Media, Polling, and Pundits

[UNLEASH IT]

In the days leading up to the elections, most media sites and polls were united on the idea of a win for Hillary Clinton. Even Fox News thought that Hillary would win. But then the election results came out and showed all of them that they were terribly wrong. I'm still confused on what caused this large discrepancy between predicted and actual results and more importantly, I wonder if the media can regain the trust of people who subscribe to them. What are your opinions on the matter?

Edit: Now that I see what people said below, I should probably change the title to "Media, Polling and Pundits". Or maybe just "Media and Pundits". Does anyone know how to do that?

Edited November 13, 2016 by UNLEASH IT

[Res]

The interesting thing about political polls is that they can cause people to change their votes at the last moment; a wavering democrat might've chosen to vote third party or not at all if they felt a democratic win was pretty safe. This definitely happened in Brexit, and the U.S. was warned of this by people studying Brexit; of course, they didn't listen.

As much as I love to read online news, I have to admit the internet is bad for the integrity of the media. It's too easy to find sources pertaining to your bias, and there's little indication of a site's veracity - and people are fooled by satirical/fake sites all the time.

[Dayni]

Well, one thing that was not reported at least explicitly were the margins of error for polls.

Which quite consistently were larger than the gap between the two. Not something to use for a celebratory whoop.

The media's reporting was a bigger problem to me, and as a result policy was pretty much not discussed beyond some of the more ridiculous ideas (see: WE'RE GONNA BUILD A WALL), and was used in their attempts to delegitemise a candidate for election for the sake of it.

[tuvarkz]

A big issue was that the polls appeared to be pick-and-choosing to make Clinton appear better, given how there was this whole attempt at demoralizing Trump voters. EG http://i.imgur.com/JAwXfxA.png

Also, to note, the Wall isn't an entirely ridiculous idea. Uneffective? Possibly. Budget? 10 to 25 Billion, going from the Trump estimate to the CNN estimate. Impossible to build? far from it, particularly because of the misinformation that the wall would go along the entire border-it's only meant to cover the parts where there aren't natural borders.

[UNLEASH IT]

Well, one thing that was not reported at least explicitly were the margins of error for polls.

Which quite consistently were larger than the gap between the two. Not something to use for a celebratory whoop.

The media's reporting was a bigger problem to me, and as a result policy was pretty much not discussed beyond some of the more ridiculous ideas (see: WE'RE GONNA BUILD A WALL), and was used in their attempts to delegitemise a candidate for election for the sake of it.

I feel like this was done out of selfish greed on the media's part because they knew that focusing more on Trump's outrageous claims would net them higher ratings than if they focused on the flaws in his economic policies, which could've actually turned away support from him. Also, the fact that their reaction to Trump winning is one of continued disbelief when they themselves had a part to play in his victory is something I'm kinda ticked off by. I also agree with Tuvarkz's point to an extent.

Edited November 12, 2016 by UNLEASH IT

[Rezzy]

I remember after the debates, many Internet polls gave Trump a victory, but the pundits dismissed the Internet polls and declared Hillary the winner. Maybe they won't be so dismissive next time.

[Sasori]

I'm sure there was bias on the part of the media, but there could have been other reasons as well why everyone was so wrong.

One of those reasons that I hear a lot is that people often gave desireable answers, instead of truthfull answers. People did't pick what they voted, but picked what they thought people wanted to hear.

Another one is the Trump voter base. A lot of those are seen as people who don't care or don't trust politics and never vote as a result. Those people did show up to vote for Trump and the media might have underestimated the size of that group since you can't exactly pick them out from a crowd.

The elections themselves might have made the polls less trustworthy as well. Polls have been made in about every election so prior experience probably gets you very far in a conventional election. This election was anything, but conventional so it was probably a harder to make accurate polls as a result.

[Nobody]

people are overstating the national polling error IMO

Hillary had on the average of polls +3% over Trump. She will probably end up beating him on popular vote by +1% when all votes are counted. A 2% error isn't that bad when it comes to polling

The polls in Michigan, Wisconsin, Pennsylvania and maybe North Carolina were messed up, but other than that the errors weren't particularly high

Edited November 12, 2016 by Nooooooooooooooooooooobody

[UNLEASH IT]

people are overstating the national polling error IMO

Hillary had on the average of polls +3% over Trump. She will probably end up beating him on popular vote by +1%. A 2% error isn't that bad when it comes to polling

The polls in Michigan, Wisconsin, Pennsylvania and maybe North Carolina were messed up, but other than that the errors weren't particularly high

But these were states that ultimately decided the election, mainly the first three.

[Nobody]

But these were states that ultimately decided the election, mainly the first three.

Well, yeah I know, that's why his victory was unexpected. But by the way people are talking, it's as if the national polling was completely wrong. Wisconsin, Michigan and Pennsylvania were probably harder to poll because of how solidly democratic those states used to be. When a state always has expected results, pollsters don't focus on them.

