Nate Silver is a Pseudoscientific Clown

"In the mid-20th century, Karl Popper put forth the criterion of falsifiability to distinguish science from nonscience. Falsifiability means a result can be disproved."

Nate Silver claims that Hillary got 80 % to win the election and Trump 20 %. In other words "political scientist" Nate Silver is actually a pseudoscientist. Because he is making a statement that clearly can't be falsified.

If Hillary wins Nate will say that he was right and everybody will say that he is a genious. If Trump wins Nate will say that Trump got lucky after all if there's a 20 % chance of Trump winning he might just win won't he?

People who listen to this clown don't understand that he is just a con artist. He put on a suit and a pair of glasses. Shave his beard. Look professional and talk mumbo jumbo and people think he actually know what he's talking about. Of course Nate is nothing but a stooge shilling for Shillary on national television. I wonder how much they pay him.

Other urls found in this thread:

fivethirtyeight.com/features/how-i-acted-like-a-pundit-and-screwed-up-on-donald-trump/
en.wikipedia.org/wiki/Dearborn,_Michigan
en.wikipedia.org/wiki/List_of_Islamist_terrorist_attacks
en.wikipedia.org/wiki/Islam_in_the_United_States
slatestarcodex.com/2016/01/02/2015-predictions-calibration-results/
twitter.com/AnonBabble

literally who?

He has a good track record for predicting elections.

>hilldog isgonna win Virginia

>if she wins literally any other swing state it's over even if Trump gets everything else

You're a fucking retard. If you want to test a probabilistic prediction, what you do is check if when Nate Silver gives an event an 80% probability of happening, those events happens about 80% of the time in the long run. It's pretty simple.

>statistician with degrees from University of Chicago and London school of economics
>pseudoscientist

Pick one

That's what I'm saying. He's a nobody!

Let's say that Hillary wins, what does that mean? Hillary could win if there was a 1 % chance of Hillary winning as well as 99 % chance of winnig. As long as something is statistically possible it can happen. So how do we test his model? We can't. So it's not science. It's pseudoscience.

Nope.

>trusting institutions

I would guess that most "real" political scientists don't like him because he is blatantly partisan. He only became famous because he called the 2008 election very early on for Obama.

Math is math

>predicted the two easiest in history to predict elections, Obama's
>has not correctly predicted anything else ever

FYI last I check Silver had Hillary at 58.7% and Trump at 41.3%

Did you overdose on spuds paddy? You really think you can make a very accurate simulation of the voting habits of 300 million people?

...

I don't know senpai. He was spot on in 2008 and 2012. That plus he won me $100 with my office's March Madness pick. I'd say even if it's pseudoscience, it's accurate.

That sounds about right. People are going to come to their senses and realize Hillary, while evil and corrupt and self serving, can still hold the country together.

Trump's base of white niggers are just cosplaying as people with respectable opinions

>political Science
>math

pick one.
Anyway no honest scientist is blatantly biased when approaching what they study. Do you actually take poly science seriously?

Trump has been going up in his models every day by 2% for the past week.

>Making a probabilistic prediction is pseudoscience
If you want to call him out on that, you should look at how he makes those in the first place

...

Economics is the modern astrology, it's complicated math but without all the scientific rigor of a real science. I have a degree in economics, and would never say this if this weren't anonymous, but we are all frauds.

He's not a political scientist.

He's a statistical analyst who focuses mainly on baseball. He just applies the baseball metrics to other things for keks.

I kinda think Nate Silver is a hack. On his site Silver also gives predictions for MLB NBA and NFL results. I once converted his percentages into odds and compared wether you would have made a profit at the bookies with those odds. If you had had bet one dollar each time there was "value" in his predictions on the
NBA at the end of the season youd had made a loss of around 70 dollars. So yeah i think his predictions are bulshit.

Really all he did was get 2008 right, there are not many cases where a sitting president can't be re-elected. How hard is it to guess that everyone would vote for cool the black guy?

...

Oh wow, I remember when it was at 20% chance

Sports and elections are totally different. The sports predictions are actually more scientific because the models can be tested.

hahahaha oh wow

Please, keep believing Trump is the one that divides us; stay on the Democratic plantation and take your handouts for being a good black voter.

Well when all you get to choose from are wealthy millionaires and in this case billionaires. Its not hard to predict who will win.

The rich!

