I was waiting impatiently for the active inference punchline, which I think might be rather profound, insofar as you've just illustrated that markets are virtual organisms.
Thanks for the writeup, Scott! I've honestly been super thrilled to see all the different use cases our community has figured out for the prediction market mechanism; I think it really validates the approach of letting people create markets on anything they want!
One more example of a bounty/action market that worked pretty well: Elizabeth & I posted a $100 bounty for some data aggregation work for Slime Mold Time Mold's hypothesis of contamination-led obesity. It got two separate takers: https://manifold.markets/Austin/100-bounty-will-someone-combine-the
Would love to figure out how to better support this natively -- if anyone has suggestions, please reach out!
Your example isn't a proper action market; it looks like it's a prediction market on someone taking the traditional bounty (which maybe increased visibility of the bounty?). Is manifold all play money?
The way you'd set this up in a real / proper market is that you'd pick the bounty B you want to offer ($100 in this case), you'd guess a probability p that this bounty is sufficient to achieve the results you want (maybe you think it's 70% likely), and then you offer N = B / (1 - p) shares of "yes" at price p (in this case with my fake numbers, 100 / .3 rounds up to 334 shares @ $0.70). You don't offer any shares of "no". If you get no takers, you're getting information that your bounty is too low / your probability is too high, you can revoke the sell offer and try again at a different point.
Note that if your probability is too low relative to the true rate, the difference will be picked up by people who can predict better than you rather than by the bounty taker.
So in this example, if I want to "accept" the bounty, I buy all 334 shares for $233.80 and do the work, so the market resolves to "yes" and I get $334 for $100.20 profit, is that right?
It doesn't seem reasonable to expect people doing small tasks to pay a deposit.
Yes, that's right. One problem with action markets is that they require having cash on hand to invest.
One interesting feature is that the less likely the market thinks the current bounty is adequate, the lower the deposit you have to pay. So it seems to me like the deposit is what you pay to stake your claim as the person taking the bounty, rather than all the other people who were likely to want it.
What's unreasonable about having to pay the deposit?
Just because the sort of people who might agree to do small tasks for small sums typically don't have cash on hand. For example, Scott mentions claiming a bounty when he was a poor medical student.
I think the equilibrium state here is that the people setting up the bounty are calibrated on how much of a deposit people can pay, and this is factored into the probability that the specified bounty is adequate. (If the deposit is too high, then the bounty setter was wrong in their probability estimate, and next time will predict a lower probability with the same bounty, which lowers the deposit. Gradient descent will keep lowering the probability until the deposit is the highest that someone is willing to stake.)
This is an interesting nuance; thanks! I didn't realize that the opening of the bounty market had three different constraints included: the size of the bounty, the size of the deposit required to claim the bounty, and the probability that someone will claim the bounty.
This is also a striking example of how prediction-action markets, like almost all other forms of increased liquidity, benefit the already rich more than they benefit the poor.
(1) "What will I (Scott Alexander) rate as most promising when I do a deep dive into the research on pregnancy interventions?"
I'm going for the "Natural birth versus Caesarian"
(2) "They’re betting on whether someone will clean up the (currently disastrous) backyard. They’ve said pretty openly that they’re hoping someone will buy a lot of “yes”, take care of the backyard project, and then take all their money. "
Could they not just pool the money and hire a landscaper or groundskeeper? When my back garden was a disaster, after a long time dithering about "something needs to be done", we finally hired a guy to come cut the hedges, cut down and take away the tree that was growing in the wrong place, and do a ton of weeding and spraying. Result! If you're waiting for someone else to do it, it will never get done - "what's everybody's business is nobody's business".
People pool the money by buying "no". If nothing happens, they recover their money. If the pool is deep enough, someone will buy "yes", take the money, use it to hire a landscaper and keep some for his administrative troubles and responsibility.
I am a simple idiot, I don't understand why, if the entire household is agreed on "the back yard is a disgrace to God", at least one person doesn't go "Right, I'm Googling 'landscapers', we'll get a quote, and if it's anyway reasonable we hire the guy to come next Tuesday. Everybody throw $100 on the table for a deposit".
Cut out the middleman of waiting for somebody else to do it? If it's "keep some for his administrative troubles and responsibility", I will very happily spend ten minutes looking up "Bay Area landscapers and garden maintenance" in return for a bung:
Well if they want to keep the money inside the group instead of paying someone else it might be an effective method to find who in the group is willing to do it for the least amount.
Or if prediction markets like this became big enough that landscapers/contractors kept an eye on them for 'jobs' it might allow them to find the landscaper willing to do it for the least amount of money. It would almost be like a reverse bidding process, just keep adding to the pot until someone is willing to bet against it, do the work, and take the money
I think the truest answer to your question is "because they are the kind of people for whom solving a problem in a cute way with a prediction market is much more satisfying than solving it in a boring way like normies." But there may be other advantages as well.
I think I buy this. I suppose I intuitively think of prediction markets as being for (in Scott's parlance above) high-% prediction low-% action; it's just confusingly low-agency to me that they already have the money to pay for someone to clean up the backyard but instead choose to create a prediction market for whether someone will do so.
Yeah, that's what I don't understand here. There's a job needs to be done, there are tradesmen who do it, and they can put up enough money between them all to hire a guy and get it done. What is the gain from putting it off even longer by "let's play a game where someone *might* or *might not* do this job for us, which will cost us more as they will want a fee for doing it on top of whatever the tradesman charges"?
Then you'd probably be the one that takes the pot, and not me according to my broken toilet and a few other stuff I've been putting off and I'd definitely pay for third parties to take care of if I know I was getting a competent job and fair price without doing any research (also, no Yelp over here nor an equivalent with reliable user feedback and ratings)
Oh, I can procrastinate to beat the band, but I have a brother who is more energetic and he was "Right, we've talked about this long enough, time to get it done!"
For small jobs like getting the immersion cylinder replaced (well, relatively small but a pain in the neck to do yourself), it was a matter of asking around work mates and friends did they know a plumber they would recommend to do it. Word of mouth is often pretty good, but you takes your chances.
It's perhaps worth noting that you can just post things like this to one of various "find a tradesperson" type sites. It probably doesn't result in either the cheapest or the best job, but it does involve minimal faff. IME, things get picked up in a few minutes.
When someone buys "yes", they can't take the money until after they succeed. When they hire a landscaper (investing their own money), then the people living on the premises who bought "no" have an incentive to tell the landscaper "we didn't call you and we don't consent to you doing the work, leave or we'll call the cops", and get the money that was put in by whoever bought "yes".
Presumably the people living in the house don't have identical agendas or values, otherwise they would just have a consensus on what to do without this market. The process certainly must account for at least one of the people "defecting" and acting in their financial self interest.
I realise that this is just basically a fun game, something instead of betting on the horses or the football like the common people do. I cannot visualise this as any practical way of decision making.
