Gallup is wahhhhhhhhing about Nate Silver: You're spoiling our spin!

[bold]Gallup is very upset at Nate Silver[/bold] [italic]The polling firm complains operations like FiveThirtyEight could spoil polling for everyone [/italic] Did Gallup just blame Nate Silver for ruining the art and science of polling? You don’t have to read too far between the lines of a statement from Gallup’s editor in chief, Frank Newport, published on Friday, to get that impression. Newport first attempts the formidable task of defending Gallup’s polling accuracy during the 2012 campaign. Perhaps he was anticipating [italic]Silver’s Saturday column, which labeled Gallup the most inaccurate pollster of all the firms[/italic] that measured voter sentiment this year. But Silver was hardly alone in wondering why Gallup regularly reported numbers much more favorable to Romney than anyone else in 2012. We deserve an explanation a little less lame than Newport’s: "what’s the big fuss? Gallup wasn’t really off by that much." But then it gets interesting: (cont at link)

http%3A//www.salon.com/2012/11/13/gallup_is_very_upset_at_nate_silver/



• [quote]But some of this will result from a variant of the venerable “law of the commons.” Individual farmers can each made a perfectly rational decision to graze their cows on the town commons. But all of these rational decisions together mean that the commons became overgrazed and, in the end, there is no grass left for any cow to graze. Many individual rational decisions can end up in a collective mess. [quote]We have a reverse law of the commons with polls. It’s not easy nor cheap to conduct traditional random sample polls. It’s much easier, cheaper, and mostly less risky to focus on aggregating and analyzing others’ polls. Organizations that traditionally go to the expense and effort to conduct individual polls could, in theory, decide to put their efforts into aggregation and statistical analyses of other people’s polls in the next election cycle and cut out their own polling. If many organizations make this seemingly rational decision, we could quickly be in a situation in which there are fewer and fewer polls left to aggregate and put into statistical models. Many individual rational decisions could result in a loss for the collective interest of those interested in public opinion... This will develop into a significant issue for the industry going forward. [bold]It is impossible to read this as anything other than an attack on Nate Silver, who is by far the most prominent aggregator and analyzer of others’ polls currently operating today. And it simply reeks of sour grapes.[/bold] During the campaign year, Silver consistently pointed out that Gallup’s results were oddly inconsistent with what other pollsters were finding. [bold]And he was right— Gallup got it wrong. [/bold] It is not inappropriate to point that out. But Gallup presumes too much when it effectively threatens to take its surveys home and just stop playing. OK, Nate Silver didn’t pick up the phone and call voters himself. And yes, it’s true, if all the polling organizations stop polling, Nate Silver will be out of a job. He might have to go back to crunching baseball stats. But he’s a smart guy, with an impressive track record. He’ll do fine. It seems foolish to blame him for crunching data that is publicly available for anyone to crunch. If he wasn’t good at what he does, he’d still be another anonymous poster hanging around the DailyKos forums. He’s been rewarded for getting things right. Gallup faces a more unhappy future. If the company keeps getting the numbers wrong, the market will punish it. [bold]At that point, Gallup won’t be able to blame its sorry fate on anyone but its own incompetence. [/bold]

  tee%20hee%21

• All this gallup backlash is making me very happy. Now they need to take a look as Rasmussen as well. Every conservative pundit and 70% of the media in general used GALLUP and RASMUSSEN's number to bolster the "horse race" meme. It was all a big fat LIE. Nate Silver saw through it from day one. He spoiled everyone's spin. I'm sorry if this sounds like a hyperbole but he's a Hero of Reason.

• Why should we care? If they can't/won't improve their methods then they should GO OUT OF BUSINESS!! Welcome to capitalism, whiners!!!

• You lost me at "wahhhhhhhhing". Op, are you retarded?

• r4, you're 3 posts late.

• Gallup was the least accurate pollster in 2008 and 2012. Any news organization that quotes their polling in 2016 risks major credibility damage, as far I'm concerned. Their head guy, Frank Newport, is a major Republican. Democratic pollster, PPP, was the most accurate pollster this year, yet the mainstream media ignored them for the most part. I'll pay attention to PPP and Nate Silver from now on.

• r4 has Gallop memorabilia decorating his home.

• Thanks for posting this, OP. When I was doing the polling threads this year I suspected several months ago that there was something wrong with Gallup's numbers.

