Sunday, October 21, 2012

NBC/WSJ Poll - October 21, 2012

Likely Voter, 816 sample size, Tied 47-47, 4% undecided.

The release of this poll was greeted with media cries of amazement, "Look it's tied!".  The only problem is that it isn't.  This poll continues to show Romney's polling lead.  The DRI of this poll is 44/38/18, of D+6.  They don't provide any information regarding independent preference between Obama and Romney.

When we adjust this result to the different models, we get the following results:

Even - Current result
R+1.4 - 2008 turnout
R+3.8 - D+3 turnout
R+5.7 - 2010 turnout
R+5.8 - 2004 turnout
R+7.1 - Rasmussen Party ID

Notice that this poll confirms the Gallup results.  Reportedly, Gallup is using an R+1 sample.  In the 2010 model, which is close to R+1, this poll would give Romney 52%, exactly the same as Gallup.

There is a lot of foolish criticism being leveled at Gallup, calling it an outlier.  People who should know better, like Rove and Cadell are saying this.  What they fail to see (or report) is that most of these polls are showing the Romney lead at about 52%, if they also used an R+1 sample.

It is going to be very interesting to see if all of these pollsters are going to ride their D+7 samples all the way to election day, or if they will finally come back to reality.


  1. It looks like the polls are for the most part are all saying the same thing, the "x" factor being what the turnout is going to look like.

    I tend to think it's going to look close to Gallup's model, which supposedly is R+1. What scares be though is Rasmussen.

    With nearly every poll saying Romney is winning independents by eye-popping margins, I have a hard time believing Obama pulls through. in fact, it's looking to actually be something of a Romney landslide.

  2. "It looks like the polls are for the most part are all saying the same thing, the "x" factor being what the turnout is going to look like."

    Exactly. These polls are all valid in that their samplings are all producing good insights into the underlying voter population. Each samples a little differently, each samples a little different region of the population space - which is what we want! - but they are producing an output which once you filter out the random noise by averaging is quite consistent, internally with each-other and with ancillary data, such as registration, fundraising, etc. There is solid information embedded in here.

    The difference between the polls, sans that wacked out IBD/TIPP one, isn't in the sampling stage, it's in the weighing as you said toward a projected voter turnout model.

    What I don't understand for the life of me is why any pollster could still be pushing a 2008 or greater democratic turnout model. You're suppose to update your priors and assumptions as new knowledge is integrated:

    We have the registration data, it's just not feasible.

  3. Sort of on-topic, but mostly tangential: I've been wondering what in the world Nate Silver is doing. I've never found him that exceptional, but you just don't go full retard - everyone knows that.

    Trying to think through and simulate his position, my best guess given his ideology and past work for the Obama campaign -- he's likely seen some of their internals and knows the state of the race.

    That said, his stupid 108.745% probability of Obama winning -- which at it's core is nothing but a silly appeal to higher authority wrapped in a reification fallacy where the model is being confused with objective reality -- is just ridiculous. Really, you're reporting a decimal point change in 'probability' nightly? Really now? Journalists are pseudo-illiterates who because they never had to take a math course, or a physics course, or a systems dynamics course, etc are amazed by the numbers and graphics. No-one questions whats in his black box, nobody cares as long as it's shiny.

    The question to ask yourself is: if Silver's model was adjusted to output the results of even a D+3 election, what would be the psychological impact on this race? Among the liberal chattering class? ...

    My guess is he sees the dynamics have shifted but is doing his part to prevent a total collapse to Obama's true ground state. He's too invested in himself to fall for the man though; he'll give O until tomorrow night's debate's aftermath to score the dramatic event necessary to upset the race's dynamics. If nothing or Romney rallies in the format, he'll point to the debate and Libya for cover of why the model broke and there's basically a discontinuity -- but that fucker will hand-tune it.

    There's no way he'll have any respect in the field, less even the NYTimes, if Obama loses after the shit he's pulled; and he's not coming back to Chicago to count baseball cards. We do this for fun, but have real jobs....

    1. Nate Silver is done, he was always just a one-hit wonder anyway.

      Basically, in 2008 the Obama campaign knew he was a fellow "comrade", opened up their polling data that they had spent millions on, and Nate just adjusted everything accordingly.

      Now that he has this undeserved "guru" status, he's decided to play to the Left "cheap seats" because that's his niche that got him to where he is today with the NYT. He parrots what Liberals want to hear.

      But he's painted himself in a corner, if he says Romney is going to win, the gnashing of teeth on the Left will be deafening and will be deemed a traitor, if he goes right off a cliff with his ridiculous models that Obama will win, he'll be a humiliated and quickly forgotten blogger after this election.