You assume all the people betting are enthusiasts. The theory of prediction markets is that rational actors in the market will recognise that a portion of the b

You assume all the people betting are enthusiasts. The theory of prediction markets is that rational actors in the market will recognise that a portion of the betters is overhyped and adjust their bets to make use of their irrational behaviour. If the rational actors are actually effective at making such adjustments I don't know, I bet there's statistics out there on how well prediction markets correlate with reality. In any case, even if the market was perfect, it wouldn't tell us if LK-99 would reproduce, which I guess is the meat of your question. It would just tell us how likely it is that an experimental result made under those specific circumstances would reproduce. And what you could do with that information depends on what your answer to the question: "How would I be affected if LK-99 would reproduce?" would be. If you're a big energy business leader, and you want to filter what topics to spend your valuable time on maybe you could set a rule that you only want spend time reading scientific papers that have >10% odds of being legit. More realistically though, I think things like prediction markets are mostly useful to traders who are trying to arbitrage things like resource markets. What's the price of copper going to do when this turns out to be true? You could adjust your futures based on that.