More broadly, I'd probably phrase it as "the backyard isn't the point, the prediction markets are the point." One doesn't make Rube Goldberg machines for the sake of turning on lights (or feeding oneself cake: https://www.youtube.com/watch?v=auIlGqEyTm8), or even just for the fun of it (although it can be a ton of fun! look at that candle sliding down that ramp by heating up a stick of butter!), but also to learn about pulleys and levers and gears and whatnot, in hope of putting them to use someday in a machine that's actually useful.
As a way of cleaning up the backyard, a prediction market is confusing and probably inefficient. But as a way of learning about human behavior and the limits of prediction markets, what's more practical than making a market for a relatively unimportant, easily solvable problem and seeing what people do with it?
They also created a situation where a literal crazy homeless person might show up to clean their yard and take their betting money, and they may realize they've made a bad mistake and tell the guy to go away. This may in fact anger the guy, and create a possibly violent incident.
Or, much more likely, some guy shows up who's not very good at landscaping cleanup and does a poor job. Or some guy shows up who has no insurance and damages something or hurts himself.
If the backyard is bad enough, a crazy homeless guy might improve it some amount. Better to do something than nothing and let entropy keep on keeping on.
Though I suppose crazy homeless guy could lug in bags of rubbish and start tyre fires etc. in the back yard, so yeah.
"We *could* take ths $1,000, look up somebody online and see what reviews they got, and hire them to come tidy up our backyard. Or we *could* take the money to deliberately lose it so that somebody else will look up a guy online, see what reviews he got from previous clients, and hire him to come tidy up our backyard, plus take a slice of the action as a middle-man, so it costs us more than if we did it ourselves.
Clearly option B is a better, faster, cheaper way to do things!"
Yea. There's also the free rider problem. But it is not strictly worse in every way than "just coordinate with everyone to pay up and then hire a cost-effective and reliable third party to do it for you".
There does seem to be a basic problem that I need to spend, say, $1,000 buying all the yes shares, then $600 hiring a landscaper, so that I can redeem the shares for, say, $2,000 and make $400 profit. But if I'm the sort of person who has $1,600 I have no particular use for, this is probably more work then I'm willing to do for $400, particularly as I take on the risk that the landscaper does an unsatisfactory job.
This is not actually a problem. If nobody is willing to deposit $1600 to make a $400 profit, then 2000 shares at $0.50 is much to high a probability estimate. This is not just a linear optimization problem: you are looking to set a bounty of $B such that there is a probability p that someone will be willing to deposit $B*p/(1-p) (+ extra fixed costs) to claim the bounty. But continuity + the intermediate value theorem guarantees that if the bounty is enough to solve the problem with no deposit required, then there is some solution (unique if the probability is monotone in the deposit).
This seems to point out the failure in prediction markets; the lovely shiny ideal is that they will harness the wisdom of crowds to find the best answer to any particular problem.
But seeking for truth is always going to take a back seat to "I could make money screwing over the other people" and so prediction markets will simply be one more way of "how do I turn a profit on this" and not "what is the best answer".
Consider the scenario as set out there:
1. The backyard is a dire mess, we want to get it cleaned up
2. We could put our money together and hire a garden maintenance guy
3. No, let's do a prediction market instead! Surely that will be better value for money and more efficient!
4. Someone *thinks* "Hey, if I screw over everyone else in the house, I could make money here!"
5. Backyard is still a dire mess, and now the money for getting it cleaned up has been scooped by a profiteer, so we're even worse off than when we started
Ironically, it's self described Rationalists who hold an impossibly naïve belief that something that can be corrupted by the profit motive/conflicts of interest, somehow won't be.
I know they're decent people with good intentions. It's just easy to bristle at the chutzpa of people who make such erroneous assumptions about human behavior and then deign to call themselves Rationalists.
Its one thing to make an argument for "this is how a group of people should behave if they want to optimize for problem solving", like a kind of moral plea, but when it's "this is actually a solution that will work for actual people", my eyebrows fly off my forehead.
5 is wrong, and 4 is probably wrong as well. So long as the people wanting the house to get cleaned up only ever sell shares of "it's clean", either their money is returned to them or the backyard gets cleaned. If they also buy shares of "it's dirty", then they are buying insurance in the case where the backyard remains dirty, and in fact make money in that case. It's only when the people who want the backyard to get cleaned sell shares of "it's dirty" (or buy shares of "it's clean") that they risk loosing their money to profiteers.
Is everybody in this conversation aware that Manifold is a play-money site, so there is no actual money at stake in this market? This whole thread reads like none of the participants are aware of that.
I'm not sure how the action market helps with the doctor situation. Isn't creating an action market for it just worse than the solution of asking a doctor to help, paying him if it does and demanding that he pay you if it does not? Your complaint with that system seemed to be that a risk-adverse doctor who was less than 100% confident might pass, but it seems like the action market would have the same issue here.
Yeah, action markets are useful when either (a) you want to fractionalize the bounty resolution and offload the impact evaluation to the market, or (b) you want to know how likely it is that the thing would have happened anyway and want the fraction of the bounty that is not required to incentivize action to go to people who are correcting your inaccurate prediction. You unfortunately can't really decentralize surgery, I think.
With respect to the Scott Aaronson thing, my understanding is that the American Math Association has come up with a solution. Or maybe some other math organization; it was explained to me a couple years ago, so one or two details may be off.
Most people who submit proofs for the Riemmann Hypothesis (or any other similarly famous problem) are are not so correct that a PhD student can't take apart their argument. So, the AMA charges about $100 per page of the proof, and then gives it to a PhD student to dissect. The student writes up a clear report on why the proof doesn't work, which is then returned to the author, and the student gets some cut of the money to use as travel funds.
Now, this may not still be totally accurate, or perhaps the person once telling me it was not accurate in their description, but it still feels like a good system.
Listen, buddy, I've been accused of a lot of things in my life, and I even did some of them. On the other hand:
1) That's just a troll.
2) Even if it were certain that I was going to defraud the market, it still doesn't explain why apparently nobody knows how to request that a Twitter account be verified.
3) Even if it were certain that I was going to defraud the market, the various people interested in whether Turkey will change its spelling could still get useful information from making comments.
> I did so bad with the hint, I think I may just steal all the money. I will think it over for an hour first.
then almost immediately after you resolved market assigning 90% to correct answer and 10% to false, which you bought and caused you to earn Manifold points from fraudulent resolution
> Even if it were certain that I was going to defraud the market, the various people interested in whether Turkey will change its spelling could still get useful information from making comments.
In such case people would likely make a separate market from than not suffering from this problem
I didn't you see commenting in those links that the accuser was a troll. That would seem to be the most relevant place to do so, because otherwise it looks like an accusation you didn't deny.
>I once joked that instead of lower courts, we should have prediction markets on what the Supreme Court would think of a given case. Same idea, higher stakes.
