|
In response to
"You're talking about a politician, which I specifically am not doing. I'm discussing methodology. "
by
David
|
"A lot of people are onto the con now" sounds like a comment on people's "true" opinions of Trump that doesn't show up in the polls.
Posted by
Inigo (aka Inigo)
Jun 1 '24, 08:38
|
Trump's con of America is what Colin Cowherd was referring to in the article you responded to. That's not what you meant?
The reason I'm responding to you in this way is because I've seen a number of your posts dismissing STers who bring up the likelihood of Trump winning/any polls that suggest the same. People are rightfully worried that Trump will win.
Yes, polling methodology has issues. Especially at the individual level. But it varies from pollster to pollster. A lot have changed their methods. A lot have an online component now (and those components have greatly improved).
The other price is that it's not just one poll. 538 and RCP do aggregates of all polling. Which is a lot. They try to factor in poll reliability and political bias.
Sure, even that aggregate has been off. But it's much more reliable than just one company like Nielsen using 1960s methods. The entertainment equivalent would be like if there 200 other companies like Nielsen out there, and other sites took all that data and averaged it. That's what we have in political polling now.
|
Responses:
|