Literally wrong about everything so far in 2016

>statistics is science
> this man has failed, thus the discipline must be wrong

It's a tool used by numerous sciences, you don't say mathematics is a science, it's a discipline of analysis of datas. When used in science, it's used to show a general tendency, to predict results or to arrange datas in a way that show their inner foundations.

>thinking that political preferences necessarily imply anything abou problem solving ability or IQ

your faulty thought process reveals you to be the brainlet in this situation.

>Statistics isn't a real Science
Oh well, I guess all those large companies that use proposed statistical data to accept/deny changes to their current systems are all crazy and should fire everyone using statistics for their requested RFCs approved.
Oh yeah, and all that jazz about statisticians being able to exploit lotteries? Pshh, utter nonsense of course.
I don't even know why Statistics is even a course in college or anything. Makes no sense.
Not like it's mathematically sound or reproducible or anything along those lines. Should be a banned practice.

...

psychologytoday.com/blog/the-scientific-fundamentalist/201003/why-liberals-are-more-intelligent-conservatives

Trump and his supporters aren't conservative though. They're fascists.

they support conservative policies

do people believe this shit?

please go back

Veeky Forums - Politically Incorrect

great rebuttal

an appropriate response for your assertion

>disprove assertion

>that shit's d..dumb!

>universal healthcare
>trust-busting
>protectionism
>pro-legal immigration
>conservative policies

do you even know what Trump supports?

great rebuttal

Somehow political discourse is worse on Veeky Forums than /pol/. I don't fucking know how they do it here.

>banning muslims
>building a wall to end immigration
>deporting millions
>slashing taxes on the wealthy

Look, ma, I can cherry pick too!

Taleb barely understands Bayesian stats, and most of what he does is repackage ideas that statisticians already knew and presents them as revolutionary.

That said it's a pretty neat way to introduce statistics to non-math people. Just don't take it as all that new.

What on earth are you talking about? Have you been paying any attention? The trump supporters and the conservatives were at each others throats during the primaries because of Trump not supporting conservative policies. Trump selected Mike Pence as his VP for the sole purpose of appealing to Christian conservatives because he was losing them due to his treatment of Ted Cruz. The republican party is completely fractured. You have absolutely no idea what a conservative is do you?

Dunning-Kruger effect + selection bias.

Veeky Forums is a board for people who devote their time to science and math, not politics (unlike /pol/). Therefore, all things equal the level of political discourse here will be worse than /pol/, just as the level of scientific discourse will be worse on /pol/ than Veeky Forums.

Moreover, the politically illiterate Veeky Forumstizenry are more likely to be self-aware and thus refrain from wading into political discussions on Veeky Forums. There are several possible reasons for this:
(a) they abide to the norm/rule of board-appropriate discussions;
(b) they are more likely to have a scientific mindset, and in the framework of science informed opinions are strongly preferred to uninformed ones, unlike politics which arguably promotes active participation by all;
(c) This is Veeky Forums, and pic related is in application.

Combined with the aforementioned Dunning-Kruger effect where the idiots are cocksure and the knowledgeable are full of skepticism and doubt, and it's no surprise that the quality of political discussion on Veeky Forums pales in comparison to other boards (hell, I've seen smarter conversations on /a/).

It's a matter of chained correlations. Trump supporters are mostly republicans, republicans are mostly conservatives, therefore Trump supporters are conservatives.

Also, the fact that conservatives don't like Trump doesn't mean that they're not voting for him. They're not as enthusiastic about him as they were about Romney or Cruz, but they're still going to vote along party lines.

>banning muslim [immigrants from large terrorist countries]
>building a wall to end [illegal] immigration
>deporting millions [of illegal immigrants]
>slashing taxes on [everyone]

sounds good to me

>chained correlations
It's just a fuggin syllogism, you minx.

>trusting a kike

>chained correlations
bullshit
you won't find this term in any statistics textbook

>Males are mostly human
>Humans are mostly female
>Therefore males are mostly female

fug forgot pic related

He was doomed to fuck up as soon as he started letting people he agreed with use him as a political tool.

Statistics is science. Have fun memorizing terms in your marine biology picture book!

Silver liked the "rational statistician triumphs over emotive pundits" narrative so much he started seeing it everywhere. Sooner or later it was bound to catch up with him.

Lies, da no lies, statistics.

Statistics demands great discipline to avoid confirmation bias and reading things into the data set which cannot actually be inferred.

If you're not careful before you know it you're doing a bunch of Wilcoxon tests or something on simulated samples in order to get the result you want.

