Set version of question here (with more options): https://manifold.markets/zcoli/date-first-peer-reviewed-manuscript
This version was created in order to be able to resolve options prior to 2029.
The reference date is the date a peer reviewed manuscript first appears online at the journal website or is otherwise published by any of these three journals after being accepted for publication. Any peer reviewed manuscript type is acceptable. If lab leak likelihood is generally accepted based on data and analysis elsewhere, inevitably this will be cited approvingly in a peer reviewed manuscript in one of these journals shortly thereafter.
Any statement equivalent to finding over 50% likelihood for SARS-CoV-2 originating in a laboratory that reflects author(s) conclusions and is published following peer review qualifies.
In the vanishingly unlikely event anyone asks, I will refuse to comment on potentially qualifying manuscripts in submission.
While this is the most objective question I could come up with to address this point, I will not participate in the market.
Update 2025-11-12 (PST) (AI summary of creator comment): The market will close on January 1, 2026 and will be reopened after resolving the first question.
1,000People are also trading
Curious about what I think is an irrationality here with 2026 having a lower likelihood than the subsequent three years and those years having similar likelihoods.
It’s an event that can only happen once. Excluding the current year that’s half over, why would this be as or more likely to happen in later years?
@zcoli Could be that the market thinks they are all close to zero, but adjusts them according to time value and the risk free rate. You can earn maybe 7% up to 2028 on the Yudkowsky UFO market, and that seems pretty close to risk free. Any long term bet is competing with that opportunity.
@zcoli Well, you did make an unlinked market for a bunch of linked questions...
All I did was short every option when the sum was over 100%, and I covered when the sum of all of them was around 90%.
If you think that the odds are actually around 0% that this paper will ever get published, then you would bid the never option to 90+% and the other options to 5-10%, depending on their time value. But then all the categories would all sum to over 100%, so someone else could come and make some risk free money. Seems hard to know what the proper price should settle towards, given those conflicting incentives.
Beyond that, of course, there are the questions of "is lab leak true" (I think less than 1%) and "how strong is the gatekeeping of arguments in those journals" (I don't really know).