Subscribe via RSS Feed


[ 49 ] November 7, 2012 |

Scott Rasmussen tries to explain why his polls were terrible. I’m reading this over and over again but I keep missing the sentence that reads “My polls suck because I’m a partisan hack who values creating a narrative that favors Republicans over accuracy.”


Comments (49)

Trackback URL | Comments RSS Feed

  1. Craigo says:

    Telephone polling works fine if you actually bother to call the 30-40% of the electorate which does not have a landline.

    • Holden Pattern says:

      Hell, we have a landline for emergencies — it rings straight into an answering machine, and we never hear it. That’s pretty common in our area.

    • Increase Mather says:

      Well, you don’t have to call them, but you have to account for them somehow (e.g., internet polling).

      From Daily Kos’ list of most accurate pollsters:

      But let’s be clear, you have to go down to number six on the list to get to someone who called cell phones. And Gallup called 50 percent cell phones and they were a laughingstock this cycle.

      • thusbloggedanderson says:

        How does internet polling correct for that? I really haven’t the first clue.

      • Craigo says:

        Rasmussen does do Internet polling of cell phone only households. Internet polling is very, very spotty methodologically.

      • Craigo says:

        And I should have said “account” instead of call, you’re right. You can reliably adjust for CPOH with dual-frame blended samples which IVR pollsters hate because of he expensive, or use stratification, which is a lot cheaper. Internet panels give me bad acid flashbacks to Zogby Interactive. The

    • DrDick says:

      You apparently do not need to do that if your sample is structured properly. The PPP/DailyKos poll, which does not call cellphones, was just about the most accurate of the election cycle.

  2. parrot says:

    data collecting and harvesting is editorial … shouldn’t be, but alas … i blame heisenberg …

  3. Joseph Slater says:

    Nonetheless, my probablistic model says that the likelihood of right-wingers in 2016 making posts along the lines of “Rasmussen, which has traditionally been one of the most accurate pollsters in the nation, says PA is tied, so SUCK IT LIBTARDS” is better than 90%.

    • Wido Incognitus says:

      Even with the fact that Joe (I wish that people could call him Uncle Joe without calling to mind one of the worst people in history) Biden lived in Pennsylvania as a child!

  4. Rasmussen points to the Obama campaign “getting his people to the polls.”

    I’d like to remind everyone that this does not mean that Democratic-leaning people turned out in higher numbers than Republican-leaning ones. It means that the Republicans’ advantage was small.

    An election that accurately reflected the opinions of the American public would be a huge Democratic win. When Republicans express their consternation about the Democratic turnout machine, they are complaining the that election was a more accurate reflection of the public’s wishes than they had hoped it would be.

  5. daveNYC says:

    Some people projected Romney would do a couple of points better than the polls and sweep those states. Instead, it was Obama who did a bit better and swept them.

    Accurate, but not exactly informative.

  6. Vertov says:

    You know who’s giving Rasmussen a hard time? Dean “unskewed” Chambers:

    Chambers himself sounds downright contrite in this piece.

    • NonyNony says:

      Heh. That seems to be the standard line.

      “Wow – we thought blacks, Latinos and kids were just voting out of the novelty of a black President in 2008. Hoocodaknown that they actually might represent a real group of voters in the long term?”

      Honestly – Democrats need to rethink their model of elections as well. Getting out the vote is going to be important for every election in the future, not just presidential ones (even if jerrymandering in the states can reduce the impact of GOTV in non-statewide contests).

      • Chatham says:

        Well, gerrymandering and voter suppression should be a wake up call that people can’t only pay attention to national races. Now’s the time for a lot of hard work, not only preparing for national races but also making sure that we’ll be ready to retake state offices from those that have been abusing their power.

        • Bob Loblaw says:

          Absolutely. State and local level races are so important for long-term viability. I would go so far as to say that the reason it wasn’t a total Dem rout last night (i.e, the Reps held on to the House) is because of Republican control of so many state houses…and that is in large part a result of their diligence in running strong in so many lower-profile and local-level races over many years, even decades.

