Poll followers with experience?

catsareniice1

Are there any people who have accurately predicted wins in the past on a regular basis?

What are they saying they expect to happen in this election as far as who will take the electoral vote? I don't know who to listen to. 

vlagrl29

I know I've posted about Rassmusen on here before and people said that it's republican biased but they were only a point off back in 2008 and had obama ahead.  right now they have O and R tied, so to me that's the one I really follow

AndrewBreitbart

I read that Pew and Rasmussen were the closest in 2008.

catsareniice1

Thanks!

CoffeeBeen

I saw a similar report - but remember that statistics is a measure of chance, so occasionally pollsters get wonky results - not due to bad methodology, but due to chance.  It's best to go with a poll of polls because that tends to cancel out the bad samples and hone in on the most accurate numbers.

AndrewBreitbart

The problem with the poll of polls is if 90% of the polls are based on a narrative that isn't true... Democrats making up a higher proportion of the electorate and Republicans less than in 2008, despite evidence to the contrary.

For example, the Marist poll had Obama with a big lead in Ohio, as could be expected when the sampling is D +9.

CoffeeBeen

I'm not convinced that 90% of pollsters are out to get conservatives. 

FWIW, a week or two ago I was listening to an NPR interview in which several prominent pollsters asserted that most legitimate polls are not biased by their leaning or "narrative".  I googled the transcript and Whit Ayres (a conservative pollster) said:

"But it's a real mistake to assume that just because a poll is put out by a Democratic pollster or Republican pollster that the numbers are somehow skewed. They are not"

He argued that whether it be good or bad, they're paid to get accurate numbers and that's what the field strives for.  It's their job and reputation on the line.  FTR, the liberal and independent guests agreed with him. 

sweepey

AndrewBreitbart:

I read that Pew and Rasmussen were the closest in 2008.

You were probably reading from the outdated, invalid Fordham "Initial Report." Try the Fordham report that was based on the CORRECT popular vote outcome (FEC verified) and you'll see that eight pollsters were more accurate than Pew and Rasmussen.

http://www.fordham.edu/images/academics/graduate_schools/gsas/elections_and_campaign_/2008 poll accuracy panagopoulos.pdf

JeniLovesNeil

Nate Silver, fivethirtyeight.com, was 100 percent correct in 2008's presidential election.  I know for a fact, as I bet on it and won.