The Polling Report [11.4.12]

Filed in National by on November 4, 2012

Team Romney is already offering excuses for their upcoming loss on Tuesday. And some campaign insiders are in full CYA mode in speaking on the record to Politico saying that Romney’s first choice for Vice President was Chris Christie, rather than Paul Ryan. I suppose they could also be trying to latch onto Christie’s recent bipartisan and excellent performance during the aftermath of Hurricane Sandy, but instead it looks like a tell tale sign of campaign already in the post mortem mode.

Speaking of Sandy:

Campaign sources concede superstorm Sandy stalled Romney’s momentum. For eight straight days, polls showed him picking up support. The campaign’s internal polling, which is using different turnout models than most public polls, had him on solid ground in Florida, Virginia, North Carolina and Iowa. He had a slight lead or was tied in Ohio, New Hampshire and Wisconsin and was in striking distance in Pennsylvania, a state Republicans hadn’t won since Ronald Reagan in 1984. …

But then came something very big: a natural disaster that left a path of death and destruction on the East Coast. Suddenly, there was little talk about small things.

Yeah, I would like to see what the Romney campaign’s turnout model is. I suppose they are like most of the right wing and assume that a 2010 electorate will show up, rather than a presidential year electorate. What morons.

PPP: Obama 50%, Romney 47%
Rasmussen Tracking: Obama 48%, Romney 48%

COLORADO–SLIM ROMNEY—(Reuters/Ipsos): Romney 47%, Obama 45%
FLORIDA–TIED—(Reuters/Ipsos): Obama 47%, Romney 47%
IOWA–The average is Obama 47, Romney 43, or SLIM OBAMA—(Des Moines Register): Obama 47%, Romney 42%; (Project New America): Obama 47%, Romney 44%
NEW HAMPSHIRE–TIED—(WMUR): Obama 47%, Romney 47%
OHIO–The average is Obama 47.5, Romney 45, or SLIM OBAMA—(Reuters/Ipsos): Obama 46%, Romney 45%; (Project New America): Obama 49%, Romney 45%
PENNSYLVANIA–LEAN OBAMA—(Public Policy Polling): Obama 52%, Romney 46%
WISCONSIN–The average is Obama 49.5, Romney 45.5, or SLIM OBAMA—(Public Policy Polling): Obama 51%, Romney 48%; (Project New America): Obama 48%, Romney 43%

About the Author ()

Comments (24)

Trackback URL | Comments RSS Feed

  1. AQC says:

    DD, I’m curious what you think of Nate Silver’s predictions?

  2. Delaware Dem says:

    He was right in 2008. I think he will be right this time too. The reason why is he has a scientific model that projects the outcome. He is also not a politico. He is a stats nerd who was interested in sports before 2008.

  3. pandora says:

    If Nate Silver is wrong then all the polls were wrong.

  4. mm2784 says:

    I like Silver because it is based on data and math, not an agenda…though the Repugs argue otherwise. Then again, they also deny other forms of science as well.

  5. Jason330 says:

    Also, Silver is the first to say that a 70% probability of an Obama win is not a prediction of an Obama win.

  6. Jason330 says:

    Speaking of Nate Silver:

    “NEW YORK—As part of his continued effort to torment the 34-year-old statistician and blogger, feared New York Times bully Derek Kriesel reportedly slapped a stack of opinion polls from Nate Silver’s hands Friday, scattering the surveys across the floor of the organization’s newsroom. “Pick them up, you little fuckin’ dweeb,” said Kriesel, who eyewitnesses confirmed kicked the papers down the hallway before Silver could gather them up. “Hey, Silverdork, I got a poll for you. It says there’s a 90 percent chance that I’m going to beat the shit out of you, what do you think of that?” At press time, sources said Silver was hiding in a supply closet and analyzing the latest electoral data as a menacing voice from across the hall called out, “Oh, Silver! Where are you, Silver?”

    You know you’ve arrived when the Onion notices you.

  7. cassandra_m says:

    Silver is doing a really remarkable thing — he is working out probabilities for winning based upon publicly available data. He was right in 2008 Presidential and pretty much on the money on the 2008 Congressional and Gubernatorial races he tracked. He isn’t the only one working out predictive models, either.

    Sam Wang over at the Princeton Election Consortium also predicted the correct outcome for the 2004 Presidential and was one EV off for the 2008 Presidential. Today he has a President Obama probability of winning at 98.2%

    Drew Linzer’s model is new to the landscape this year and he is at . He has Obama with 332 EVs today.

    Scott Elliot is a conservative and he runs Election Projection. His track record is good, but not as good as others. He too, currently has President Obama winning with 290 EVs (but Romney winning the Popular Vote) as of today.

