Probability of the 11.8% discrepancy in the NY exit poll is 1 in 236,000 according to mathematician Richard Charnin

Around the world, exit polls have been used to verify the integrity of elections. The United States has funded exit polls in Eastern Europe to detect fraud. Discrepancies between exit polls and the official vote count have been used to successfully overturn election results in Ukraine, Serbia, and Georgia."

Yet when the same thing happens here, the results are considered valid. So frustrating.

I really hate how we can't talk about how much this cluster fuck lost Bernie votes without be labeled a sore loser or a conspiracy nut. Not only were thousands of voters unable to vote, but even the ones that could vote faced other obstacles such as broken machines, or being told to go to different precincts. Think of all the voters who couldn't afford to wait around. I'm not saying Bernie would have won, but this cluster fuck definitely stole points away.

There used to be a time when these discrepancies immediately raised alarms around the world. Something stinks, or humans naturally become worse at polling science.

I really hate how we can't talk about how much this cluster fuck lost Bernie votes without be labeled a sore loser or a conspiracy nut.

This is carefully orchestrated. As soon as you mention something, the astroturfers will be there and tell everybody in the thread/comment section what a loonie you are.

Hijacking this top comment because this is a very important point of Charnin's blog that we aren't talking about enough:

The exit polls this primary cycle, not only in NY, but also AZ, MI, WI, MO, and MA, have been so different from the reported vote count that is almost statistically impossible for the reported vote count to be true and that the best explanation for the reported vote count results is election fraud, fraudulent vote counts, and manipulated votes.

And, it just so happens every reported vote has heavily favored Hilary.

Even in states like MI and WI that Sanders won, the exit polls indicate that he should have won by even more than the reported vote count.

Why is all the statistics and math that Charnin shows important? Because as of two weeks ago, Charnin estimated that if the reported primary results mirrored their exit polls, Bernie would gain at least 100 more delegates.

We cannot sit idly by and ignore this.

Of course, the only folks willing to even look at this possibility are the JFK conspiracy sites, so it will immediately be dismissed by just about everyone.

From OP's link: https://richardcharnin.wordpress.com/2016/03/11/mi-primary-bernie-did-much-better-than-the-record...

There's a difference between preference polls of potential voters and exit polls of people who actually voted. That the former were fucked up in MI is the reason it was considered an upset.

exit poll showed 52-48 base percent he used was the exit poll results for men. women comprised 59% of the vote (41% men), and women voted 61%-39% hillary.

combine the two (bernie: .39 * 59 + .52 * 41; HRC .61 * 59+.48 * 41) and you get bernie: 44.33% of overall vote, HRC 55.67%.

which is pretty close to the actual results of the primary (58-42) and as you can see, lies within the margin of error (44.33 - 2.6 margin of error = 41.73%), which gives more than a 16% chance that these results are accurate. this is assuming normal population distribution curve, which i have no idea is true, but charmin uses it so i'm going to use it as well.

so yeah. charmin got bad observed data from a dumbass reporter https://twitter.com/mj_lee/status/722590996131889152

the real exit poll results are here http://www.nytimes.com/interactive/2016/04/19/us/elections/new-york-primary-democratic-exit-polls...

This stopped around the time BushCo started stealing elections

Please don't let this fade away like AZ or any of the other states or let the astroturfers discourage discussion of this. This has quickly become the most important issue of the election.

It also says a lot that Arizona still is the only primary state to not release any exit polls, from any media outlet.

It absolutely isn't a stretch. I ran election protection in Ohio in 2004. Ohio was stolen.

Also remember this is not an isolated incident.

We’ve been keeping track of how Clinton’s and Sanders’s actual results have compared to initial exit polls, and the differences have been pretty random over the course of the campaign.

With that said, sometimes a candidate whose supporters are more enthusiastic can be overrated by exit polls because of response bias — this was sometimes an issue for Obama during the 2008 campaign, and it’s plausible we could see something parallel happening with Sanders. Furthermore, pre-election polls showed a wider lead for Clinton than the exit polls did, and it’s usually worth taking a blend of exit polls and pre-election polls even when you have exit polls in hand.

