Nate silver

Is coming to my university. Any questions I should ask him?

Other urls found in this thread:

projects.fivethirtyeight.com/2016-election-forecast/#plus
fivethirtyeight.com/features/how-i-acted-like-a-pundit-and-screwed-up-on-donald-trump/
twitter.com/natesilver538/status/730251094614528000?lang=en
myredditnudes.com/
twitter.com/NSFWRedditVideo

ask him why he is always wrong

Why the combover?

Dont ask him shit, instead yell out "JUST FUCK MY POLLS UP SENPAI" and make sure to record it.

This but if I ask him like that everyone will just dismiss me as >alt-right. I want to ask in a way that he gets btfo

F A M not senpai obviously

Ask him why he's a relentless faggot

Ask him how much he's being paid by your university and why that should come out of your tuition and for what purpose.

im with aussie bro
be sure to periscope it

Ask him how he can make a living from being consistently wrong.
Also ask some basic math questions. Hilarity will surely ensue

Tell him you're really bad with numbers and if that means you also might one day run an aggregate polling website of your own.

One way might be to ask him why he didn't factor in the size of Trump's rallies (which were always so much larger than Clinton's) as well as the google analytics data showing Wikileaks was by far the most searched topic during the election.

For instance, there are companies which now say they can determine which movies will smash the box office based on social media metrics like tweets. If going by that, surely he should have known Trump had much better odds than what every poll was claiming?

Basically, point out that he ignored all the evidence and is a partisan hack.

this

ask him how he was so wrong on the election polls and why he still has a carrer after that

So far I like this question best.

Ask him who are you

ask him: "Why are you wasting our time and yours by coming here to talk about anything? You're obviously either inherently wrong about everything or trying to manipulate people by being grossly incorrect."

Trump ruined him.

Bonus points if you call him Mr. Copper or some other element during the course of the question.

Hey Barbados Slim.

Ask if he's changing his name to nate bronze

Ask him why in the fuck he would not take into account that polling institutions were using the models from the Obama elections when they were so clearly flawed as a methodology.

>Give Trump a 20% chance of winning
>Trump wins
>Somehow this means all math is wrong

Do you also tell everyone that the casino is lying about the Evens or Odds squares being more likely when 00 comes up in roulette?

More information: it's hosted by the computing and idata science department and is being described as an "intimate chat" and Q & A session. It's not open to anyone so there won't be many people and prominent members of the department will be in attendance. That's why I want to ask him in a way that makes him look like a retard.

why do you wear a combover?

>give trump 20%
>20%
he had trump at around 6% first 3 maybe more polls, he knew hilliary didnt have that kind of support after seing the trump rally. the whole point of the question is to make him look like a complete fucking ass. he knew hilliary didnt have this kind of support yet pushed this bullshit to try demoralize people. he knew what he was doing. this kike has been doing this for a while and he is good at it. you are seriously trying to argue this kike faggot didnt know that trump had this kind of support. he not only looks at the opposition support but also his own party's support to get the most accurate readings. heres yout (you) shill.

projects.fivethirtyeight.com/2016-election-forecast/#plus

>28.2% chance on election day
>At no point was Trump ever at 6%

WOW A NATE PLASTIC DEFENSE FORCE SHILL

"are you still allowing your personal political preferences to skew your results like you did in the 2016 presidential election?"

or, if you want to be less confrontational

"After the the glaring inaccuracy of almost all polling except the LA Times during the 2016 presidential race, how can professional polling regain any semlance of leitimacy or crediblity in the minds of the American electorate?"

FPBP

Everytime you ask him a question, change his last name to another element.
Mr. Gold, Mr. Aluminum, Nate Tin, etc.

this

The casino is far less predictable, the fact is it's not just random chance trump was elected and would've been predictable with better measurements

>july 30 through november he was below 20%
>around 6%
learn to read
>comparing election day to first day of campaign
wow, literally not an argument
kys shill, your (((prophet))) is wrong and is done.

>gaslighting this might

We were all there, why are you trying to convince us that your fiction us more accurate than what we witnessed. Week after week of "This is the end of Blumpf." Suddenly, Nate dindu nuffin wrong.

Nate get back to work.

kek, this damage control at being proven wrong

You didn't make your point at all. There wasn't a single one of Nate's models that predicted Trump winning the states he did.

HAHAHA

LOOK AT THE TOP OF HIS HEAD

AHAHAHAHAHAHAHAHA

It's alright to say Nate got blown the fuck out. Every pollster did.

...

>this faggot doesn't know what probability is

You need to be 18 to post here.

If you give something a 60% probability, and it happens 100% of the time, that means you were wrong. If it doesn't happen sometimes, that means you were right.