Related idea I've had whenever there's a SCOTUS vacancy: a major qualification for appellate court judges to be elevated should be the extent to which they have exercised their delegated authority faithfully, which is to say, when cases they've decided subsequently make it before the highest court, do the Justices agree with them? 5-4 splits would have the least weight either way, 9-0 agreeing/disagreeing with the judge would have the greatest favorable/unfavorable weight; more emphasis given to cases in which the judge was in the dissent at the Circuit.
Isn't the main difference for incentivizing actions via the stock market vs. prediction markets is that stock markets are highly regulated and large transactions such as placing a major short on Tesla the day before Elon Musk is assassinated would be easily tracked by the SEC/law enforcement. The companies running prediction markets could presumably track these types of things and inform law enforcement about them, but would they necessarily have the infrastructure/processes in place to do so? And would they all voluntarily do that? Surely some privacy libertarian style prediction market would advertise how it doesn't track any user data or supply it to the govt. And what about blockchain prediction markets using anonymous untraceable cryptocurrencies like Monero?
Most other major markets where you could make large sums of money by taking a huge position and purposely affecting change in that market has some form of regulatory overview that presumably helps deter these kinds of things.
Iirc the design, many people can contribute to a pool for a given person to be killed, and to win the pool you have to guess the exact day and put down some fairly large amount of money. This encourages the pool to grow, and the down payment helps make it unlikely the assassin will need to share the winnings.
For some entertaining reading about a strange crypto-anarchist, check out Jim Bell's Wikipedia page and the links to primary sources. I first encountered his writing on a cyperpunk mailing list in the late ⁰⁰'s, I think while he was momentarily between prison sentences.
IANLAL (I Am No Longer a Lawyer), but I think Daniel's claim is maybe a little too sweeping, but I suspect he means either:
The result of a case can turn on a number of different questions of fact and law and it is typically only questions of law that are subject to appeal. So the higher court is answering a different, narrower subset of the question. But you'd still like the lower court to have come to the same conclusion on the specific legal questions.
And/or
Lower courts are more bound by more precedents than higher courts. So it might consider itself required to reach a conclusion other than the one it would reach from first principles because of an existing decision of another equal or higher court, while a high court might feel free to overrule that decision. This is all true, but a relatively rare species of overruling, AFAIK.
Disclaimer: I am an English lawyer. I imagine that the system is similar in the US, but I'm not sure.
Most of the work done by courts of first instance is determining the facts of cases, for which purpose they hear evidence from witnesses who are cross-examined. Appellate courts do not hear evidence and take the facts as found by the first instance court. This is why Supreme Court hearings are often much shorter than the first instance hearings to which they relate. Somebody needs to make findings of fact in order for the Supreme Court to make a decision, so you can't short-cut the process in the way proposed.
This is all true, Robert, but it's still my understanding that the lower court will also address the specific questions of law which are ultimately appealed in their decision, and optimally their decision ought to match that reached by the appellate court, unless there are issues of differential levels of deference to authority.
Note that, while it's obviously not for money, fantasyscotus.net is quite real and could quite easily be monetized in a different regulatory environment.
For many of these applications, I think one of the biggest challenges for prediction markets will be to beat the crude approximations to prediction markets that we already self-organize on an ad-hoc basis. To take an example from this post where I have some firsthand knowledge, I indeed don't read most of the P=NP proofs, refutations of quantum computing, and other revolutionary developments that cross my field of attention -- but that can be overridden by colleagues I trust staking some of their reputation that, hey, this is interesting, this is worth looking at, etc. Which means that a prediction market wouldn't just have to beat nothing whatsoever, it would have to beat THAT.
In the meantime, though, if it's really important to optimize my use of time, then there's surely lower-hanging fruit to be had, like me not spending hours per day laying in bed with my phone scrolling through social media like now... :-)
Speaking of weird papers, I read an interesting peer reviewed paper once purporting to show that quantum mechanics is a consequence of special relativity. (https://doi.org/10.1088/1367-2630/ab76f7 ) I know "why quantum mechanics" is an interest of yours, so it might not actually be a waste of your time to look at it. ;)
The super short version of the argument:
1) Assume that special relativity is correct.
2).Assume that, in particular, special relativity still makes sense when you allow reference frames to move faster than light with respect to each other.
3) In order for superluminal reference frames to make sense, particles need to act in certain ways consistent with QM but not classical mechanics.
I know enough modern physics to understand what the paper is talking about, but not nearly enough to know if it's clever, trivial, wrong, or not even wrong.
The phrase "the market is wrong" makes no sense to me.
Consider a market created by punters on a horse race. If the longshot going off at 40-1 odds wins, it does not mean the market was wrong. That horse *should* win about 1/40 races. OTOH, if the longshot wins 40 races in a row, that market probably isn't functioning. More important to note, though, is that if the favorite wins 40 races in a row, that also means the market isn't functioning.
If you want to know how well a market is functioning, give the market a Brier score across at least dozens of predictions. The EMH is true by tautology if you have a true market.
This sentence has a minor error where the order of 2 words is reversed, but it happens to be an error that tricked my internal parser into going down a garden path that actually took me a nontrivial amount of time to back out of:
"I throw these out all unread the time when I get them in the mail"
I considered doing the fix Valinor's backyard myself thing, but (a) it was unclear to me that they were actually angling for that and I thought it might be intrusive if I just showed up and started touching their stuff, and (b) if I'm going to do manual labor I want to be paid in real actual money and not Monopoly money ¯\_(ツ)_/¯
On why we don't have more geniuses: "His summer following his father around learning medicine was probably good for him, but not outside the bounds of what still happens today (I followed my father around learning medicine)."
On whether small prizes are great for thoughtful, complex: "I put in 20 hours of research and won. If I’m typical, this is proof of concept that [...]"
This is maybe not the most effective argument that parental tutoring is less than very effective, or that medical students can be cheaply incentivized to do good lit reviews.
Thirded - I am about to start making children, and need that guy stat! I did read the old one and adjusted accordingly, but I assume new facts will come to light since several years ago that the old one got made.
In regard to the historical prediction market, I wonder how you could use them to judge the consensus on past events. There are many things that happened in the past that are still unclear, with papers being published explaining that this and this happened and two months later a counter-paper that says it totally did not. It is often hard for non-experts to
1/ judge what is the most plausible explanation of what happened,
It will be probably be too obscure, too niche, too complex and too irrelevant for it to gain traction (let me know if you got better ideas for a disputed event, which should be easy to find). I thought of using the same technique in my previous experimental market, as in "Convince me that XXX is true and I'll resolve the market this way" (https://manifold.markets/TaoSumer/are-there-more-doors-than-wheels-in), but I feel like such a world-changing event is too complex to ask people to debate constructively in the comments. I guess the best way to make it interesting would be to invite historians who wrote papers on Stalin deaths to bet money and debate/argue, and then get the crowd to decide who got the best arguments - tho rethoric would probably come in the way, with people voting for the best speaker/writer.
Do you have any ideas on how we can make such markets better? I can implement them for the Stalin market - or create a more interesting one.