The Democratic exit poll — which has been re-calibrated to reflect Clinton’s larger-than-expected margin of victory — now has her winning 75 percent of the black vote in New York, along with 63 percent of the Hispanic vote. Clinton and Sanders split the white vote in New York almost evenly.

It’s true: the exit poll results are re-calibrated as the night goes along to match the actual vote count. In fairness, however, the people who conduct exit polls — and the networks and newspapers who pay for them — are quite insistent that exit polls are not intended to project election results and instead are mostly meant for demographic analysis after the fact.

Nate Silver

And...

If you want an idea of why the exit polls were off on the Democratic side, look no further than the 15th district. The 15th, which is the most Hispanic in the state, is favoring Clinton by over 40 percentage points. The exit poll had Clinton winning Hispanics statewide by 18 percentage points.

Harry Enton

FOOD FOR THOUGHT

In many states (North Carolina, Ohio, Illinois, NY...) the exit polling data was significantly different from the actual tabulated results (hurting Sanders).

JUDGE: "Have you reviewed at all the election results in Ohio?

WITNESS (computer programmer): "No, I haven't."

J: "Given the availability of such vote rigging software and the testimony you've been giving under oath of substantial statistical anomalies and gross differences between exit polling data and the actual tabulated results, do you have an opinion whether or not Ohio presidential election was hacked?"

W: "Yes, I would say it was. I mean IF YOU HAVE EXIT POLLING DATA THAT IS SIGNIFICANTLY ABOVE FROM THE VOTE, THAN IT'S PROBABLY HACKED."

J: "And your testimony is under oath."

W: "Yes, Sir."

J: "And the testimony you'd given is true."

W: "Yes, Sir."

Polling before != exit polling. Per the article, in MI Bernie's exit polling was actually right on point to a tad bit better than what he actually got.

the exit polls were consistent with Bernie's win. There's a difference between polls done in advance of an election and EXIT polls. Exit polls are polling people who actually just cast their votes. Exit polls are very reliable and if they're this far off, the election should be investigated.

I'm just playing devil's advocate here, but what did this guy have to say when Bernie won Michigan?

Because if you say otherwise, you are a sore loser conspiracy theorist. It is an American fact that questioning what happens behind the curtain means you're crazy. It is also a fact that wealthy people with vested interests never, ever collude against the public benefit. What are you, crazy? Take off your tinfoil hat and stop questioning things.

Look into Ohio. It's not a stretch.

Yet people will, this election is literally being stolen by the establishment, and no buildings are on fire. Good job team.

Sadly this is true. No reputable sources will dive into this, and therefore most of us will never be willing to share this. I sure won't, no matter how I feel about it.

Then they will start complaining about how the GOP establishment supresses Trump.

Arizona was not polled at the state level. Charnin wrote:

“The National Exit Pool of six major media conglomerates funds exit pollster Edison Research. The NEP decided not to poll AZ. It’s as if they knew they would have to match the unadjusted poll to a bogus recorded vote; the massive discrepancies would be too obvious. But the networks called it for Hillary with less than 1% of the votes in. How did they know this if they did not exit poll?

“Luckily the Yavapai County Daily Courier did an exit poll – and Bernie led by 63-37%. Hillary won the county by 54-43%- an impossible 37% difference in margin. But the evidence of fraud goes much further than this one poll.…”

https://richardcharnin.wordpress.com/2016/03/24/last-train-to-yuma-the-arizona-2016-democratic-pr...

Interesting fact: The CIA invented the term conspiracy theorist to discredit and ridicule what used to be known as government skeptics.

Well, there's a 1 in 236,000 chance. How much money would you be willing to lose using those odds?

The GOP establishment does suppress Trump. That doesn't mean this isn't fishy. Even Trump has called out the fishiness on the DNC side.

That makes zero sense. There were a number of polling places in Buffalo that didn't open until Noon and Buffalo is nowhere near representative of NYC, the most important district in the state by far.

That sounds a lot like an excuse from media to help explain why the statewide fraud caused such a huge discrepancy.

