Main Menu

Polling conspiracies

imageI saw in passing a bit on the Rachel Maddow show on MSNBC that there was a story circulating that all the polls are being manipulated (even the Fox News Poll1) to show Obama ahead, when he’s actually not ahead. I wouldn’t always trust Maddow to report a story representing the best possible case from the other side, but what she said was that these conspiracy theorists say that the polls are biased because they polled more Democrats than Republicans, and that they should have polled more representative numbers of each.

Let me give an example. Let’s say that there are 100,000,000 Democrats and 1,000 Republicans, and to simplify matters that all the Democrats are supporting Obama and all the Republicans Romney. In a random poll of 1,000 people (a typical size for such polls) one would get 999 Democrats and 1 Republican, and the polls would show Obama winning with 99.9% of the vote. The objectors say that a fair poll would survey 500 Republicans and 500 Democrats and the poll would show both candidates at 50%. Maybe the objection is a little more sophisticated than that.

You can read it from the horse’s mouth at UnSkewedPolls.com. Here are their assumptions:

The QStarNews poll works with the premise that the partisan makeup of the electorate 34.8 percent Republicans, 35.2 percent Democrats and 30.0 percent independent voters. Additionally, our model is based on the electorate including approximately 41.0 percent conservatives, 20.0 percent moderates and 39.0 percent liberals.

Republicans are 68 percent conservative, 27 percent moderate and 5 percent liberal. Among Democrats, 14 percent are conservative, 36 percent are moderate and 50 percent are liberal. Independents include 39 percent conservatives, 36 percent moderates and 25 percent liberals.

Our polls about (sic) doubly-weighted, to doubly insure the results are most accurate and not skewed, by both party identification and self-identified ideology. For instance, no matter how many Republicans answer our survey, they are weighted at 37.6 percent. If conservatives are over-represented among Republicans in the raw sample, they are still weighted at 68 percent of Republicans regardless. This system of double weighting should insure our survey produces very accurate results, not skewed either way for the Democrats or for the Republicans.


1In the latest UnSkewed Poll, Fox News shows Obama up by +2, the only UnSkewed poll showing Obama ahead.

, ,

64 Responses to Polling conspiracies

  1. G September 28, 2012 at 2:24 pm  (Quote) #

    Or the “Unskewers”, as their laughable site denotes… Yeah, I too would agree that they fit the definition of the first truly “new” Obama Conspiracy Theory meme that we’ve seen in a long time…still birthed out of an extreme need for denialism….but a new and spreading conspiracy amongst the dispirited right, nonetheless.

    Scientist: Doc, would you please consider a thread on those who claim that there is a giant conspiracy to skew the polls in Obama’s favor? I even have a catchy name for them-skewers.

  2. Scientist September 28, 2012 at 3:50 pm  (Quote) #

    G: Or the “Unskewers”, as their laughable site denotes… Yeah, I too would agree that they fit the definition of the first truly “new” Obama Conspiracy Theory meme that we’ve seen in a long time…still birthed out of an extreme need for denialism….but a new and spreading conspiracy amongst the dispirited right, nonetheless.

    Imagine what those same people would be saying if the polls had Romney ahead by 5.

    Meanwhile, Mitt is at 21% on Intrade (and below 20% on some other sites). So, if Mitt really believes the polls are wrong, he could buy some shares and make a 400% profit in 5 weeks. That beats even the best deals he did at Bain. And, yet, there are very few bids for Rmoney futures even at these fire sale prices.

  3. John Potter September 28, 2012 at 4:43 pm  (Quote) #

    “…they should have polled equal numbers of each.”

    *facepalm*

    Quite an assumption, that there are equal #’s of each.

    Not all Reds are birthers, but they all seem to think like them.

    Denial, denial, denial …. “we lost not because stupid doesn’t sell, but because the fix was in.”

    Well, I wouldn’t want to admit I was stupid, either.

    The first step is admitting there might be a problem …

    Seriously, we’ll be hearing cries the election was rigged any day now.

  4. Dr. Conspiracy September 28, 2012 at 4:50 pm  (Quote) #

    In a recent survey, 76% of Americans said they didn’t trust public opinion polls.

  5. Scientist September 28, 2012 at 4:54 pm  (Quote) #

    Doc: Thanks for putting this thread up. Skewerism is indeed an Obama conspiracy theory, just as much as birtherism, and no less silly.

    Just to expand on what you wrote, party identification, unlike age, gender, race, etc. is fluid. While some people are life-long Rs or Ds, many people are fairly flexible as to which party they support at a given moment, So, you could call a random group of 100 people today and get 45 Ds, 35 Rs and 30 Independents and next year, the same 100 might be 35 Ds, 40 Rs and 35 Is. All the pollsters do is record`what people tell them. They don’t adjust or apply a fudge factor, So, if Obama is ahead, you would expect a higher D identifiication than if Romney were ahead, and vice versa. The D skew is not causing Obama’s lead, it is reflecting it.

    In case you want to read someone who knows more about it than me here is Gallup’s Editor in Chief, Frank Newport http://pollingmatters.gallup.com/

    Or, as Nate Silver says, in a poll of Mitt Romney voters, Mitt Romney is way ahead.

