Is this an outlier? The MJ crowd as well as the show on the Clinton News Network are busy this AM conducting Trump's funeral. pl
http://www.reuters.com/article/us-usa-election-poll-idUSKCN10G2BQ
RSS- Iran launches air attacks on Israel
- ISW on the developing Ukrainian defense industrial base and Russias strikes on the Ukrainian energy infrastructure
- Britain? Kaput! By Walrus.
- Declaration “Dignitas Infinita” on Human Dignity
- The situation and some possible courses of action … A Pat Lang post reposted from 26 February 2022
- HARPER: COL. W. PATRICK LANG R.I.P.
- A Note On Our Folly By Walrus
- Israel attacks Iranian embassy in Damascus
- He is risen…
- The Polish Proa Fleet
A few thoughts:
-one elite Republicans on TV aren’t the same as the Republicans in even Prince William County. TV Republicans get more media time, but they represent a class of appointed Republicans who don’t know their place or even if they have a place in the Trump order of things. Trump’s only real competitors have been Cruz and Carson both fairly religious guys from movement conservative backgrounds. Recent elections (Romney and McCain) have shown Republicans come home. They are still a shrinking party. As far as Hillary appealing, my belief is the GOP id is dedicated to despising Democrats. They simply can’t crossover.
-the permanent GOP political class is awaiting a paycheck Trump isn’t producing. They are annoyed.
-voters are partisan, much more partisan than they would care to admit. They know little about issues a day rely on a mix of nostalgia and dog whistles. The ignorance for the voting behavior reinforces their stubborn nature. If they leave now, why didn’t they leave then when they should have known how awful the party was?
-The Clinton Democratic Party is also shrinking. There isn’t much demand for a faux multicultural neo Republican Party given the GOP already exists. The popular revolt that brought Dean to the head of the DNC reversed the trend as Democrats ran and won on popular issues (imagine that) and took advantage of a new voter boomlet going through the population, but Obama jettisoned this popular revolt as soon as he was elected in favor of Clinton Inc elites (Rahm Emmanuel). Fear, nostalgia, and blind faith (ex. Obama is workin’ hard when he doesn’t golf) hold the Democrats together hence the apocalyptic language about Trump so early. These factors don’t draw in new voters, and Democrats have demonstrated the 2006 and 2008 voters might not be there. Fear worked in 2012 but failed miserably in 2014. The old adage about flies and honey is still true, and Hillary will never be able to offer honey because she is perceived as untrustworthy, anyone entering her orbit is deemed untrustworthy.
Hillary isn’t going to soar, and Trump isn’t going to collapse. If Hillary was an even tolerable candidate, she would already already be President.
The election can still be won or lost because the margins in states that matter aren’t very large and weren’t very large in 2012 despite reports of Obama’s commanding electoral college victory.
The recent polls that have shown Clinton with a large lead over Trump had poor sampling methodology.
Surveys conducted over the past few years have been pretty consistent about the proportion of Democrats to Republicans in the country. The resulting being that there are 1% or 1.5% more Ds. However, in the recent polls, the samples have consisted of 10% more Ds. That alone can explain the Clinton “lead”.
There are other interesting confounds in the recent polls; one such being a high “Undecided” proportion. One of the major polls had it 19%. With 19% undecided, things could go either way in November.
Some of the other confounds are real, but a little technical. The above should sufficiently illustrate why the polls should not be believed. I have not seen the methodology behind the most recent Reuters/IPSOS poll that contradicts the previous week and has Clinton only up by 3%, which is within the margin of error and therefore a statistical tie. Maybe someone screwed up and did the sampling right for once.
I don’t understand what the MJ thinks it’s accomplishing by quoting poorly done poll numbers, in August nonetheless. Is it to demoralize? Is it mindlessness and stupidity? Childishness? My more paranoid side thinks it’s setting the stage for manipulation of the voting machines in November (i.e. why are you surprised Hillary won? She was always ahead in the polls).
If the best cherry-picked poll for Trump shows Clinton +3, Trump is in trouble. Compare:
http://projects.fivethirtyeight.com/2016-election-forecast/
Colonel LANG
Apparently Trump’ notice of obituary was twitted this morning by no other than your favorite morning TV personality, JOE S. I think Borg is getting worried and don’t want to miss/ leave/ give any slight possibility/chance to Trump’ election.
” Citing “multiple sources” in a series of posts to Twitter, MSNBC host Joe Scarborough was the first to announce McMullin’s bid. He said McMullin will kick off his campaign Monday and has “the backing of key $$ contributors in the Republican Party.”
“Joe Scarborough ventured into Bill Kristol territory Monday morning, saying on MSNBC’s “Morning Joe” that unnamed sources tell him an unnamed independent presidential candidate will enter the race at this late stage to try to stop Donald Trump.”
https://www.washingtonpost.com/news/the-fix/wp/2016/08/08/joe-scarboroughs-mysterious-new-independent-candidate-is-evan-mcmullin/
If it doesn’t gel, it isn’t aspic, and this ain’t geling.
http://www.ipsos-na.com/news-polls/pressrelease.aspx?id=7333
Col.,
Poll averages have it at about 7%. Looks like it’s within one stddev, but probably a bit low.
The electoral map is more interesting (and, in fact, should probably be the only serious way of discussing the likely outcome). 538 has, IMHO, the best projections. http://projects.fivethirtyeight.com/2016-election-forecast/
Even with a modest national lead, Trump is still less likely than Clinton to win. Right now it looks like an electoral vote landslide.
Two things going for Trump….First he is not a hypocrite Second he does not crack under pressure. He is mentally stronger.
Presidential elections is all about mental toughness when everything else is equal.
Zero Hedge has an interesting graph with Clinton lead peaking in the latest Reuters poll. For six days Trump numbers have been going up and Clinton’s downward….
http://www.zerohedge.com/news/2016-08-06/peak-hillary-reuters-baffled-clintons-lead-over-trump-suddenly-evaporates
Next
I have written off this election. We’re fucked either way.
You can spend hours trying to figure out what got us here, but where we sit is that we are a country making a choice between the worst two candidates ever to grace the center ring of this quadrennial circus. Neither major party candidate is worth a bucket of warm spit.
In a nutshell, no matter who wins, we are going to have a clown in place for the next four years, with an idiot running the Administrative branch for the next four years. The odds of the country actually improving will be minimal, the chance that things will either not improve or get worse is increased.
So, what comes in two years when the next election is held? Who will take over the house and how? What happens in four years when someone competent captures the anger of the Trumpeteers? What happens when the Democratic party continues spitting on the poor, who will rise out of that?
Look, we are fresh out of problems here in the US. All we have left is predicaments.
What happens next?
I wrote this over at my place (http://mightaswellliebackandenjoyit.blogspot.com/)
IMO, HRC is one more mini-stroke from campaigning from a hospital bed. If and when WikiLeaks whips out their October surprise, who knows if her narcissistic self will be able to physically handle the shock? Libertarian, anyone?
Noise. Signal noise. How can it be otherwise when the choice presented by the ruling classes is between bad and worse – “would you rather be hit on the right side of the head or the left?”
The whole spectacle is depressing. I have work to do. Actual work actually making something for my family and clan’s future. I’m sick of these worthless parasites and their filthy lies on behalf of their paymasters, creating chaos and misery for personal power and wealth.
May I suggest Prof. Sam Wang for accuracy in polling and predictions? His methodology is based on state (not national) polls.
http://election.princeton.edu/
Cheers!
Sir,
You gotta look at the metrics in these polls. A lot of the are oversampling Dems by double digits, undersampling men and independents, and geographically focusing on Dem strongholds.
Meanwhile 15k people show up to his Florida rally while Hillary can’t fill the floor of a HS gym.
Polls only gain predictive power a couple of weeks out. Nicholas Nassim Taleb amusingly ripped Nate Silver for ignorant use of stats to go from predicting a Trump win at 55% to 20% 10 days later.
I think it will come down to the debates, Trump has more charisma, far superior policies and it will be his opportunity to present himself without the distorting impact of the media. Until then he just needs to be in touch and most polls show that is the case, allowing for the rough patch.
Col: Following up LondonBob’s comment, Nassim Nicolas Taleb has been addressing the risks of–and inaccuracy of–this constant polling. See https://twitter.com/nntaleb/status/762048739334885380.
Beyond the jargon, it’s pretty clear that the more pollsters granulate their polis, the more arbitrariness they inject into them.
And, frankly, how do you poll these profound differences from 2012? (1) There is significant revulsion at both candidates, which one would think would imply lower voter turnout; (2) There is much more obvious excitement among Trump supporters; and (3) the rapid rise of non-affiliation with either party.