Because the elections ARE fraudulent.

Some small town news papers conducted their own. They were off by 10%.

It makes sense. The people voting for Bernie with affidavits didn't get counted yet. The question is will they ever get counted? The answer is we will only know if the results change. It is not clear how many of those affidavit voters had registered correctly but been screwed by the system.

he stole Ohio in the second election

We're already there.

That's not true, there were multiple exit polls from different news sites (CNN and CBS) that put the race at 52-48. Look at about 9 PM from 538's Live Blog to see the discussion about it: http://fivethirtyeight.com/live-blog/new-york-primary-presidential-election-2016/

From that feed:

Harry Enten: "The preliminary exit polls have Clinton’s lead at just 4 percentage points."

Nate Silver: "That would project a finish of Trump 57, Kasich 25 and Cruz 17 on the GOP side, and Clinton 53, Sanders 47 for Democrats. The latter race is close enough that it isn’t safe to assume Clinton will win."

and there are others discussing them as well. The discrepenacy is that the exit polls have since be retroactively reweighted to match the final vote counts, which is the whole criticism of Richard Charmin.

100% agree. no shortage of info on this subject.

But this implies that the major news sites are reputable.

Hijacking this hijacking, there is probably going to be another Arizona in Rhode Island- http://wpri.com/2016/04/20/ri-will-only-open-1-in-3-polling-places-for-primary/

No, the reporter is not a dumbass. Shortly after 9 pm, CNN said that its exit polls showed 52% for Clinton and 48% for Sanders. It's mentioned here, here, here, here, and in many other places online.

Flashbacks... I was one of those college kids in rural ohio and was in line 8-10 hours to vote that day.

No. It started way before Bush and he only stole the first election.. There has never been proof of him stealing the second one.

no one is willing to believe that election fraud happens in this country, which is why it is being done so extensively and will continue. Pretty soon, no one but the corporate endorsed candidates will ever have a chance to win an election in this country,. That's where we're heading

Let's say you a roll a six sided dice. If the average roll after 10 rolls is 5, that's not that strange (the average should be 3.5). If the average after 100 is 5, that's really unlikely. If after 1000 rolls rolls the average is still 5, that's exceedingly unlikely, unless there is something else going on ("the chance of this happening randomly" is so small that it can't be random).

Two things give "power" to stats: a significant difference from the expected value, and a large sample (dice rolls, people interviewed after voting, etc).

In police investigations and in exit polls, thousands of people don't generally conspire to lie, and when they do, it makes for an interesting story.

"How can they call it already with thousands of people standing in line to vote and 1% in?!"

"Exit polls man! Statistics don't lie!"

::blink::

Just like those 'crazy climate scientists' crying about global warming and then accepting grants to study global warming. Clearly just a scheme to make money.

/s

because the United States isn't a Democracy. we're an Oligarchy.

So I googled him, and right next to where he says this, and actually you can see the same if you follow the link, but he also blogs about JFK conspiracies. Not sure he is exactly trustworthy.

Early and absentee ballots in New York are very complicated unlike every other states. You need to fill in a form to request early/ absentee ballot, that will be sent back and only can be used specifically by you and you only. So Hillary people can't go and hand out en-mass like any other states.

It's much much much easier to go out to vote, hence the amount early/ absentee ballots should be fairly low

CNN. It was being cited and posted by other agencies all around the web as soon as the polls closed last night. They've since tried to burry it

Heard the exit poll was from buffalo which they considered a good case for the overall result, obviously not though

Don't confuse exit polls with forecasts. Election forecasts try to predict how many people of what demographics will show up for an election ("among likely voters") and try to figure out how that demographics vote. They get demographics wrong as well as likely voters.

Exit polls have none of those problems. There are really only two ways exit polls can go wrong: not randomly sampling people as they leave (only asking white women, for example; there are strong and easily methods for not doing this that interviewers are trained on, such as interviewing every 10th person leaving), or else if people are voting for a candidate that they don't publicly want to admit they're supporting (such as Trump, early on in the election cycle, before people were really admitting to supporting him but were voting for him).