>Lies, da no lies, statistics.
Dat autocorrect.

mental illness

Silver has said himself that most of his fuckups were because he didn't actually believe the results of his models.

His models said that Trump would win the whole time, it was his own flawed judgement that was telling him to say he didn't have a chance.

>Psychology/10
/BTFO

>the effect of childhood intelligence on adult political ideology is twice as large as the effect of either sex or race
>blacks have an average IQ score 20 lower than that of whites
seems like they're getting mixed results

As a follow up, no science as conducted, just conjecture.

Valid but unsound, if I remember my logic class correctly

>Literally wrong about everything so far in 2016
His model follows the polls, which up until mid last week were very clear. The race is tightening, and consequently the prediction of model is too. Pic related. Let's not forget that the only prediction that truely matters is the one the model makes on the day of the election.

As for the primaries: yes he was off initially. That was due to the fact that primaries are not a singular event but a series of non linearly interacting chain of events. In addition, the model takes a Bayesian prior that specifies the expected chances for any candidate winning the nomination. That variable is picked by hand, because you have no data to go on, and he has admitted that this is where he went wrong.

Lastly, the fact that the predictions that the model makes are probabilistic doesn't mean that the model can never be wrong. It only means that the model is expected to be wrong about as often as its predicted probability for the least likely outcome. So in this case, a 32.2% chance for Trump winning the election means that the model is going to be wrong 32.2% of the time, which is a substantial proportion. Silver doesn't deny that, in fact he explicitly states this often. People simply misinterpret probabilistic predictions. And going by this thread, that includes people here.

>the level of political discourse here will be worse than /pol/
kek

>descartes
The picture represents Voltaire not René

The quote is also completely made up. It's a joke picture, genius.

>“we have to take out their families“
>taxplan which adds trillions to national debt because they believe less taxes and larger military somehow equals more money for the government
>torture "even if it doesn't work" and "do a heck lot worse than waterboarding"
>"we will open up libel laws"
>build a really expensive wall against Mexico even though there's a huge fence with thousands of sensors, activist border guards, border guards, etc, etc, and have Mexico pay for it (war?)
>why can't we use nukes why can't we use nukes why can't we...
>a deportation force and camps to get rid of all illegal immigrants which will be really expensive not to mention aliens contribute billions in taxes
>vague policies and constant contradictory statements, yet people proclaims he "tells it as it is"
>failed businessman who was a billion dollars in the hole at one point according to leaked documents, multiple failed businesses, literally under investigation for fraud with regards to Trump University (not to mention will be on trial for child rape), yet his business senses will help this country because he will default on the national debt (as if that's a thing)
>Mr. Anti-Establishment even though he begged for bucks in the beginning of the campaign only to be served this Anti-Establishment idea by a representative from a huge superPAC who just wanted to leave, and now he takes money from a Jeb Bush pac he specifically criticized during the primary
>obviously a malignant narcissism with low self-esteem who is only concerned with himself, yet he clearly cares and will help the people
I swear I can go on. And I swear I can make a similar list for $Hillary at least as long.

Fuck me people are dumb. I hate everyone. Nuclear war when? Because I can't fucking wait!

This is why I hate politics. Now get the fuck off my board with this asinine prattle about everything that's wrong with humanity. We deal with facts here. Kill yourselves.

this kind of cognitive dissonance belongs on /pol/ or /x/

> 32.2% chance for Trump winning the election means that the model is going to be wrong 32.2% of the time
you mean is going to be RIGHT 32.2% of the time, right ?
We all know who's the winner now.

>tfw you have to decide between either corrupt politician or corrupt businessman
>neither side make no effort at all to hide said fact and simply bank on the pure hate of the people to win the election

...

Friendly reminder that this hack got btfo by Nassim on multiple ocassions already.

>Taleb barely understands Bayesian stats
I hope you meant to say that the book isn't quite rigorous, otherwise terminate your existence, please.
>repackage ideas that statisticians already knew and presents them as revolutionary
He never does that. In fact, he's pretty explicit about his intent of revealing the blatant lies and ignorance of hacks like Silver, and that the techniques he uses have been known for a very long time.

Go to bed, (((Nate))).

> mathemtica
Trash.tif

refer bias to self

BTFO
T
F
O

clinton = 25|sin(x)|+50
trump = -25|sin(x)|+50

what is this even complaining about

significant digit anal aggravation

muh tenths decimals

So the whole shitstorm is basically because one guy did statistic and gambled for high probability while the coin flip went for low probability?

He did that many times, though, and made predictions a year in advance about things that are unpredictable.

Sounds like he is just cocky shit that loves attention. How's that related to statistics as a whole?