  6. John Potter September 28, 2012 at 4:54 pm  (Quote) #

    The cry for fairness has quite a Harrison Bergeron feel to it.

    Also, are we talking self-identified Blues and Reds, or registered Blues and Reds, or those intending to vote Blue or Red?

    …or just whatever it takes to show rMoney in the lead?

  7. G September 28, 2012 at 4:58 pm  (Quote) #

    😉

    Dr. Conspiracy:
    In a recent survey, 76% of Americans said they didn’t trust public opinion polls.

  8. Scientist September 28, 2012 at 5:03 pm  (Quote) #

    John Potter: Also, are we talking self-identified Blues and Reds, or registered Blues and Reds, or those intending to vote Blue or Red?

    The pollsters ask “affiliation”, not what are you are registered as. They of course have no means nor desire to verify the answers. And the registered Independents will almost all vote for Obama or Romney. So, one who plans to vote for Obama this time might genuinely consider his affiiliation D for now and the same wiith one who plans to vote for Romney.

    Also, if Johnson is included in the polls, as he will be on the ballot, Obama’s lead increases. So the polls, many of which do not name Johnson, may be underestimating Obama’s lead, not overestimating it.

  9. G September 28, 2012 at 5:11 pm  (Quote) #

    Agreed. Well said on how you explained the issue too.

    This is certainly a new Obama Conspiracy Theory…and one that has the potential to even set up the next “conspiracy”, once Obama wins re-election.

    The folks who are seeking comfort in their self-induced cocoon bubbles, by taking up “Skewerism” (or “Unskewerism”, depending on how you define it) are not just seeking to keep their morale up enough to vote – No, they are also laying the insidious groundwork to decry ANY election result for Obama, no matter how vast the margin of victory, as “fraud”…because they simply won’t be able to accept that the results differed from their “unskewered” pet polls…regardless if everyone else’s predictions are pretty much in line with the actual results.

    So it cues them up to invent new “ACORN” myths and claims of voter fraud and a stolen election…because in their minds, there is simply no way that America can elect that black democrat with the funny name…not only once, but twice!

    Which comes back to Birtherism – because that in the end is really what their need to invent those Birther myths are all about – sore-loser denialism to pretend that the 2008 election didn’t really happen and that they can haul him off in chains and press a magic reset button to make the last 4 years and those election results simply “go away”.

    They HAD to invent Birtherism in the last cycle… because Obama’s margin of victory in all measurable terms was otherwise UNDENIABLE…and couldn’t not be seriously contested or explained away as an accident of counting or margin of error.

    So too, this cycle is shaping up to be a rout…and therefore they need to start laying the groundwork now, with their own fake “unskewed” polls and Baghdad Bob style prounouncements by Dick Morris, in order to build that alternate fictional reality, in which Romney really won the election and was ahead the whole time….

    Scientist:
    Doc:Thanks for putting this thread up.Skewerism is indeed an Obama conspiracy theory, just as much as birtherism, and no less silly

    Just to expand on what you wrote, party identification, unlike age, gender, race, etc. is fluid.While some people are life-long Rs or Ds, many people are fairly flexible as to which party they support at a given moment,So, you could call a random group of 100 people today and get 45 Ds, 35 Rs and 30 Independents and next year, the same 100 might be 35 Ds, 40 Rs and 35 Is.All the pollsters do is record`what people tell them.They don’t adjust or apply a fudge factor,So, if Obama is ahead, you would expect a higher D identifiication than if Romney were ahead, and vice versa.The D skew is not causing Obama’s lead, it is reflecting it.

    In case you want to read someone who knows more about it than me here is Gallup’s Editor in Chief, Frank Newporthttp://pollingmatters.gallup.com/

    Or, as Nate Silver says, in a poll of Mitt Romney voters, Mitt Romney is way ahead.

  10. G September 28, 2012 at 5:12 pm  (Quote) #

    In most polling methods, this is nearly always as a result of self-identification. Which is why such things are quite fluid.

    John Potter: Also, are we talking self-identified Blues and Reds, or registered Blues and Reds, or those intending to vote Blue or Red?

  11. John Potter September 28, 2012 at 5:30 pm  (Quote) #

    Scientist: So the polls, many of which do not name Johnson, may be underestimating Obama’s lead, not overestimating it.

    I also note that in most polls (that I have tracked anyway), the sum of Obama’s and Romney’s number has continued to approach 100 (i.e., undecideds are deciding), at the same time Obama is pulling away (i.e., they’re deciding for Obama).

    As rMoney so eloquently put it, the candidates fight for the sliver in the middle. He lost it.

  12. donna September 28, 2012 at 5:38 pm  (Quote) #

    G: In most polling methods, this is nearly always as a result of self-identification. Which is why such things are quite fluid.

    if you look at the crosstabs, pollsters ask other questions used in their analyses

    did you vote in the 08 election? for whom did you vote?

    did you vote in 2010? for whom?

    Fox poll: Obama up 5

    The new Fox News poll finds Barack Obama with a stable, 5-point national lead over Mitt Romney, drawing 48 percent of the vote to the Republican’s 43 percent.

    That’s the exact same result that the previous Fox poll, conducted Sept. to Sept. 11, found in the race.

    the most recent WSJ poll: Obama +5

    of the 10 most recent polls, romney ONLY LEADS in one: Rasmussen Tracking ….. 9/14 – 9/16 ….. Romney +2

    Fox’s Chris Wallace Snaps at Radio Host:

    Wallace tells radio host Mike Gallagher that criticizing the polls is an example of what he terms “craziness,“ and that ”no self-respecting pollster” weights their polls to make party ID even.

    “I actually did some research on this today, which is more than you’ve done,” Wallace said sharply

  13. G September 28, 2012 at 5:48 pm  (Quote) #

    Although that assumption makes sense, I have not seen enough polls that have included him in order to validate that is what is happening.

    In fact, there were a few polls taken from out west that did include Johnson and they seemed to indicate that he was pulling down the numbers from BOTH main parties… but those results did seem to agree that the pulldown was greater on Romney’s side.

    So I would be a bit careful, as Johnson is correct that some of his positions are “to the left of Obama”, while others are “to the right of Romney”…so his appeal is not limited to only one side of the fence here…

    The Constitution Party candidate, Virgil Goode, also tries to make an argument that he appeals to folks in both parties (due to his personal background of having served in office as a Democrat (1996-2002_ and then switching over to be a Republican in 2002-2008). That may turn out to be true in his home state of VA, where he is well known and liked. However, his personal views and platform seem to align much more closely with the far-right of the spectrum…so expect most of whatever support he draws to come from the right.

    Both Rocky Anderson (Justice) and Jill Stein (Green) would draw votes from only the far left, so they are more a concern for Obama.

    Then there are a bunch of smaller minor-parties that only impact the vote in certain states. These range across the spectrum. I don’t expect much of any impact from them this cycle, nor from any of the true “Independents” running this time around – which range from extremists, to clowns, to no-name folks. But once all the state ballots are printed, I may look to analyze and assess the net-total effect of “alternative options” may have on each particular state and its polling trends…

    Scientist: Also, if Johnson is included in the polls, as he will be on the ballot, Obama’s lead increases. So the polls, many of which do not name Johnson, may be underestimating Obama’s lead, not overestimating it.

  14. G September 28, 2012 at 5:54 pm  (Quote) #

    True. And if I were a pollster, I too would want to ask about how people voted in the past, to get a better “true sense” of their voting preferences.

    However, as voting is a “secret ballot” process, we are again left with simply trusting what people self-report as HOW they voted in the past …so I find it a helpful metric, but as with all polling, I realize that what people tell pollsters and what they may actual do or have done are not necessarily the same.

    So I tend to look at all those results as more of “mood trend” indicators and not true hard data.

    donna:
    G: In most polling methods, this is nearly always as a result of self-identification. Which is why such things are quite fluid.

    if you look at the crosstabs, pollsters ask other questions used in their analyses

    did you vote in the 08 election? for whom did you vote?

    did you vote in 2010? for whom?

  15. donna September 28, 2012 at 6:02 pm  (Quote) #

    another interesting factoid were the exit polls after the walker recall from june:

    the question: If the presidential election were today, for whom would you vote?

    51% obama

    44% romney (a 7 point spread, although republican walker won the recall)

    if you look at the rcp aggregate today for wisconsin:

    Wisconsin: Romney vs. Obama Obama +7.8

  16. Scientist September 28, 2012 at 6:10 pm  (Quote) #

    donna: Fox’s Chris Wallace Snaps at Radio Host:
    Wallace tells radio host Mike Gallagher that criticizing the polls is an example of what he terms “craziness,“ and that ”no self-respecting pollster” weights their polls to make party ID even.
    “I actually did some research on this today, which is more than you’ve done,” Wallace said sharply

    That sounds a bit like people here, arguing with birthers. And it will likely have about as much effect.

    G: So too, this cycle is shaping up to be a rout…and therefore they need to start laying the groundwork now, with their own fake “unskewed” polls and Baghdad Bob style prounouncements by Dick Morris, in order to build that alternate fictional reality, in which Romney really won the election and was ahead the whole time….

    Sounds like Doc will be in business for the next 4 years…

    G: In fact, there were a few polls taken from out west that did include Johnson and they seemed to indicate that he was pulling down the numbers from BOTH main parties… but those results did seem to agree that the pulldown was greater on Romney’s side.

    IMO Johnson is the only one of the minor candidates who potentially could get enough votes to swing a state. Some polls say he could get as much as 10% in his home state of NM, but that is an Obama state so he would only increase the margin of victory there. Maybe in CO, if the race tightens, he could swing the state from Romney to Obama. Conversely, if Obama’s lead widens a bit, MT and even AZ are in play and he could be the difference for Obama there.

  17. Scientist September 28, 2012 at 6:31 pm  (Quote) #

    This is the guy behind unskewedpolls.com. http://2012.talkingpointsmemo.com/2012/09/dean-chambers-unskewed-polls.php?ref=fpb

    We can think of him as the Phil Berg of skewerism, doubtless soon to be supplanted by those even more crazy. Who will be the Orly Taitz of skewerism? Will skewerism split into a deliberately skewed wing (like the born in Kenya birthers) and a skewed by mistake wing (like the Vattelist birthers)? Like birthers, I suspect skewers will believe both.