The more the elites from both parties attack Trump the more obvious it is that he is the real change candidate. I have no idea how this election will turn out, but then neither do the “pros.”
bks, Nate Silver at 538 has sold out to the NY Times and Bill Kristol. He contracts with both. Yes, yes, he appears to have been pretty good in 2008 and 2012. That was before selling out to Borg Propaganda Central.
FYI –
http://www.dangerandplay.com/2016/08/06/hillary-clinton-stroke-seizure-coughing-fits/
The technician who repairs the large format plotter in my office says that the voting machines are very secure. He had training on them when he was thinking about taking a job with that company. I don’t remember the specifics but he’s a very smart fellow and I trust his judgment.
Trump was also ahead in the GOP primary polls. In fact, the pollsters did very well in predicting the state results during the primaries for both parties. There were a few mistakes, but that’s what one expects of real data. The idea that the polls are slanted and the election rigged is tinfoil-worthy.
https://www.yahoo.com/news/evan-mcmullin-independent-presidential-campaign-000000187.html
Ex-CIA officer launches presidential campaign aimed at thwarting Donald Trump
A former CIA case officer who has served as top policy aide to House Republicans is launching an independent campaign for the presidency on Monday with the backing of veteran GOP strategists and donors determined to block Donald Trump from getting anywhere near the White House.
Feeding off mounting discontent within GOP ranks over Trump, Evan McMullin — who is resigning today as chief policy director for the House Republican Conference — will announce his campaign with an open “Letter to America” aimed at rallying independents and conservatives to his cause…
The polls (including the much accoladed and repeatedly incorrect this year, nate 538 silver) and the media were overwhelmingly in agreement with themselves one day out on Brexit). I think voter turnout is key and also honesty with poll takers, particularly when the media trash one side as ignorant buffoons.
The recent rash of publicized opinion poll results that say that Donald Trump is behind Hillary Clinton are described in terms of results only, that Hillary is ahead by ‘X’. To evaluate an opinion poll includes obvious questions, such as: how many people were in the sample? what were the characteristics of the people in the sample? where were the people in the sample located? was the survey by phone, in person, or by computer? what were the questions asked? what statistical formulas were used to arrive at the prediction? And so on.
There is no way to know if the polling results are valid within the norms of opinion research, or not. Unfortunately, with the present sad state of media conduct, one cannot rule out that some of the “results” are just disinformation.
Related to the recent topic of health exams for the presidential candidates, there is additional information of interest.
Hillary appears recently to have a black man near her who may not be in the U.S. Secret Service. Particularly intriguing is the behavior of both of them at a recent rally. This brief clip from ABC News down the web page a little shows the black man going up to the podium and he is saying, “Are you OK?” “Keep talking.” “You handled it.” “We’re not going anywhere.” She then starts talking again. The three men who also go up to the podium look like Secret Service in their job of scanning for threats; but not the black man–
http://www.thegatewaypundit.com/2016/08/oh-hillary-handler-carries-diazepam-pen-seizures/
Further down the web page is a photograph of what looks like the same man near her on the day of her speech at the Democratic Party national convention. In his left hand is an object that has created speculation that it may be a type of medical drug dispensing device.
This next link contains a photograph from Reuters and Getty of Hillary getting assistance going up some stairs, taken from the back. The same black man may be on her left helping her to walk up the stairs. The Drudge Report headline aggregator has that photo, and in a more circumspect manner, he links to four stories with his lines being: “2016: Hillary conquers the stairs… 2012: Falls at home, blood clot… 2011: Falls boarding plane… 2009: Falls going to White House, broken elbow…”
http://theamericanmirror.com/shock-photo-grandma-hillary-helped-stairs/
what is it with you people. I have foolish left leaning friends that parrot the exact same thing. Either the sampling was done right or not. It’s obvious to anyone with a moderate understanding of basic logic that the samples were wrong. There are 1% more Ds than Rs in the country, yet the big polls were sampling with 10% more Ds.
Is this too confusing for you? Why?
bks
Nate Silver completely botched the Republican primary forecasts. I would take their projections with a grain of salt.
https://mishtalk.com/2016/04/20/piss-poor-analysis-by-financial-times-nate-silver-others-on-trumps-chances-of-winning/
Degringolad, As Col. Lang has pointed out, the big difference is the Supreme Court. With Trump’s appointees, the country can always bounce back. With Clinton’s we will never be the USA again.
A Borg financed candidate to peel some Trump voters is not gonna work in this election. Many in the Borg don’t get this election and the attraction to Trump by a segment of our population.
IMO, there’s not a lot of undecided voters. Everyone is locked and loaded. The question is which side has more motivated voters who will show up.
Excellent summary, particularly on Obama/Immanuel destroying the successful organization that Howard Dean had built.
Normally I would agree however it is clear everything, including the kitchen sink, is being thrown at Trump, from multiple candidates to split the vote to censoring by twitter, so why not manipulated polls? There is clearly something wrong with polls that clearly overstate Democrats and when a highly reputable pollster like the LA Times/USC Dornsife has Clinton one point ahead, or others that show the battleground states tied, and some have her thirteen ahead. Well outside an acceptable margin of error.
First, don’t pay attention to polls until after Labor Day.
Second, don’t pay attention to individual polls, look at polling aggregates.
Third, any time someone talks about “unskewing” polls remember that they failed horribly four years ago.
New poll today shows Clinton has a 13% lead. Have at it:
http://www.realclearpolitics.com/epolls/latest_polls/elections/
The Borg is sure throwing everything they got at Trump in the hope something sticks. My good friend who lives in Nevada (a battleground state) has told me that ads attacking Trump are running on all channels and especially on Spanish language channels including radio. They’re already at saturation levels. But it is only increasing the resolve of Trump voters to make sure they vote.
I think polls can be valuable if one discards the outliers and build a moving average weighting higher the more recent ones. That should provide a sense of the trends. A big problem with the polls this year are their focus on historical data to construct their models. Those models don’t work as well in years when there are changes to underlying social trends. And the big one is the anti-establishment mood right now all across the world.
After Labor day when the campaigns get going in earnest the classic question of “are you better off now compared to 8 years ago?”, can’t be answered in the affirmative for 90% of the voters and in particular voters in flyover country. If Trump does what Carville did for Slick Willy and emphasize “its the economy, stupid” and then hammers home the Borg Queen’s judgment on foreign affairs he could get much more competitive than where he seems today. There’s a lot of time left in this election and it will be interesting to see what the situation is at the end of September.
Then the plotter technician is naive. See http://blackboxvoting.org, THEE go-to site.
Bev Harris, the head of blackboxvoting.org, has educated Congress, Election boards, and does not report anything that cannot be proved with her massive data collection. Over ten years ago, she and her volunteers dumpster-dived Diebold’s garbage to retrieve thousands upon thousands of line of code that made up the election machines.
If you’re going to trust anyone’s judgment, trust hers.
Nate Silver’s forecasts are IF THE ELECTION WERE HELD TODAY. If you notice the graph on his site, the big change came in the last two weeks. That could change, could reverse, or get worse. Who knows?
Actually, it was VP candidate, Tim Kaine, Hillary’s running mate, who was put in charge of the operation, Howard Dean’s “50-State Strategy” (contesting even in states that were deemed “too red” to have a chance). Ol’ Tim promptly took it around behind the barn and put one between its eyes.
The Democrat party (and the Republican party, for that matter) are highly ambivalent about voter recruitment. Best not to give the dirty, unwashed masses any hope that their voices may be heard and responded to. This would only potentially disrupt the progress [sic] toward that globalist, corporatist, neo-liberal utopia that both parties’ elites want to see come into full efflorescence. Populist outbursts such as those occasioned by the candidacies of Sanders and Trump are to eschewed, or destroyed by any means necessary. Hence the closing of the ranks of the .01%, the NeoCons, and the NeoLiberals in opposition to Trump, as he in some ways embodies the abominated populist sentiments.
BabelFish, you were perhaps thinking of Obama for America, an organization that was quickly euthanized when its intended purpose of fooling Obama voters into believing that their progressive views might be listened to after the election had run its course. The last thing that Obama, or especially Rahm, wanted to see was the perpetuation of a forum for “f’ing retards” being encouraged.
Then, when OWS arose and seemed to be gaining some traction, Obama coordinated with federal, state, and local authorities to shut it down.
Hmm, do I detect a pattern here?