Political """"science"""" isn't science.

It's all about trying to predict the future using highly debatable ""models"" which are about 90% bullshit. You'd be better off sacrificing a goat and checking it's liver.

Nate actually admitted to practicing pseudoscience himself in his latest article

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

>Our early forecasts of Trump’s nomination chances weren’t based on a statistical model, which may have been most of the problem.

Given the amount of states which literally never vote differently like Texas, California, etc its not that hard to predict all states right.

>listen and believe

No. Fuck you, you disgusting, corrupt hatchet wound.

My degree is in MIS so I've had to take I think 6 economics courses. I would agree that it seems to be mostly shoddy algebra but I figure there are people much more qualified than me to say otherwise.

I don't think you understand, his prediction has swung 20% in a little over a month. Norpath's model gives Trump a success rate in the high 90's, it's based on primary turnout and electoral cycles. The argument could be made before that the high turnout in the primaries wouldn't benefit Trump because he's an outsider, but the Pence pick will likely bring the party together.

Kek
I'm voting for Hillary because I hate trailer park Americans. At least niggers make good music.

I don't even understand what you're talking about. If you didn't want to have pro-Trump/anti-Hillary rhetoric shouted at you, you came to the wrong image board.

and whose fault is that Mr. Silver?

You know he's a statistician right? Have you ever taken a statistics course?

OP you're probably one of those people who asks "what? How can they call it???" when 100% of the vote is not completely counted yet.

Oh, you're still in College. That explains a lot.

don't worry famalam, one day you'll grow up and either
1) get a job, become Republican
2) don't get a job, need handouts, become Democrat

But look at what he just said in his latest article

Were you too young to vote back in 2012? It was one of the closest elections in a long time. Certainly everybody here though Romney would get it and Obama would be out until pretty late in the election. It's not that he knew Obama would win. It's that he accurately predicted which states he would win with.

True. But he is still accurate even with the non-scientific population sampling.

that's really not that impressive. most states are solid blue or red. How many aren't, less than 10? And of those, how many could be considered ambiguous when you look at months of polling?

If it's that easy, draw up a map of how the election will turn out. Post it so we can all laugh at how terribly wrong you are in November.

States going one way or another is based off of hundreds of smaller districts. So sure you can say oh it's just two choices on whether Florida goes red or blue, but it's more than that. You have to consider the polling in those different regions and how likely they are to change based on previous elections.

If he was predicting Trump winning, you guys would be making memes about him and calling him based.

Face it people, Hillary will win. And i will be here

Why is this greenest? Are you quoting someone?

Nobody cares about you.

Your entire continent is already lost to the Jihadis.

When the Allied forces (U.S., U.K., Russia, China) finally make our landing on the beaches of the Islamic state, Portugul will be one of the first fronts that opens up.

Your country will be reduced to rubble, all because your fellow countrymen were too cucked and just surrendered to the muzzies

That is why this pic exists.

Hard Working, Real Americans, Democrats that work tirellesly to provide for trailer park trash white niggers retuglicans

it's the
>implying
meme

>someone says something implying something false
>you point out what they are implying whit green text.

His "probability model" is a con
He tricks the reader into thinking it's a prediction of an outcome, when it is merely a way of restating the prediction of the underlying poll. His probability model is the probability of the poll being correct. It amounts to saying "if this poll is correct it will be correct". He adds nothing to the underlying polls, and just represents their results in a way that tricks the reader into thinking it is making a future predicition.

It is not a prediction of future results, that it punditry, although it is understood by it s readers to be, that is the con. Predictions of future resukts is punditry, and Nate is a very bad pundit.

It's a false equivalency

I assume you are saying that
since: southern states vote red
then: southern states are representative of the Republican party

The same logic could be used to say that

since: southern states have the highest concentration of Blacks
then: southern states are representative of Black people

Geographical distribution of federal aid (by state vs. other state) has nothing to do with it. What you really look up is the Geographical distribution of federal aid (by rural vs. urban)

>Your entire continent is already lost to the Jihadis.
We have less muslims per capita then you

en.wikipedia.org/wiki/Dearborn,_Michigan

lel, you guys count muslims as white

>When the Allied forces (U.S., U.K., Russia, China)
it's crazy that alll 4 have more muslims then us in %.