  18. John Potter September 28, 2012 at 6:40 pm  (Quote) #

    Thanks for adding the justifications of the purportedly UnSkewed. Yes, indeed, they have a perfectly logical artifice, guaranteed surefire, …. so long as you accept / can trust their base assumptions …. a long string of unsourced %ages.

    This system of double weighting should insure our survey produces very accurate results, not skewed either way for the Democrats or for the Republicans….

    … but won’t. Any takers at +1,000 that “Dunscrewed” will be the most accurate predictor of the election? C’mon, that a 10:1 payoff if they are most accurate …. 😉

  19. Slartibartfast September 28, 2012 at 6:41 pm  (Quote) #

    I actually think that this was a very strong recall effort—people tend to vote against recalls for a political purpose (rather than scandal, for instance). If you looked at the “average” recall (for which there is insufficient data), and made a metric like the measurements of the partisan bias in the states, I think you would find that the Walker recall was far closer than would have been expected generically.

    I would also point out that the recall effort was successful—it flipped the state senate, something I doubt would have happened if Governor Walker’s recall wasn’t on the ballot.

    donna:
    another interesting factoid were the exit polls after the walker recall from june:

    the question: If the presidential election were today, for whom would you vote?

    51% obama

    44% romney (a 7 point spread, although republican walker won the recall)

    if you look at the rcp aggregate today for wisconsin:

    Wisconsin: Romney vs. Obama Obama +7.8

  20. donna September 28, 2012 at 6:58 pm  (Quote) #

    Slartibartfast:

    here’s the link to edison research who conducts exit polls:

    Wisconsin Recall Exit Polls: How Different Groups Voted

    http://www.nytimes.com/interactive/2012/06/05/us/politics/wisconsin-recall-exit-polls.html

  21. US Citizen September 28, 2012 at 7:13 pm  (Quote) #

    In a recent poll of people searching online for snow mittens, 62% wish Romney would drop out of the race so they could find what they’re looking for.

  22. Slartibartfast September 28, 2012 at 7:26 pm  (Quote) #

    Thanks donna 🙂

    I noticed that the last question on the survey was: “Do you think recall elections are appropriate?” with the answers being:

    For any reason (27% of voters—9% Walker; 90% Barrett)

    Only for official misconduct (60%—68% W; 31% B)

    Never (10%—94% W; 5% B)

    That seems like a pretty stiff headwind to a recall effort to me…

    donna:
    Slartibartfast:

    here’s the link to edison research who conducts exit polls:

    Wisconsin Recall Exit Polls: How Different Groups Voted

    http://www.nytimes.com/interactive/2012/06/05/us/politics/wisconsin-recall-exit-polls.html

  23. Dave September 28, 2012 at 7:46 pm  (Quote) #

    Some of the discussion above gives the impression that pollsters just call people at random and report what they get, so that the counts of D’s and R’s they report simply reflect reality.

    Ideally, that would be true. But there are two problems pollsters have to deal with: 1) the selection of respondents slightly fails to be random in a few ways, which introduces selection effects, and 2) pollsters want their population to be people who will vote, so they each have a secret sauce for determining that. Both of these things can introduce bias into the poll. Each pollster has their own method of reweighting the poll, based on one or more demographic factors, to correct for this bias.

    So because polls inherently include possible bias followed by an attempt to correct that bias, there is the possibility that they get it wrong. Probably the best answer to these concerns is the studies Nate Silver has done, comparing polls to each other, and comparing polls to election results. His results could be summarized as, most pollsters get it right to within a few percent.

  24. Rickey September 28, 2012 at 7:47 pm  (Quote) #

    Slartibartfast:
    Thanks donna

    I noticed that the last question on the survey was: “Do you think recall elections are appropriate?”with the answers being:

    For any reason (27% of voters—9% Walker; 90% Barrett)

    Only for official misconduct (60%—68% W; 31% B)

    Never (10%—94% W; 5% B)

    That seems like a pretty stiff headwind to a recall effort to me…

    People I know in Wisconsin whose information I trust told me that many people who oppose Walker nevertheless voted against his recall because they oppose recall elections in principle.

    I’m sure that many, many voters feel the same way, because successful recalls have been very rare.

  25. G September 28, 2012 at 8:22 pm  (Quote) #

    Johnson is definitely the #3 contender in this year’s cycle, for whatever that is worth.

    However, expect Virgil Goode to be #4 and to have an effect, at least on Virginia.

    What will also be interesting is what the Ron Paul devotees do… support the Libertarians and Johnson or write-in Ron Paul’s name… so he might actually be #5, even though he’s not a candidate.

    Scientist: IMO Johnson is the only one of the minor candidates who potentially could get enough votes to swing a state. Some polls say he could get as much as 10% in his home state of NM, but that is an Obama state so he would only increase the margin of victory there. Maybe in CO, if the race tightens, he could swing the state from Romney to Obama. Conversely, if Obama’s lead widens a bit, MT and even AZ are in play and he could be the difference for Obama there.