  2012%20Poll%20Troll

• [quote] It’s much easier, cheaper, and mostly less risky to focus on aggregating and analyzing others’ polls Isn't this a research meta analysis? They're done all the time in science

• Gallup's complaint is a fair observation. It's essentially the equivalent of the traditional media companies saying: "we go to the trouble of paying reporters to hunt down stories, have someone check them, pay editors to put them in readable format and then we publish for sale - - - only to have others republish our content for free and/or write opinion pieces based on what we produced. If everyone keeps this up, we'll have to take our efforts home (i.e. "go out of business") and then the only thing left will be FOX.

  ECON 101

• [quote] Organizations that traditionally go to the expense and effort to conduct individual polls could, in theory, decide to put their efforts into aggregation and statistical analyses of other people’s polls in the next election cycle and cut out their own polling. Well who needs inaccurate polls? [quote]we go to the trouble of paying reporters to hunt down stories, have someone check them, pay editors to put them in readable format and then we publish for sale - - - only to have others republish our content for free and/or write opinion pieces based on what we produced. Would you prefer journalists to ignore your polls? (If they keep being so biased, that may happen in any case.)

• I actually know Frank (Newport). A decent fellow from Fort Worth but I have always suspected him of being a right winger. He says he's impartial but he comes from a family of far right wing lunatics.

• Rasmussen is making its own excuses too:

  http://www.politico.com/blogs/charlie-mahtesian/2012/11/rasmussen-explains-149240.html

  2012%20Poll%20Troll

• [quote]It’s much easier, cheaper, and mostly less risky to focus on aggregating and analyzing others’ polls Nate Silver is not a pollster and never claimed to be. He's a statistician and an analyst. [quote]Gallup's complaint is a fair observation. No, it isn't. Polling is not like hunting down a news story. It's a process and some organizations do it differently (and better) than others. Statisticians and analysts always look at the methodology to evaluate a process. It's more like writing a science article based on research. A university and its staff may put a lot of money into doing a study. The study is then published and subject to review.

• Sounds like Gallup has been drinking a bit too deeply of the butthurt whine.

• Oh Please r10. Gallup's "complaint" is like saying, "If we're all held to some objective standard... then.. then.. ALL POLLS WIL DIE!" It's ridiculous to say that examining a pollsters standards and repeated skewed results is going to kill the process. Polling shouldn't be an like an artform, it should be a science. It should stand up to scrutiny. Polls have a way of skewing perceptions. Politicians love to use them to justify their campaigns. Having no standards for them is as bad as having no polls.

• Tweeted today:: Nate Silver ‏@fivethirtyeight Polls shouldn't bounce around as much as Gallup's, given their large sample sizes. Something is designed badly and introducing noise. Nate Silver ‏@fivethirtyeight Gallup's demographic weighting algorithms probably contain a few bugs. They need to hire better statisticians.

• Gallup blaming Nate Silver for a reverse tragedy of the commons. Classy. Take your poll & shove it, Gallup. That way polling averages like Silver's will be even more accurate without your skewed Rethug numbers. There will still be hundreds & hundreds of national & state polls available for Silver, the Princeton Election Consortium and TPM.

• Gallup needs to take ownership of the fact they have been very inaccurate and pledge to work to improve their model. Silver is right, they need to hire better statisticians. I'm glad Nate has balls and is standing up for himself, you saw it with Scarborough also. Don't want to see him get cocky but it is nice to see he won't let people run all over him.

• Yup, they need to get better and compete with those who predicted with more accuracy- simple as that.

  charlie

• Miss Ann is so angry at all the pollsters that, now when she rides me very fast, she refuses to call it a "gallop."

  Rafalca

• Chuck Todd did a pretty good job this year in reading the polls correctly. He didn't jump on the Romney momentum bandwagon like so many others did. And the NBC/WSJ/Marist polls, which Todd loved and touted, were very accurate in the end.

• Gallup is headquartered in the ruby red state of Nebraska. 'nuff said.