> let me know if you got better ideas for a disputed event
I love this comment! I guess a market on "did Jesus physically resurrect from the dead?" would be a prime example. The question has a lot of significance (historically, and still today for millions of people), has been debated at length, yet each side has people who fervently argue for it.
I would be very skeptical though that this marked could be resolved... much more so than for the Stalin case. Maybe one day when we get wormholes and space/time loops ;-)
I thought that question was rather settled in the positive?
https://en.wikipedia.org/wiki/Historical_Jesus claims that "Virtually all scholars of antiquity accept that Jesus was a historical figure, and attempts to deny his historicity have been consistently rejected by the scholarly consensus as a fringe theory."
I agree that it would nevertheless make for an interesting market, and would definitely be easier to resolve than the resurrection.
> Do you have any ideas on how we can make such markets better?
Idea: reverse the role of market-maker and participants. Participants are *only* allowed to sell shares, and the automated market-maker (or the creator of the market) only buys shares. Perhaps implement by mandating that a given user can have open sell orders only on one side of the market at a time, and submitting a buy order on one outcome requires submitting a buy order on the other outcome simultaneously?
Theory:
The "obvious" prediction market resolution condition for, e.g., "did Joseph Stalin die of natural causes?" is something like "does the market reach $0.95 for yes within a month?". But by the prediction-market-analogue of Lob's theorem, this is equivalent to a market betting on the question "will this market reach $0.95 for yes within a month?" without any reference to Stalin. The problem, I think, is that it's so much cheaper to shift the market than to make an accurate prediction of the outcome of the Keynesian beauty contest that the market just becomes an action market and the market maker is incentivizing shifting the equilibrium in whichever direction is easier to shift, so you don't actually get information out.
Alternate analysis: Buying "yes" shifts the probability of "yes" higher, so all you have to do to win is flood the market with enough money that the price shifts in your favor.
But if you flip the roles of participant and market maker, then the market-maker is actually disincentivizing action that shifts the market.
What's the incentive, here?
In this scenario, you only make money when you're wrong. When you're right, you have to pay out money (in proportion to how uncertain you were that you'd be right). So your incentive (at least if you're risk-averse) is to find the markets where you're most confident in your prediction (high probability), but also where you are the least meta-confident in your confidence. (Huh, this is interesting. It rhymes with the thing that I think brains do in predictive processing, which is attempt to maximize the amount of surprise they can eliminate/explain/understand/resolve.)
This is also cute: in normal prediction markets, you make money by being *more right* (buying the correct outcome, with more accurate predictions than the market), while in these reversed prediction markets, you make money by being *less wrong* (selling the incorrect outcome, but at a price which indicates that you think it's less likely to happen than the market does).
Coming back to the point, as long as there are no secondary markets on these "reversed prediction markets", I think there's no way to make money by influencing the outcome within the market. (Someone please do a better analysis than me, though!)
> let me know if you got better ideas for a disputed event
The US government declassifies some of its classified material on a certain schedule. I recommend looking up e.g. the FBI's automatic declassification schedule, then making a market set to resolve after a relevant FBI document release occurs.
By the way, I made a prediction in your Stalin market. Let me know whether the comment I attached to the prediction takes a reasonable approach.
Oh man, you wrote the blog post I wanted to write! (But you got some things wrong.)
> But it’s strictly worse than a regular bounty system.
Nope. In a normal bounty system, you set a bounty and if anyone is willing to do it for that price, they do the job and get the bounty. If they were willing to do the job for less than the bounty you set, they get the excess as profit.
Action markets have one advantage and one difference (which might be an advantage).
The advantage is that they allow decentralized claiming of the bounty:
Someone can buy up all the "yes" shares, clean half the backyard, then offer to sell them at the the halfway mark between the price they bought and 1, and thereby claim half the bounty. Moreover, in a sufficiently liquid market, the price that they get for their shares will reflect the market consensus on how much their action contributed to the goal (so you'd get less than half the bounty for cleaning the easier half).
The difference is that in normal bounties, the excess bounty (over what people are willing to do the job for) is paid to the person taking the bounty. In action markets, that excess is paid to people who are making actuate predictions and giving information to the market. I actually think this is pretty cool (but maybe you'd prefer to tip the people who perform your bounty rather than paying that money out to the people who are teaching you that you're wrong about how much it costs to get something to happen in the world?)
> For one thing, if I suspect someone else will take care of the backyard, then just by registering that prediction I can share the winnings with them, even though I didn’t do any work
You can, but only if you believe that sharing winnings with them will not push the bounty below what they'd be willing to take. And if you do that, you're providing information to the market.
For example, say there's 1000 shares of yes that can be bought for $0.70. This means there's a $300 bounty on cleaning the backyard. If you think someone is about to clean the backyard and try to preempt then by buying half the shares, you decrease the bounty to $150. If that's still enough for them to clean the backyard, that means the bounty-setters were miscalibrated on the bounty required to clean the backyard, and you're claiming the excess. If it's not enough, then you're betting incorrectly, and the price should not rise.
> (since this is at 70% (rather than 100%), someone must be doing something like this).
This is not true. If the probability is at 100% there's no bounty to be had. (Maybe you meant "rather than 0%"?). Additionally, *you can't decouple the action from the prediction*! You're supposed to set up an action market at a price $p and volume N shares such that you believe $(1-p)N is the minimum bounty required to make the action happen with probability p. If you differ from this constraint at all you will lose money (in a sufficiently liquid action market) to people who are better than you at predicting.
> For another thing, it incentivizes people to bet “no” and then sabotage cleanup projects.
Only if you set up your market wrong. If you're trying to make "yes" happen, you're not supposed to offer any shares of "no".
Importantly, none of the money that you are bountying towards "yes" can be extracted by the people who are trying to sabotage you. If the the yard gets cleaned, you pay out the bounty to the bounty-taker and the people who predicted better than you. If the yard doesn't get cleaned, you get compensated.
So insofar as there is an incentive to "bet no" and sabotage the yard cleaning, it must come from something other than your bounty (maybe it comes from the people playing the prediction game on your bounty market? maybe it comes from making the outcome and the current prediction more legible?)
> Banned because Wall Street developed a financial instrument that let them short ConTrackeds, then tried really hard to prevent bridges from being built
Ha, but this isn't an argument against action markets, it's an argument against inadequate equilibria. Why couldn't Wall Street come up with a financial instrument that lets them short bridge building directly and then go interfere in bridge building? If the shorting method involved literal borrowing of ConTrackeds, then this is an argument for the stupidity of the lenders, who should have realized that lending this token incentivizes the lendee to sabotage the project. At worst, you're not supposed to ban the action market, you're supposed to ban the derivative markets that incentivize antisocial behavior.
> Commenters brought up that you get weird incentives as soon as more than one person owns ConTrackeds
Link? The incentives, as I understand it, are:
1. If you bought the token to claim the bounty, you have an incentive to make partial progress on building the bridge, and to sell as soon as the market adjusts to account for the partial progress you believe you will make.
