Quote:
Originally Posted by elphenor
I'm pretty confused about the Nate Silver thing because can you simply never be wrong as long as you say there's a chance the other guy could win
|
Well maybe he's not "wrong" in a strict sense, but if the outcome is far off from your original prediction, then one would be inclined to ask the statistician (Nate Silver) what was wrong with his polling methodology/forecast models/etc. that resulted in the disparity between the prediction and what happened in reality. I'm guessing Nate Silver and Fivethirtyeight have released articles about things they didn't account for in the 2016 election, but I haven't read them. And even if I had, they could just as well give you a list of other things they didn't account for when Trump wins again this November.
It's a bit like a weatherman. Okay, you get the forecasts slightly wrong a couple times a week, it's understandable - but if you get the foreceasts completely wrong - and/or you do it repeatedly, folks are not going to trust you - and assume your team of meteorologists got their degree from a cereal box.