Hacker News new | past | comments | ask | show | jobs | submit login

I don't think that prediction markets are useful in situations like this.

As it's not about a situation where a large enough group of people can influence the outcome, there's nothing that the pools can make to deduce a more accurate prediction.

Instead you have basically gambling with large volatility every time somebody post a positive or negative news item.




Your statement that placing bets does not affect the outcome is true but I don't think you draw the right conclusion.

If markets don't capture current sentiment accurately then bet against every news item.

Let's say there's a market for a coin flip, and periodically, news comes out that makes the market lopsided (one way or the other). You should trade against it. Every time. If you are correct, this strategy will win on average, until the true breakthrough news item, which you'll be on the wrong side of. Just don't keep doubling down, because the final trade will be the one you lose.

This will demonstrate who has "deduced a more accurate prediction". Everyone else put together, as events unfold, or the opposite of that.

*changed to will win on average.


I interpreted the parent comment to mean the bets don't have an influence in supply/demand behavior as they could in other cases. E.g if the markets heavily favor presidential candidate A to win, that may cause more people to vote for A, or for their opponents to stay home. Nothing like that can happen in this case because it's the laws of physics. It doesn't mean the bets can't instantiate some crowd sourced wisdom about physics or scientific research.


I wonder if there is any quantum theory that can be applied to prediction markets. The act of observation can change a system. https://en.wikipedia.org/wiki/Schr%C3%B6dinger%27s_cat


That's what makes prediction markets work — if you (or anyone else, for that matter) feel the market has priced the expected outcome wrong, you stand to make a fortune!

Great article on the economic theories behind it here: https://astralcodexten.substack.com/p/prediction-market-faq


Metaculus is just forecasting without any betting

Despite that Manifold market having the most traders ever on Manifold, there's still roughly only a thousand dollars to be made on it, and the winnings can only be donated. Hardly a fortune to be made on either of those sites


The markets still favor people who read and understand the papers. Admittedly, that will be a few solid-state physicists only.


Yes, people with knowledge of the topic can make money from that if the prediction market trends towards the wrong answer.

But that is not how those prediction markets are used. You see arguments like "the prediction markets give only 20% for this to be true" and people take that as "it is untrue". Without any qualifier (it also doesn't help that most people don't understand statistics).


actually, people in large groups tend to estimate correct answers quite well when you average their responses




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: