A Warning: Anonymous

[Ingomike]

Yeah, but in the post-2016 political world, polling is a bit of a illusion - or at least until it proves itself again.

Awkwardly, I always hated the use of proprietary modifiers to shift raw polling numbers one way or the other. It just smacked of confirmation bias. However, with Trump, I think any raw polling numbers need to be shifted in his favor. Except for Rasmussen, which probably goes just a bit too far in his favor.

If NBC has him -10 for job performance, and Rasmussen has him at -2, then the latter is probably more accurate when it comes to voting.

Especially with the current crop of Dems.

And in the case of polls saying we got 2016 right or close, look at their polls for the month before the election. The media use polls to show their candidate winning to create momentum then in the last few day before the election they slide them a little closer to claim accuracy while they are manipulating them again in the lead up to the next election. Viscous circle...

 

[jamil]

Yeah, but in the post-2016 political world, polling is a bit of a illusion - or at least until it proves itself again.

Awkwardly, I always hated the use of proprietary modifiers to shift raw polling numbers one way or the other. It just smacked of confirmation bias. However, with Trump, I think any raw polling numbers need to be shifted in his favor. Except for Rasmussen, which probably goes just a bit too far in his favor.

If NBC has him -10 for job performance, and Rasmussen has him at -2, then the latter is probably more accurate when it comes to voting.

Especially with the current crop of Dems.

Depends who actually votes. That was the problem with polls leading up to the 2016 election. In the states that mattered they thought their standard models were good to go. The failure was not knowing how pissed working class people were at Democrats. I think the Trump rhetoric might cool their enthusiasm to vote depending on what the atmosphere is next November. But put a bunch of stories out about 7 year old kids being conned into gender assignment surgery and I think that will trump the stupid tweets.

 

[jamil]

And in the case of polls saying we got 2016 right or close, look at their polls for the month before the election. The media use polls to show their candidate winning to create momentum then in the last few day before the election they slide them a little closer to claim accuracy while they are manipulating them again in the lead up to the next election. Viscous circle...

Maybe. The outcomes make you wonder. But the outcome alone doesn’t prove the means. Polls may change because people get more serious about who they’re voting for. In other words there are competing reasons why the outcome was what it was. You catch a polling org with their finger on the scale, you have something.

 

[T.Lex]

Depends who actually votes. That was the problem with polls leading up to the 2016 election. In the states that mattered they thought their standard models were good to go. The failure was not knowing how pissed working class people were at Democrats. I think the Trump rhetoric might cool their enthusiasm to vote depending on what the atmosphere is next November.

I can go along with that. But that also exposes the problem. Polls - historically - were pretty accurate about revealing people's mood about something. People would respond to the pollster that they are pissed off at the Dems, and it would get counted that way.

There's something about Trump that causes people to mask that kind of thing, at some level.

To put it another way, I can see the models being wrong about people favoring Trump. That is, the Dem candidates are so bad, it might be more palatable for people to say they'll favor Trump, even if they vote differently.

Basically, other than broad brushstrokes, I think the 2020 polls will be about as (in)accurate as the 2016 polls, but there's no way to predict which direction they'll be less accurate.

 

[Somemedic]

Politicians looking like idiots?

Are these the same guys who have closed the government, the buildings in Washington DC, the VA, then can't get a budget together? And they're worried about looking like idiots?

I realize they're senior officials and not Congress but they're from the same ilk. I think if the media and their liberal agenda can't overcome Trump bye name-calling then they will stop at nothing to discredit him by any means necessary

 

[nonobaddog]

The general public did not know about JFK's escapades. I imagine if they did, he would have gotten much worse than Clinton.

Are you making this stuff up? There were lots of rumors. Kennedy was so popular that I bet every person in America at the time had heard them.



ETA - now for something completely different - JFK was a lifetime member of the NRA.

 

[Ingomike]

Maybe. The outcomes make you wonder. But the outcome alone doesn’t prove the means. Polls may change because people get more serious about who they’re voting for. In other words there are competing reasons why the outcome was what it was. You catch a polling org with their finger on the scale, you have something.

It it not the outcome that is the basis of the accusation, but rather the internals of the poll are often different on Election Day or very near it than in the lead up polls. Similarly, many of the polls indicating support for impeachment are so weighted to what the big media wants it would be impossible not to find it...

 

[josh64]

Yet another "anonymous" (read, cowardly and probably completely fictional) source with "warnings" about Trump. How is this not considered libel? No proof, no evidence, just spewing out negative nonsense. Of course Rachel Madcow and others of her ilk laps it up.

https://www.nbcnews.com/politics/do...-s-decision-making-eroding-over-time-n1078586

We have seen that pretty much everything they accuse President Donal J Trump of, they themselves are factually guilty of.