  26. Slartibartfast September 28, 2012 at 9:13 pm  (Quote) #

    Dave,

    You’re absolutely right about sampling bias (like landline v. cellphone polls) and likely voter models being necessary to consider. They don’t make the polls wrong, but they make it a lot easier to make misleading conclusions (or conclusions which depend heavily on modeling choices rather than the data). You should also look at what a particular poll is doing with respect to its long-term average. If one poll makes a big jump it’s not really a big deal, but if all of the polls are a couple of points off of their long-term values in the same direction, it’s time to start taking them seriously. I wonder if unskewed will keep tinkering with their formula if things keep moving towards President Obama… (that’s the problem with putting your finger on the scale—you’ve got to do it more and more just to cover up the first time).

  27. Slartibartfast September 28, 2012 at 9:19 pm  (Quote) #

    That sounds like about what I would have guessed (I have a cousin in the Milwaukee area, but he would have been voting for Walker in any case). Unless you’re caught snorting coke with gay hookers in your office while taking a massive bribe, you’re pretty much never going to be recalled (this might not be the case in recall-friendly states like California, but I suspect even there it is hard to do).

    Rickey: People I know in Wisconsin whose information I trust told me that many people who oppose Walker nevertheless voted against his recall because they oppose recall elections in principle.

    I’m sure that many, many voters feel the same way, because successful recalls have been very rare.

  28. Northland10 September 28, 2012 at 9:21 pm  (Quote) #

    I would definitely be in the “Only for official misconduct” group and even that is a high bar. Living up on northern Michigan (that is the northern part of the lower peninsula for those who think the UP is northern Michigan), I saw small town politics made worse by the constant use of the recall. I am relatively certain that some were planning a recall petition before the newly elected took office. Watching a corporation back an effort in Acme due to losing a zoning fight was especially horrifying.

    We have scheduled elections and schedule changes in government which provides stability to our system of government that could so easily fall into anarchy. Of the 8 years living there, I can remember only one recall I may have supported and the person had actually been convicted of wrongdoing in office but refused to resign.

    Not being in Wisconsin by a 25 minute drive, I chose to avoid a full opinion on their effort but I did have concerns. A polling sample might have put in in the column of supporting Wilson when, actually, I despised what he did but I also highly disagree with recall elections.

    Slartibartfast:
    Thanks donna

    I noticed that the last question on the survey was: “Do you think recall elections are appropriate?”with the answers being:

    For any reason (27% of voters—9% Walker; 90% Barrett)

    Only for official misconduct (60%—68% W; 31% B)

    Never (10%—94% W; 5% B)

    That seems like a pretty stiff headwind to a recall effort to me…

  29. gorefan September 28, 2012 at 9:39 pm  (Quote) #

    “If you torture the data long enough, it will confess to anything.” Ronald Coase, Economist

    When I read polls I’m reminded of this October, 2008 memo from the McCain campaign.

    http://blogs.wsj.com/washwire/2008/10/28/in-memo-mccains-top-pollster-sees-tighter-race/

  30. Slartibartfast September 28, 2012 at 10:07 pm  (Quote) #

    I certainly appreciate the distinction—I’ve been going “up north” every summer all my life (my family has a cottage on Higgins Lake [it’s right here *pointing at the top knuckle on the middle finger of my right palm*]). In person it’s not a problem—I would just look at which hand you are holding up and pointing to. Online I just use the simplifying assumption that no one lives in the UP (come on, have you ever really seen a yooper? 😛 ).

    I haven’t seen recalls abused personally, so I’m more in the “another tool in the toolbox is good” camp, but I can respect your position. Especially in this environment where the right seems bound and determined to find a way to abuse any tool they can (look what they’ve done to the Senate with the filibuster).

    Northland10:
    I would definitely be in the “Only for official misconduct” group and even that is a high bar.Living up on northern Michigan (that is the northern part of the lower peninsula for those who think the UP is northern Michigan), I saw small town politics made worse by the constant use of the recall.I am relatively certain that some were planning a recall petition before the newly elected took office.Watching a corporation back an effort in Acme due to losing a zoning fight was especially horrifying.

    We have scheduled elections and schedule changes in government which provides stability to our system of government that could so easily fall into anarchy.Of the 8 years living there, I can remember only one recall I may have supported and the person had actually been convicted of wrongdoing in office but refused to resign.

    Not being in Wisconsin by a 25 minute drive, I chose to avoid a full opinion on their effort but I did have concerns.A polling sample might have put in in the column of supporting Wilson when, actually, I despised what he did but I also highly disagree with recall elections.

  31. richCares September 28, 2012 at 10:34 pm  (Quote) #

    predictions by 538 Nate Silver (not a poll)
    Obama 82.7% Romney 17.3%
    http://fivethirtyeight.blogs.nytimes.com/

    predictions by Intrade (not a poll)
    Obama 78.5% Romney 21.8%
    http://www.intrade.com/v4/home/

    from Meet the Press:
    Nate Silver of 538, puts Obama’s chance of winning re-election at 80 percent, has gained a reputation for his political prognosticating. In 2008, he correctly picked 49 out of the 50 states in the presidential race, as well as the eventual winner in all 35 senate races. His new book, The Signal and the Noise, looks at, as the subtitle suggests, “why so many predictions fail – but some don’t.”

