Hacker News Reader

Comment by peterkshultz | Hacker News Reader

I had been following the likelihood of a Brexit on Bloomberg. Before polls opened today, they said there was a 25% chance of the UK leaving. Seems they were quite, quite wrong.

http://www.bloomberg.com/graphics/2016-brexit-watch/

View on HN

I don't think it's terribly surprising. When people are demonized for their political beliefs (i.e., you're stupid and/or racist if you support Brexit), the truth will only come out with anonymity. Hence, polling becomes much more unreliable.

This bodes well for Trump ...

The polls were way closer for Brexit than Trump

http://www.economist.com/blogs/graphicdetail/2016/06/britain...

http://elections.huffingtonpost.com/pollster/2016-general-el...

And the stigma of being trump supporter is way higher.

Why? 25% is a very significant chunk.

You'd expect the 25% option to happen more often than just now and then.

25% of the time, actually.

25% chance of leaving is not the same as "they are definitely going to stay". I don't see how you can call a percentage based prediction "quite wrong" if it doesn't end up happening.

Agreed. They'd be wrong if they made a bunch of 25% predictions and the 25% predictions happened more than 1/4 times.

ah, thank you, that's what I was trying to say but couldn't quite figure out how to say

Fair enough. Perhaps my surprise comes from the fact that the prediction markets got it wrong as well.

You're right nevertheless--25% is a meaningful probability.

Yes, interesting that the prediction markets got this quite wrong as well.

Reply on news.ycombinator.com