Inaccuracies and errors are part of the nature and essence of election polling, it will be off almost as much as it's on. But that said, we can do a better job reporting results and putting them into context.
Election forecasts report a margin of error of typically 3 percentage points. That means that 95 percent of the time the election outcome should lie within that interval. However, the true error is actually much larger than that.
Democratic pollsters working in top races across the country say they don't expect Obama to run up the score like he did in 2008, but Romney's window is closing quickly. Most I talked to pointed to the upcoming debates as Romney's last chance.
Because they do tons of polls and the same kinds of questions over and over, pollsters have a very good idea in advance of writing their questions for any given poll what phrases or even individual words will trigger a more positive answer.
"I mean, asking survey questions is our bread and butter. I always felt folks enjoyed giving their opinions about all sorts of things. But now, to find out that they hate what we do -- well, as you might think, it's depressing."
Thanks to a hit piece by one of those Beltway pseudo-"bipartisans" we can now state conclusively what many of us have long suspected: Occupy Wall Street speaks for the American majority. We've got the polling numbers to prove it. We now know where the real center lies.
Our country is in the midst of a clash between two competing moral visions, between those who believe in the common good, and those who believe individual good is the only good. It's time our leaders in Washington listen to someone other then themselves.
The White House and Democrats want this election to be a "choice," but midterms are rarely ever "choice" elections -- that's what Presidential elections are for. Instead, midterms are a referendum on the President and the party in power.
Will this be the year that "cell phone only" voters wreak havoc on the results of pre-election polls? And does the cell phone only problem doom pollsters that depend on automated, recorded voice methodologies?
For better or worse, FiveThirtyEight's prominence makes pollster ratings central to our conversation about how to interpret and aggregate polls, and I have some serious concerns about the way these ratings are calculated and presented.