For examples;

Killary/dnc/0bama/brennan/clapper/comey/muelie conspired with UK int officer Steele, a foreigner, as well as task Ukraine with digging up fake dirt on Trump, as well as tasking an Australian int agent and others to conspire with them against the Trump candidacy.

Also, President barak soetoro sobarka 0bama is on video telling the Rucssian Vice President Medved "I'll have more flexibility after my reelection" to wich Medved replies "I will relay this information to Vladimir", this comprises Russian collusion.

The 0bama regime gave 1.5 billion US Dollars to the Iranian regime in cash, flew the money right to Tehran.
Something about aiding and abetting the enemy and giving him comfort applies here.

Killary is guilty (several thousands of instances) of the espionage act for passing official state dept secret and above secret comms via her private server so as to bypass federal accounting regulations regarding secret comms. The fbi/doj declined to prosecute, each instance of wich would get killary and inhuma and the rest 5 years in club fed. If you or I had done, this, or heaven forbid President Donal J Trump, we'd be in prison for the rest of our lives.

On quid pro quo, Vice President Joe Bidet told the Ukrainian gov that if they didn't fire their prosecutor (who was investigating Joe Bidet' son's company for corruption) that they'd lose out on a billion US taxpayer dollars, and they had 6 hours to do so. If they didn't believe him they were to ask President 0bama. Vice President joe bidet is on video admitting to the above, yet President Donald J Trump is under impeachment proceedings for hearsay. Also, tomorrows star witness in the impeachment proceedings/stalinist show trial just alluded to gave Ukraine a list of people and organisations they were not to investigate for corruption or they would lose US taxdollars, quid pro quo.

 

[Kutnupe14]

We have seen that pretty much everything they accuse President Donal J Trump of, they themselves are factually guilty of.

For examples;

Killary/dnc/0bama/brennan/clapper/comey/muelie conspired with UK int officer Steele, a foreigner, as well as task Ukraine with digging up fake dirt on Trump, as well as tasking an Australian int agent and others to conspire with them against the Trump candidacy.

Also, President barak soetoro sobarka 0bama is on video telling the Rucssian Vice President Medved "I'll have more flexibility after my reelection" to wich Medved replies "I will relay this information to Vladimir", this comprises Russian collusion.

The 0bama regime gave 1.5 billion US Dollars to the Iranian regime in cash, flew the money right to Tehran.
Something about aiding and abetting the enemy and giving him comfort applies here.

Killary is guilty (several thousands of instances) of the espionage act for passing official state dept secret and above secret comms via her private server so as to bypass federal accounting regulations regarding secret comms. The fbi/doj declined to prosecute, each instance of wich would get killary and inhuma and the rest 5 years in club fed. If you or I had done, this, or heaven forbid President Donal J Trump, we'd be in prison for the rest of our lives.

On quid pro quo, Vice President Joe Bidet told the Ukrainian gov that if they didn't fire their prosecutor (who was investigating Joe Bidet' son's company for corruption) that they'd lose out on a billion US taxpayer dollars, and they had 6 hours to do so. If they didn't believe him they were to ask President 0bama. Vice President joe bidet is on video admitting to the above, yet President Donald J Trump is under impeachment proceedings for hearsay. Also, tomorrows star witness in the impeachment proceedings/stalinist show trial just alluded to gave Ukraine a list of people and organisations they were not to investigate for corruption or they would lose US taxdollars, quid pro quo.

You're clear that Obama was president when he said that right? How saying "I'll have more flexibility after my reelection" counts as collusion, for a sitting president, is beyond me. I'll ignore the rest of the issues I have with this post, as the most glaring issue, I've addressed.

 

[DoggyDaddy]

You're clear that Obama was president when he said that right? How saying "I'll have more flexibility after my reelection" counts as collusion, for a sitting president, is beyond me. I'll ignore the rest of the issues I have with this post, as the most glaring issue, I've addressed.

Seriously? How would it not? He was up for re-election, not running uncontested. Talk about quid pro quo... In other words, "make sure I get re-elected and I'll help you out."

 

[amboy49]

Well, it isn't really that simple, right? When people come out publicly, there are death threats, public harassment, and all sorts of downside. It is completely reasonable to try and avoid that.

But, in this day and age, I think it pretty unreasonable to think anonymity can be achieved long-term. It'll come out at some point.


Have you read the book already?

"All the President's Men" wasn't exactly sworn testimony.

I can see how some folks would be reluctant to publicly provide accusatory information. Why, just look at many of the Clinton family associates who have mysteriously died or, due to to extreme pressure, committed “suicide.”