    There are a few others out there (like Electoral Vote) but Silver is getting beat up on largely because he is the most visible (and he is on a book tour). All of these sites are a real education in statistics and polling methodology. But all of these sites — provided they are correct come Tuesday– are a threat to the punditry who want you to think that they come by their horse race BS from some superior knowledge and better sources.

  8. Rusty Dils says:

    New poll out of Michigan today has race all tied up in Michigan(Actually Romney with less than a percentage point lead)

  9. pandora says:

    Rusty, please provide links to all your claims. I’m tired of digging through right wing sites looking for your proclamations.

  10. Jason330 says:

    Poor chap.

  11. geezer says:

    Rusty: A single poll has little meaning by itself. Romney’s internal polling has him ahead in almost all the swing states because it uses a different likely voter model than other polls.

    People began averaging polls before Nate Silver did. Silver went beyond that by factoring in data from election-prediction models based on economic factors and trends — basically a second set of election predictors independent of the more subjective data generated by polling (polls ask people to tell the truth, which they don’t always do, and always have a margin of error to allow for sampling variations). This form of election prediction has a long history that the media routinely ignores; its crudest application is that the unemployment rate can predict the election (unfortunately for us all, the GOP bought into that one and so have worked at keeping unemployment high).

    He runs statistical analysis on all the data using programs of his own design. He calibrates the program by running past election predictors and results until he has the formula fine-tuned. This worked well enough in 2008 that every Senate race and all but one state’s presidential vote correlated with Silver’s highest-probability result.

    That’s the key: He isn’t saying one guy will surely win. It’s a statistical probability, not a prediction.

  12. levotb says:

    PLEASE don’t insult Conservatives by referring to drudge, humanevents, townhall, Newsmax, Foxnews et al as “right-wing sites”. There is NOTHING “right-wing” about them! They aren’t conservative. They are Establishment (RINO)-loving sites run by what are known as “faux conservatives”. FYI, many Conservatives loathe Romney and aren’t voting for him as he moved to the left at the end of January when he saw he had the nomination in the bag and was the most famous flip-flopper in modern American history. Now he’s pushing his Grand Amnesty-Guestworker Plan (mentioned in Debate #2) that he intends to shove down our throats next year. Romney’s in trouble in NC, MN and has lost Iowa due to Evangelicals’ (Christian Conservatives’) dislike of Mormonism. Remember how the Evangelical vote in the Iowa Caucuses went solidly (38%) for Santorum, a Catholic, and knocked out Bachmann? Romney received just 14% of Evangelical support. If you’re going to slam “right-wing sites”, go after a site like mine, a law-and-order Independent Conservative site. And get your facts straight–a neocon or RINO is a liberal war hawk/Establishment Republican like John McCain, Newt Gingrich and Rick Santorum. A Conservative is someone like Rep. Steve King, Sen. Rand Paul or Sen. Mike Lee who know the Constitution and (with the exception of Sen. Paul’s belief in legalizing marijuana) are law-and-order Conservatives. There are very strong differences between these three men and the Republican Establishment even though Romney endorsed King in September to try to get Conservative support. The GOP is run (and has been run for as long as I can remember) by “country club Republicans”, very wealthy Chamber of Commerce types, not Conservatives.

  13. geezer says:

    Uh, please don’t bring your inbred-family fights to a liberal forum. You’re all pieces of shit from here, you law-and-order pants-pissers most of all.

  14. Davy says:

    Voter turnout in 2012 is unlikely to resemble voter turnout in 2008 or 2010.

  15. cassandra_m says:

    So what?

    Or maybe you’d like to claim that Fannie and Freddie are responsible for *that*, too.

  16. Davy says:


    I’m not taking your bait; you can be the child.

  17. cassandra_m says:

    It isn’t exactly childish to ask WTF that comment about prior year voter turnout had to do with anything going on in this thread.

    It *is* childish to not just admit that you were saying just any old thing.

  18. pandora says:

    The winter of 2012 is unlikely to resemble the winter of 2008 or 2010. 😉

  19. cassandra_m says:

    Oh, right! I get it now….

  20. pandora says:

    High school graduates of 2012 are unlikely to resemble the high school graduates of 2008 or 2010.

  21. pandora says:

    I could do this all night. 😉

  22. Davy says:

    The initial post comments on voter turnout models. And, I noted that voter turnout in 2012 is unlikely to resemble voter turnout in 2008, the Democrats’ preferred model, and in 2010, the Republicans’ preferred model. Any poll that relies on either model is crap.

  23. pandora says:

    Really? Why are these models crap?

    And if that’s what you were saying why did you write it in the fortune cookie language.

  24. cassandra_m says:

    Good grief.


    ps — he has no idea P. Especially since at least one of the models was also used in 2010 to predict pretty accurately the R wave.