2. If you bought the token because you thought the market was wrong and you know better, you have an incentive to sell as soon as either the market adjusts or you learn otherwise.
3. You are incentivized to buy only if you either think you know better than the market or you are planning to make progress on the bridge building.
> I think you could solve this problem by subsidizing a prediction market at 25% chance of recovery, and letting doctors bet on other odds.
Yes, you can. This is what my formula above says in general.
> This would still have some problems - if someone saw a famous doctor betting, they could bet too, and “steal” some of the “winnings” from the doctor - but it seems to be kind of the right idea.
No, this means that either you were dumb and set a higher bounty than necessary (the doctor still buys enough shares to make it worth it to treat you) or that these people are dumb and will lose money to the market (there isn't enough bounty left to incentivize the doctor).
> Different prediction markets “leak” into being action markets at a different rate: one about when a distant star will supernova is 100% prediction; one about when a President will die is 99% prediction; one about when the forecasters’ backyard will be cleaned is maybe 80% action.
Important question: can you read off the leakage rate for a given market from market data? I think solving this problem is required for solving AI interpretability, based on the connection you mentioned.
Important follow-up: If you can't read it from market data but you can read it from brains / predictive processing / action-inference, what does this say about the design of brains, and can we port this design back over into the marketplace to make better prediction-action markets?
And finally, I want to note that this issue is already present in existing AI and prediction markets. The book Surveillance Capitalism talks about "behavioral futures markets" that Google and all the other big Internet companies run (I'm not sure how much this is a literal thing and how much it's a model), which are used for advertising. The book also talks about how they want to control our actions to make them predictable. (I think this book is secretly about misaligned super-human AI that is actually already here, and how we really better get our act together and align it before it's too late (unless it already is), except the author doesn't quite realize this. (I want to write a book review on it, or collaborate with someone on writing a book review on it.))
Final thoughts: there are five directions of incentives that action markets will incentivize:
1. They incentive better predictions. This is the standard incentive from prediction markets.
2. When there's much more money on selling "yes" (/ buying "no"), they incentivize "yes"
3. When there's much more money on selling "no" (/ buying "yes"), they incentivize "no".
4. When there is the same amount of money on both sides ("yes" and "no"), they incentivize whichever direction is easier to influence, i.e., they incentivize making behavior more predictable. This is the issue talked about in Surveillance Capitalism.
5. When the amount of money on each side, after being adjusted for the marginal cost of influencing the outcome in that direction, is the same, I *think* the market disincentivizes any action that shifts the likelihood away from the current prediction.
So beware that playing market maker and adding liquidity to prediction-action markets is not morally neutral. You are incentivizing whichever outcome is easier to cause. Since destruction is generally easier than creation, this says that we should tax market makers of action markets at the rate of destruction their market-making is liable to induce. (This is a great use case of being able to read the leakage rate off of market data! If we can read off the leakage rate, then we can accurately tax market makers!)
While I'm here, I want to share an interesting tangent about how action markets are cool that didn't make it into the above comment.
I want to investigate an interesting incentive / financial product around ConTrackeds. If you buy ConTrackeds intending to claim the bounty, and then a natural disaster hits making bridge building much more expensive, you're out a lot of money. The solution to this, I think, is *more prediction-action markets*. If you set up a market in "I will invest at least $X into building this bridge", you should be able to derisk your ConTrackeds by transferring your investment to this market. Namely, you buy a bunch of ConTrackeds, declare your intent to build the bridge, and then offer to buy up shares of "yes" in this market (equivalently, sell shares of "no") using the proceeds of selling off the ConTrackeds. You should pick actions that are sufficiently hard to interfere with yet sufficiently correlated with actually building the bridge that the bounty you're placing on sabotage isn't actually significant. For example, "I will place an order for this amount of materials with this provider at this time, and earmark them for the bridge building project, provided the price is at most this amount" is better than "I will have this amount of materials by this date" because it's much harder to sabotage "placing an order conditional on price" than "receiving materials".
If there's enough liquidity in the market and you set it up right, I think you can derisk pretty quickly. I think this is pretty cool!
(It does show that we're going to need something like contract lawyers and bug finders to get action market incentives set up right.)
(I don't know anything about finance, but) maybe he meant that bombing a plane might temporarily reduce public confidence in planes for mid-distance travel so demand for train travel as an alternative increases, boosting train company stocks? Or something like that?
“The agreed-on solution for this is lots of conditional prediction markets.” - a downside here is that each individual market ends up being lower confidence, as bettors have to split up their time across many questions. But if you think that there are diminishing returns to time spent on a question, this is probably am overall win?
For the pregnancy market, you are assuming women are already taking prenatal vitamins, right? (Otherwise folic acid or iodine or something will be best).
I think those are the low-hanging fruit, certainly if you are planning to get pregnant then every resource including your doctor tells you to supplement with folic acid, stop smoking, cut back on drinking, eat healthily, get enough sleep, etc.
That's why I'm punting on the "natural birth versus Caesarian". Caesarians are a procedure that are becoming increasingly common and there's a lot of pro- and anti- debate around them. There's the "too posh to push" bit, where the claim is that when it is not medically necessary, women are asking for Caesarians in order to cut down time in labour. There's the "vaginal delivery puts stress on the newborn, including physical risks" claim, where Caesarians are said to be better for both mother and baby.
Like any medical procedure it has risks, including if you're intending to become pregnant again:
There should be enough data now, since it is becoming more common, to see what the effects (if any) on newborns of natural versus surgical delivery are, as well as fertility effects (see paper linked above).
Sure, there are papers like this that show caesarian-born children are at higher risk of asthma, obesity, and respiratory tract infections: https://www.mdpi.com/1660-4601/17/21/8031
However, I don't know of any randomized controlled trials, and the confounding factors seem pretty significant here (at-risk births are more likely to end up as c-sections). I don't even think the above link restricted the analysis to elective c-sections (it's not mentioned in the abstract; I didn't read the paper).
> when I was a poor medical student, Zvi Mowshowitz’s company ran a contest for the best literature review of mineral supplementation with a $5000 prize. I put in 20 hours of research and won. If I’m typical, this is proof of concept that big enough bounties can incentivize poor medical students to do good lit reviews
Is there a website which aggregates links to these sorts of contests? (It could attract entrants who are indifferent to anything beyond the semblance of accuracy, so I wouldn't be surprised if biomedical companies wouldn't actually use it.)
Had you expected a large number of entrants, would you still have entered the contest? If so, you were probably motivated by more than just the money.
Start-up idea: a repository for unusual bounties of all-kind / a prediction market platform adapted for conditional markets (for example with the ability to visualize conditional markets in a tree-like structure of some kind)
About the idea with organising market to decide which papers should be sent to e.g. Scott Aaronson: isn't it just a non-expert version of peer-review process?
The question about the dead drummer is not a question about past events it is a question about a future event -- i.e. the contents of the autopsy report. If the report is not made public, it may actually be about what the ME says about the report.
Yeah, the natural assumption is that drugs were involved, but it could be due to a completely unrelated health problem. We won't know for sure until the report.
