Hacker News new | ask | show | jobs
by FergusArgyll 33 days ago
It was 30% in the end, before the nomination it famously gave him a 2% chance of getting nominated. All the talk about 30% is disingenuous

https://en.wikipedia.org/wiki/FiveThirtyEight#2016_U.S._elec...

2 comments

They had him at 30% at a time when most reputable media had him in single digits at best. You might not remember now, but the result of that election was a real shock.
From the article:

> Being dismissive of Trump’s chances early on in the Republican primary was the biggest analytical mistake of my career, and I think I deserved a lot of blame for that.

He went on to write a whole retrospective, which is exactly what you'd want to see:

https://web.archive.org/web/20160527093831/https://fivethirt...

So unless your standard is "pollsters must be perfect at all times", I dunno what else should be asked of Silver here.

I'm fine with him[0]. I don't like the apologists who explain patronizingly that 30% odds happen 3/10 times. He gave Trump much much lower odds early on and only put it at 30 in the last few weeks.

[0] I'd say the better defense of him is: the whole model is built on polls, the pollsters stink, people should complain about pollsters, not their aggregators.