>Dean Chambers, the man who garnered praise from the right and notoriety on the left for his "Unskewed Polling" site, admitted today that his method was flawed.
>"Nate Silver was right, and I was wrong," Chambers said in a phone interview.
>Chambers' method of "unskewing" polls involved re-weighting the sample to match what he believed the electorate would look like, in terms of party identification. He thought the electorate would lean more Republican when mainstream pollsters routinely found samples that leaned Democratic.
>But as it turned out, the pollsters were right — self-identified Democrats outnumbered Republicans by 6% in election exit polls.
>updated their polls this morning >had Clinton winning >site goes down, likely for maintenance >B..b...but Trump was winning yesterday!
You are retarded
Adam Cooper
shill thread
sage
Wyatt Rivera
They call for a tight race when a realistic assessment has Clinton winning by a good margin. They called for a Clinton loss when a realistic assessment had Clinton winning by a small margin. They're delusional and I hope I you post on Nov 8th.
Noah Lewis
She will never be their Pres. She is damaged goods and will melt.
Juan Murphy
>a retarded kraut who's probably never been to America is telling me how the American election will go
Jaxon Wright
I've been to America 2x, and I hope to go back soon. I liked it. Yours is a beautiful and diverse country, with exceptional beauty and exceptional ugliness.
I don't believe I have any particular insight into this election. I'm looking at your pollsters and poll aggregators though, who DO have one, because it's their fucking job, and your free market would punish the fuck out of them if they weren't good at it, and they're telling me Trump is gonna lose.
Luke Brooks
Free market is an illusion. The pollsters are all owned by six media conglomerates. Five have donated to the Clinton campaign and were revealed to be physical arms of the Democratic party which collude with each other through leaked emails. The sixth doesn't want Trump to win because it has financial interest with neocons who want more wars in the middle east. It is probably also colluding with the other five because it changes methodology in unison with the others
The polls actually are skewed this time because these conglomerates have full interest in Clinton winning this election. They have changed their methodology three times so far. Each time, Hillary takes a huge lead. Their current methodology is to oversample people aged under 30 and to not give them the choice of selecting a third party candidate. They're acting as if this age group will be 30% of the electorate when usually it's about 15% and that turnout will be much lower this year because millennials hate both Trump and Clinton. But when offered a choice between the two, they overwhelming choose Clinton.
You can see this in polls like NBC, which identifies everyone as registered voters rather than likely voters and puts in their methodology that they ask for the youngest member of the household. They give Clinton a 10+ pt lead.
Meanwhile places like LA Times and People's Pundit Daily which are still independent and are trying to accurately create a likely electorate rather than issue propaganda have Clinton only winning slightly by 2-4pts.
If you lived here and talked to the average person, you would realize this election is still very close as the genuine polls indicate.
Nolan Ramirez
>The polls actually are skewed this time >this time it'll be different
>If you lived here and talked to the average person You have spoken to approximately 20 more Americans in real life about politics this year than I have.
Sure, that's almost 1000% more. But it's still a laughably tiny sample.
Michael Scott
>Longroom new delusion
Adam Martinez
except the actual polls themselves are done by small privately held businesses, not by the big media companies
Gavin Phillips
>must protect muh redpill circlejerk
Josiah Wright
>literally tell you exactly what they're doing and why >literally tell you what the legitimate polls are >still won't listen
>approximately 20 more Americans about politics You don't need to physically talk about politics. It's the mood of the nation. A huge portion of the population thinks the same way as Trump and is tired of Washington. People are actually so angry at the government establishment that there's an unwritten rule in America that you shouldn't bring up politics in conversation because everyone will just go on an angry rant about how shit the government is
Do you know how they work? >contact polling agencies >tell them what you're polling and what methodology to use >agencies carry out orders >give you results back
Anyone can have these agencies do anything for you
Zachary Jenkins
>It's the mood of the nation What frequency must I tune my spirit crystals to to feel the mood of the nation?
(You're 300 million people spread over one of the largest countries in the world.)
>A huge portion of the population thinks the same way as Trump and is tired of Washington How do you know that? What methods have you used to figure this out?
You know there are groups who 1. have a lot more money to spend on this than you, 2. have a lot more expertise with this than you, 3. are spending all of their skills and resources on figuring this out, and they are - on aggregate - coming to the opposite conclusion of you?
Maybe your crystal is tuned wrong.
Isaiah Ross
>opposite conclusion to you
We're literally both saying Clinton is winning. You are retarded
Isaac Sullivan
ok
i have 2 options listen to nate silver listen to you
which should i pick
note: ur both americn!!