BabelFish,
That was done long before Bernie figured out the system is “rigged”. We experienced that in Michigan in the 2008 race when the state was essentially sidelined by party operatives:
“… the Democratic National Committee determined that the date of the Michigan Democratic Primary violated the party rules and ultimately decided to sanction the state, stripping all 156 delegates and refusing to seat them at the convention.”
https://en.wikipedia.org/wiki/Michigan_Democratic_primary,_2008
My fellow democratic party members in this state conveniently forget this fact.
Molehill is certfiable. Not just tin foil hat crazy.
bks,
Romney believed all his polling too.
NotTimothyGeithner,
Many of the sort of Clinton supporters who show up on blogs speak real contempt and resentment towards the Sandervoters. The Clinton voters appear to be sore winners at the very least. Their behavior towards Sanders people deepens and renews and refreshens the ever-flowing springs of ill will flowing right back from the Sanders people.
The Clinton people make very clear their feelings that we OWE Clinton our votes because First Woman President EVER and redeeming the insults of 2008. They also make very clear their feelings that they don’t NEED our Sexist Bernie-Bro White Privilege College-boy Liberal votes anyway. How this will affect the final voting decisions of Sandervoters on Election Day is unknowable to me. I only know myself that I will not vote for Clinton. I don’t know what I WILL do.
Half those people at rallies won’t vote anyway.
Mark Gaughan,
I don’t believe that. If a vote is not cast on a Legal Paper Ballot, that vote can be faked, flipped, reversed, erased any which way. Digital fraudmasters can change, delete, erase, discount votes anywhere in the process from touchscreen casting to digital tabulation and announcement.
Digital MEANS fraud. The fraud goes in before the name goes on. Can any True Hackmaster here tell me I’m wrong?
BabelFish,
If this is true, the question arises as to how physically body-fit and stamina-healthy Trump is. If he is fit and strong and could campaign and campaign and campaign, perhaps he and his planners could schedule the kind of campaign appearances which place the heaviest possible physical burden on Clinton as she tries to keep up. Having his own private plane and being accustomed to traveling in it could give him an advantage in doing that. Early morning appearance in NYC, then very early morning appearance in Ell Ayy or San Frisco, then mid-day appearance in Texas, then evening appearance in Minnesota where the days are still longer, and back to California again. Doing phyically long-term-sustainable versions of that would pressure Clinton to either keep up and die trying, or not even try. Either would look bad for Clinton.
Did I think that up all by myself? Not really. I remember hearing at a talk somewhere about the Brezhnev-to-Andropov succession that Andropov knew how sick and weak Brezhnev was. So Andropov started doing the Communist Inner Party version campaign appearances all over the USSR to force Brezhnev to wear himself out trying to keep up and die sooner. The lecture-giver said it may have kind of worked. We will never know if
Andropov literally ran Brezhnev to an earlier-than-otherwise death, but Andropov was certainly trying to do that. In the same way, Trump could get Clinton running running running in the hamster wheel . . . and see what happens.
bks – I have determined that you are not a serious person; rather, a childish agitator.
I will try one more time. Polls can say anything based on methodology. If you had taken the time to read the link that you posted you would have seen that it suffers from EXACTLY the methodological issues that I have already explained (and that Tyler did elsewhere on this thread).
The Monmouth poll states that it small sample of 803 registered voters consisted of 26% Rep, 39% Ind and 35% Dem – that is not a representative sample and it gave Clinton an automatic artificial advantage of 8% or so simply by inflating the proportion of D to R by 9% when it should have only been 1%. Furthermore, the 39% independent is highly questionable.
If you want to extrapolate a Clinton win from a skewed sample of 803, have fun and enjoy. But don’t be surprised in Nov
Here is a new signed “poll” Former Republican serving Intell officials:
http://www.nytimes.com/interactive/2016/08/08/us/politics/national-security-letter-trump.html?_r=0
In any case the hope is that the notion and process of democracy embodied in the Constitution/preparation of will survive the voters 100 days from now and what passes for the Dem/Rep political parties entering the 21st century.
Related is the honest question of trust within the US government. Ie, what is the reason she and apparently other SecSTATE’s communicate via “private” servers rather than use fully functional in place White House servers?
My question is what was she or those that either advised her or directed her attempting to ‘avoid’? Eg, internecine (DOS/DOD) or other smoke-stacking games? If internecine, what are we looking at vis-a-vis eg, civilian vs military/MIC?
That virtually all electronic communication (civilian/gov/military) is vulnerable if not in minutes then in days or weeks — from within the government and outside domestically/internationally — is surely a KNOWN. So pray-tell what silliness/gross incompetence/cataclysmic whatever is the server drama cover for?
Well said. The people who run the polling organizations think about potential and real problems with the polls for a living. Every criticism that I read above had already been discussed, ad nauseam, when I worked in survey research at U.C. Berkeley in the late 1980’s. While I’m sure there are frauds and mistakes, there are also a lot of people doing serious political science who are trying to get it right. Trump has has a bad fortnight, but there are 6 more fortnights before the election.
All,
I will admit to being totally sick of this election cycle like some here. I am 67 and first voted in a presidential election in 1972 when I was first eligible. I have voted in every election since then. I found this piece very apt, since I too can recall the Eisenhower campaign in 1956, although I was only seven. See Col. and Prof. Andrew Bacevich on this election and the last 60 years: http://www.tomdispatch.com/blog/176172/
To all,
There is much about Donald Trump that reminds one of Justin Trudeau (Canada’s current Prime Minister in 2016).
Take a look at this chart depicting “the evolution of voting intention” during our 2015 Canadian elections:
https://en.wikipedia.org/wiki/Canadian_federal_election,_2015#Leaders.27_debates
Trudeau is represented in RED.
—
August 2015…he trails the leading New Democratic Party by around 6 or 7 points.
The “analysts” agree. Trudeau suffers because he does not follow standard operating procedure. He insisted on skipping parliament (and hence, television). Rather, he spends his time “on the ground” meeting various groups & rallying the faceless plebs.
What television coverage he does receive is mockery for his gaffes. He praises the efficient Chinese system of government, visits mosques & embraces unattractive bearded men, has campaign events touting his physical attractiveness to women, etc.
—
September 2015…he narrows the gap to 1 or 2 points.
The televised debates are occurring. This, and not summer, is when people really start paying attention.
—
October 2015…he starts to pull ahead by around 5 points, ending with an almost 10 point victory over his nearest rival.
My suspicions were correct (I posted here long ago on this matter, won’t rehash). My early warnings to the NDP & Conservatives were not heeded. Trudeau was always a national phenomenon. The polls & models have trouble capturing such men in the early days. The quality of early respondents & context of blanket media hostility conspire to blind many “analysts”.
—
August 2016…the SST commentator ‘Freudenschade’ points out that Donald Trump trails by around 7 points in the RCP.
The “analysts” agree…
Best,
Paul
To all,
The correct link to the voter intention polling cited above: https://en.m.wikipedia.org/wiki/Canadian_federal_election,_2015#/media/File%3AOpinion_Polling_during_the_2015_Canadian_Federal_Election.svg
I read this the other day and passed on to several acquaintances. Bacevich is one of the best commentators out there. Earlier this yr I viewed him on Charlie Rose–he had to go up against 2 neocons on the panel of 3. Thus unfortunate, Rose gave him little opportunity to discus the topic at hand.
I am not sure if any polls can be believed, in terms of their overall totals. Just a week ago, LA Times/USC and McClatchy/Marist had national polls conducted over identical periods. One had Clinton leading by 15%, the other by 1%. The goings on under the hood, so to speak that are captured in these polls are fairly consistent: Clinton does well with certain demographics, Trump with another, and some people are jumping all over the place. This leads me to believe that what looks to be the noise is the real data, not the averages: there are many people who really are unsure about how to vote that happen to be casting about somewhat unpredictably (ironically, this group appears to be mostly college educated whites of above average wealth, especially women–who have been fairly predictably Republican even as of 2012.) I do think Trump is probably a bit behind, but the potential for weirdness is such that the chance of an “unexpected” outcome is quite high.
Correct. Romney believed all of the polling his people had done. He did not believe these polls and the ones done by people like Silver. How did it turn out? Who was correct?
Romney believed his secret internal polls. The aggregated, public polling nailed the Electoral College. Nate Silver, e.g., correctly called 51 out of 51 states (D.C. counts). A couple of astrophysicists got 50/51 just using the medians of the final 3 polls in each state:
http://www.colleyrankings.com/election2012/
Nate Silver messed up his predictions about the 2016 GOP primary because he couldn’t believe that Trump was going to win and he second-guessed his own data.