• [quote] Gallup's complaint is a fair observation. It's essentially the equivalent of the traditional media companies saying: "we go to the trouble of paying reporters to hunt down stories, have someone check them, pay editors to put them in readable format and then we publish for sale - - - only to have others republish our content for free and/or write opinion pieces based on what we produced. If everyone keeps this up, we'll have to take our efforts home (i.e. "go out of business") and then the only thing left will be FOX. It isn't like that at all. What is really happening is they are saying, We go to the trouble of paying people to hunt down information. We skewer the results to our own liking, only to have someone go out and dare to report that we were wrong. Then the rest of the media republish content about and/or write opinion pieces based on our sloppy inept manipulations If everyone keeps calling us on our inferior work and lies we'll have to take our efforts home (i.e. "go out of business"). Which is okay because after 2008 and 2012, no decent reputable news organization should ever use gallup results. But at least they'll still have fox news.

• I love how the reich wing is getting slammed with their comeuppance in 2012. From Romney and his entitlement, to KKKarl Rove's head on the chopping block, to Paul Ryan being bitchslapped by his hometown, to the total fucking meltdown of Victoria Jackson and all those other loonies, to now GALLUP being called out for the fraud that they are... If this doesn't prove that Republicans DID conspire to steal the election, then I don't know what does. However, the AMERICAN PEOPLE triumphed in this election, and we couldn't be bought or sold, and we wouldn't allow our voting rights to be taken away from us despite their vigorous attempts to do so, and we wouldn't be intimidated. WE WON FAIR AND SQUARE, despite a shitload of opposition and dirty tricks. All these things make this victory in 2012 SO MUCH SWEETER, and very delicious!

  Congratulations, Nate

• Gallup needs to rethink its models.

• Gallup needs to go bankrupt.

• Is Gallup behind the #DrunkNateSilver feed?

• r24 - I agree with a lot of what you posted, however it is "skew", not "skewer". And not to defend Fox News, but their polling was pretty good.

• Remember the loon who was obsessed with finding fault with Silver? Where is that loon today?

• [quote] Where is that loon today? His credit card got cancelled on the election night, as soon as it became clear that Romney was going to lose.

• r 31 has it.

• Dan Levitan is credited as the typhoid Mary of this lastest meme. Gawker, BuzzFeed and others followed the tweet-trail to his Wednesday night communique that read: "Drunk Nate Silver is riding the subway, telling strangers the day they will die."

  http://www.scpr.org/blogs/news/2012/11/08/10996/drunk-nate-silver-tweets-math-meme-la-twitter/

• But...but...but...isn't Nate Silver a short, thin, [childish epithet posted by a bigoted tool] voiced little boy-man....how can such a wimpy, little queer take down a big tough manly pollster like Gallup - it just don't compute.

  Kansas%20boy

• Oh, the Gallups are leaving? Bye.

• Sober Nate Silver predicts demise of Gallup polls!

• I can understand a bit of what Nate Silver went through during the election because I was attacked on a number of the Polling Threads by the freepers. They told me that I was deliberately excluding polls that were favorable to Romney and that his real level of support was much higher. They said that Romney was in a position to win states like Pennsylvania, Wisconsin, Minnesota, etc. However, unlike Nate Silver, I am basically anonymous and so there wasn't much they could do to me. I'm glad to see Nate Silver redeemed, and I'm glad that my own analysis of the polls proved to be sound as well.

  2012%20Poll%20Troll

• They need to shut the fuck up.

• Poll Troll, there's still a lot of residual love for you!!

• [quote]Chuck Todd did a pretty good job this year in reading the polls correctly. He didn't jump on the Romney momentum bandwagon like so many others did. And the NBC/WSJ/Marist polls, which Todd loved and touted, were very accurate in the end. This is so inaccurate. Chuck Todd kept pushing the narrative that the race was too close. Although Scarborough was more blatant, Chuck Todd essentially said that anyone who thinks he can predict the outcome is delusional. This is the second time I've read a post like r22. I wonder if we're watching the same Chuck Todd? This guy is a hack.

• yes

• I have to chuckle about those two-faced pundits like Scarborough who is now jumping on attacking the right wing media who pushed inaccurate polling data. Yet just days before the election he was jumping on the Romney surprise making references to the Reagan revolution. He was talking about what a great polling weekend Mitt had had going into Sandy. It was all bullshit. He is part of the problem but now points fingers to save face.

  Joe%20Scarborogh%27s%20dead%20intern%20problem

• Nate Silver is an economist, NOT a statistician. He has an undergrad degree in economics from the University of Chicago -- and that's ALL. Everybody who had three stat courses in college fancies himself a statistician. That tells me more about him than any prediction he comes up with.

• LOSER. r43.

• [Quote]....and I'm glad that my own analysis of the polls proved to be sound as well. OMG.