  32. jayHG September 28, 2012 at 10:37 pm  (Quote) #

    Oh please….even conservatives are making funn of unskewed.

  33. jayHG September 28, 2012 at 10:41 pm  (Quote) #

    John Potter:
    The cry for fairness has quite a Harrison Bergeron feel to it.

    Also, are we talking self-identified Blues and Reds, or registered Blues and Reds, or those intending to vote Blue or Red?

    …or just whatever it takes to show rMoney in the lead?

    There’s your answer!

  34. Wile September 28, 2012 at 11:23 pm  (Quote) #
  35. Thrifty September 29, 2012 at 2:51 am  (Quote) #

    I’ve been trying to wrap my mind around the sheer stupidity of this ever since I heard of this yesterday morning. So polling is not to be trusted because, when they polled self identified Democrats, who statistically make up a majority of voters, Barack Obama comes out ahead? Because the guy who’s party matches the majority gets the vote of the majority, that’s wrong?

  36. Thrifty September 29, 2012 at 2:56 am  (Quote) #

    Scientist: Who will be the Orly Taitz of skewerism?

    Orly will probably get bored of losing in court once she can’t make any more ballot challenges, so I nominate Orly. Besides, she keeps trotting out the claim that there are millions of illegal immigrants voting fraudulently. There’s no doubt she’ll claim voter fraud if Obama wins a second term.

  37. The Magic M September 29, 2012 at 7:26 am  (Quote) #

    G: because in their minds, there is simply no way that America can elect that black democrat with the funny name…not only once, but twice!

    Especially since most of them have been deluding themselves for at least a year that a vast majority is “disappointed with Obama” and that therefore Romney will win by a landslide.
    Projection, projection, projection…

    Dr. Conspiracy: In a recent survey, 76% of Americans said they didn’t trust public opinion polls.

    According to a poll I just made up, 57% of Americans believe poll results are just made up anyway. 😉

  38. James M September 29, 2012 at 11:10 am  (Quote) #

    Abstention decides nearly every election. If I had a lot of money to pour into a campaign, I would definitely work to create a sense of complacency among the opposition with the objective of increasing voter apathy and helping to skew the abstention in my favor.

  39. Thrifty September 29, 2012 at 11:39 am  (Quote) #

    The Magic M:
    According to a poll I just made up, 57% of Americans believe poll results are just made up anyway.

    Abraham Lincoln famously said “You can’t trust everything you read on the Internet.”

  40. Thrifty September 29, 2012 at 12:50 pm  (Quote) #

    Well I doubt that old fashioned “omg he’s not a natural born citizen” Birtherism will die until President Obama is out of office. Even then, who knows? Birthers have their belief in the magic reset button, so they might be fighting even after he leaves office.

    Cries of voter fraud will be in there too. These voter ID laws have plenty of folks up in a frenzy about the perception of rampant voter fraud. These latest claims about polls being skewed are just further fuel for the fire. Orly will of course claim voter fraud, as she is constantly whining about how she was cheated in the Senate primary, and about how there are millions of fraudulent ballots. Mark my words; in a 2nd Obama term, Orly will add “he used millions of fraudulent ballots to get re-elected” to her litany of conspiracy theories.

  41. Joey September 29, 2012 at 1:12 pm  (Quote) #

    For the record, 2008 general election exit polls found that 10% of registered Democrats voted for John McCain (“Reagan Democrats”) and 9% of registered Republicans voted for Barack Obama (“pro-choice Republicans- in- name- only”).

  42. US Citizen September 29, 2012 at 1:13 pm  (Quote) #

    One ironic point is that if Romney loses, he’ll be known as the candidate who couldn’t even get *himself* a job.

  43. Plutodog September 29, 2012 at 2:58 pm  (Quote) #

    The thing that I think needs paying attention to is the bandwagon amongst some of the playas on the right — Rove, etc., to claim polls are skewed to Obama. Take that along with the fact of all the hard work Reethugs have done in the run-up to this election to suppress the Dem vote through tighter voting windows, new ID requirements. Also throw in what will happen in terms of “poll watching”, and other skullduggery. Assuming the Reethugs are able to thus dishonestly “win” this election despite pervasive polls suggesting a loss, they are going to want to be able to cite some “proof” the election returns are “fair”.

    If the polls continue to show Obama decidedly ahead in popular and (especially) the electoral college computation but Obama “loses”, there are going to be a helluva lot of pizzed off people and a stink to high heaven. They need some coloration of rational reason for a crooked win.

    I don’t know that there’s ANY possibility of an aggressive investigation that reverses an election based on finding proof positive, smoking gun evidence of Reethug fraud, but they sure don’t want to go there if they can avoid it. And understandably, there would be sentiment against such riotous conditions amongst a lot of people who were neutral or Obama voters too. The whiff of possible violent rebellion reasonably scares a lot of folk.

    Bottom line, I think the Reethugs have reason to want to introduce doubt to losing polls before, during and after election night. Dems/Obama have reason to debunk this as thoroughly as possible as soon as possible, make it clear that only crazy fringe would buy it for a second.

