[DennisP]

I'm having a hard time seeing a way assassination markets are actually a good business model for assassins.

If the prediction is tightly bounded (e.g. "Mr. X will be killed on Sept. 3, 2018") then you warn the victim.

If the prediction is loosely bounded ("Mr. X will be killed within the next year") then the odds won't be nearly as good, since many non-assassins may bet in favor. The assassin will have to put up a larger bet for a given payoff. Having done that, the assassin wins even if a different assassin does the work, so there's a public goods problem among assassins, and they all have some incentive to stay home playing Call of Duty instead.

[lostctown]

Additionally, not all bets on a person's or persons' life will be made with malevolent intentions. A global non-profit can use prediction markets as an exotic form of donating. The donator takes the "bad" side of the market, and if the non-profit is successful at saving lives, or doing some other form of reportable good, they receive a payout from their bet. Now, if the non-profit pays its employees absurd salaries, operates inefficiently, etc then they stand to lose some amount of money in the prediction market.

If you approach it this way there is no reason all bets on the "bad" side have to have "bad" intentions.

[DennisP]

An idea I had a while back was to set up bets on breakthrough technologies, where you can fund X-Prizes by betting against them. The original X-Prize essentially did this by buying insurance against a payoff, which allowed them to significantly increase the prize amount; my idea was basically to crowdsource the insurance side of that.

https://www.climatecolab.org/contests/2015/shifting-attitude...

[CodesInChaos]

One possible way to implement this: Have on betting pool per person. When betting cryptographically commit to a time interval for predicted death time. After death bettors have limited time to reveal the time interval they bet on. Split pool among correct bets weighted by 1/interval-length.

Making it easy to hire an assassin anonymously is why I find privacy preserving coins like zcash scary, despite valuing my privacy highly.

[DennisP]

Well crap that does seem to solve the problems I mentioned. I don't think that could be set up on Augur, but it could be implemented as a new contract.

[the_cat_kittles]

make a market, buy up all the yes shares when the price is low, kill the person, collect your reward.

[camjohnson26]

Why would anyone take the other side of that bet?

[flyingfences]

Because they don't think you're gonna go out and kill someone.

[camjohnson26]

Still they would get low odds and the risk/reward ratio makes that a terrible bet since it would only take one accident to lose all your money. For like a 1% return.

[the_cat_kittles]

if you think the odds of something happening are lower than the share price, you sell / short / buy "no". if you think they are more likely, you buy. the odds are 100% if you plan on killing someone, and surely some people will think its less than 100%. they will buy "no" shares because they think that its less than a certainty that someone is going to be assassinated.

[DennisP]

Of course. My point is, that wouldn't actually work very well, for the reasons I mentioned above.