• Actually R43, you're telling us a lot about yourself right now.

• [quote]Nate Silver is an economist, NOT a statistician. And yet his stats were spot on while a company whose ONLY job is to do accurate polling failed miserably. Or faked all their alleged polling. LOL!

• I thought drunknatesilver was funny. In his first blog entry after the election was called for Obama, he said he was going to grab a beer. I thought they just extrapolated on that. People are so sensitive. I don't see how anyone could find #drunknatesilver insulting.

• [quote] Everybody who had three stat courses in college fancies himself a statistician. Baseball Prospectus considers him a statician. They bought his PECOTA program for predicting the future of young ball players.

• What he has a degree in r43, is irrelevant. Jesus, do you even READ his column? That's like saying a person who has a license to be a sommelier could never be a great bartender. What he DOES and DID was aggregate the polls as well as other factors like the economy, the weather, the history and methodology of each poll- and weigh their impact based on those factors- to precisely predict the election. If I happen to be a great chef, whether or not I went to cooking school, I'm still a great Chef.

• [quote]Nate Silver is an economist, NOT a statistician. So he's an even BIGGER geek than we imagined. Well, no wonder he fucking schooled the pay-per-analysis "statisticians". Geeks rule. (#DrunkNateSilver is hilarious, BTW)

• r43 = pwned

• R40 thanks for your post. I totally agree with you - r22 is dead wrong. On the Friday before the election, I watched Chuck Todd proclaim he was "feeling" NH would go to Romney. He also vigorously maintained that the Romney people 'knew something' about PA and that other states the President won handily would be super close and very likely go to Romney. He pushed the Mitt-mentum meme long after Nate Silver and others demonstrated it was over. In years past, I felt about Chuck Todd like r22 still does, but this year he ripped off the mask and pushed Romney for all he was worth. He looked visibly disgusted and pushed back hard anytime a guest contended the President would win re-election.

• LOL at R43. Are those words just gibberish you tell yourself as a Republican to make yourself feel better?

• 2012 Poll Troll @ R8 said what I'd been thinking throughout most of the general-election period (which essentially started when it was clear Mitt Romney would win his party's nod; before he had the delegates). Rasmussen Reports, we know, is a Republican outfit. And four years ago, they were last on board to claim a Barack Obama (over John McCain) lead in the bellwether state of Ohio. (We were literally into the month of October until Rasmussen finally said that Obama had the lead.) Gallup was constantly playing down the president's prospects for re-election while playing up the challenger's. After the "47 percent" speech scandal, they got on board. The job-approval percentage for Obama shot up. But an aggregation of the polls from [italic]Real Clear Politics[/italic] also reflected that, as well as the gap closing of "right track/wrong direction," clearly pointing to re-election of President Obama. I eventually stopped paying attention to Gallup. I knew, prior to Election Day, that either Rasmussen and Gallup would end up trumping their competitors -- as if they knew better -- or they would become exposed as not faulty but fraudulent pollsters. The latter happened. And whatever credibility they had is now gone.

• I'm not a big fan of Chuck Todd, but he stated several times on his show and on Meet the Press that Romney was not going to win PA. He also repeated over and over that the path to 270 EVs was very challenging for Romney. He said many times that the most important polling question to look at was: "Which candidate better understands the issues that affect me?" Todd said that Obama was in a good place because he always led on that question. Lastly, Todd believed that the NBC/WSJ/Marist polls were the gold standard, and every one of them showed Obama either leading or tied.

• friend of chuck's, r56?

• I watcched him pretty regularly on NBC and MSNBC and when he movied those states around on his big Electoral College map, it alwways came up Obama.

• Nate Silver tweeted from a location in my town very close to where I live about a year and 1/2 ago. I was so excited that I posted it on Facebook but all of my friends were like, "who?" I wish I had gone over to say hello to him.

• ...

• How dare he take his job so seriously, be so damn good at it and show up us and others who aren't as proficient. Who the fuck does he think he is?

• Correction. Rasmussen was not the last to have polling show Obama with an Ohio lead in 2008. Even their final Ohio poll in 2008 showed a tied race.

• The Repukes are unbelievable!

• HEY! WE'RE not the ones who are uppity homosexuals!! We're proper repressed/uptight closeted homosexuals who will do anything to keep their secret.

• Even more than Chuck Todd, John King seemed to be pushing for a Romney win.