          • NonyNony says:

            In the Franklin County Ohio area we had a Republican state Senator who ran unopposed. Completely unopposed – not even a name on the ballot to register as a protest vote against him. Sure the district has been jerrymandered beyond belief but come on – you can’t even put for the effort to put a damn name on the ballot? What if there had been a giant scandal that made voting for him toxic? What if he’d had a heart attack the week before the election? You can’t win if you aren’t even trying.

            That’s part of the reason that the Republicans have taken over state government here in Ohio – they’re always able to find someone willing to put their name on the ballot everywhere, and not just in places where they’re likely to win.

            • ralphdibny says:

              Look at the guy who ran on the Dem ticket in TN against Corker this year, or the SC guy two years ago. AL now has zero high-ranking Dems. Southern Dems are practically an endangered species. It’s gonna take a LONG time to build the party back up in the South.

              • Pee Cee says:

                SC voter here. On the (three-page) ballot in my town, there were exactly four Democrats.

                Two of them were Biden and Obama, and one of them was the rather weak Democratic candidate for SC’s new House seat. Most of the ballot was filled with Republicans running unopposed for various offices (including seats in the state legislature).

                Democrats will never win here if they can’t be bothered to even field candidates.

                • Chatham says:

                  Agreed – then again, if everyone is waiting for someone else to start, we’re not going to get anywhere. Someone’s going to have to take the plunge and run for a local race, meet other Democrats in the area, network, etc.

                  It something I hope we focus on in the next year. It’s much, much more important than the “Boehner said something stupid” “Don’t vote for Jill Stein” stories.

                • cpinva says:

                  they don’t bother running there (SC) because they know, before they even register, that they’re going to lose. they know that the DNC knows this, and they will get zero financial/organizing help from them.

                  given that scenario, it would take a person with a ginormous ego, who knew they were going to lose going in, and was willing to be the sacrificial lamb, to at least break the ice for the future. because, it would also affect that person’s ability to make a living in SC afterwards.

                  that’s asking a hell of a lot of someone.

            • Major Kong says:

              I forget which Christian Coalition leader said that they would have a candidate on every ballot even it was for “a dogcatcher in Montana”.

              • cpinva says:

                i think that was ralph reed. and that’s what the democrats need, and don’t have. they also don’t have the deep pockets zillionaires, ready to shell out the millions necessary to build those organizations.

    • some other guy says:

      He’s still an a-hole, though. Let’s not forget that his principle argument against Nate Silver was basically:

    • dan says:

      I enjoyed reading that. I really enjoyed reading that. In fact, I’m enjoying today more than I enjoyed four years ago, just for the pleasure of seeing the conservative bubble being (temporarily) popped by reality. This guy, at least, seems to realize he was being told what he wanted to hear. Some of the cons at my work might be catching on to the same thing.

  7. catclub says:

    Also, because you can make good money telling the rubes what they want to hear, rather than the best version of the truth, so why stop?

  8. DrDick says:

    I love the scent of wingnut tears in the morning.

  9. Jim Lynch says:

    More like: “My polls suck. However, I’ll remain a well paid, partisan hack who values creating a narrative that favors Republicans over accuracy because that’s where the money is.”

  10. Midwest_Product says:

    Where does Scott Rasmussen live? If it’s Colorado or Washington, he can at least smoke some weed about it.

  11. somethingblue says:

    Hadn’t thought of it in this light, but a certain amount of the weird Nate Silver hate is perhaps due to the Rasmussen phenomenon. The Right just assumes that Silver must be a kind of anti-Rasmussen–doing what Rasmussen does, but for the other side. The idea that somebody might run the numbers to find out the actual state of affairs is not something that’s even occurred to them. Why would anyone do that?

    • ralphdibny says:

      I think the anti-choice arguments follow this pattern as well. All those (R) candidates kept bringing up rape because they think that Dems are gaming the system, and it drives them crazy. They think that Dems keep bringing up rape, incest and health of the mother as distractions, loopholes that we plan to exploit so that everyone can get govt. subsidized birth-control abortions. The idea that pro-choice votes are actually concerned about the health of women doesn’t seem to have occurred to them.

