The Polling Industry Is in Crisis

[Ivytalk]

Jill Stein is horrified by theses results...

I’ve read some pretty bad theses myself.

[kalm]

Jill Stein is horrified by theses results...

No armpit hair? What a fraud...

[Chizzang]

I still want the list of republicans that voted for Trump.

You're pretty slow to notice that nobody gives a sh!t about what you want

[CID1990]

I still want the list of republicans that voted for Trump.

Joe McCarthy claimed he had a list of Communists in the State Department.

It would have been easier to have a list of non-communists at State


Sent from my iPhone using Tapatalk

[ALPHAGRIZ1]

I still want the list of republicans that voted for Trump.

You're pretty slow to notice that nobody gives a sh!t about what you want

I dont care list them ass hat.

Sent from my SM-G975F using Tapatalk

[Chizzang]

You're pretty slow to notice that nobody gives a sh!t about what you want

I dont care list them ass hat.

I want to hear you beg...

[JohnStOnge]

Ok. It ain't so. The Michigan example from the 2016 Democrat primary is an example where the polls were indeed way off. But such examples are rare. The polls are correct WAY more often than they are not.

The polls suggested that Hillary Clinton would win the popular vote. Hillary Clinton won the popular vote.

The polls suggested that the Democrats would take the House in 2018. The Democrats took the House in 2018.

In the 2016 Presidential race, if you look at the State by State averages complied by RCP, the candidate that led in the RCP average of polls won in 42 of the 46 jurisdictions for which averages were reported.

It's absolutely ridiculous to say that the polls don't tell you anything. It's the "in" thing to to. But it's absurd. If you predict the winners of political races based on the polls you are going to be right WAY more often than you're going to be wrong. Like 90% of the time at least. And if you take random sampling error into account before you make a call you're going to be right a lot more often than that.

Triggered!

Also correct. If a person doesn't understand that the polls provide the best available insight into what's going on that person is an idiot.

[CID1990]

Triggered!

Also correct. If a person doesn't understand that the polls provide the best available insight into what's going on that person is an idiot.

I think you need to expand on that a bit -

the value of polls that turn out to be wrong


Sent from my iPhone using Tapatalk

[JohnStOnge]

Also correct. If a person doesn't understand that the polls provide the best available insight into what's going on that person is an idiot.

I think you need to expand on that a bit -

the value of polls that turn out to be wrong


Sent from my iPhone using Tapatalk

If you understand survey sampling you understand that it is not going to be correct 100% of the time. If you look at political polling by credible pollsters the candidate that's ahead on election even the candidate that is ahead by the point estimate is going to win more than 90% of the time. And that's being conservative. It's probably more than 95% of the time.

Also that is not considering margins of error.

[CID1990]

I think you need to expand on that a bit -

the value of polls that turn out to be wrong


Sent from my iPhone using Tapatalk

If you understand survey sampling you understand that it is not going to be correct 100% of the time. If you look at political polling by credible pollsters the candidate that's ahead on election even the candidate that is ahead by the point estimate is going to win more than 90% of the time. And that's being conservative. It's probably more than 95% of the time.

Also that is not considering margins of error.

I still don't see how that explains anything.