>were too cucked
>again with the cuck fetishes

This is the typical american post, completely insulated from reality, and uses only memes.

yeah but the muslims we let in share our values (mostly)

you guys just bring them in by the boatload hahaha

let's be honest famalam, you're getting all the wrong kind of muslims

He presents poll data in a misleading way to make it seem like a probability prediction of a future result happening.

It's a con. The underlying polls may be valid, butbthey are not predicitions of a future event. They just represent what respondants have told the pollster AT THAT PARTICULAR POINT IN TIME.

Nate takes that poll, and then presents the data as if it were a strict probability predicition of a future result. It's not, it's a con.

You can average poll data and be exactly as accurate as Silver, it isn't hard. Generally eyeballing it by favoring newer and better studies is enough. Polls are very telling by themselves, Silver doesn't much, if any, value.

This is correct

Think of it as a probabilistic snapshot of time. Based on the aggregate polling and voting history. If the election were held today. Hillary would win in 80% of all simulated contests. If the polls shift the probability shifts. A month from now it could be completely different.

This.

What Silver does is con the reader into thinking he is adding something to the poll data through stats magic, and producing an objective odds prediction of the future event occurrance.

Readers think his "odds" are the odds of the actual future outcome, not just a way of restating what the poll reports people responded at that time.

When Nate tries to be a pundit, he is hopeless. He doesn't understand poltics. All he knows is how to represent poll data in a way that misleads lay readers into thinking it is saying something more than it is. He's a phoney and a charlatan.

>yeah but the muslims we let in share our values (mostly)
And the same applies to european nations, no matter how much Fox News says the otherwise

en.wikipedia.org/wiki/List_of_Islamist_terrorist_attacks
>ctrl+f
>type "States"
>19 results
>half of them are of terrorrist attacks
>then type "Portugal"
>0 results

>you're getting all the wrong kind of muslims

Becasue the somalis ones are better, amirite?

en.wikipedia.org/wiki/Islam_in_the_United_States


Yet another typical american post. BAsed solely on memes and never on reality

Which is saying nothing other than that the polls show Hillary is ahead at this point in time.

Nate adds nothing to the poll data. His odds prediction is adds nothing to what the underlying poll has already stated, and is deliberately misleading. Readers think it is a probability of the future outcome, like a magic odds handicapper that is infallible because science and statistics (that they don't understand).

Nate = charlatan

Yet another typical EU post. Based solely on the list their un-elected leaders tell them.

Seriously, you might as well be a fucking monarchy again, you serf

What reader think Nate's odds are:
>Odds handicap of the future event by an infallible statistical science magic

What Nate actually does
>Represent the likelihood of the underlying poll being correct by way of odds, a meaningless measurement choosen only because readers confuse it with an odds handicap

>Nate adds nothing to the poll data

I think his model weights the different polls

Statistical factors should be considered, but far more prominent are intangible things like mass psychology, history, and a deep understanding of the American people.

Jews have none of this because they are rootless cosmopolitans.

Statistical models are falsifiable, idiot.

>Based solely on the list their un-elected leaders tell them.
>he think EU regulations and directives are binding

Each individual state chooses of they want to follow EU regulations. There is always heated negotiations betwen member states and EU. The influence of each individual country dictates the outcome. Britain was pretty much already on a brexit status inside the EU. Yet another american who doesn't have a single clue of internal EU politics.

ANd not to mention the numerous times member states choose to completely ignore EU regulations, wich are incredibly numerous.

But yes, Eu parliament should be elected

>you serf
>coming from a american

You better not strike, or you could get fired ;)

Yet another american post, solely based on memes supplied by Fox News

As long as he continues to make predictions with probability values for the outcomes, it is actually possible to measure how accurate he is.
For an example of what I mean, see slatestarcodex.com/2016/01/02/2015-predictions-calibration-results/
Absolutely right.

>Each individual state chooses of they want to follow EU regulations
hahahahahahahahahhaha

Well he just based it on the saying "monkey see monkey do"

Hold on, I need to get better reaction images for this. I'll BRB

It's this condescending shit that will drive undecideds into Trump's camp. Trump's message is about lifting everyone up, while Hillary is obsessed with having us lower our heads in shame. You can't win an election with this rhetoric. Even Obama knew not to explicitly shame whites with this shit during his campaign.