Edited November 12, 2016 by Nooooooooooooooooooooobody

[UNLEASH IT]

Well, yeah I know, that's why his victory was unexpected. But by the way people are talking, it's as if the national polling was completely wrong. Wisconsin, Michigan and Pennsylvania were probably harder to poll because of how solidly democratic those states used to be. When a state always has expected results, pollsters don't focus on them.

Well, isn't that a fault of the pollsters themselves, to assume something without actually checking it?

Edited November 12, 2016 by UNLEASH IT

[Nobody]

Well, isn't that a fault of the pollsters themselves, to assume something without actually checking it?

yes

[Rezzy]

The polls weren't that far off. People were just overconfident in the very slim lead Clinton had, which was within the margin of error in most polls. They shouldn't have been giving Hillary a 99% chance to win.

[Dark Holy Elf]

Came into post exactly what Nobody and Rezzy did. The polls didn't actually miss by much; the problem was pundits failing to understand what the polls meant, and a lack of understanding of polling errors. Polls are not perfect; they must make assumptions about the composition of the electorate and some of those are going to be slightly off. You should always expect some slight error. In this case, the slight error was in Trump's favour, and that was enough for him to win.

[Tryhard]

Yes, they were in the margin of error, but Hillary did underperform compared to Obama (where Trump performed about the same as McCain and Romney in terms of total votes).

It came as a surprise because not many were saying how Trump winning could have much of a chance even though the polls show before their election it was a real possibility.

Edited November 12, 2016 by Tryhard

[UNLEASH IT]

Came into post exactly what Nobody and Rezzy did. The polls didn't actually miss by much; the problem was pundits failing to understand what the polls meant, and a lack of understanding of polling errors. Polls are not perfect; they must make assumptions about the composition of the electorate and some of those are going to be slightly off. You should always expect some slight error. In this case, the slight error was in Trump's favour, and that was enough for him to win.

This, along with the above posts, has cleared up a lot for me, thank you. But what made pundits so stubborn to consider a Trump presidency that they only looked for data that supported their view?

[Rezzy]

Yes, they were in the margin of error, but Hillary did underperform compared to Obama (where Trump performed about the same as McCain and Romney in terms of total votes).

It came as a surprise because not many were saying how Trump winning could have much of a chance even though the polls show before their election it was a real possibility.

Nate Silver was saying he had about a 30% chance of winning and the rest of the media gave him serious crap for saying he had even that chance, even though Hillary was still the favorite to win.

This, along with the above posts, has cleared up a lot for me, thank you. But what made pundits so stubborn to consider a Trump presidency that they only looked for data that supported their view?

This pretty much.

[Lord Raven]

I remember after the debates, many Internet polls gave Trump a victory, but the pundits dismissed the Internet polls and declared Hillary the winner. Maybe they won't be so dismissive next time.

one of those internet polls stated that "this isn't a metric that determines who won but in the end it determines who has the most rabid fanbase on the internet." There's a lot of 4chan threads that state ways to vote on the polls multiple times. The polls were basically brigaded. Edited November 12, 2016 by Lord Raven

[autismoe]

Polls oversampled Democrats across the board to create a masterful illusion of Clinton leading. This was to demoralize Trump voters and encourage Hillary voters. The only polls that showed a chance of Trump winning was the LA Times poll and the USC poll out of all the other polls. If you compared these 2 polls compared to all the other polls, you could easily dismiss them as outliers with faulty methodology as most media sources did. It's all confirmation bias from the media seeing Clinton with a 99% chance of winning on almost all the polls.

Just exposes how rigged the system is.

[Rezzy]

one of those internet polls stated that "this isn't a metric that determines who won but in the end it determines who has the most rabid fanbase on the internet." There's a lot of 4chan threads that state ways to vote on the polls multiple times. The polls were basically brigaded.

Yeah, it's hard to tell if it was merely troll voting, but at the very least, it shows voter enthusiasm, which is something Clinton sorely lacked.

It seems that people really hate "establishment candidates" in recent history in the USA.

1980: Outsider Reagan sweeps Carter

1984: Establishment candidate against incumbent

1988: Not really an "outsider" candidate

1992: Outsider Perot has the most successful 3rd party run in recent history. Clinton was a young up-and-coming governor

1996: Establishment candidate against incumbent

2000: Like 1988, no real outsider

2004: Establishment candidate against incumbent

2008: Obama wins the primary over establishment candidate Hillary and easily beat the Republican establishment candidate McCain

2012: Establishment candidate against incumbant

[UNLEASH IT]

Nate Silver was saying he had about a 30% chance of winning and the rest of the media gave him serious crap for saying he had even that chance, even though Hillary was still the favorite to win.

untilThis pretty much.

That video represents something that has been in the back of my mind for some time now but I didn't really want to talk about it until after the election, because I figured that the only people who'd actually listen to me would be on the opposite side and not the people who actually needed to hear it. I felt that, over the past few years, liberal and progressive ideology had grown to the point where it considered itself morally infallible and anyone who disagreed with their views were irrevocably wrong and evil. In a way, I thought that it became like the same extremist parts of religion we constantly derided for being oppressive, with the only difference being that our ideas were endorsed by the media. But I decided to drown those thoughts out because I believed that it was for the greater good. Hopefully this election can open people's eyes, but judging at how the media has been reacting to this and continuously proclaiming racism and sexism as the cause, it does seem unlikely.