There are a lot more points of failure and assumptions than that:
1. The autopsy could fail for reasons beyond anyone's control. E.g. the body may have been contaminated or too decayed, or the cause of death may be beyond the ability of curreent technology to detect it.
2. The pathologist could be incompetent or dishonest.
3. The report could be lost, garbled, or miscommunicated to the coroner.
4. The coroner may alter the report or hold it back from publication to save the family embarrassment. Ever heard of auto-erotic strangulation?
5. The press may misunderstand, misreport, or lie about the corner's statement.
We assume that the autopsy will produce reliable results which will be correctly and competently reported to public officials who will accurately disclose their contents to the communications media. who will report on them accurately.
in 2017 someone bombed bus of top German soccer team, having “allegedly bought options to short sell 15,000 shares of Borussia Dortmund stock for 78,000 euros [£65,000]".
It seems like some of the moral hazard issues with prediction markets would be well addressed by limiting the players and the potential stake of any players. Even if we assume that someone might be willing to do antisocial things for money, the effort involved should at least make low-stakes bidding possible. ... until you get multiple, redundant markets, that is.
1. Market-making in a prediction-action market is not morally neutral. It incentivizes whichever outcome is easier to make happen.
2. As far as I can tell, no one yet knows how to determine the point at which prediction markets leak into action markets. (Unlike what Scott implies, it's not a fixed rate, but a function of how much money is flowing through the market in each direction.)
2. As far as I can tell, people don't currently understand the moral implications of the four ways of participating in a prediction-action market. If cleaning the backyard is prosocial, then selling "clean" and buying "dirty" are both prosocial (the former increases the bounty on "clean" and the later decreases the bounty on "dirty"), buying "clean" is conditionally antisocial (you are decreasing the bounty on "clean", but if you are planning to clean it yourself that's fine and if you think the bounty is higher than it needs to be than you are claiming the part of the bounty that's actually for getting information about what the minimum bounty is), and selling "dirty" is actively antisocial (you are increasing the bounty on "dirty")
How about an attention market where the currency is attention of every participant?
Every person upvotes content that they think deserves your attention and downvotes content that will waste your attention. They have some starting balance of your attention tokens and with every upvote/downvote they place a bet. If you upvote some content - you reward everyone who upvoted it by giving them 100 tokens of your attention (divided equally). And you penalize those who downvoted it by taking away some percentage of your attention tokens they hold.
You do the opposite when you downvote content - penalize those who upvoted and reward those you downvoted.
Then in such system you can rank content based on the amount of your attention token bets and the content that is ranked at the top naturally gets your attention.
This creates a feedback loop where you give your attention to those who have earned your attention in the past.
I am building such a system at linklonk.com and would appreciate any feedback.
Note on crackpots: you do not, in fact, need to be Scott Aaronson. In my experience, any cosmology PhD student who's put anything on arXiv will receive a good amount of "refutations" of all of modern physics!
I was waiting impatiently for the active inference punchline, which I think might be rather profound, insofar as you've just illustrated that markets are virtual organisms.
Thanks for the writeup, Scott! I've honestly been super thrilled to see all the different use cases our community has figured out for the prediction market mechanism; I think it really validates the approach of letting people create markets on anything they want!
One more example of a bounty/action market that worked pretty well: Elizabeth & I posted a $100 bounty for some data aggregation work for Slime Mold Time Mold's hypothesis of contamination-led obesity. It got two separate takers: https://manifold.markets/Austin/100-bounty-will-someone-combine-the
Would love to figure out how to better support this natively -- if anyone has suggestions, please reach out!
Not even kidding, I spend half my day joining annoying tables, if you have more bets or something like this, please link/post somewhere!
Your example isn't a proper action market; it looks like it's a prediction market on someone taking the traditional bounty (which maybe increased visibility of the bounty?). Is manifold all play money?
The way you'd set this up in a real / proper market is that you'd pick the bounty B you want to offer ($100 in this case), you'd guess a probability p that this bounty is sufficient to achieve the results you want (maybe you think it's 70% likely), and then you offer N = B / (1 - p) shares of "yes" at price p (in this case with my fake numbers, 100 / .3 rounds up to 334 shares @ $0.70). You don't offer any shares of "no". If you get no takers, you're getting information that your bounty is too low / your probability is too high, you can revoke the sell offer and try again at a different point.
Note that if your probability is too low relative to the true rate, the difference will be picked up by people who can predict better than you rather than by the bounty taker.
So in this example, if I want to "accept" the bounty, I buy all 334 shares for $233.80 and do the work, so the market resolves to "yes" and I get $334 for $100.20 profit, is that right?
It doesn't seem reasonable to expect people doing small tasks to pay a deposit.
Yes, that's right. One problem with action markets is that they require having cash on hand to invest.
One interesting feature is that the less likely the market thinks the current bounty is adequate, the lower the deposit you have to pay. So it seems to me like the deposit is what you pay to stake your claim as the person taking the bounty, rather than all the other people who were likely to want it.
What's unreasonable about having to pay the deposit?
Just because the sort of people who might agree to do small tasks for small sums typically don't have cash on hand. For example, Scott mentions claiming a bounty when he was a poor medical student.
I think the equilibrium state here is that the people setting up the bounty are calibrated on how much of a deposit people can pay, and this is factored into the probability that the specified bounty is adequate. (If the deposit is too high, then the bounty setter was wrong in their probability estimate, and next time will predict a lower probability with the same bounty, which lowers the deposit. Gradient descent will keep lowering the probability until the deposit is the highest that someone is willing to stake.)
This is an interesting nuance; thanks! I didn't realize that the opening of the bounty market had three different constraints included: the size of the bounty, the size of the deposit required to claim the bounty, and the probability that someone will claim the bounty.
This is also a striking example of how prediction-action markets, like almost all other forms of increased liquidity, benefit the already rich more than they benefit the poor.
(1) "What will I (Scott Alexander) rate as most promising when I do a deep dive into the research on pregnancy interventions?"
I'm going for the "Natural birth versus Caesarian"
(2) "They’re betting on whether someone will clean up the (currently disastrous) backyard. They’ve said pretty openly that they’re hoping someone will buy a lot of “yes”, take care of the backyard project, and then take all their money. "
Could they not just pool the money and hire a landscaper or groundskeeper? When my back garden was a disaster, after a long time dithering about "something needs to be done", we finally hired a guy to come cut the hedges, cut down and take away the tree that was growing in the wrong place, and do a ton of weeding and spraying. Result! If you're waiting for someone else to do it, it will never get done - "what's everybody's business is nobody's business".
People pool the money by buying "no". If nothing happens, they recover their money. If the pool is deep enough, someone will buy "yes", take the money, use it to hire a landscaper and keep some for his administrative troubles and responsibility.
I am a simple idiot, I don't understand why, if the entire household is agreed on "the back yard is a disgrace to God", at least one person doesn't go "Right, I'm Googling 'landscapers', we'll get a quote, and if it's anyway reasonable we hire the guy to come next Tuesday. Everybody throw $100 on the table for a deposit".