Unsure whether the following linked piece may be of interest to your Committee, Colonel Lang, so no problem for me if you decide it is worthless and don’t post:
http://wallstreetonparade.com/2016/08/sanders-supporters-stunned-by-sudden-death-of-38-year-shawn-lucas-who-served-the-lawsuit-on-the-dnc-and-wasserman-shultz/
Dead right, different clue. A security expert by the name of Spoonamore describes it in a series of youtubies exactly as you say. Spoonamore is/was American Express’s go-to guy for security.
Spoonamore said that only a paper ballot is secure and suggested the right way to vote was at a bank. Have the tellers and security guy’s stay late to count; tellers are used to counting paper. Citizens could congregate within the bank to verify the vote. Spoonamore said it would be cheaper and more accurate than what we have now.
D,
HRC is worse
https://consortiumnews.com/2016/05/11/neocons-and-neolibs-how-dead-ideas-kill/
JJ, who gave him the task? Did that come out of the DNC?
Thanks for the link. (very interesting) My conversation with the plotter tech started out with me saying that electronic voting machines are not secure and can be tampered with. He then explained that they’re secure. It had been a while since I had read about these machines, so he got me thinking that maybe they are now secure. I put my comment out there to get responses like yours. Thanks again!
Obama gave OWS a little time, a few months, to see whether it could be turned into a democrat astroturf wing similar to the Tea Party for republicans at that time. When it became apparent OWS people weren’t going to be ‘Bama supporters, Obama and the government shut it down.
Half the people at Hillary rallies are on the pay roll.
CBS Early Morning Show is throwing everything but the kitchen sink at Trump.
Ex “acting” DCA Morell spewing hawkish attitude which is hard to stomach. Charlie Rose….well he is Charlie Rose!
Didn’t everyone botch the Republican primary forecasts? Nate Silver did do a post-mortem about what he got right and what he got wrong about the primaries: http://fivethirtyeight.com/features/how-i-acted-like-a-pundit-and-screwed-up-on-donald-trump/
Mr. Sliver is using his lessons learned moving forward.
“It’s a reminder that we live in an uncertain world and that both rigor and humility are needed when trying to make sense of it.” — Nate Silver.
Will Evan just shiven off some republican votes that would have gone to Hillary?
The Trump that showed up in Detroit on 8 Aug managed to show a bit of discipline. He studiously ignored the hecklers, he did not engage with them at any level, and he kept on message. If Trump can avoid the distractions and self-inflicted harm the polls should narrow again (IMHO).
All,
Here is a good break out of party affiliation proportions: http://www.gallup.com/poll/15370/party-affiliation.aspx
Any poll that is to be taken seriously should tell you in the methodology section the % D to R to Ind. If it doesn’t generally conform to what appears at the link, then the poll is no good. So far the polls that have Clinton way ahead have been no good.
Of important note – and a wild card when it comes to these polls; Independents are the largest single affiliation in the country these days. However, if asked how they lean with regards to the traditional D or R parties, they are about evenly split. There is a lot of opportunity for mischief in the polls by selecting atypical Independents that lean more heavily, proportionally, to either D or R.
A Republican polling organization, Susquehanna Polling and Research, shows Clinton +10 in Pennsylvania. NBC has a new poll today with Clinton +10 nationally.
http://www.realclearpolitics.com/epolls/latest_polls/elections/
But there are still 90 news cycles before election day.
Will Trump’s appointees help overturn the Citizens United decision?
At the beginning of 2009, Tim Kaine replaced John Dean as the head of the DNC. He favored the idea, along with a bunch of other Democrat party alphas, to instead of continuing with the 50-State Strategy to shift to trying to only selectively “flip” seats, and basically only those for national office.
Dean’s general idea was that, unless you started to consistently put in an appearance in every state and advocated for policies in tune with Democrat sensibilities, you will never give yourself a realistic chance of gaining political agency in so-called irrevocably Red states. Crucially, the focus in Dean’s plan was NOT strictly on national offices, but also on local, and state offices. In some states, this was indeed truly a lost cause; but in others, it was yielding results. To my mind, and to those of others, allowing a total wipe-out of Democrat state senators and representatives, and an essentially unobstructed path to the governorship for Republicans would guarantee that things would only get worse for Democrat chances when the decennial census provided a golden opportunity for gerrymandering in heavily Republican states. Voting districts would be contrived in such a way as to construct “safe” Republican seats on a massive scale, while Democrat voters would be “red-lined” into ghettos. And so it proved to be once any genuine challenge on the state level was abandoned by the DNC under the rule of Kaine.
To my way of viewing the situation, this was an almost explicit acknowledgement that challenges on the basis of principled policies were out the window, to be replaced by a mere division of spoils between the two wings of the Uniparty, which realistically would not be disadvantaged by this at all. Rather the contrary, as unchallenged fealty to its underlying policy goals would be reinforced, and the anti-democratic tendencies of the political culture would be massively reinforced. The Uniparty is another name for the Borg, an agglomeration of corporatists, globalists, and in general a vast unchallenged swamp of NeoLiberalism; i.e., an inverted totalitarianism run by the “credentialed” for the “credentialed”.
The original populist revolts against the consolidation and hyper-empowerment of the Borg have all been subverted and/or crushed: the original Tea Party overwhelmed by the Koch-funded astroturf Tea Partiers, and brought to heel to serve the interests of the Republican wing of the Uniparty (as observed by poster Christopher Fay); the disarticulation of OWS when (as also noted by Mr. Fay), it proved unsuitable to be transformed into an astroturf organization serving the Democrat wing of the Uniparty through a nationally-coordinated police suppression; the Sanders movement, done down by a combination of media neglect and mockery, and the active connivance of the DNC (as clearly seen in the recent leaks); Trump’s winning of the Republican nomination being greeted with the full power of both wings of the Uniparty deployed against this movement.
So it appears to this poster, anyway. But YMMV.
steve,
methodology matters.
Maybe you should gather in a circle and start a fire. Sing some songs. Maybe that will help. Or, maybe hire a shaman to perform a Trump dance.
Bks,
Did you even read what he said or is your idiocy on autopilot?
I believe the Donald dismissed this cohort as a bunch of “failed Washington elites”, which when looking at most of their track records, appears to be a largely accurate assessment, this list being a veritable Who’s Who of Foreign Policy F*&% Ups. I’m not so sure that the crowd responsible for the Iraq War is going to have a great deal of sway with undecided voters, and they sure as shit aren’t going to peel off any already-committed Trump voters. Moreover, this simply codifies the notion that Hillary is the war candidate of choice for the Neo-Con establishment. Love War? Vote for Hillary!
I’m quoting facts. Mr. Newhill is spouting nonsense. Does he honestly think the folks who have been doing survey research for decades don’t think about sampling? To a first approximation, that is what people in survey research think about. Does he think that a survey organization run by Republicans is chosing samples to make Clinton look good? Why would they do that? Tinfoil-worthy.
bks,
sigh….so you just believe anything you are told by an “official” source and you chose to not think for yourself. The perfect Democrat.
You are not considering – refusing to consider, actually – the pollsters are either deliberately lying to you (why is open for speculation) or are just sloppy. Since, as you point out, they are experts, I opt for the former.
However, in fairness, there are reasons for releasing sloppy work other than lying or lack of skill. For example, Rs may be at work when the calls are made, while Ds are more likely to not be and are available to answer the phone. Rs may tend to have unlisted phone numbers (both have proven to be true in studies the past). There is pressure to get these polls out the door and get paid. Some of the sampling issues are very difficult or impossible to resolve. Do you think a company that makes its profit getting surveys out the door is going to stop doing them b/c they can’t get it right without investing considerably more resources into the process? No, they go with what they have. The media is sloppy for not ready the methodology section and commenting on it. So are you. Plus most people are statistically ignorant and a good story means more to them than good solid data.
I’ll bet you haven’t even looked at the links I sent that show the proportion of D to R in the general population and then taken the time to read the polls you love to quote and compare the proportion of D to R in them.
This is really simple. It doesn’t take expertise. It is common sense. I could give you my credentials (which include having passed actuarial exams and working in a field where I apply those skills every day where 100s of $millions are dependent on my analysis), but that’s just, you know, crazy statistical stuff and I’m probably lying; something that Nate Silver or other corporate shills in the polling industry would never do. Right?
Enjoy your bubble.
Or Gary Johnson.