Edited November 12, 2016 by UNLEASH IT

[Dark Holy Elf]

This, along with the above posts, has cleared up a lot for me, thank you. But what made pundits so stubborn to consider a Trump presidency that they only looked for data that supported their view?

There are a few possible explanations, and I imagine all were factors.

1. Some (not all) pundits strongly preferred the idea of Clinton winning and thus interpreted numbers in a way that made them feel better. I certainly watched some of my own friends do this. Of course the flipside of this is that some pundits strongly preferred a Trump win and some actually overestimated his chances (predicting a Trump landslide). We saw this from both Romney and Obama supporters in 2012, too, so it's not some new thing.

2. For all that they're all aware of the electoral college, only a few analysts really considered who had an advantage in the electoral college given a popular vote tie (which was certainly Donald Trump). If Trump had needed to win the popular vote to win this election, his chances would have been much worse, but he didn't have to.

3. The vast majority of people struggle with probability. They see a "90% chance" (about what it was at the height of Clinton's post-debate swing) and they assume it's a sure thing, when in fact it isn't. We've all missed on an 80+ hit in Fire Emblem.

[UNLEASH IT]

There are a few possible explanations, and I imagine all were factors.

1. Some (not all) pundits strongly preferred the idea of Clinton winning and thus interpreted numbers in a way that made them feel better. I certainly watched some of my own friends do this. Of course the flipside of this is that some pundits strongly preferred a Trump win and some actually overestimated his chances (predicting a Trump landslide). We saw this from both Romney and Obama supporters in 2012, too, so it's not some new thing.

2. For all that they're all aware of the electoral college, only a few analysts really considered who had an advantage in the electoral college given a popular vote tie (which was certainly Donald Trump). If Trump had needed to win the popular vote to win this election, his chances would have been much worse, but he didn't have to.

3. The vast majority of people struggle with probability. They see a "90% chance" (about what it was at the height of Clinton's post-debate swing) and they assume it's a sure thing, when in fact it isn't. We've all missed on an 80+ hit in Fire Emblem.

And we've also lost a unit to a 2% crit. That makes sense

Edited November 12, 2016 by UNLEASH IT

[Rezzy]

There are a few possible explanations, and I imagine all were factors.

1. Some (not all) pundits strongly preferred the idea of Clinton winning and thus interpreted numbers in a way that made them feel better. I certainly watched some of my own friends do this. Of course the flipside of this is that some pundits strongly preferred a Trump win and some actually overestimated his chances (predicting a Trump landslide). We saw this from both Romney and Obama supporters in 2012, too, so it's not some new thing.

2. For all that they're all aware of the electoral college, only a few analysts really considered who had an advantage in the electoral college given a popular vote tie (which was certainly Donald Trump). If Trump had needed to win the popular vote to win this election, his chances would have been much worse, but he didn't have to.

3. The vast majority of people struggle with probability. They see a "90% chance" (about what it was at the height of Clinton's post-debate swing) and they assume it's a sure thing, when in fact it isn't. We've all missed on an 80+ hit in Fire Emblem.

I fully expect to see a bill introduced to have all future elections use True Hit instead of single RNG.

[Tolvir]

I know part of their issue, and they even admitted this, was that they polled a lot more dems than republican. That alone can skew polls to appear the way they want.

Honestly, I never put any faith in polls. People can change their views at the drop of a dime, going off of why the media thinks out of maybe a few thousand doesn't represent the millions in America. Even if the polled all 13 million people in New York, it would accurately represent the views of the country.

I honestly hope the media is having a wake up call right now. Their job is supposed to report everything from a non biased pov. Report the facts, and let the reader decide. For the past 10 or more years though, they have been in the pockets of the highest bidder, and in the case of this recent election, Hillary. They spread a lot of lies and false narratives about Trump that has made him seem so terrible, while leaving out a lot of details. The media is entirely responsible for the fear and panick of a Trump presidency, and these protests have started as a result. People are liking to point at the dems for that, but in reality it is entirely the media ( though in the case of Bernie, that was entirely the Dems) If they had just reported the facts, used unbiased questions for the debate, and didn't try to favor Clinton because she was paying the most, then maybe things would of been different.

They even demonized the voters, the same people they should try and convince, but instead painted them as morons, bigots, or a mix of the two. Maybe Trump still would of won, but with a much more clear picture of what his vision is like, not the fear mongering fictional story the Media threw out there. Or perhaps we would of had an entirely different line up. Or maybe Hillary would of won, who knows. Either way, the media is responsible for this. Both a Trump presidency and people truly scared for their life where there truly shouldn't be any. The media needs to snap out of its current stance, because they are supposed to be giving the facts and letting the people decide, not hide the facts and decide for the people.

Edited November 12, 2016 by Tolvir