3 ms·

An interesting observation about the HN thread that you linked to, is that the most substantive reply, that pointed out that Trump was sitting at a real, 25% ch

• Natsu 10y ago
  A good and fair point. But neither was the post saying "I have visions of Bill, passed out in a hammock on the White House lawn, surrounded by empties and cheeto packets." voted down, so that would tend to weigh against it being purely a matter of being a comment with no substance. I'd say that it appears that people believed it was the type of political falsehood we're talking about here. I mean, it did say 'more than likely' so anyone who believed the 98-99% estimates should have considered it laughably false based on the evidence they knew. The other comment made a more specific, factual claim that couldn't be laughed off, though I saw many stories aghast that 538 would break ranks. I can't find a source, but in part the consistency of the polls reminds me of the stories of how the measurements of some physical constant (g?) drifted over time because everyone felt beholden to the 'consensus' value, which was wrong. I wish I could find that, I think I saw it on HN a long time ago. Obviously, it's 'safest' to use the same assumptions everyone else did. When 538 broke from the herd, a lot of people questioned them on it. Based on the results, they were at least 30+% more correct than the 99%ers. And I think all of them were operating on assumptions that have been invalidated.