing thats holding onto Romney's 16% chance of a win on his index. Mr. Silver is about the numbers, and the odds are clearly in Obama's favor. Its possible they are wrong, and we'll know on Tuesday if thats the case. But they have been very accurate over the last several elections, its very unlikely, about a 16% chance, that they are all wrong.
Pretty good analysis. When it comes down to it, if you have one or two polls showing a candidate with a small lead, its one thing to argue well the other candidate could be ahead. However, when its basically every single poll, the chances of them all being wrong are quite slim. 12 separate pollsters have released polls in Ohio during the last week, one shows a tie, one shows a 1 pt Obama lead, four show a 2 pt Obama lead, two show a 3 point Obama lead, two show a 4 pt Obama lead, one shows a 5 pt Obama lead, and one shows a 6 pt Obama lead.
The RCP average is 2.8, a 2.8 average lead when you are looking at 12 polls is pretty clear evidence that Obama clearly has the upper hand.
yep the one I used showed a lead too. 2 points within the margin of error. You do understand there isn't a poll outside the margin of error. Pretty good for the pollsters. they are right no matter who wins
Margin of error doesn't translate to "could be anywhere within the interval". There's a much higher likelihood that reality is close to the computed result than it being near to the full MOE off in either direction.
Well..by that logic, there are more democratic leaning polls, so therefore they must be right? There are more democratic news agencies, so they must be right too? Come on!
Cant we all just agree that polls are skewed one way or another.....either the turnout favors one party or the other. It should be obvious to us all that we will choose the poll that favors our party. Rassmussen gets 2000 and 2004 right cause it was a R+1 turnout or whatever. Nate Silver gets 2008 right due to D+7 turnout. I remember a saying about "a broken watch is right twice a day"....
I choose to believe my eyes which tell me this election will look nothing like 2008. Of course if Dems somehow turn out with a D+9 advantage, it wont matter much as Romney wont have a chance.
One thing that is benefical, is those people on this forum are all a bit smarter this go around for having particpated in this discussion (especially about polls), and I am thankful for both sides input, even if we dont all agree on politcal issues.
Till tommorrow.....
When it comes down to it, its always better to look at the consensus of all polls. When one poll is showing results different from every other pollster out there, chances are much greater than the one poll is the outlier and wrong, then every other poll being outliers and wrong.
Pretty good analysis. When it comes down to it, if you have one or two polls showing a candidate with a small lead, its one thing to argue well the other candidate could be ahead. However, when its basically every single poll, the chances of them all being wrong are quite slim. 12 separate pollsters have released polls in Ohio during the last week, one shows a tie, one shows a 1 pt Obama lead, four show a 2 pt Obama lead, two show a 3 point Obama lead, two show a 4 pt Obama lead, one shows a 5 pt Obama lead, and one shows a 6 pt Obama lead.
The RCP average is 2.8, a 2.8 average lead when you are looking at 12 polls is pretty clear evidence that Obama clearly has the upper hand.
But don't tell Republicans that, its all about the latest poll, and what that shows.
I've said, from the beginning, you take the high poll, you take the low poll, for both candidates and throw them out. Average out the lead in every other poll left standing, and whatever that average is is most likely the real deal.
But, as I said, they could all be wrong. Very unlikely, but it is a possibility.
Well, if the Philly area if having problems because of the storm...that's even worse news for obama.
The point IS: PA should not be in play, or MI, MN, WI, IA.
Other point - if all these BLUE states are this close, OHIO is already in the RED column.
The Philly area is pretty much fine, it was during the storm and immediately after the storm in which there were some power issues, etc which could make polling quite difficult. Its pretty much back to normal there now.
The consensus of all polls is always a much better metric than just picking out a single poll, that is good for your side. In most of what you mentioned the consensus is quite clear, as is the consensus in Ohio, where in the last 12 polls Obama has an average lead of 2.8. Its kinda hard to win a state when you aren't even leading in any polls...
Also dem registration and their votes are not all for 0bama as many dems in this forum seem to portray from what I can tell.
I was out campaigning last week and met a Democratic Committee member who told me he was voting for Romney. Not because he agreed with Romney, but his exact words were
"Obama isnt the type of Democrat I want to lead my party"..
Please register to post and access all features of our very popular forum. It is free and quick. Over $68,000 in prizes has already been given out to active posters on our forum. Additional giveaways are planned.
Detailed information about all U.S. cities, counties, and zip codes on our site: City-data.com.
Please register to participate in our discussions with 2 million other members - it's free and quick! Some forums can only be seen by registered members. After you create your account, you'll be able to customize options and access all our 15,000 new posts/day with fewer ads.
