Posted on 11/05/2010 10:45:22 AM PDT by WebFocus
Every election cycle has its winners and losers: not just the among the candidates, but also the pollsters.
On Tuesday, polls conducted by the firm Rasmussen Reports — which released more than 100 surveys in the final three weeks of the campaign, including some commissioned under a subsidiary on behalf of Fox News — badly missed the margin in many states, and also exhibited a considerable bias toward Republican candidates.
Other polling firms, like SurveyUSA and Quinnipiac University, produced more reliable results in Senate and gubernatorial races. A firm that conducts surveys by Internet, YouGov, also performed relatively well.
What follows is a preliminary analysis of polls released to the public in the final 21 days of the campaign. Our process here is quite simple: we’ve taken all such polls in our database, and assessed how accurate they were, on average, in predicting the margin separating the two leading candidates in each race. For instance, a poll that had the Democrat winning by 2 percentage points in a race where the Republican actually won by 4 would have an error of 6 points.
We’ve also assessed whether a company’s polls consistently missed in either a Democratic or Republican direction — that is, whether they were biased. The hypothetical poll I just described would have had a 6 point Democratic bias, for instance.
The analysis covers all polls issued by firms in the final three weeks of the campaign, even if a company surveyed a particular state multiple times. In our view, this provides for a more comprehensive analysis than focusing solely on a firm’s final poll in each state, since polling has a tendency to converge in the final days of the campaign, perhaps because some firms fear that their results are an outlier and adjust them accordingly.
(Excerpt) Read more at fivethirtyeight.blogs.nytimes.com ...
Not trying to be a smartass, but did Rasmussen cost you a lot of money, or did you cost you a lot of money? Granted, it's possible that Scott Rasmussen held a gun to your head and forced you gamble. Just saying.
Where do you think people who bet on INTRADE get their information?
I’d like to see a scientific analysis of the polling this time around. Rasmussen has received a lot of praise from Fox News for having nailed the 2008 election, but they booted some of these races, badly. Whenever the polls are so wildly different than the actual outcome our first inclination is to cry “vote fraud!” but the simpler explanation is that the polls were simply wrong, and we allowed ourselves to believe them.
The lesson is not to trust the “likely voter” models if they are off 10 points from the registered voters.
Kind of suspicious that the New York Times/CBS News polls aren’t listed for accuracy.
Hmmm. I wonder why.
Can you folks at FR please stop screening all of my posts?
Union/urban voter fraud/intimidation of 3-4% for dems should be factored into any polling data to obtain any amount of accuracy.
This.
Silver is a leftist with an agenda. Rasmussen's polls were fairly consistent with most public polling, including those of SurveyUSA, who Silver praises. Accusations of bias are ridiculous and his method of poll evaluation is poor.
Polls are a snapshot of public opinion at a point in time for the population surveyed, and should be evaluated as such.
We lost NV 100% because Angle was the wrong choice and then ran the completely WRONG campaign. She did not nationalize that election. She made it about her and Reid. If it was all about the national scene or all about Reid we win. Yet, she made it about her Vs him. We lost. Stupidly so.
DE. We lost because O'Donnell was a terrible candidate from the word go. She is just immature first and foremost. Plus it is a blue state.
WV. This one does surprise me. The silly "hick" ad the RNC tried to help out with I do believe is what cost us here though...
Um, I don’t think this shows a bias in polling. I really don’t.
I think it shows the measure of voting fraud perpetrated by the Dems. No joke...
The professional polling crowd has always had an axe to grind with Rasmussen. His method are not conventional, and therefore considered “unproven”. Most years he does quite well, but this was not one of those years. This will bring out his critics in droves. All but one of the studied polls had a “R” bias according to the NYT. This is understandable as all of the buzz was in the Republican direction this year. They didn’t perform as well as expected in Democrat strongholds probably because the dead are seldom polled, and the polls weed out unregistered voters who also tend to show up on election day or vote absentee.
Rasmussen needs to continue tweaking his methodology, but one bad year is nothing to get too worked up about.
“People who look to Intrade for insight are idiotic.”
Evidence? Even skeptics of political markets concede that the predictive power of sites such as Intrade GENERALLY is superior to reliance on political polls:
“Indeed, three days out of four, a poll will be less accurate than the vote-share market price at predicting the election outcome. Someone who played the vote-share market based on the expectation that the division in the latest polls would translate one-to-one into the final vote division would lose decisively in the long run.”
http://www.edwardtufte.com/files/MarketsandPolls.pdf
Good call. The simplest explanation is ALMOST ALWAYS the correct explanation.
I took a look at Rasmussen’s polls. By using 21 days, Silver got to throw in a couple of that were way off, but that Ras (and others) had fixed as election day neared.
Why didn’t Silver go back 4 weeks or 2 weeks. Because it would not have helped his already-made-up thesis.
This is a bogus hit piece.
If it’s a hit piece, it is definitely backed up by the election results.
Rasmussen overestimated the Republican performance in just about every race.
“Rasmussen overestimated the Republican performance in just about every race.”
Wrong...Rasmussen UNDERESTIMATED democrat fraud.
Send me $75.00 and I’ll remove the screen. I hear if you donate to FR fundraising thread the screening stops quicker.
My first thought when I saw the header, like so many above, was the Rasmussen probably has no fraud factor.
It's not predictive. It's reflective. Intrade simply reflects conventional wisdom as published by sites such as Cook Political Report or RCP. It's thus a lagging indicator of external data. Except that it's easily subject to market manipulation.
In summary, it's crap.
“It’s not predictive. It’s reflective.”
Yet it PREDICTS outcomes better than polls do 3 out of 4 times. As the paper suggests, polls are subject to a lot of volatility based on breaking news, hence they won’t necessarily be an accurate indicator of what voters actually will do a few days later in the polls. The prediction markets, OTOH, can take such volatility into account based on past experience and do a better job of predicting a future vote (as opposed to predicting how people would vote on the day a Gallup poll is conducted etc.).
If you know of a superior prediction tool, please advise. But until you weigh in, I’ll regard political markets as superior to reliance on polls: they may not be perfect, but neither are they crap, IMHO.
Yeah, that was real inaccurate polling. /s
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.