>a psychic animal literally predicted the election outcome accurately where you couldnt. do you think future pollsters like you will be completely replaced by such animals?

wtf I hate math now

>pollsters are never wrong unless they give something a zero percent chance of happening

Nate got blown out. He even admitted he acted like a pundit and screwed the numbers up.

Not a single one of his models predicted the outcome, or even came close. The 28% chance was based on Trump barely winning with Ohio, Florida, North Carolina, and Virginia.

>all this butthurt

He did the best possible job with the data at hand. The odds were against Trump, but not outrageously so. In the end, Trump won.

There is nothing contradictory or deligitimizing about all this. In fact, it's called statistics. Go to if you got any more questions.

>Any questions I should ask him?
Ask if his model accounted for meme magic

Thank Ahmed

>proven wrong
all your doing is proving us right by how this faggot purposely lied and gave Hillary overwhelming chances of winning when Hillary couldn't even attend rally's let alone fill them. you are trying to say that trump had lost massive amounts of support after debates he did fucking great in and that he gained more support from. that he started off at 30 some % and stayed below almost all the way till the end of the election and still btfo hilliary and her 3+ million illegal votes and votes from voter fruad. over whelming evidence proving she had no chance yet everyone including nate silver saying she had 80-90% chance to win and only got some 230 electoral votes. she got fucking destoryed, he knew she had no chance yet continued to lie.
>using a straw man
stop, you already lost

Question:

Considering the accuracy of modern political polling, that is to say - frequently wrong, what separates you from Romainian gypsies who also promise to tell us the future by looking at a crystal ball?

He even admitted he acted like a pundit instead of a pollster after the nomination was over. He then turned around and made the same mistake during the general. The data was all there, ask the autistic Portuguese poster in PTG who called every state right.

fivethirtyeight.com/features/how-i-acted-like-a-pundit-and-screwed-up-on-donald-trump/

form this post into some sort of question

Nate isn't even a pollster. He makes statistical models based on other people's polls.

>He even admitted he acted like a pundit and screwed the numbers up.

Someone didn't actually read what Nate wrote and is basing his entire argument off of a headline.

But go on, keep talking about things you have no idea about. It makes you look really smart, trust me.

>gave Hillary overwhelming chances of winning
>60%

mate, its no use arguing with them

>e even admitted he acted like a pundit instead of a pollster after the nomination was over.

Every day I wonder if Trump supporters can actually read, or if they just fake it like Trump does.

Let's take a look at the actual article, eh?

>The big mistake is a curious one for a website that focuses on statistics. Unlike virtually every other forecast we publish at FiveThirtyEight — including the primary and caucus projections I just mentioned — our early estimates of Trump’s chances weren’t based on a statistical model. Instead, they were what we “subjective odds” — which is to say, educated guesses.

Oh, he's not talking about his models at all, he's talking about the opinion pieces he wrote, that weren't part of his models.

>HE DID THE BEST HE COULD
no he didn't, he lied the entire campaign to try and get hilliary more sway with undecided votes. any literal brain dead person would have to do is look at the rally's they had and could make a better prediction. like i said earlier, hilliary couldn't even attend rally's let alone fill them because she had so little support.

as if it wasn't delibrate

They were carrying the water for Hillary through the entire election cycle. They were using (((massaged))) polls to promote one particular candidate. Do you even remember their sampling methodology? Democrats were purposely over represented in each and every poll except the LA Times. I guess they prefer to be viewed as incompetent rather than frauds who were trying to rig public opinion.

>pre-election days
>based barbarossa of statistics and accurate polling parting the clouds and bringing hope
>PA red
I didn't believe him. I didn't think you could get tired of winning.

You can play with numbers all you want. Here are the undisputed facts:

A 71.8% chance of Hillary Clinton winning the election
Hillary Clinton: 301 electoral votes, Trump 235

Actual Result:
Hillary Clinton: 232 electoral votes, Trump 306

Nate got it wrong.

Just call him Nate Copper or something and pretend it was a slip

>60%
using your link to his website this is the first thing that pops up, thats on election day. 1 sec and ill get the other links nad pics from when her predicted her at much higher

Then how come he lost the popular by the(or one of the) largest margins in history?

He did the best he could. The problem is that state polling was so much worse than national polling. If you go by national polls, the average is pretty close to what we got in the end. Thanks to the electoral college, however, this doesn't matter.

I won a scratch ticket once, even though the back of the ticket said I only had a 1 in 3 chance of winning.

Statistics BTFO am I right? Why did they lie on the ticket?

>MUH RALLIES
wouldnt that be more of an indicator of the popular vote at best?

but even then it only accurately gouges the most enthusiastic supporters and not the whole base

>Overwhelming chances
>70%

WOW at mid october he had her at 85% OMG!!!!!
thats called overwhelming chances shill.