    • cpinva says:

      then they are even dumber than we normally give them credit for:

      The Right just assumes that Silver must be a kind of anti-Rasmussen–doing what Rasmussen does, but for the other side.

      because nate silver runs no polls at all, zero, nada, zip. apparently, the concept of taking polls, run by others, and statistically averaging them, is so far beyond their thought processes, it must amount to black magic in their world view.

  12. greylocks says:

    With wingnuts, it’s all about validation.

    They see validation in winning. But they also see validation in silencing/discrediting their critics and opposition. They go after people like Nate Silver and Michael Mann because in the wingnut mind, you “win” and are proved “right” if you can get people who don’t agree with you to STFU.

    Rasmussen is not a wingnut, but he’s a hack who polls the way he does because there’s a market for a pollster willing to cater to the wingnut need to be validated. Wingnuts aren’t going to pay for better polls because better polls show them losing too often. They need a pollster on whom they can pin their hopes. If he’s right and everyone else is wrong, they are validated. If he misses and everyone else is right, the election must have been stolen.

    Rasmussen’s explanations of his errors may read like excuse-making to rational minds, but to a wide of wingnuttia, they will be heard as “the damned negros and wetbacks got off their normally lazy asses and voted”, and Rasmussen will still be in business in 2014, getting it wrong again, because he was close enough that wingnuts can believe the election was stolen by the mud people.

    • cpinva says:

      actually, you’ve just pretty accurately described high level wingnuttia: they’re in it for the cash. most of them either know, or sense, that they’re full of shit. they don’t care, because truth isn’t the issue at all. what’s important is how much they can scam the rightwing rubes for, by telling them what they so desperately want to hear and believe.

  13. thebewilderness says:

    Well, yanno, that reality won’t create itself!

  14. Eli Rabett says:

    With polls you look at the trends not the numbers.

  15. O. D. Jones says:

    N. Silver = J. Christ…I would elaborate more but grey men in hoods are muttering something that sounds like “benghazi” outside the bunker door…

  16. Anon21 says:

    I think this just shows that in the way that Rasmussen departed from methodological consensus and made a name for himself, he was wrong. Party ID remains an attitudinal, not a demographic variable, and weighting for it just screws your model up.

    I basically reject the notion that Rasmussen was trying to flatter the biases of Republicans. I think he intended to call this race correctly, not produce a bunch of GOP agitprop. The test of whether I’m right or whether the rest of you are will be in whether this embarrassing miss prompts him to re-examine and change his bad methodology, or simply to double down.

    • actor212 says:

      I’d buy this, except Rasmussen has been consistently wrong the past three elections. He got lucky in 2004 in calling the race in the right direction, but he had Bush winning going away comfortably, which was clearly not the case.

      • Anon21 says:

        I think Rasmussen did fine in 2008, didn’t they? 2010 was a big miss, but sometimes even honest pollsters will have them.

        Again, I think this election in particular should be a wakeup call, assuming Rasmussen is actually interested in getting it right. There was a lot of loose talk flying about how 2008 voters demos were some kind of bizarre aberration that could never be recreated.

        Now that we’ve dispelled that, now that we’ve seen that a good GOTV effort can turn out traditionally unlikely voting blocs, a conscientious pollster is going to build a likely voter model that relies mostly on objective indicators, not attitudinal variables or past voting behavior. We’ll see if Rasmussen takes that approach.

  17. actor212 says:

    To this day, Rasmussen claims his greatest professional thrill was hosting a 50th birthday party.

    OK, for Gordie Howe while he was still playing for the Whalers, but still…

  18. jeer9 says:

    Rasmussen actually underplayed the deficit in Missouri, citing an 8 point advantage before the election when McCaskill was actually in double digits.

  19. […] Rasmussen: Erik Loomis This entry was posted in Potpourri. Bookmark the permalink. ← Reader Feeder Bits for (Tue. 6-Nov-12 1630) […]

  20. […] Loomis: Rasmussen: Scott Rasmussen tries to explain why his polls were terrible. I’m reading this […]

  21. Levi Olivos says:

    Bed Bugs Chicago, Treatment of Bed Bugs Chicago, Removal of Bed Bugs Chicago We can exterminate those bed bugs now! Contact

Leave a Reply

You must be logged in to post a comment.