We've been there since at least Reagan. Maybe Carter was the last real progressive US prez.

I haven't heard much talk of absentee ballots in these discussions. Could they account for any significant discrepancy in exit polling and final vote numbers?

Astroturfers started talking about how those who favor Sanders would complain about New York having similar issues to Arizona, days before there were any allegations.

Is there anything wrong with people writing books about an issue that they are studying on?

Where did you hear that? Why would Buffalo be a good measure of support in NYC? Makes no sense.

Reputable =/= trustworthy. Not in the context of the media in 2016. They have storied histories and well-established reputations, and for the most part those reputations were earned through legitimate, good journalism in the past. Hell, the whole reason CNN is the powerhouse that it is now is because its coverage of the Gulf War was so groundbreaking. But in 2016, they are little more than tools that use their reputations as weapons to make their spin and obfuscation seem objective. Folks like my parents have a hard time accepting that. They see The New York Times and by default assume it's trustworthy content. That's what makes them dangerous.

not true. cbs conducted and posted their exit poll than pulled them. a scream shot was captured by a poster on dem underground bernie firum. it had bernie winng comfortablly.

Many conspiracies are real. Many ridiculous conspiracies are purposely created to water down the real ones and pee in the pool, so to speak. Even within 911 conspiracies, there are many deliberately fake trails designed to divide and distract. Once you realize how its done, its really really easy to see how the psychology of the public is manipulated.

the exit poll was not from Buffallo! If you're going to post something like that, please provide a source

That sounds like a shitty conspiracy of its own.

My main grip is that you would expect Bernie to do better in areas with student housing. Many thousands of students in a concentrated area. But no.

http://www.wnyc.org/story/map-ny-primary-vote-nyc-2016/

watching TYT yesterday they said NY and RI are the most corrupt states for elections. so your link doesn't surprise me :(

Watching Hillary's smug ass face during her victory speech was sickening, especially since there are now at least two states that fucked ovr Den voters(AZ and NY)

Source?

two different exit polls showed Clinton winning by 6 points. The odds that they would both be so off are incredibly small. And it's EASY to change votes with voting machines. A teenager could do it. It doesn't need to be a mass conspiracy, it could be one Super Pac coordinating it with a handful of people.

Exit polling should include a survey of the population at large to figure out how those who used an absentee ballot voted. The total number of such ballots is usually publicly available from elections boards and so their distribution can be incorporated into in-person exit poll results from election day itself.

I really doubt that the exit poll takers didn't think to include "did you vote by affidavit" in their questioning when it was commonly known there would be a lot of that today.

There are lots of ways preference polls can go wrong (misidentifying potential voters and potential voter demographic being the major ones).

There are really only two ways exit polls can go wrong: failing to randomly select people leaving the poll (this basically never happens; picking eg every 10th person leaving is really basic stuff), and bias introduced by a candidate who people do not want to publicly admit they are supporting (this was happening with Trump early in his campaign in exit polls).

A large exit poll in a fair election will not be off by more than 1 or 2 percent, and each percent after that becomes exceedingly less likely as it moves out further onto the tails of the normal curve. Thousands and thousands of people don't conspire to lie about who they voted for.

Because of their accuracy, in many countries, exit poll results are used to determine the election and the casting of ballots is a precautionary measure used to detect for election fraud, or, if you're so inclined, as a measure of the corruption/honesty of the government.

We've already seen Bernie win recounts, and the NY comptroller raising flags over the issue is a really good sign that we'll see favorably revised numbers soon enough.

With enough momentum to make local delegates keep their promises of support, though? We'll have to wait and see.

Not only were thousands HUNDREDS OF THOUSANDS of voters unable to vote

ftfy

This guy sells books to people who want to hear about how elections are fraudulent and then runs around saying the election was fraudulent...

He runs a Wordpress blog about the jfk assassination...

The preference polls were bad because they couldn't target all demographics. exit polls can. His argument states that. Yet to say all the exit polls were fucked up? Nobody knows how to do them? That's hard to believe.