Ahhh! The old Clinton body-count just a keeps on growing …
bks,
A final thought – Given your “reasoning”, how is it that anything goes wrong in this world? With all the experts running things we should have achieved utopia.
bks,
No, he’s talking facts. You’re talking inanity and burying your head deeper in the sand. In this case, the fact its a Republican organization is laughable as if all the mainstream Republican organizations are against him.
Furthermore, your denial sounds more of that peculiar “LA LA LA” nonsense that floats around here lately. All one has to do is look at the demographic metrics, in which it obviously shows an oversampling of Democrats in these surveys by double digits.
Or for tldr: you’re a damn fool of a shill, and a bad one at that. Sad!
I’m just a computer programmer who spent five years working in political science survey research, and I have followed it, on and off, for thirty years. If I’m a shill, I wish someone would tell me where to pick up my check. The pollsters put Trump in the lead for the GOP nomination starting in July 2015:
http://elections.huffingtonpost.com/pollster/2016-national-gop-primary
and yet suddenly, according you folks, the exact same set of organizations are in the tank for Clinton? Not seeing it.
BabelFish,
That last I read, all she can do is pray to recover from the next disclosures.
Now, will the MSM report them?
I do see an Independent stepping forward but don’t know enough about him.
Dr. K,
Keeping telling yourself that.
Herb,
Shouldn’t you be writing another long winded post about how Trump is doomed while hoping Big Grandma doesn’t keel over before Labor Day?
HOM,
Bullshit. (((Nate Silver))) basically engaged in Talmudic talking of black into white to explain that he actually right in spite of being totally wrong over and over again. Humility? Rigor? lmbo. Silver didn’t learn anything except to double down on his shilling.
All,
Was this man invited to get a reaction from Trump to divert attention from something else?
https://www.google.com/amp/abcnews.go.com/amp/Politics/father-orlando-nightclub-shooter-attends-hillary-clinton-rally/story%3fid=41231649?client=safari
I guess Scott Adams nailed it:
http://blog.dilbert.com/post/148692199141/how-to-identify-the-brainwashed
Tyler
Just a heads up.
Use of triple parentheses is known as an (((echo))), an antisemitic symbol that has been used to highlight the names of individuals of a Jewish background. The practice originated from the far-right blog The Right Stuff; the blog’s editors explained that the symbol is meant to symbolize that the historic actions of members of Jewish ethnicity had caused their surnames to “echo throughout history”.
I’m sure you wouldn’t want people to think you might be an anti-Semite.
https://en.wikipedia.org/wiki/Triple_parentheses
A couple of weeks ago, 538 had Clinton and Trump at 50-50. 538 has HRC way ahead now, post conventions and post-Khan family smearing. My opinion is that Silver is probably close to correct, but that those numbers are not solid at all, that they are still fluid. I don’t think opinions will be finally formed until after the debates.
The same arguments that are being made now regarding bad polls are the same ones being made back in 2012. If we dip into the wayback machine ( https://web.archive.org/web/20121031001154/http://www.unskewedpolls.com/ ) we can see the unskewed results showing Romney up by 5 to 9 points across the board a week before the election.
Having said that I will readily admit that I may be suffering from the Dunning-Kruger effect ( https://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect ). With that admission of the way; I agree with bks on this one.
No need Tyler, but thank you for the suggestion. I’m fine with reality, and even more pleased by your boorish reaction to it.
Actually, I think you are the one needing hope that Clinton keels over. It is the only possible way Trump avoids the worst ass-kicking in modern history. On the other hand, Pence would probably beat him worse, not having Clinton’s obvious baggage. See, I’m no Clinton-lover, just a Trump hater, and I am really happy with what I’m seeing.
But what about you? How’s it going for you? Events, my dear boy, are certainly not something by which you can be pleased.
Is 3% down an outlier? Yes, looking like.
Edward Amame,
Tyler can speak for himself, but I think if you believe that he gives a damn about being considered an anti-Semite you simply have not grasped the way the world has changed.
I write as someone whose culture might well be described as ‘Anglo-Jewish’ – who has had close Jewish friends and colleagues throughout life, and has personal reasons to be very conscious of the immense contribution made by Jewish refugees from the disasters of continental European history to British culture.
However, over the past years I have seen American, and British, foreign policy, essentially hijacked by dotty Zionist agendas – with catastrophic consequences. And I have seen people who tried to resist this process relentlessly smeared as ‘anti-Semites.’
The catastrophic sequence of events thus unleashed has ended up with the candidate for President the vast majority of influential American Jews prefer apparently all-too-happy to risk not simply empowering jihadists but war with Russia in order to destroy the ‘Shia Cresent’ which Zionist folly was responsible for creating.
As this extraordinarily dangerous crisis unfolds, I see even many ‘liberal’ Jews in the United States and Britain with their heads firmly buried, ostrich-like, in the sand.
So I learn from Peter Beinart that it is acceptable, in ‘almost any synagogue’ in America, to explain that you do not believe in God, but not to say that ‘I think what Israel is doing is immoral or I have questions about Zionism.’
Moreover, it seems that in Beinart’s own view, Jews are ‘a people’, who ‘deserve self-determination’, and that ‘to reject that, I do think gets you closer to antisemitism.’
(See http://mondoweiss.net/2016/08/criticize-israel-beinart/ .)
Obviously, in this situation, I am very interested in who, among people holding influence in the United States, and Britain, is Jewish – and what kind of Jew they are.
If people are ‘Beinart’ Jews, then from my point of view, I am happy for them to live where they like, but do not want to see them in positions where they can have any influence on American or British foreign policy.
If Jews want to belong, wholeheartedly, to the American or British ‘peoples’, that is a quite different matter. We have had much very much ‘good counsel’ from such people in the past.
Likewise, if people are caught between conflicting loyalties, I can sympathise and understand that – so long as they also recognise the loyalties others of us have.
However, the hijacking of American and British foreign policy by people who are essentially ‘tribal’ Jews was always bound to produce an antisemitic backlash.
And, unsurprisingly, one of the places one finds this most strongly is among people who, like Tyler, have fought in unwinnable wars, which were ‘cheered on’ by Zionists who had never regarded it as possible or desirable that they or their only children would fight in them.
Get used to it: Tyler is one face of the future.
David Habakkuk
I like Tyler. It is a tribal thing. He is one of my people but the reason I put up with his Trumpian quickness to retaliate is precisely what you have said. He is a voice of the people for whom Trump holds the key to hope in a country turning against them. When he tells you that armed resistance is coming, people should listen to that. Even the MJ crew are beginning to talk about the possibility. pl
That’s not really true: there are MANY undecided voters, just that they don’t look like the usual undecided voters. Many formerly reliable Republican voters, college educated and affluent whites, especially women, are all over the place from poll to poll. Clinton campaign is working hard to gain their support. We shall see how that pans out.
The shorter version of Taleb is that the predicted probabilities themselves are shrouded in so much uncertainty that they should come with a big margin of error next to them. The predicted probability of Trump win might as well be 30%+/-30% (I’m making this one up). The chance that Trump would lose is very high, but things are so volatile that we don’t really know, and the way polls are being reported on discounts massively how weird this election is.
The jumpiness of Silver’s predictions is itself a clue about the inherent uncertainty in this election. A lot of data analysis people think that uncertainty is just “errors” that will go away if you run the numbers just right. These people are stuck in 19th century thinking. Sometimes, the uncertainty just exists–people don’t know what they want exactly. The most amazing thing to watch in the polls this year has been how wildly the “usual Republican voters” have been jumping around. If Trump can expand the working class white electorate a good bit (which Romney could not but Trump has an excellent chance at) and keep the usual Republicans (which Romney did extremely well, but Trump has obvious trouble with), he wins a narrow victory. This picture has stayed constant for months, even longer (Bob Dole, of all people, was quoted last year making roughly this point!). Whether the “usual Republicans” would ultimately support Trump, however, is the big question raised by all these polls.
David Habakkuk
What about all those Southern Baptists and Conservative American Christians and christian members of Congress who have so wholeheartedly embraced Bibi and the Israeli right and the Zionist project? Do they get the (((treatment))) too? For all I know, Nate Silver is one of the ranks of the many Jews here in NY who are vehemently opposed to Bibi and neocons. Spare me the lecture. You can get used to that face of the future if you want, I don’t have to.
HawkpofMay,
I’m glad you recognize the meaning of skewed polling samples.
Here’s a different interpretation of what it means re; outcomes.
In 2012 the sampling was indeed skewed. Unskewed, Romney was actually ahead. However, it didn’t mean anything on election day b/c he didn’t excite people. He didn’t realize his potential advantage. The people stayed home.