 

[jamil]

It it not the outcome that is the basis of the accusation, but rather the internals of the poll are often different on Election Day or very near it than in the lead up polls. Similarly, many of the polls indicating support for impeachment are so weighted to what the big media wants it would be impossible not to find it...

When you say internals of the poll, what do you mean? Are you saying you have evidence that the polling models change very near to election day? If that's the case, then there are few reasons for that to change. And about the impeachment polls, what evidence do you have that suggests that the weighting is artificial. I think that's what you're saying, but correct me if I'm wrong.

And that's what I mean by outcome. I'm not talking about election outcomes or whatever. I'm saying it more generically, kinda like a function. You have inputs, function, and output. You pump the inputs in one end, you process the function, and you get an output. In polls, your inputs are the results from all the people questioned in the poll. The function is "means part" (tallying, turnout models, weighting, calculations, etcetera). The outcome is the percent this way or that way. So what I was saying is that the outcome isn't evidence of what the function is. We don't typically have enough information to figure that out. So if you're asserting that because the poll numbers were what they were leading up to the election, and then changing right before the election, that's not proof that the function changed.

The same goes for the polls about impeachment. The poll numbers isn't proof that the model is skewed one way or the other. Of course the pollsters may have their thumbs on the scale. There's probably some advantageous psychological effect of having polls to make people think something is more popular or less popular than it is. And the media has demonstrated a willingness to lie about other things, I don't see why polling should be any different. But that's just me being partisan. I don't really have any evidence that this is the case other than the output. And that's not evidence by itself.

There are reasons other than manipulation why polls might change just before the election, which I listed.

 

[jamil]

You're clear that Obama was president when he said that right? How saying "I'll have more flexibility after my reelection" counts as collusion, for a sitting president, is beyond me. I'll ignore the rest of the issues I have with this post, as the most glaring issue, I've addressed.

I suspect that if Trump said it, you'd be saying something else.

 

[jamil]

I can go along with that. But that also exposes the problem. Polls - historically - were pretty accurate about revealing people's mood about something. People would respond to the pollster that they are pissed off at the Dems, and it would get counted that way.

There's something about Trump that causes people to mask that kind of thing, at some level.

To put it another way, I can see the models being wrong about people favoring Trump. That is, the Dem candidates are so bad, it might be more palatable for people to say they'll favor Trump, even if they vote differently.

Basically, other than broad brushstrokes, I think the 2020 polls will be about as (in)accurate as the 2016 polls, but there's no way to predict which direction they'll be less accurate.

My theory: I think during turbulent times the polls probably aren't accurate. In these polls they sample a few hundred to a few thousand and then they have to figure out how that represents a much larger population. That's done in their turnout models and demographic models and so on. In turbulent times I think the models can be off by a lot more than they think. The people they'd normally count on to vote one way, because that's the way they always tend to vote, end up voting differently than they expected because they're less predictable. Same with the turnout. A lot more working class people who just show up for a paycheck and don't give a **** about politics, gave a **** about politics in 2016 and they voted for Trump.

 

[DoggyDaddy]

I suspect that if Trump said it, you'd be saying something else.

Damn skippy he would.

 

[DoggyDaddy]

My theory: I think during turbulent times the polls probably aren't accurate. In these polls they sample a few hundred to a few thousand and then they have to figure out how that represents a much larger population. That's done in their turnout models and demographic models and so on. In turbulent times I think the models can be off by a lot more than they think. The people they'd normally count on to vote one way, because that's the way they always tend to vote, end up voting differently than they expected because they're less predictable. Same with the turnout. A lot more working class people who just show up for a paycheck and don't give a **** about politics, gave a **** about politics in 2016 and they voted for Trump.

It doesn't help when their "sample" is taken from oh, I dunno... Cal Berkley students. "We polled 1000 rabidly liberal college students, and 100% said they would NOT vote for Trump!"

 

[nonobaddog]

It doesn't help when their "sample" is taken from oh, I dunno... Cal Berkley students. "We polled 1000 rabidly liberal college students, and 100% said they would NOT vote for Trump!"

I'm not sure they even need a 'sample'. They can simulate that. They already know the outcome they want so it is easy to tweak the simulated sample and the function and voila - there is the outcome.

 

[jamil]

It doesn't help when their "sample" is taken from oh, I dunno... Cal Berkley students. "We polled 1000 rabidly liberal college students, and 100% said they would NOT vote for Trump!"

Do you have evidence that this happened with a poll where they misrepresented the demographics of the poll?

 

[Blackhawk2001]

Heck, I never knew that private intelligence dossiers could make one subject to libel. Guess ya learn something new everyday.

How about when they're used to get a FISA warrant to spy on your election campaign?

 

[jamil]

How about when they're used to get a FISA warrant to spy on your election campaign?

That would only be bad if Trump did it.