Cut out the middleman of waiting for somebody else to do it? If it's "keep some for his administrative troubles and responsibility", I will very happily spend ten minutes looking up "Bay Area landscapers and garden maintenance" in return for a bung:
https://www.yelp.com/search?cflt=gardeners&find_loc=San+Francisco+Bay+Area%2C+CA
Well if they want to keep the money inside the group instead of paying someone else it might be an effective method to find who in the group is willing to do it for the least amount.
Or if prediction markets like this became big enough that landscapers/contractors kept an eye on them for 'jobs' it might allow them to find the landscaper willing to do it for the least amount of money. It would almost be like a reverse bidding process, just keep adding to the pot until someone is willing to bet against it, do the work, and take the money
I think the truest answer to your question is "because they are the kind of people for whom solving a problem in a cute way with a prediction market is much more satisfying than solving it in a boring way like normies." But there may be other advantages as well.
I think I buy this. I suppose I intuitively think of prediction markets as being for (in Scott's parlance above) high-% prediction low-% action; it's just confusingly low-agency to me that they already have the money to pay for someone to clean up the backyard but instead choose to create a prediction market for whether someone will do so.
Yeah, that's what I don't understand here. There's a job needs to be done, there are tradesmen who do it, and they can put up enough money between them all to hire a guy and get it done. What is the gain from putting it off even longer by "let's play a game where someone *might* or *might not* do this job for us, which will cost us more as they will want a fee for doing it on top of whatever the tradesman charges"?
Then you'd probably be the one that takes the pot, and not me according to my broken toilet and a few other stuff I've been putting off and I'd definitely pay for third parties to take care of if I know I was getting a competent job and fair price without doing any research (also, no Yelp over here nor an equivalent with reliable user feedback and ratings)
Oh, I can procrastinate to beat the band, but I have a brother who is more energetic and he was "Right, we've talked about this long enough, time to get it done!"
For small jobs like getting the immersion cylinder replaced (well, relatively small but a pain in the neck to do yourself), it was a matter of asking around work mates and friends did they know a plumber they would recommend to do it. Word of mouth is often pretty good, but you takes your chances.
It's perhaps worth noting that you can just post things like this to one of various "find a tradesperson" type sites. It probably doesn't result in either the cheapest or the best job, but it does involve minimal faff. IME, things get picked up in a few minutes.
When someone buys "yes", they can't take the money until after they succeed. When they hire a landscaper (investing their own money), then the people living on the premises who bought "no" have an incentive to tell the landscaper "we didn't call you and we don't consent to you doing the work, leave or we'll call the cops", and get the money that was put in by whoever bought "yes".
Presumably in this case people living in the house bought No not to profit, but to fund the Yes cause.
Presumably the people living in the house don't have identical agendas or values, otherwise they would just have a consensus on what to do without this market. The process certainly must account for at least one of the people "defecting" and acting in their financial self interest.
I realise that this is just basically a fun game, something instead of betting on the horses or the football like the common people do. I cannot visualise this as any practical way of decision making.
More broadly, I'd probably phrase it as "the backyard isn't the point, the prediction markets are the point." One doesn't make Rube Goldberg machines for the sake of turning on lights (or feeding oneself cake: https://www.youtube.com/watch?v=auIlGqEyTm8), or even just for the fun of it (although it can be a ton of fun! look at that candle sliding down that ramp by heating up a stick of butter!), but also to learn about pulleys and levers and gears and whatnot, in hope of putting them to use someday in a machine that's actually useful.
As a way of cleaning up the backyard, a prediction market is confusing and probably inefficient. But as a way of learning about human behavior and the limits of prediction markets, what's more practical than making a market for a relatively unimportant, easily solvable problem and seeing what people do with it?
They also created a situation where a literal crazy homeless person might show up to clean their yard and take their betting money, and they may realize they've made a bad mistake and tell the guy to go away. This may in fact anger the guy, and create a possibly violent incident.
Or, much more likely, some guy shows up who's not very good at landscaping cleanup and does a poor job. Or some guy shows up who has no insurance and damages something or hurts himself.
If the backyard is bad enough, a crazy homeless guy might improve it some amount. Better to do something than nothing and let entropy keep on keeping on.
Though I suppose crazy homeless guy could lug in bags of rubbish and start tyre fires etc. in the back yard, so yeah.
Please stop, my head is exploding.
"We *could* take ths $1,000, look up somebody online and see what reviews they got, and hire them to come tidy up our backyard. Or we *could* take the money to deliberately lose it so that somebody else will look up a guy online, see what reviews he got from previous clients, and hire him to come tidy up our backyard, plus take a slice of the action as a middle-man, so it costs us more than if we did it ourselves.
Clearly option B is a better, faster, cheaper way to do things!"
Yea. There's also the free rider problem. But it is not strictly worse in every way than "just coordinate with everyone to pay up and then hire a cost-effective and reliable third party to do it for you".
There does seem to be a basic problem that I need to spend, say, $1,000 buying all the yes shares, then $600 hiring a landscaper, so that I can redeem the shares for, say, $2,000 and make $400 profit. But if I'm the sort of person who has $1,600 I have no particular use for, this is probably more work then I'm willing to do for $400, particularly as I take on the risk that the landscaper does an unsatisfactory job.
This is not actually a problem. If nobody is willing to deposit $1600 to make a $400 profit, then 2000 shares at $0.50 is much to high a probability estimate. This is not just a linear optimization problem: you are looking to set a bounty of $B such that there is a probability p that someone will be willing to deposit $B*p/(1-p) (+ extra fixed costs) to claim the bounty. But continuity + the intermediate value theorem guarantees that if the bounty is enough to solve the problem with no deposit required, then there is some solution (unique if the probability is monotone in the deposit).
Clearly I am too brainless to be a rationalist.
This seems to point out the failure in prediction markets; the lovely shiny ideal is that they will harness the wisdom of crowds to find the best answer to any particular problem.
But seeking for truth is always going to take a back seat to "I could make money screwing over the other people" and so prediction markets will simply be one more way of "how do I turn a profit on this" and not "what is the best answer".
Consider the scenario as set out there:
1. The backyard is a dire mess, we want to get it cleaned up
2. We could put our money together and hire a garden maintenance guy
3. No, let's do a prediction market instead! Surely that will be better value for money and more efficient!
4. Someone *thinks* "Hey, if I screw over everyone else in the house, I could make money here!"
5. Backyard is still a dire mess, and now the money for getting it cleaned up has been scooped by a profiteer, so we're even worse off than when we started
See, this is a rational analysis.
Ironically, it's self described Rationalists who hold an impossibly naïve belief that something that can be corrupted by the profit motive/conflicts of interest, somehow won't be.
They're good people who try to do right and treat everyone nicely.
Grumpy cynics like me who've been burned once or twice naturally see the flaws in proposals that depend on "assume human nature is basically nice".