This year we have the same skew issues (there is something inherently more difficult about getting Rs to answer polls, than Ds. A lot of research showing it’s true and suggesting why). However, this year is different. The supporters of the guy on the wrong end of the skew, Trump, are highly excited and motivated. It’s elect Trump or lose our country. Win through politics or, just maybe, pick up a rifle and try to win through armed conflict. So the true potential that was hidden by the skewed sampling will show up at the polls and Trump will win.
That is another valid way of looking at this thing.
Colonel Lang,
I like Tyler, always have.
But there are also other aspects of this, to some of which you point.
In my television producer days, I saw a key part of my job as being, as it were, to try to ensure that different voices were heard – to try to help people to some understanding of the complexities of their own society.
(That was also part of the fun of the job – I get bored out of my mind, talking to people who have the same views as me.)
Sometimes, there may be good grounds for trying to silence people. But a lot of the time, the attempt ends up blowing up in people’s faces: as has happened, massively, with the ‘Powellite’ sections of opinion in this country.
Also, the old-fashioned cynical Tory strand in me ‘kicks in’ here.
If, from a purely Machiavellian point of view, I had decided that Tyler was liable to be an enemy – which, believe me, I have not – I would want a realistic appreciation of what would happen if I got into a conflict with him and his like.
And this would include calculating what my prospects were of, as it were, ‘rubbing him out.’
It was very clear to me, from the start, that he came from a ‘fighting culture’.
If people who do not come from ‘fighting cultures’ choose to pursue courses of action which, in effect, encourage those who do to indulge their taste for combat, they must make some attempt to calculate the likely outcomes.
Having done so, they can decide whether to go ahead and confront, make some attempt to conciliate, or whatever.
When I reads Mike Morrell explaining talking about ‘some number of uneducated white Americans who fear the browning of America’ it seems to me absolutely clear that he is an absolute cretin, who has no idea of the kind of fight into which he may be getting himself.
At heart, I am an old Anglican – my sympathies, in both our Civil War and yours, are with those who tried to prevent polarisation happening and then mend the wounds after the conflict.
So, obviously, I would like to encourage Tyler to be more diplomatic and calculating – and in particular, to grasp how important it is to attempt ‘butcher’s cleaver’ moves.
But: my blood boiled at that remark by Morrell.
David
Our society is fracturing. The “white trash hillbillies” to use JD Vance are getting to the point of hopelessness. The many Scots-Irish among this segment are fighters. They have fought many a war for America and now see how they’re getting shafted. The urban and coastal elites and the “salary” class as the Archdruid calls them who aspire to join the ranks of the elite have only contempt for the hillbilly.
As you note if we get to serious societal conflict with armed rebellion the effete elites will be reliant on state forces peopled by the very folks rebelling. That could get very interesting when the grunts rebel against the “officer” class by refusing to kill their brothers.
If the Borg Queen triumphs in November, the hillbillies that voted for Trump will believe with even more certainty that the system is rigged against them and will become more susceptible to the siren call of a demagogue that violence is the only recourse.
That is a valid point. Traditional college educated Republicans particularly women are vacillating. Do they vote as they normally do or sit out the election or swing to the Democrats will in many ways decide the outcome. Similarly, how traditional union labor Democrats vote will be important too. Its not a given that Hillary has their vote.
Jack,
What frightens the living daylights out of me is that the ‘imbecile clerisy’ which governs us cannot see what is happening.
After the ‘Remain’ victory, my sometime television colleague Peter – now Lord – Mandelson produced a post-mortem in the ‘FT’, entitled ‘How the struggle for Europe was lost.’
It included the revelation that ‘Our pollsters assured us that economic concerns trumped those about immigration.’
The extent of the distrust, particularly outside London, caused by immigration policies which have led to a situation where only 45% of the population of the capital are white Britons, was clearly beyond the capacity of Peter and people like him to imagine.
(See http://www.dailymail.co.uk/news/article-2950401/How-one-three-Londoners-born-abroad-areas-live-in.html .)
A situation where for many in the country as a whole, London is essentially a foreign city is fraught with very ugly potentialities.
I’ve always thought (and I think I mentinoned this in a reply to another post before) that the problem is that the society is going “post racial,” but in the wrong direction. The plight of the so-called “white trash hillbillies,” especially as they are perceived by our social elites, is reminiscent of the Invisible Man, by Ralph Ellison. In Ellison’s novel, it’s the African Americans, especially those who were poor, that were invisible. But the ranks of the invisible have expanded to include the poor whites as well. In a sense, this is not an accurate characterization–the poor whites of the Appalachia, especially, were a minority that were held in contempt since the days of the Whiskey Rebellion and earlier, were literally targeted for slow extermination via Eugenics movement and forced sterilization, and so forth, and still get tarred with accusations of “white privilege” which most of them had never even seen. But their numbers have been expanding as the social inequality (not just economic inequality) has been increasing exponentially last couple of decades.
The great thing about Tyler is that he is extremely sharp witted, has a great sense of humor, and doesn’t respect the conventional sensibilities–a little bit like a George Carlin or even a Chris Rock, except with different political sensibilities. Yes, he is a bit too sharp-tongued and gets a bit carried away sometimes, but, if he didn’t, he would not have half the impact he has–sharp, direct language cuts through salubrious nonsense like a samurai sword through tofu, so to speak, and that I respect, a lot.
When numbers are jumping like they have been (and when numbers are so completely varied across multiple polls), that means the world is too messy (or uncertain, or has too much variance, or whatever) inherently and we literally don’t and can’t know. The pollsters are being dishonest, knowingly or unknowingly, when they say we can reliably predict X or Y with actual numbers when we see things like that.
Well, I brought up the Republicans because their vote, I think, will be much more decisive.
Union voters, I think, are rather less important now: there aren’t too many of them left and they don’t vote, and many of those who do have already gone over to the Republican camp. Now, Sanders did show that there are enough of them who are still Democratic enough to vote for a right kind of Democrat and they made up a significant minority of the Sanders voters–and I think they are the voters who are most resistant to the Clinton appeal. Still, I estimate that if they swing over to Trump in any large numbers, Trump could survive losing a decent (but not many) traditional Republican voters. These odd groups of “swing voters” are the most fascinating aspects of this campaign.
kao_hsien_chih
I’m absolutely certain there are websites out there where Tyler’s political sensiblities are pretty conventional. Parading them around mainstream websites in a comic Mike Hammer-esque voice may seem refreshing to you, but to me it sounds like the same old thing in brand new drag.
His conclusions probably are, but that would be same thing as saying that Aristarchus understood modern physics because he believed in heliocentrism, and it’s not as if “certain websites” are a reflection of the real world–and if you think they are, then you have bigger problems than Tyler.
Quite frankly, I don’t really care what his opinions are: they are his and he is entitled to them. I also know that, like Carlin and Rock, he is quite earnest and honest when he makes his points which are, whether I like it or not, an important component of the public opinion in this country today and understanding their reasoning is important. In this sense, I appreciate that Tyler is direct and forthright.
If, as you conservatives and liberals seem eager to do, all the people whom you disagree with are to be treated with ignorant contempt (literally–in that you don’t care to know why they disagree with you other than that they do), what chance do you have to understand what makes them tick? Without understanding what makes them tick, how would you expect to win whatever you hope to win? By faith? Because “God/Righteousness/Reason” is on your side, that you know because you have faith?
Edward,
It seems this has proven (((very effective))) in triggering people who have been anti-gentile, anti-western and anti-American. Somone on the alt-right should be given cudos for driving the left mad with some brackets.
I’ve just gotten back from a trip, and all I can say is that a lot of you are abusing each other over a topic of which you have very little understanding. Let me try to explain the whole party identity vs. voting thing in as simple and non-partisan a way as I can.
Party identity is attitudinal data. It’s about how you feel about something at a particular point in time. By itself, attitudinal data is noisy, fluid, and does not correlate strongly with outcome. In market research, it’s useful but should never be used by itself if you can avoid it. In polling, where you usually have more objective data, you should simply ignore it, which most pollsters do. The idea that you should correct relatively objective data with attitudinal data, which many here are suggesting, is just plain wrong.
To see why, let’s imagine you have a town of 15,000 inhabitants with a Walmart and a Target. One day, a company does an independent survey asking whether people consider themselves Walmart shoppers or Target shoppers. Not surprisingly, 70% identify with Target, and 30% with Walmart. This is your attitudinal data.
A few days later, a different company does a second survey. It first asks people how much money they spent at each store in the last week. Then it asks whether they are Walmart shoppers or Target shoppers. The answer to the first question reveals that 80% of the money was spent in Walmart and only 20% in Target. In addition, 60% of all shoppers in that survey say they consider themselves Walmart shoppers and only 40% Target shoppers.
What many of you are saying is that such an outcome “proves” that the second company deliberately over-sampled Walmart shoppers to produce a favorable outcome to Walmart. Obviously, this is nonsense. Rather, when asked in the abstract, people had one identity, and when reminded of their actual behavior, they had a different one.
In addition, attitudinal data is not suggestive of outcome. If we did a third, purely attitudinal survey that revealed the identity split was now 75/35 in favor of Target, we should not expect any meaningful difference in customer behavior or store revenue.
Pollsters have long recognized that answers to questions about party identity, counterintuitive as it may seem, do not correlate enough with voting behavior to be useful in predicting outcomes. That’s why they don’t account for it and certainly don’t adjust for it.
I’ve run into the same disdain for using party id for stratifying polling data among commercial pollsters and I’m a bit confounded by this. Even in the Target vs. Wal-Mart example you mention, those who say that they like Target more than Wal-Mart are more likely to shop at Target, ceteris paribus. While insisting that the sample must balance Target and Wal Mart supporters is silly, as there are many other factors, insisting that it must be excluded as a factor seems equally silly: companies do spend billions trying to build up their name brands, after all, in recognition of this. In electoral politics, the correlation between party id and vote choice has always been considerable and has become much greater last few decades. Accounting for party id, if it can be done accurately, does improve the predictive power of polling analysis greatly and is, furthermore, consistent with our understanding of American party politics today. (as can be seen in this paper http://www.stat.columbia.edu/~gelman/research/unpublished/swing_voters.pdf) Now, there are multiple problems with accounting for parties in polls, but they can be dealt with relatively easily. Rather than trying to balance out the sample, party id information can be used in the post stratification process along with other demographic data, as per Gelman et al. While the conditional probabilities of vote choice by each partisan-demographic group is uncertain, we do have data for that, literally. Of course, the composition of the voters who turn out on the election day, including the partisan makeup, is unknown, you can take your best shot at it, again from the data that you have. All in all, why shouldn’t take party id into account, even if not as the decisive factor, knowing all that we know about relationship between voter behavior and party affiliation?
KHC,
I’m with you. I am not buying what Shepard says – I am not a poll/survey expert by any means, but I do have a vast amount demographic and statistical analysis experience. Shepard’s views are counterintuitive. I am open to being educated. It would require some serious scientific/scholarly explanation.
Agreed that sample doesn’t have to equate to the population. If not, results must be weighted – and they are not in these polls.
I, like you, can accept that party identification is not a perfect predictor of voting behavior. However, it has to be pretty strong. Again, avoiding all of the shop talk gobbedlyguk and working off basic common sense, are there actually Democrats that will vote for Trump? Are there Republicans that will vote for Clinton? In both cases, the answer is probably, “Yes, a small few”.
Otherwise,Ds will vote Clinton and Rs will vote Trump and independents are up for grabs. The % of Rs, Ds and Is in the poll will determine which candidate will lead the poll if the poll results are not weighted to reflect the population proportions.
Bottom line – If you have a sample that is not representative of the population, you will have a poll result that is not representative of the population.
I’m probably risking a scolding from Richard, but saying that party affiliation is “attitudinal” seems, to me, like a fancy way of saying nothing – or a misapplication of theory. Political party affiliation is not nearly as ephemeral as which junk store one shops at. As Kao says, brand loyalty is important and nowhere is brand loyalty as strong as in politics (my perception anyhow)
But at least shepard’s explanation – assuming he is familiar with polling methodology – answers some of the questions as to why experts would be cavalier about sample proportions.
kao_hsien_chih,
I’d forgotten about the Gelman paper. It was very provocative and a lot of fun when it was released, but my understanding was that no one has been able to (or maybe tried to) use its findings successfully in a live setting.
For what it’s worth (and I’m sure you know) the paper has a very restrictive data set–45 days of a single election. I remember reading about it and thinking, this is really interesting, and I was hoping to see its findings confirmed elsewhere. And then I never heard of it again.
Do you know of any research that follows up on Gelman? I don’t mind being proved wrong.
New Rasmussen poll has Clinton +3 in a four way race. The reuters poll referenced above is not an outlier nor is the one below.
http://www.rasmussenreports.com/public_content/politics/elections/election_2016/white_house_watch
jld,
How does Scott Adams know what anyone else “can’t know”? Does he know about trade policy and its effects his own self? If he doesn’t, how is he in a position to assess whether someone ( anyone) else’s stated positions and understanding of trade policy is information-based and logic-achieved or not?
Before NAFTA was passed, numbers of people predicted it would lead to job losses in America. Some people even predicted it would lead to a rise in illegal immigration from Mexico into America. And those predictions turned out to come true as predicted. If the people making those predictions did not understand about the effects of the proposed NAFTA trade policy they were discussing, how did they make predictions which ended up coming true just as predicted? How would Scott Adams explain it?
( And by the way, if Trump ends up losing, what happens to Adams’s self-burnished reputation as a master persuasionologist and master political predictor? He’d better hope Trump wins, or I might not be tempted to buy his book).
shepard and Kao,
I’m probably obsessing about this and should have let it go several comments back, but I find this discussion very disturbing. Here is a topic I know a lot about – not polling or surveying per se, but research and statistical analysis and there are only two people who have any idea what end is up; Tyler and myself. Kao so tepidly responds that what he says loses force.
I come to SST for the high quality of the posts and comments. A lot the material is outside my real of knowledge or experience. So I stay quite and learn. Now I am worried. If a topic that I understand can be so terribly misrepresented, what is happening on those topics where I am trusting others to be knowledgeable.
Party affiliation is highly correlated with what candidate you vote for, In the general population, If there are 26% Rs and 26% Ds and the rest identify as Independents, then your poll sample MUST reflect those population proportions (or you MUST weight the results of your sample) BECAUSE party affiliation IS so highly correlated with candidate choice on election day. Party affiliation could probably even be said to be causational in a normal election cycle. I can’t believe that a discussion full of grown men that are supposed to be intelligent and with some who are claiming research credentials can’t grasp this basic concept.
No? Then what if I told you there is a poll with a sampling consisting of 80 Rs and 10% Ds and 10% Ind and that Trump is killing Clinton in it 71% to to 11% (balance goes to Johnson and Stein). You wouldn’t point to sampling as a criticism?
abkao_hsien_chih
What makes Tyler tick?
I don’t know, but I do know that anti-immigrant groups have been around since the beginnings of this country, so there is probably no “winning,” at least in the short run. Hispanics, blacks, Asians, Jews, women and gays all have lobbies now to protect themselves/advance their interests. That suggests to some that white nativism has been, or will be, the response and that violence might be the result. I don’t. The days of the Posse C and militia groups are probably over since the Oklahoma City bombing. Since that time, there’s been an effort to mainstream nativism via the Tea Party. Candidate Trump has my heartfelt thanks for putting aside the dog whistles and speaking directly and forthrightly to/for nativists. There will be a referendum on that in November. There’s talk that they’ll resist majority rule if a President Clinton is elected, there are threats that there will be insurrection. If it comes to that, those threats would have to be addressed. There are economic concerns of whites with only high school education that will need to be addressed, but congress….
Meanwhile, this is how immigrants won back in 1990s California and it’s instructive. There was a GOP backlash against immigration there that led to passage of a slew of ant-immigrant propositions, most of them specifically targeting Latinos — the same Lations who had voted out 50% for Ronald Reagan as governor. Latinos organized and along with enough white voters alienated by those propositions, eventually flipped California eventually from being a GOP leaning state with a GOP governor and legislature to having Dems control every statewide elected office and being heavily pro-immigrant by 2016.
bks,
Yes you are a shill, and yes, they are in the tank for Clinton, as they would have been in the tank for any other Republican candidate.
When they screw with their methodologies so massively in order to create a “Hillary inevitable” narrative, what else can you call it but what it is?
Well, unless you’re a progressive borg drone, then you invent new terms in order to pretend that your girl isn’t doing horribly when she can’t fill a high school gym while Trump stuffs 10k+ into an auditorium. Another question of who am I to believe? You or my lying eyes.
EA,
Mr. Habakkuk nailed it, so there’s not really a whole lot for me to say except to add onto Fred’s “Are you triggered?” question.
We tried the Tea Party, and a bunch of old people who liked to LARP as colonial era actors got made into a group of crazy racists.
Now you get the Alt Right Nationalists in the form of Trump, who you are trying to suborn as crazy racists.
