Hacker News Reader

Comment by nathan_phoenix | Hacker News Reader

They got 5/8 correct for the last year, so basically a bit better than a random guess. Seems like the future is still hard to predict...

Edit: As some people have pointed out, around half weren't binary choices (which I didn't notice) so 5/8 is actually good!

View on HN

These aren't binary choices, so 5/8 doesn't seem too bad to me.

2022's bets: https://www.economist.com/the-world-ahead/2021/11/10/the-exp... https://archive.ph/bam31

I read Tetlock's Expert Political Judgement many years ago, and though I can't guarantee my memory of it, I think one upshot of some pretty detailed empirical work was that no-one's any good at predicting political and economic futures. Foxes (in Isiah Berlin's sense, ie. who approach problems without an overarching conceptual framework) were marginally better than Hedgehogs (who have a central big idea), but no-one was up to much.

I think the trick, if there is one beyond luck, involves the ability to draw conspicuous attention to the occasion one is "right", while distracting from the all the other off-target pronouncements.

Quite. Also drawing heavily on the general population's limitless capacity for misunderstanding probability.

The incorrect 3 were related to the Omicron variant -- not bad for armchair* analysis!

* My take from reading Tetlock's book is that superforecasting is essentially painstaking analysis by laypersons based on common rationality followed through diligently. If among the only things this process fails to predict is mutations then this is actually very encouraging.

If you guess 5/8 dice rolls correctly, that's way better than a random guess

Not for 8 rolls.

Why not? We're talking dice, not coins. (Unless you're saying the superforecasters are doing coins and not dice, which makes sense.)

Wouldn’t the expected number of correct answers for random guesses of rolling dice a number of times be (1/sides)*rolls?

I don’t generally place much stock in forecasts but… I’m unaware of what the 8 questions were last year but this years includes a few non binary outcomes. If this was the case last year then their performance was a fair bit better than random.

"Republican will control the house, Democrats the Senate" is a really, really impressive prediction.

Reply on news.ycombinator.com