    Mock and ridicule and out-fact the Gallaghers and their ilk mercilessly and repeatedly. Kill this dead. Kill it deader than dead. Whack it, slice it, dice it, grate it, grind it, sear it, cremate it, piss on the ashes, picket the funeral, give the funeral a funeral. Nuke any return by skewed poll zombies. Allow no memorials, no grieving, no “fair and balanced” retrospectives.

  44. Slartibartfast September 29, 2012 at 3:01 pm  (Quote) #

    This. Then rinse and repeat.

    Plutodog: Mock and ridicule and out-fact the Gallaghers and their ilk mercilessly and repeatedly. Kill this dead. Kill it deader than dead. Whack it, slice it, dice it, grate it, grind it, sear it, cremate it, piss on the ashes, picket the funeral, give the funeral a funeral. Nuke any return by skewed poll zombies. Allow no memorials, no grieving, no “fair and balanced” retrospectives.

  45. Paul Pieniezny September 29, 2012 at 10:47 pm  (Quote) #

    Thrifty: Orly will probably get bored of losing in court once she can’t make any more ballot challenges, so I nominate Orly.Besides, she keeps trotting out the claim that there are millions of illegal immigrants voting fraudulently.There’s no doubt she’ll claim voter fraud if Obama wins a second term.

    And the way she proved that was by gathering “evidence” (orlyvidence) that in past elections, dead people had voted.

    Dead people in Orly’s eyes meant everybody over the age of 100.

    But most of the people on that list were far older than 100 when they voted for the last time.

    I recognized one name as definitely Belgian and started to look on genealogical sites and googling for certificates and affidavits. I soon got lucky: the lady had died during the last two years – a local paper announced the name of the executioner of her will, and yes she died in the USA and her property was in the USA. Shirley if that lady had died at the age of 180, that would have been in the paper too?

    In other words, Orly’s data was garbage data. Like the document she uses to prove the real Barack Obama was born in 1890. Even though the same document also says born in 1990. (one wonders why Orly never used it to claim Obama is not old enough to run for president).

    It is not like the “detective” who gathered that garbage should not have known that it was garbage. All those people who should have been dead strangely only started to vote during the last 20 years of their lives. The Belgian lady was even older (probably late naturalization).

    Funny thing is that being Belgian, rich and not living in Illinois or Michigan she almost certainly voted Republican. Your typical illegal immigrant, what.

  46. Scientist September 30, 2012 at 7:44 am  (Quote) #

    Nate Silver with what should (but probably won’t) be the last word on this topic.

    http://fivethirtyeight.blogs.nytimes.com/2012/09/29/poll-averages-have-no-history-of-consistent-partisan-bias/

    The parallel with birtherism is very clear-I don’t want Obama to be eligible, so he isn’t; I don’t want Obama to be winnng re-election, so he isn’t. Facts? We don’t need no stinkin facts.

  47. JPotter September 30, 2012 at 1:44 pm  (Quote) #

    Great series of cartoons by Cagle touching on this topics:
    http://www.cagle.com/2012/09/my-romney-cartoons-things-not-looking-good-for-mitt/?utm_source=daily&utm_medium=newsletter&utm_content=cartoon-comment&utm_campaign=093012

    Also, this morning on Face the Nation, in a roundtable involving a left- and rightwing debate consultant and a polling guy, the wingnut dismissed the poll guy’s recap with a passing comment about how “these polls are being weighted” …. and then proceeded to tell the pollster what he was seeing in his polling. Still stuck in the denial stage…

  48. JPotter September 30, 2012 at 1:52 pm  (Quote) #

    JPotter: Great series of cartoons by Cagle touching on this topics:

    And piling on the cartooning:

    http://www.cagle.com/news/romney-poll-numbers/

  49. Dr. Conspiracy September 30, 2012 at 9:53 pm  (Quote) #

    Following up on the YouGov widget on my sidebar, here’s how they explain it:

    YouGov trackers are carried out online using the YouGov US panel of 1.2 million respondents. Respondents were selected from YouGov’s panel using sample matching.

    The sample was weighted using propensity scores based on age, gender, race, education, news interest, voter registration, and non-placement on an ideology scale.

  50. Thomas Brown September 30, 2012 at 11:00 pm  (Quote) #

    So… What do the Skewerists say about the very same pollsters consistently find Romney ahead in bright-red States?

    What, are these polling companies only skewing the results in blue states? Or just the competitive States, but not blue States? My head hurts, trying to follow their stupid.

  51. JPotter September 30, 2012 at 11:09 pm  (Quote) #

    Skewer the skewerites.

    Not intended as an endorsement of political violence. Symbolically, conceptually skewer them. I in no way desire to see a reenactment of the final scene from Spartacus. With stupids skewered from sea to shining sea. No, sir, nope, not me.

    Ahem.

  52. The Magic M October 1, 2012 at 11:32 am  (Quote) #

    JPotter: Skewer the skewerites.

    I first read “Screw the skewerites”. As in “attached to another object by an inclined plane, wrapped helically around an axis”. 😉

    One may wonder: were similar arguments made during the 2008 pre-election times? Or when Clinton became President?
    Or is this a typical symptom of “oh noes, the black guy cannot possibly be RE-elected, can he?”?

  53. G October 1, 2012 at 12:37 pm  (Quote) #

    Pretty much, yea.