*ahem*

AHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHA

Thats even worse though. A poll stands on its own merits, on the validity of its own methodology. It is either right or wrong. You don't get more accurate results by blending a good poll with bad ones. Either a poll is correct, or it is wrong. And there is no way of knowing until election day which it is. By blending polls together all you do is dilute good polls with bad ones, and elevate bad polls with good ones. You don't get a better result, you just add chaff to wheat and blend it together in a way that prevents you guessing which one was wheat and which one was chaff.

A poll of polls is more than meaningless, it is by definition misleading.

>memes are more important then reality

See folks, in here we have opurtunity to watch a american having a pathethic try at banter

Yeah, but how many times will Hillary run against Trump? Or do we count all times he has given something an 80% chance even if the context is completely different?

...

(((Karl Popper)))
(((Nate Silver)))

I literally haven't read any of your posts past the first line, and i'm pretty sure nobody else has either

...

Hilarious. Trump winning was obvious to anyone who saw Trump speaking on immigration and really understood what drove Republicans. An innate honor and love of country being violated en masse being given a champion?

And Nate's response is to become more autistic.

2016 has been completely dominated by memes becoming reality. Are you even paying attention?

He's biased towards the Democratic or 'establishment' candidate. See: this year's primaries. He shilled hard for Rubio.

>ebin

Well, i will have to keep my posts short and simple, with plenty of images, so it stays interesting to americans, since you have the atention span of my autistic cousin

NATE BRONZE

Never forget the day that Nate Bronze admitted that he was wrong on Trump.

since 2012 he's been wrong about pretty much everything, especially the UK election.

topkek, why doesn't Nate understand, you can't stop something like Trump, all this resistance is doing is making him even bigger, they should have never given him a platform to begin with, but the outrageous stuff he says sells too much papers and generates too much clicks, so they are forced to report on it, but they hate it, so they only report negatively on him, which only confirms the notion among his voters that they all are biased fucks, again making him win even harder, at first I thought this some kind of fortunate, but random positive reinforcement thing he stumbled into with the political climate being as it is, but I don't thing so anymore, the guy is either a fucking genius or really has the support of some kind of deity or higher being, he is like a tsunami wave that gets bigger and bigger the closer it gets to the shore and in November it's going to break, at which point you will either ride it or get crushed by it.

...

...

Lol why do you need a statistical model to figure it out? Trump obviously has a 50% chance of winning because he either will or will not.

It's going to be harder than the primaries though because now it is head to head (with a bunch of candidates they split their 50% odds of winning).

amazing

Silver is smart and although he is a liberal he is aware of his own biases and tries to make his projections fair. Tbqh for E!SPN he's basically far right.

I'll find you a coin among the thousands that also has a great track record of predicting elections

The entire field of computer modelling and "xxxmetrics" is a gross tumour on science.

Statistics are supposed to be a tool that empowers scientists... not an independent epistemology that makes a mockery of the scientific method and masquerades as science.

You must have a hypothesis that explains the phenomena you're looking at, THEN design an experiment or observational study that tests your hypothesis, THEN you can use statistical methods to analyze your results. You can't flip that on its head and try to extrapolate outcomes by pure number-crunching. That's not the scientific method.

Silver was kind of good but then he started making sports predictions. He ended up betting on sports and lost a lot of money and started drinking. He's just a mess now.

He's a moron, there is only 50% chance of Hillary winning, and 50% chance of Trump winning, because you either win or lose it's a 50/50.

I'm pretty sure Karl Popper didn't come up with falsifiability, I could be wrong, I know he dealt with demarcation in science, but it was more whether or not all philosophy should be included or no philosophy be included.

the big problem with science is we have one philosophy, Naturalism, that rules over all of them, even though the premises of all philosophies are technically equal at the outset.

No, it was a shared disgust at Cruz's childishness that brought the party together. There were a *lot* of republicans, top to bottom, who now loathe Cruz. The Texas delegation - who were, let us not forget, Cruz's delegation - booed him during his speech that night and his speech the next morning at breakfast.

Nope, that was Popper. Before that the accepted wisdom wss thet theories were proven true or false by experiment, Popper showed that all you can do is fail to prove false, or prove false. There is no proving true.

This demonstrated the fallacy of the aristotelian theory of inductive reasoning. It's all deduction, the closest you can come to Aristotelian induction is to do your deduction Memento style.

lol... youre a tard

Wait is he Michigan State fan? What a loser. RTR.