The polling industry is out of business

He'll have to find work as the Cub's mascot now.

projects.fivethirtyeight.com/2016-election-forecast/?ex_cid=rrpromo

Great laughs

Technically, Nate Silver gave Trump a roughly 30% chance of victory. He didn't say he'd lose outright. Moreover, a number of states he predicted to be for Hillary, that Trump won regardless, he had very little polling data for. In the days leading up to the election, he was noting that Trump still had a path to victory, and that the reason why he still gave Trump a chance was because his model could not produce strong conclusions based on the limited polling data in those Rust Belt states.

anyone have a link to this guy's twitter? I have to fucking see it

he gave him 30 percent like a week before the election to cover his ass when the FBI started investigating hillary's shit again

I hate to sound like I'm defending Silver, but he doesn't conduct polls, he's essentially running a metapolling service, and he's quite transparent with his methods. The problems clearly lie with the ways the actual polling is done.

...

The elections over, you can go back now.

(((Nate Shillver))) exists to shill and deceive, not inform. There's no defending that.

Then don't. Get the fuck out.

He's a garbage can. Garbage in, garbage out.

(((Polls))) are meaningless garbage for the goyim to slop over, but no longer. Fuck off back to where you came from and pray you do not find yourself on a list.

He's not lying. Nate Silver doesn't pick and choose the numbers on his site. He takes polls, feeds them into an algorithm that runs a bunch of simulations based on those polls, and reports back the percentage of victories in those simulations for each candidate. In 29.1% of the simulations right before the election, Trump wins. But regardless, Nate never held any polls himself. He merely drew conclusions from existing polls from as many credible sources as he could. As it turned out, the polls were a poor predictor of actual state voting behavior.

...

Then he's even worse, he's basically a parasite that is feeding off of other people's work, and basically a hack.

...

He's a statistician. His job is not to write the polls, but to run statistical analysis on them. To effectively conduct statistical analysis, he needs to pull from as many sources as possible.

Yes, he made a bad call, but that isn't to say that the nature of his work is wrong. Statistics is one of the most useful branches of mathematics we have, due to its applicability to most facets of life. He's not a parasite; he's just doing what is necessary to conduct statistics as effectively as he can.

It is entirely possible that Nate is either unaware or unconvinced of poll rigging in the numerous sites he has pulled data from. I would not peg him as deliberately misleading people.(wir mussen die juden ausrotten)

I'll stick up for your defense of statistics. Nate was just a hack who couldn't see any of the problems with the polls this time.

Fucking Kek

Reported.

That's all the numbers you needed, Nate. Even I figured this out last September after about 2 minutes of thought.

This is a classic case of jews jewing jews, the (((pollsters))) rigged their own polls to set a narrative and blackpill Trump supporters, which in turn fucked up Nate Pewter's models.

...

You're essentially right, and anons here are giving you more shit than they should for this factual statement. However, Silver had every tool available to unskew the polls, because other people were doing it too.

He just didn't want to get his hands dirty with unpleasant facts about minority demographics. And he didn't want to alienate all his liberal friends.

He should have known better. This was his profession. Many amateurs understood the polls were skewed and corrected them competently. He was beaten badly by hobbyists and part-time alt-right pundits.

well they were right in the end lmfao

...

DURING BREXIT I DID SOME RESEARCH ON UK POLLSTERS

RWDS

This is one of my favorite articles.
=HUFFPOST PUNDITS ON SUICIDE WATCH=

archive.is/ipDGm#selection-3223.0-3229.183

2 days before the election huffpost attacked 538 claiming that giving Trump 35% was way too high and their brilliant model was putting Clinton at 99%. They were saying that Nate was just scaring democrats.

I regret not archiving it earlier as I think this may be a poor addition but it has some great quotes.

COULD NOT RECOGNIZE THE POLLS THEMSELVES AS BEING GARBAGE

(interestingly I followed 538 a bit. All of the Google customer survey polls picked it for all the states I have seen… maybe someone can confirm? looked really spot on to me and the predictions on Holla Forums)

Another good one –
archive.is/I2q15

=They drank their own koolaid=

I wonder if forecasting the landslide Clinton wins it dissuaded democrats to even bother voting?

I've only seen ONE poll who's actually been good. And that's LAtimes.. Which also was closest last election.

Augury BTFO

Remember that fucking ABC poll with Clinton up by 13%? Holy shit they're so full of it.

I am retarded in math. The highest math I ever took was stats 101 at a community college and I got a C-. I could see the bullshit in the polls. This sniveling kike let his own perverse lust for the destruction of the White Race cloud the validity of the numbers and he ran with it. With glee he looked out from the leftist circlejerk of awesome he was to spout lies he wasto blind, or too stupid to see.

Let him reap what he has sewn. Hint: it's the end of a rope

The IBD/TIPP poll also had Trump leading and it's been the most accurate of the last four elections now

...

Then don't be in the business of hard numbers, or stay in something inconsequential like baseball. This is the problem, it's what Thomas Sowell calls "the anointed". He's semi-intelligent/successful in one area and assumes it will carry over into other areas as well, and when it doesn't, well gosh, my "formulas" must have been "wrong". Meanwhile, entry-level political novices could have figured out Hillary was not going to get anywhere near the turnout Obama did, and that's kind of a big deal when your entire strategy hinges on people voting as they're "supposed" to.

His methodology reflected what he wanted and hoped would happen, and he was unable to adjust until the 11th hour because he doesn't know enough about politics to realize this was coming for a very long time. It didn't take a ton of perspective and insight to realize this was essentially a role-reversal of the 2008 race.

It may wonder you all, but huffpopo is accurate and there is a scientific method to determine exactly the accuracy of their publication and journalistic work.

They are 1,7% accurate. More than I thought. Who knew?

web.archive.org/web/20161109220752/elections.huffingtonpost.com/2016/forecast/president

Is this photoshopped?

Just fuck my entire professional field up fam.

Just do it.

Some of them even tried to defend him

Why. Literally what the fuck has he ever done to earn anyone's respect.

Sure but Nate wasn't conducting raw statistical analysis, he was using it to make predictions, of which there is the assumption that the statistical analysis reflects reality. He never even attempted to reflect reality, and so his predictions were completely worthless.

The Trafalgar Group had Trump winning as well. Their projections in FL and PA were also extremely accurate.

There's lots of asians against affirmative action, etc.

Good-bye, (((poll))).

Hello, Holla Forums.

The polling industry is not out of business.

Media faggots are still, without a trace of self-awareness, saying that Obama has high approval ratings.

Hey Nate.
GO CUBS GO