The snap back keeps building as the Left tries to stuff anarcho-tyranny down the throats of people who just want to be left alone.
At least you’ll finally be right about the “crazy racists” part when the RWDS kick in your door.
Eric Newhill,
While it’s amusing to be called dangerous, I’ll have to decline the title.
I’m sorry, but you’ve made a basic mistake in terminology, which I think drives a lot of the confusion in this discussion. Party affiliation, i.e. which party you belong to, is an important metric. It is highly correlated to how you vote and important in polling. You have to account for it, and polls do. In that, you and I are in complete agreement.
Party identification is something entirely different and the beloved metric of conspiracy theorists and those behind in the polls. It refers to how you answer a question about which party you feel you identify with in the course of a survey. It’s not whether you are a Democrat or Republican, it’s whether you feel like one at that moment.
Obviously, this is a problematic question. Think of this election cycle, where a lot of registered Republicans may or may not feel like Republicans based on what Trump said yesterday. Or they may feel like Republicans in a general survey, but not when reminded of Trump’s leadership of the party and their intention to vote for Gary Johnson. The same person will answer this question in different ways at different times, not least depending on the context provided by the survey itself. Thus, using one survey’s results to judge the accuracy of another, which you suggest, doesn’t work. Nor is the metric stable enough normally to use in predicting results.
Kao brings up a really intriguing paper that calls this basic premise into question. It analyzed the last month and a half of the Obama/Romney campaign. Its conclusion was that the electorate had become so partisan that party identification had become consistent. Thus, it was something from which you could derive meaningful, predictive insight. However, to my knowledge the paper has not been peer reviewed, has not been widely cited, and has not been confirmed by other research. Stay tuned.
Kao and I are not tepid or fanciful. I think we’re both practitioners in our own ways and come at this from very different angles.
EA,
You are a historical illiterate.
The “backlash” occurred before Prop 187 passed, because of REAGAN’S AMNESTY. This is more liberal “memory holing of history”.
Mind you, Prop 187 was voted in and failed because Gray Davis refused to appeal when the Ninth Circus’ unelected federal mandarin decided that it was “rayciss”.
And you don’t think the already resident population has any right to object to their standard of living and the cohesiveness of their communities being undercut by mass immigration? You think they should just, for want of a better phrase, bend over and take it? They’ve been bending over and taking it for a long time and now they’ve had enough. How can you blame them?
EA,
“What makes Tyler tick?”
So many pithy answers here. The old Conan line about “driving your enemies before you” comes to mind, or maybe “Make America Great Again”.
While those have their appeals, the honest answer is that progressive policies lead to tyrannical, violent, hellhole and the destruction of the human spirit over and over again for nothing other than the pride that progressives have in being able to virtue signal their good thinker status.
So that’s part of it. There’s also the truth, another big part of it. And have others have pointed out here: I love a good fight.
Shepard, Utter nonsense. The polls are of typically *registered* Ds and Rs and Indies. There is no language in the questions about how they feel they identify at the moment. I myself have been polled recently and the questions were “are you a registered voter?” “Yes.” “What party?” “If the election was today, who would you vote for?” This is line with exactly what the polling methodology says it asks.
You’re just making stuff up so it fits some marketing approach you think you understand – or you’re confusing the polling with the idiotic bathroom/gender issue.
Physician says Hillary has Parkinson’s disease:
http://www.dangerandplay.com/2016/08/12/hillary-clinton-has-parkinsons-disease-physician-confirms/
Make of that what you will. Supposedly written by a board-certified anesthesiologist, but no names given.
Tyler
“And have others have pointed out here: I love a good fight.” Finally it appears we have something in common.
EA & Tyler
Fight clean lads. Fight clean. pl
irf520
First off, since 2013 China and India have beat Mexico as the most common immigrant countries of origin who’ve been in the US for a year or less. More Mexican immigrants have returned to Mexico than have migrated to the US since 2009.
If they are here legally, why not? Back in the day, my Irish Catholic and Italian ancestors undercut the cohesiveness of communities, so on that point… As to standards of living, that’s tougher. It’s been shown that immigration has hit families w/o college diplomas by lowering their incomes from some 7-8%. So that needs to be addressed, though I doubt congress as it is currently made up really gives two s*^$s about that.
Tyler,
It was a response to Reagan’s amnesty? Thanks, I didn’t recall that. I DO remember that Pete Wilson was running a tough re-election campaign and he embraced it in a big way to try to save his ass.
Col Lang
Will do.
Fred
How the %$#@ do you know that Nate Silver is anti-gentile, anti-western and anti-American? Because he’s a Jew?
Shepard, the LA Times poll – which performs the weighting I insist in necessary – has the poll results too close to call.
http://graphics.latimes.com/usc-presidential-poll-dashboard/
That is a poll I am more inclined believe.
BTW – I never said or implied you are “dangerous”. Are you trying to be? I guess if someone is making important decisions based on your “thinking”, you could indeed be dangerous.
Edward,
Wilson embraced it as a Johnny come lately when he originally opposed it. Much like how Gov. Brewer was against HB17 in Arizona until her poll numbers tanked and then she became a national darling after supporting it with a full heart.
Edward,
Or we can just stop importing them all.
This isn’t a contest of where you go “OH YEAH BUT THESE CHINESE”. They’re ALL foreigners.
Eric,
I think it’s time we looked at some data, because we’re talking about different things (and my apologies, I tend to write sloppily on blogs). So far, we’ve established (and correct me if I’m wrong) that you believe polls do not ask about feelings or what someone is thinking at the time. You also believe that current pollsters are oversampling Dems. You believe this because of an additional set of polls that show a different breakdown for party affiliation.
Well, let’s look at the Gallup poll that now shows an even split between Democrats and Republicans: http://www.gallup.com/poll/15370/party-affiliation.aspx.
For what it’s worth, this is a trend line poll. Its purpose is not to establish exact numbers for affiliation at any one point in time; rather, it is intended to demonstrate over time whether the country is becoming more of one or the other (that’s why it presents its results in historical context). The data it gathers is both attitudinal (how you feel) and, as I predicted, highly volatile. Both of these things are plainly obvious from the survey page itself.
First, here’s the question it asks: “In politics, as of today, do you consider yourself a Republican, a Democrat, or an Independent?”
In other words, it is asking about how you feel about your affiliation today (not tomorrow, traditionally, or anything else). It does not ask about registration, membership, voter intention, or anything remotely approaching that. It specifically asks you to think about what you consider yourself right now.
In your thinking, political affiliation is a highly stable concept. I have stated the opposite, and this is the context in which I meant that. To see why, let’s look at the data.
The fluctuations from week to week in this survey may seem to a layman like a couple of points here and there. But you’re an actuary, so you have to recognize that they are wide swings. To use this data set to normalize other polls, you’d have to believe that it would be possible not only for Republican party membership across the entire country to change as much as 10% in a two-week period, but for this poll to accurately reflect that fluctuation. Maybe you liked the last iteration of the poll, which showed a 50/50 split. But two weeks earlier, there seemed to be substantially more Democrat-leaning voters than Republicans. Did the world really change that much in such a short period of time? And two polling periods before that, it’s the Republicans who had a slight edge.
Although it may not seem logical, if you ask about current attitudes towards a party, this is what you typically get. This survey should be looked at over time, with deep suspicions as to accuracy of any one data point. It cannot be used in the moment to support a claim of fraud.
shepard, too many confounds and too many opportunities for lousy sampling technique or just plain old random error. I would believe you if they called the same people at different times and then got variation in party affiliation. Mrs. Kowalski polled in January say she’s *feels* like a democrat and favors Clinton. Asked in March, she feels like a republican and favors Trump….in June a democrat again, etc.
You are putting way too much faith, IMO, in these pollsters having good technique and their samples being both truly random and truly representative. You are asserting if the polls end up with a 36%/26% D to R ratio, then that is reflecting the mood (attitude) of the country. Prove it! I am saying it reflects bad sampling; either on purpose or due to sloppiness and a desire to get the poll out to the media and get paid. I do not believe that party affiliation is so subject to how you are feeling.
Anecdotally, I identify as Republican. I do not think Trump is a great candidate. I will vote for him, though. I did not like McCain at all and I voted for Obama in that election. Yet I identified as a republican voting for Obama. I did not identify as a democrat. So that goes against your theory (and a theory is all it is).
I think you are dancing past the fact that most of these polls are targeting registered voters and asking what they are registered as. That question eliminates your attitudinal theory; unless you want to tell me that they’re running down to the county building and changing their registration every couple weeks or so.