    The Magic M: Or is this a typical symptom of “oh noes, the black guy cannot possibly be RE-elected, can he?”?

  54. John Potter October 1, 2012 at 1:14 pm  (Quote) #

    The Magic M: “Screw the skewerites”.

    Works for me! 😀

    The Magic M: One may wonder: were similar arguments made during the 2008 pre-election times? Or when Clinton became President?

    Yes, always. Always denial. When was the exact moment the public voice of the Reds succumbed to infantilism? I think it was a process. Started somewhere in the Bush I years, and was cemneted by the election of Clinton. The brains let the nuts in during the Reagan Dream Times and the rot commenced. For years now, listening to the right has been like playing Chuck Yeager with a indestructable plane (Pull up! *BOOM* Pull up! *BOOM* Pull up! *BOOM* …). Now the stupid has been exported; did you catch Netanyahoo channeling the spirit of Bush/Cheney at the UN? Why isn’t Bibi wearing a cowboy hat yet?

  55. Wile October 4, 2012 at 9:34 am  (Quote) #

    So, it appears last night’s CNN snap poll showing the President getting walloped….only polled white southerners over fifty.

    A conspiracy to try and tighten the race?

  56. Wile October 4, 2012 at 9:55 am  (Quote) #

    Wile:
    So, it appears last night’s CNN snap poll showing the President getting walloped….only polled white southerners over fifty.

    BTW, this appearance appears to be misleading.

  57. John Potter October 4, 2012 at 10:11 am  (Quote) #

    Wile: So, it appears last night’s CNN snap poll showing the President getting walloped….

    Listening to it reminded me of the discussions here. Realists have the handicap of being limited to reality. Nut are free to say “nuh-uhhhhh!” are continually rattle off their preferred laws of physics.

    My irony meter finally blew when rMoney insisted he was used to people telling him something repeatedly in an effort to make him believe it.

    That one cuts both ways … speaking to the persistence of reality, and the doggedness of denial.

  58. James M October 4, 2012 at 10:24 am  (Quote) #

    John Potter: Listening to it reminded me of the discussions here. Realists have the handicap of being limited to reality. Nut are free to say “nuh-uhhhhh!” are continually rattle off their preferred laws of physics.

    My irony meter finally blew when rMoney insisted he was used to people telling him something repeatedly in an effort to make him believe it.

    That one cuts both ways … speaking to the persistence of reality, and the doggedness of denial.

    He is used to his own *sons* being liars.
    He won this debate by calling his five sons liars and by saying he loves Big Bird enough to kick him out of the nest.

  59. John Potter October 4, 2012 at 10:36 am  (Quote) #

    James M: He won this debate

    I hope you’re not conceding that. In 1980 and 2000, the people went for the pie in the sky. Such pies are very expensive.

  60. donna October 4, 2012 at 11:24 am  (Quote) #

    best quote: romney’s appearance was “DEBATE & SWITCH”

    his OWN SPOKESPERSON came out 10 minutes after the debate and SAID ROMNEY WON’T BE COVERING THOSE WITH PRE-EXISTING CONDITIONS

    snap polls on after the bush/kerry 1st debate:

    cnn poll after the first debate in 2004?

    53 kerry/ 37 bush

    newsweek: 61 kerry.19 bush/16 undecided

    latimes: 54 kerry/15 bush

  61. James M October 4, 2012 at 11:36 am  (Quote) #

    donna:
    best quote: romney’s appearance was “DEBATE & SWITCH”

    his OWN SPOKESPERSON came out 10 minutes after the debate and SAID ROMNEY WON’T BE COVERING THOSE WITH PRE-EXISTING CONDITIONS

    snap polls on after the bush/kerry 1st debate:

    cnn poll after the first debate in 2004?

    53 kerry/ 37 bush

    newsweek: 61 kerry.19 bush/16 undecided

    latimes: 54 kerry/15 bush

    Hundreds of millions of people see the debates or hear snippets of them.
    Hundreds of thousands of people hear the retractions and clarifications.

  62. James M October 4, 2012 at 11:37 am  (Quote) #

    John Potter: I hope you’re not conceding that. In 1980 and 2000, the people went for the pie in the sky. Such pies are very expensive.

    Of course I am. The question is, did the debate result in changing the expected outcome in PA, CO, NV, OH, FL, MI, etc?

  63. The Magic M October 4, 2012 at 11:47 am  (Quote) #

    John Potter: Now the stupid has been exported; did you catch Netanyahoo channeling the spirit of Bush/Cheney at the UN?

    Obviously, since “they have WMD, we have to stop them” has worked once, it will work again. Even if it again should turn out to be untrue. I wonder why Schicklgruber never thought of that…

    What always strikes me is why those countries still believe that playing along will make their position *stronger*. Saddam could’ve easily played the innocent victim, yet he chose to obstruct the investigations so bluntly that every idiot would get the message “I have something to hide”. Now Iran is doing pretty much the same and Syria with its alleged chemical weapons looks poised to follow suit.

  64. John Potter October 4, 2012 at 12:28 pm  (Quote) #

    James M: Of course I am.

    Reviving shades of Nixon/Kennedy … I listened to it on the radio. With nothing to look at, only words to listen to, I got a very different impression!

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.