https://nypost.com/2022/11/09/sam-bankman-frieds-16b-crypto-fortune-shrinks-to-1-billion-in-hours/
The year isn't over, but I'm calling this market as "NO". SBF should have used his money while he still had it.
@IsaacKing Surely the odds of the correct resolution of this market being YES are well below 0.1% at this point.
@MartinRandall If I were to create a market called "Will Isaac King publicly donate at least $5 million to buy solar panels by the end of 2022?", would I be justified in resolving it to NO whenever I feel like it, just because I don't currently have 5 million dollars?
@IsaacKing This was an "action market", in the hopes that it might make the outcome more likely. Now that SBF is possibly a fugitive from the law, and is certainly bankrupt. I have no interest in that action. It probably would not be a crime to entice him to spend any remaining ill-gotten gains in this way, but I still am unwilling to do it.
As far as "improperly resolved" - if SBF does somehow follow through on this, I am sure the Manifold staff will find a way to let me re-imburse the "YES" bettors twofold.
@IsaacKing I think you could resolve it NO upon deciding that you would not take that action, and that would be honorable as long as you stuck to your decision. Resolving NO would be equivalent to a promise to YES bettors. If you broke the promise you would provide compensation or take a reputation hit.
@IsaacKing I made a follow-up market for anyone still interested in this question.
https://manifold.markets/MartinRandall/will-sam-bankmanfried-publicly-dona-a766863fc8cf
As far as "improperly resolved" - if SBF does somehow follow through on this, I am sure the Manifold staff will find a way to let me re-imburse the "YES" bettors twofold.
You wouldn't need the staff for that, you could just reimburse them yourself via tips/manalinks.
think you could resolve it NO upon deciding that you would not take that action, and that would be honorable as long as you stuck to your decision. Resolving NO would be equivalent to a promise to YES bettors. If you broke the promise you would provide compensation or take a reputation hit.
I think my example was poor. Say the market was about my neighbor Bob rather than me. Bob has expressed no particular intentions around solar panels. Would it be ok for me to resolve such a market to NO just because Bob doesn't currently have 5 million dollars?
@IsaacKing I think you'd be taking a very small risk of having mis-resolved the market, but you haven't actually done so until and unless Bob donates $5,000,000 of solar panels.
I'd we can't resolve markets until we have 100% confidence in the outcome then as good Bayesians we can never resolve markets. There has to be some cutoff, well beyond reasonable doubt, at which a market can resolve.
@MartinRandall Would you also be comfortable with this market resolving?
https://manifold.markets/Duncan/will-russia-control-the-majority-of
@IsaacKing I wouldn't resolve it while the probability is at 2%, because I wouldn't take a 2% chance of mis-resolution and needing to compensate a bunch of YES bettors. But I wouldn't call it a mis-resolution. On the other hand if it resolved N/A I might.
On another thread I said that I'd want to see the probability quiesce at 0.1% before I resolved early based on "never going to happen". Does that seems reasonable?
@MartinRandall Sounds like your system would make Manifold unable to host markets about low probability events.
@IsaacKing I think it already can't, low probability events are already overwhelmed by opportunity costs / interest rates. Arguably a problem with prediction markets in general or even prediction in general.
Amplified odds markets don't solve that problem and I mostly bet on them for the lulz.
@GeorgeVii oh it says "donate .. to buy" so i guess maybe only the bunker would count on that one
@MartinRandall Yeah, I was thinking the same thing. I'd have to check, but I think most of the carbon offset technologies would have a better per-dollar return on carbon reduction than solar panels. Although, I suspect that buying off the federal agencies blocking about 18 GW worth of offshore wind projects in the US might easily beat out those, too, depending on what it costs to buy a regulator these days.