>lost the popular
she had 3+ million and still counting illegal votes from illegal immigrants and voter fruad. and popular vote doesn't mean shit in a constitutional republic.
>using star trek over and over as a pop culture reference
your showing your shill side to much

I go to Charlotte too, Don't be a cunt to balding nate. He probably wont sign my printed tweet if you do.


twitter.com/natesilver538/status/730251094614528000?lang=en

Nate's prediction wasn't binary. He attempted to guess the outcome of each state and failed miserably. Look at He can blame the reliability of state polls, but Nate has professional pollsters on staff, surely he should have known state polls were unreliable before the results came in.

If I am not completly mistaken, this was at the time when the hollywood access tapes surfaced, right?

If the elections would've been taken then and there, then yes, I'd have also said that Hillarys chances were overwhelming.

probably because polls at the time reflected that? Unless you go into whats wrong with his methodology you cant really prove him wrong

Supporter enthusiasm is an important metric, for each person who goes to rallies, there are many more willing to talk about and defend their candidate
A common complaint among liberals was how during this election everything become about politics, and how it drove a wedge through their social circles. Supporting Hillary wasn't as fun and interesting as supporting Trump.

Ask him why he let personal opinions influence his statistic bias during the election.

but you used actual crowd size as a measure of trump's support, in that case popular vote would fucking matter

What was the transition like?

>wouldnt that be more of an indicator of the popular vote at best?
yes, for a individual states which wins you electoral votes. electoral voters vote on the majority vote, trump had majority so electorates voted for him because they see from popular vote of their state that's what the people of that state want.

>Some states with a 40% chance to go to Trump went to Trump
>States with a 99% chance to stay with Trump stayed with Trump

Wow what are the odds?

>she had 3+ million
Right.
>and still counting illegal votes from illegal immigrants and voter fruad.
Baseless claim.
>and popular vote doesn't mean shit in a constitutional republic.
Re-read what I said, I did not dispute this. In fact, it is my whole point. State polling was shit and the model suffered for it.

the tapes idk, but if i remember right that was about the time of the third debate where nigel had trained him on debating and even places like cnn admitted he did a amazing job at the debate.

My favorite thing about the crowd size meme was that it was the exact argument Bernie supporters used, and Trump supporters mocked them for.

Ask him if he's going to change his name to Nate Bronze now since he was big league wrong on the World Series and the Election.

>aggregates polls with statistical models
so, hes a pollster. and a very poor one at that. why shill for him? he was wrong and btfo by every conceivable metric.

no not all polls reflected that, just the ones he wanted to look at and try to use as evidence to back up his claims.

The metals have to get lesser in value each time. First time Mr silver, then bronze, copper, aluminum, ore.

didnt trump get bigger crowds no matter where he went though? sounds like a shit metric to me desu. If you made predictions based on it i bet ud get like 90%+ trump victory which is straight up retarded

No, the difference is that Bernie pulled large crowds in large liberal areas. But Trump pulled large crowds everywhere he went.

What's next on your shill schedule? Say that CNN, MSNBC, CNBC, etc etc. Election month polls were legit?

actual crowd size proves something, that he is wanted, welcomed and supported in a state. it doesn't prove his entire popularity in the country just the state. and winning the state popularity wins you the electoral votes needed to win.

obligatory

Ask him what incentive do pollsters have to collect accurate data now that their main source of income(newspaper articles) have been syphoned off by aggregate services like his?

Take that shit to Kinko's right now!!

You should become a lecturer.

you would yes, but what i was stating by that was by even a retard looking at rally sizes would give trump more than a 20% chance of winning like nate did. that is my point. by rally size alone he should have know that he had way more support that 20%

The close states are the ones he's paid to get right. Everyone know Texas is going red, he doesn't get credit for getting that right. Take a look at how he called the close states.

The problem isn't Nate, it's the entire polling industry. Nate did the best of the pollsters, but he should still catch alot of hell for not seeing obvious bias in the polls.

Yeah, large liberal areas like Texas.

Rallies were painted as a big deal with Obama had the crowds. The media dismissed them completely for Trump.

ask him if he was too stingy to afford the last name gold.

>Well within the margin of error in all of them

Good to see you've never taken a stats class in your life.

But played Hillary's as big as they could by moving the cameras up and curtaining off half the building. Everyone knew it was her turn and everyone said fuck that.

>baseless claim
>what are the dnc leaks proving democrat voter fruad.
state polling was shit because his own or hilliarys people would only go to states that SHE OWNED like CA and do polls. they lied to themselves intentionally to try and make it look like they would win

where has his hair gone