It's not clear to me what the author is getting at. 538 is good at aggregating polls and showing some probability. The author complains that's not enough becaus

It's not clear to me what the author is getting at. 538 is good at aggregating polls and showing some probability. The author complains that's not enough because a non zero probability of the less probable outcome gives 538 too much cover when the less probable outcome occurs. Isn't that a shared characteristic of any other model dealing with probabilities? Is the author trying to say Nate Silver gets too much spotlight considering he doesn't guarantee outcomes? I didn't sell in 2008 and kept buying. I have the same strategy today. Not sure if that also makes me a guru if that's the bar (as the author applies it to Taleb). Disclaimer: not a mathematician