[happytoexplain]

It's childish and destructively hostile to paint a prediction of the outcome of an election giving high odds to the result that turns out wrong as some kind of catastrophic failure or humiliation. It boggles my mind how many apparently otherwise smart people commit this obvious fallacy out loud.

[mitchdoogle]

Its even more childish to do this before final results are even known. Let's wait until votes have been completely counted and then discuss the matter.

[dragonwriter]

> It's childish and destructively hostile

That's kind of Greenwald's schtick, particularly recently.

[leftyted]

I agree the language is a bit childish but it is quite remarkable how -- for two elections in a row now -- we've seen these models get so much so wrong. Almost everyone was predicting a landslide for Biden and it's dead heat right now.

It seems that there's something wrong with the polls themselves, how the models interpret the polls, or both.

There's lots of talk about "stealing the election" coming from both sides. Lopsided predictions from polls and models can increase the perception that the election was "stolen" if the vote defies the predictions.

[happytoexplain]

I agree that in the Trump case the models/polls are missing something, and we should try to figure out what it is, but I just don't think that's a bizarre thing for models/polls, and, more important, I think it's crazy how it's portrayed as some kind of ignorance of the elites in their political bubble, or some moral fault of arrogance.