Sebastian Lewis
Nate Silver is just as disconnected as Dean Chambers. He just picks the Democrats and hopes for the best. He's even altering these polls to give Clinton a greater lead. He's praised when the Democrats win but got BTFO when the Republicans won the house and senate against his model. He's so delusional that he thinks Clinton has a greater chance at winning Texas than Trump does of Florida
You're not helping your cause at all
Noah Reyes
so you're saying i should listen to you over nate silver
nov 8 will be hilarious
Gavin Campbell
>over nate silver (((Silver))) has been consistently wrong about Trump for basically a year now.
Tyler Collins
I don't even know what you're trying to argue. We're both saying Clinton is winning
Read that. It's an in depth explanation on how different polls are weighted, or in the parlance of a trumptard: "unskewed".
I guess that doesn't address your initial speculation that some shadowy cabal of influence peddlers is working in concert, but I don't have any steel mesh gloves to peel away that many layers of tinfoil
Nicholas Lewis
Do you also agree Nov 8th will most likely be hilarious due to all the Trump fans who believed Longroom et al suddenly being confronted with the reality of a male first lady?
It's spelled >(((shadowy cabal of influence peddlers))) actually
Luke Collins
Either the polls are being manipulated, in which case Trump is winning and keep doing what he's doing.
Or the polls are true, in which case who fucking cares about polls? The goal stays the same no matter what.
Carson Phillips
Also fuck Nate for not putting his model on Github.
Matthew Mitchell
>leaked emails prove there's media collusion >polls are very open about changing methodology and put it right in the report >still calling it tinfoil wew lad
Your link is just bullet points and tells nothing about how he weights his polls. Longroom of all places gave more information
It will be hilarious to see everyone who thinks Trump will get 300+ electoral votes. Clinton's presidency is still far from guaranteed however. Trump is still kicking within striking distance and Clinton is only being propped up by the media and an elaborate web of lies to hide her crimes. A lot could go wrong in her campaign
But Trump's max is still around 280 electoral votes and he has an uphill battle til November
Austin Hughes
Because he has no model. He arbitrarily weights polls to his feelings, much like Dean Chambers
Brayden Robinson
As an FSF member, I'm okay with this. Nothing compels him to share his source code unless he distributes some sort of locally run binary, and nothing says that you can't develop software for personal use while sharing the applied results of that program.
Anyways, I think it's possible that even Nate Silver's models have a slightly blue house effect, or even that the aggregate polls are on an extreme end of the margin of error or are even a statistical outlier. It's pretty unlikely, obviously, but the fact that we use words like "margin of error" and " outlier" gives an aura of plausibility.
I'll even allow that the guy you're arguing with is correct and thinking the "true" number is the extreme floor of 4%. That's still enough for Hillary to win and Dems to take the Senate.
Anyone who thinks 4% and likely more isn't a commanding lead ought to take watch some videos about statistics
Ayden Collins
>Because he has no model. He arbitrarily weights polls to his feelings ... you really believe this?
In that case, aren't you really impressed by how his feelings allowed him to correctly call the last 99 state votes correctly?
Also, do you have any evidence for it?
>As an FSF member, I'm okay with this. Nothing compels him to share his source code unless he distributes some sort of locally run binary, and nothing says that you can't develop software for personal use while sharing the applied results of that program. Nah I'm not really hating on him, I'd just really like to see it.
I heard originally it was just a bunch of excel spreadsheets.
Dominic Smith
>99 state votes correctly That's not as impressive as you think. The only one people couldn't see from a mile away was Indiana in 2008, which he got wrong. Meanwhile he got the 2014 senate race, the 2015 UK election, and the 2016 Republican primary so laughable wrong
He just picks the Democrats or the Democrat equivalent in foreign nations and hopes for the best. That is his model, which is why he gets any election where the conservatives win wrong.
Chase Howard
Rare
Matthew Morris
He also missed the Nebraska CD Obama won in 2008. Another hard one to predict that he couldn't
Brody Edwards
Checking
Carter Flores
>Literally every competent campaign has a media coordinator >Polls are conducted by professional pollsters who are incentivized to produce accurate polls for their clients
Longroom offers a paragraph where they claim to balance the numbers against voter registration data, but this claim is complete bullshit since:
Not all states collect party identification Even reliable partisan voters don't choose an affiliation when registering
Pollsters get the number of Dems, Reps, and Inds as part of the interview. Rarely, they will get additional demographic information (age, race, etc) and model those factors to determine the "likely voter" scores to ascribe final predictions. It's not rare to control the outcome; it's rare because it takes employees twice as long to ask the extra question and the polling firm wants as many calls possible in the shortest time possible.
