Comment by smeej
1 day ago
> The size of the bounty isn't defined by the price of the contract, but the total upside available in the order book.
But one person doesn't get the whole thing. ALL the people holding that side of the contract split the payout, in proportion to the size of their holdings in that side of the market.
I think if I use hypothetical numbers, it will help me explain how I think it works, and maybe this will help someone figure out where my error is.
Let's imagine the market is about whether I will die by the end of the day. So far, there are $500,000 in total bets in the market, and there are 5,000 shares in this market. Let's say it's currently sitting at only 10% odds that I'm going to die. I think that means 4,500 shares, or $450,000, is on the "No" side and 500 shares, or $50,000 is on the "Yes" side. Do I have that right so far?
If nothing changes about the market and I'm still alive at the end of the day, everyone who holds a "No" share splits the $500,000 pot, correct? There are 4,500 of them, so they each get $111.11 per share.
But suppose someone has a solid plan to kill me by the end of the day. They decide they want to dump $50,000 in on the "Yes" side. That's not going to buy them 500 shares, because they would need someone willing to sell 500 shares at the current price. They'll actually get well under 500 shares, and probably not even half that many, and they'll still be splitting the pot among the other people who already have the 500 shares on the "Yes" side. So they're still at not even half the "Yes" side of the market. They can probably double or triple their money, but we're talking about making another $50-100k on top of getting their own $50k back. It's not like they get the whole $500k.
That's what I mean when I say it's "not a bounty." A "bounty" makes it sound like, "If you're the one who kills smeej, you get $500k," but that's not what's happening here.
Lots of people might be willing to try to kill me for $500k. A heck of a lot fewer are going to be willing to try to kill me for 2-3x whatever capital they can come up with right before the hit.
Am I at least understanding this part of it correctly, how the payouts actually work? If I'm not, that would go a long way toward helping me figure out what I'm missing.
> I think that means 4,500 shares, or $450,000, is on the "No" side and 500 shares, or $50,000 is on the "Yes" side. Do I have that right so far?
No - there's always an equal number of contract outstanding on both sides of the bet. A contract is a promise from the person who sold "no" to pay the person who bought "yes" a dollar if the outcome happens. These contracts can trade from anywhere between 1 cent to 99 cents corresponding to a 1% chance to a 99% chance that you would die*. The odds the market reports is just whatever price the last contract traded at (or alternatively whatever price sits between the current open offers to buy/sell contracts. In liquid markets these tend to be the same).
> If nothing changes about the market and I'm still alive at the end of the day, everyone who holds a "No" share splits the $500,000 pot, correct? There are 4,500 of them, so they each get $111.11 per share.
They each get $1 per share. Their profit is $1 minus how much they paid for the share. It's not (meaningfully) a shared pot which is divided up, it's a fixed amount per share.
> They decide they want to dump $50,000 in on the "Yes" side. That's not going to buy them 500 shares, because they would need someone willing to sell 500 shares at the current price.
Ignoring the numbers at this point - you're generally right that they need to find someone willing to sell them the contracts. The existence of a large number of outstanding contracts doesn't guarantee this - they might be held by someone who is holding them to minimize the payout a hitman could get for killing you for instance.
The most direct guarantee is the order book The order book is the collection of open offers "I'm willing to sell X yes-contracts at Y price" that the market has for potential purchasers. The hitman can look at this and snatch up all of these simultaneously (up to some race conditions in the market - we can mostly pretend those don't exist but they do introduce some risk on the hitmans side). This can be thought of as the size of the currently available bounty.
There's a chance the market will continually over-price these yes contracts - and the hitman will never kill you as a result. That would be a huge mistake on all the financially motivated holders of yes contracts though - their positions go from worth something (if they sell to the aspiring hitman) to worth nothing if they don't price them low enough. In general you should expect the market to find the price at which a hitman will carry out the contract - so long as there's enough money in the market in the first place.
* Ignoring transaction fees and the time value of money, it's close enough for this discussion.
But the hitman still does not get the entire value of the contract. The hitman gets the value of the number of shares he can afford to buy, but that's not the whole contract by any means.
I think I understand what you're saying about the pricing. Am I correct in saying, then, that if the odds are 90% in favor of my living through the contract, the "No, smeej won't die today" price should be close to $0.10 (again, ignoring fees and the time value of money)?
If the hitman tries to buy in with 10% of the total funds already in the market, the odds/price are going to shift hard. It's going to devour a huge chunk of the order book. Any market that suddenly has someone come in at 10% of the whole market value is going to get a massive trading wick. So yeah, he'd get some shares at $0.10, but he's probably going to eat the open order book to a much higher cost. He can 10x some very small portion of his money (however many shares are on the book at $0.10), but he can only 5x his money at $0.20, or 3x at $0.33.
Even if we assume he does have $50k to dump into the market, I still don't see how he's going to more than triple his money, which is a heck of a lot less than taking the entire market's value as though it were a bounty.
A risky bet can give back much more than 3x the money. Take this absurd one for example, it seems like it's offering up to 9x on a yes bet right now: https://polymarket.com/event/us-civil-war-before-2027
(I don't know how much volume is available, nor do I endorse betting on either side of things like this)
Yes - we agree on how the pricing and odds work now :)
The hitman shouldn't expect to capture the value of the entire open interest. The market here is serving to negotiate the bounty with speculators betting that too much was offered taking the rest (a privilege they pay for by buying contracts that only pay out if they don't take too much). It's a curious form of negotiation since the people paying for the murder don't participate... but should (in a very theoretical efficient market) come to a "fair" (large enough to get the job done, and no larger) payment for the hitman.
2xing your money in a night is a huge payout, I think you're overestimating how high the multiplier on the capital requirement needs to be. That said, if you aren't, and you need a 5x payout to find a hitman then no rational speculator would purchase contracts for more then $0.20...