John Dickerson @jdickerson
Kid just looked at our bowl of candy and said the sample size was too small. #dc #unskew

The Fox News poll released today continues the trend of skewed polls that over-sample Democratic voters to produce results favorable for the president. The poll reports President Obama tied with Mitt Romney at 46 percent. But this is based on a sample that includes five percent more Democrats than it does Republicans. This five percent over-sampling of Democrats shows the race tied in survey data that should show a five percent lead for Mitt Romney.
...
When the data from the Fox News poll is unskewed by weighting their reported percentages between Romney and Obama to the partisan affiliations data from the latest QStarNews poll, the expected partisan make up of the electorate expected for this year's election is 34.8 percent Republicans, 35.2 percent Democrats and 30.0 independents. Unskewing by those ratios, the results calculate to Romney leading 49 percent to 44 percent over Obama. A large majority of the undecided voters, who usually break for the challenger in a presidential race involving the incumbent president, can be expected to swing toward Romney and lead to a likely Romney majority on election day according to this polling data. If 75 percent of those undecided voters break for Romney by election, the voting patterns indicated by this poll would be consistent with a 54 percent to 46 percent slim majority for Mitt Romney.

jerseyhoya wrote:

phdave wrote:

jerseyhoya wrote:RCP R+0.9
TPM R+1.0
Pollster R+0.3

Nate O+1.7

Makes a big point about how 2 point leads tend to hold up (re: Ohio) as his model is 2%+ off from all the big national polling aggregators.

New averages from today (and change since Friday):

Nate O+1.9 (O+.2)
RCP tied (O+.9)
TPM O+1.3 (O+2.3)
Pollster R+.8 (R+.5)

Nate with the most stable numbers. I'm not sure what's up with Pollster. Don't they know that they should be leaning towards Obama to help set the narrative? I wonder if Republicans are going to start making the Huffington Post their polling aggregator of choice now.

I just noticed that they don't seem to be including RAND. What's wrong with RAND? The RAND poll is cool (in my unbiased opinion) because it is a panel of people who are interviewed repeatedly and rate their likelihood of voting for a candidate as well as their likelihood of voting in general.

Panels are really interesting for seeing how things change, but if you start with a bad sample you're stuck with it. Seems like it's been consistently more pro-Obama than just about any other national poll.

phdave wrote:

pacino wrote:were the polls randomly not skewed after the first debate?

I have been predicting for this entire year, since he won the nomination, that Mitt Romney is going to defeat President Obama in the election this year. Currently I project Romney winning 54 percent of the popular vote and as many as 359 electoral votes. Even when all those polls shows Obama holding big leads I said Romney was really winning and the polls were skewed. Then soon after I illustrated exactly how skewed the polls were and the polls suddenly began to be less skewed, more accurate and started showing something closer to the real Romney lead in the presidential race.

td11 wrote:

i think this guy also made a dunham response video

mozartpc27 wrote:Nate's projected numbers now "hidden" by a "Subscribe to the Times" ad that I assume goes away if you do so.

Fuck you, NYT.

mozartpc27 wrote:Nate's projected numbers now "hidden" by a "Subscribe to the Times" ad that I assume goes away if you do so.

Fuck you, NYT.

Silver’s smug defenders assume that it is his detractors who can’t grasp the concept — but Silver’s critics know that the general public doesn’t understand the nuances at play here. And that’s part of problem.

Silver comes out of the baseball statistics world, and his defenders like cite sports and gambling analogies when defending him. But there is a key difference. If Silver says the Giants have only a 5 percent chance of winning the World Series again next year, it is highly unlikely that would impact the outcome of games. Umpires won’t begin making bad calls, the fans won’t stop attending games, etc.

But when the public sees that a prominent New York Times writer gives Barack Obama a 70 percent chance of winning, that can become a sort of self-fulfilling prophesy. It has consequences. It drives media coverage. It dries up donations. Whether Silver likes it, or not, people do interpret his numbers as a “prediction.” They see this as election forecasting.

jerseyhoya wrote:

Drudge hyping a sex scandal

Apparently it's about BOB MENENDEZ