A max of 280 is pretty dangerous, and by itself implies that Trump must carry a blue states like PA, NH, and VA simultaneously. Barring a happening or the greatest debates in years, Trump is already doomed and it will have real consequences for down ballot Rs
Mason Wright
>consequences for down ballot Rs
Even if Trump wins, the Republican party is dead
Caleb Gray
Fine, let's put it this way:
Which media group or personality has a better record than Nate Silver? Is there a verified single instance of more accurate predictions?
>Meanwhile he got the 2014 senate race, He predicted red >the 2015 UK election dunno about this one >and the 2016 Republican primary Yep, he didn't trust the model.
Jaxon Phillips
This isn't Longroom, Longroom accurately predicted Romney's loss (and was almost spot on with the percentage, too).
You'd have known that if you weren't a shill/
Landon Reed
Longroom.com has a perfect track record prediction every past presidency, including everyone which precedes Longroom.com, which is every single one of them
If you disagree, that's proof you're bought by Soros and/or ISIS.
>predicted I see your usage of the past tense, and disagree.
Parker Thomas
fucking what?
Luke Rogers
>... you really believe this?
NS has not provided any evidence to the contrary. Either he makes his secret open or we can just assume he's making shit up.
He can prove his worth.
Evan Richardson
You said "predicted". I say, the only election longroom has ever predicted is the current one.
The nature of time prevents you from predicting stuff preceding your own existence.
Jaxson Thomas
If you look on their website it explains how they've predicted all of the races since 2004 (I think) within .3%.
I'd be be more confident in saying it was 2004 if it wasn't for that fact that now I can't access Longroom. Is it just me or did someone shut it down?
Isaac Miller
>literally every competent campaign has a media coordinator They had the media pass stories through them before publishing to be sure they weren't damaging to the party
>polls are conducted by professional polsters who are incentivized to produce accurate polls no they're incentivized to poll the way their clients tell them to so that they're clients come back. The campaigns have insider pollsters that are more accurate that are not published
>is there a verified single instance of more accurate predictions No because elections are nearly impossible to predict correctly. Silver got lucky because his bias was in line with who was winning
>he predicted red no he didn't. His model did, but he said it was wrong. After this, he started dramatically altering his model
>bought by Soros and ISIS fuck off. No one's talking about boogie men
Jackson Lopez
>this thread is literally two shills talking to each other
Adrian Phillips
>If you look on their website it explains how they've predicted all of the races since 2004 (I think) within .3%. There you are using the past tense again,
>no he didn't. His model did, but he said it was wrong To be honest, I could care less for his opinions. But his models are spot on.
Nicholas Garcia
That's stupid. Me and nZzKJ5YH are both true believers. I believe in (((data))), he believes in a (((liberal))) conspiracy.
Liam Jenkins
You're either a shill or a retard.
Michael Richardson
It's not a liberal conspiracy. There are conservatives involved. Again you can't fucking read
It's the media conglomerates trying to protect their financial interests by using their platforms to get Clinton elected. Turn on any media platform right now, I guarantee they're telling you Trump will destroy the world and Clinton's a saint. With the exception of Fox which is saying Trump will destroy the world and Clinton's pretty bad so you should probably abstain from voting
Soros has nothing to do with the polls.
He's retarded
Jack Stewart
Is just me or longroom has been taken down?
James Diaz
Maybe they're as incapable of running a web server as of building a predictive model.
>It's not a liberal conspiracy. There are conservatives involved. Ok, then (((elite))) conspiracy. That better?
Hey, I hope you'll be checking in on the 8th.
Cameron Perez
I gotta be honest here. I have seen one hillary bumper sticker on a car in town here. Meanwhile ive seen five trump signs and one guy has a badass trump flag. If people are actually going to vote for her, they're certainly not proud of it.
Eli Morales
>"elites" spend years lobbying government to pass extremely favourable laws > man who won't take their money with one vote away from highest office threatens to repeal all those laws
You're a fool if you think anons narrative isn't plausible
Henry Ward
>I gotta be honest here. I have seen one hillary bumper sticker on a car in town here. Meanwhile ive seen five trump signs and one guy has a badass trump flag. The interesting thing is that Sup Forums is full of people who put more faith on stories like this than on polls.
They put more stories on one single individual looking at bumper stickers and yard signs, vs. the concerted efforts of large, extremely professional operations competing to get the best data possible.
Chase Perry
welcome to the selective reality distortion filter that is Sup Forums
