Message from @pratel
Discord ID: 483912856354553876
9_6
Which actually crosses the line into the "never trust a unanimous consensus" territory.
HOW are they more careful.
Well for one, no one is putting odds above 80% without lots of evidence to back it up.
Lots of?
And most are closer to something like 538 in 2016.
Around 60%.
How much is lots of.
Which was widely criticized at the time.
Months of consistent polls.
And are you just taking people's word on it?
Lots of independent models.
OK. What makes you think it's all wrong.
Have you, as a person, ever been polled for something like that?
I've never been polled.
No. But I've done poll work so I know how easily you can manipulate things if that's where you're going.
How many people have you met in your life that have been polled?
But polling in the US is reasonably good.
How are we so certain these pollsters don't just hang around shopping malls?
Well for one, I get the phone calls.
The bias in phone responses tends to over-emphasize the elderly.
For that reason.
Not an argument.
I'd like to print this and frame it.
Right
The argument is: 2016
IF you want.
But simply because it happened in 2016 does not imply it happens in 2018.
Your faith is amusing.
So saying "Trump won 2016 and therefore Republicans win 2018" is a little fallacious.
You think everyone with a job is doing it well.
No. I think there's a number of people who are.
To the best of their ability
im with scribble on this one, the polls are unreliable
Pollsters tend to try and be objective.
PRFF
Watch the Penn and Teller episode on NUMBERS
Both parties rely on it for strategy and contract it all out. And you don't deceive yourself if you're talking strategy.
I believe it's the numbers episode anyway