I know they're decent people with good intentions. It's just easy to bristle at the chutzpa of people who make such erroneous assumptions about human behavior and then deign to call themselves Rationalists.
Its one thing to make an argument for "this is how a group of people should behave if they want to optimize for problem solving", like a kind of moral plea, but when it's "this is actually a solution that will work for actual people", my eyebrows fly off my forehead.
5 is wrong, and 4 is probably wrong as well. So long as the people wanting the house to get cleaned up only ever sell shares of "it's clean", either their money is returned to them or the backyard gets cleaned. If they also buy shares of "it's dirty", then they are buying insurance in the case where the backyard remains dirty, and in fact make money in that case. It's only when the people who want the backyard to get cleaned sell shares of "it's dirty" (or buy shares of "it's clean") that they risk loosing their money to profiteers.
Is everybody in this conversation aware that Manifold is a play-money site, so there is no actual money at stake in this market? This whole thread reads like none of the participants are aware of that.
Is the literature review of mineral supplementation public?
Also interested in this
I found it!
https://web.archive.org/web/20120228040434/http://www.medicineispersonal.com/contest/1st
I'm not sure how the action market helps with the doctor situation. Isn't creating an action market for it just worse than the solution of asking a doctor to help, paying him if it does and demanding that he pay you if it does not? Your complaint with that system seemed to be that a risk-adverse doctor who was less than 100% confident might pass, but it seems like the action market would have the same issue here.
I was also unclear on the purported advantage of the market in this scenario.
Yeah, action markets are useful when either (a) you want to fractionalize the bounty resolution and offload the impact evaluation to the market, or (b) you want to know how likely it is that the thing would have happened anyway and want the fraction of the bounty that is not required to incentivize action to go to people who are correcting your inaccurate prediction. You unfortunately can't really decentralize surgery, I think.
With respect to the Scott Aaronson thing, my understanding is that the American Math Association has come up with a solution. Or maybe some other math organization; it was explained to me a couple years ago, so one or two details may be off.
Most people who submit proofs for the Riemmann Hypothesis (or any other similarly famous problem) are are not so correct that a PhD student can't take apart their argument. So, the AMA charges about $100 per page of the proof, and then gives it to a PhD student to dissect. The student writes up a clear report on why the proof doesn't work, which is then returned to the author, and the student gets some cut of the money to use as travel funds.
Now, this may not still be totally accurate, or perhaps the person once telling me it was not accurate in their description, but it still feels like a good system.
Most of them are so incorrect it doesnt make sense to write anything but "this is gibberish"
Two markets this week:
An action market - if I keep dumping money on it, hopefully somebody can make it happen: https://manifold.markets/AlexPower/will-thestalwart-have-a-verified-tw
And a "no-action" market, more specifically a prediction market where it seems nobody knows the answer: https://manifold.markets/AlexPower/will-the-country-of-turkey-be-recog
Why anyone would use market that will be resolved by a known fraudster?
You are accused of fraudulent resolution in comments under those links.
Listen, buddy, I've been accused of a lot of things in my life, and I even did some of them. On the other hand:
1) That's just a troll.
2) Even if it were certain that I was going to defraud the market, it still doesn't explain why apparently nobody knows how to request that a Twitter account be verified.
3) Even if it were certain that I was going to defraud the market, the various people interested in whether Turkey will change its spelling could still get useful information from making comments.
In https://manifold.markets/AlexPower/letter-frequency-in-the-march-25th you explicitly commented
> I did so bad with the hint, I think I may just steal all the money. I will think it over for an hour first.
then almost immediately after you resolved market assigning 90% to correct answer and 10% to false, which you bought and caused you to earn Manifold points from fraudulent resolution
> Even if it were certain that I was going to defraud the market, the various people interested in whether Turkey will change its spelling could still get useful information from making comments.
In such case people would likely make a separate market from than not suffering from this problem
OK, I'll bite. https://manifold.markets/AlexPower/will-the-phrase-resolution-risk-app
If you call "fraud" a "resolution risk" it still remains a fraud.
(in this case it is both "fraud" and "resolution risk")
I didn't you see commenting in those links that the accuser was a troll. That would seem to be the most relevant place to do so, because otherwise it looks like an accusation you didn't deny.
> Niels Bohr supposedly said that “prediction is very difficult, especially about the future”.
I would have guessed that was a Yogi Berra quote:
• "When you come to a fork in the road, take it."
• On economics: "A nickel ain't worth a dime anymore."
• "Always go to other people's funerals. Otherwise, they won't come to yours."
• And last but certainly not least: "It's deja vu all over again."
He also gets a fair amount of stuff wrongly attributed to him. As the man himself put it, "I really didn't say half the things I said."
My personal favorite:
Friend: "Hey Yogi, what time is it?"
Yogi: "You mean now?"
He ostensibly did say it, so I'd guess Scott is joking.
https://www.goodreads.com/quotes/261863-it-s-tough-to-make-predictions-especially-about-the-future
Oh, okay. It sounds very much like Yogi. I had done a web search and saw a reference to Bohr too. Probably just missed the joke.
Here’s the Bohr reference:
https://blogs.cranfield.ac.uk/cbp/forecasting-prediction-is-very-difficult-especially-if-its-about-the-future/
Ha! It's pretty marvelous that a quote would be be attributed to both Bohr and Berra; I hope they'd both be amused :)
Clearly we need a prediction market to resolve this...
>I once joked that instead of lower courts, we should have prediction markets on what the Supreme Court would think of a given case. Same idea, higher stakes.
Related idea I've had whenever there's a SCOTUS vacancy: a major qualification for appellate court judges to be elevated should be the extent to which they have exercised their delegated authority faithfully, which is to say, when cases they've decided subsequently make it before the highest court, do the Justices agree with them? 5-4 splits would have the least weight either way, 9-0 agreeing/disagreeing with the judge would have the greatest favorable/unfavorable weight; more emphasis given to cases in which the judge was in the dissent at the Circuit.
Robin Hanson has also proposed gambling over court cases:
https://www.overcomingbias.com/2007/10/double-or-nothi.html
Isn't the main difference for incentivizing actions via the stock market vs. prediction markets is that stock markets are highly regulated and large transactions such as placing a major short on Tesla the day before Elon Musk is assassinated would be easily tracked by the SEC/law enforcement. The companies running prediction markets could presumably track these types of things and inform law enforcement about them, but would they necessarily have the infrastructure/processes in place to do so? And would they all voluntarily do that? Surely some privacy libertarian style prediction market would advertise how it doesn't track any user data or supply it to the govt. And what about blockchain prediction markets using anonymous untraceable cryptocurrencies like Monero?
Most other major markets where you could make large sums of money by taking a huge position and purposely affecting change in that market has some form of regulatory overview that presumably helps deter these kinds of things.
Indeed there has been at least one assassination market set up on Tor, using Bitcoin for payment: https://en.m.wikipedia.org/wiki/Assassination_market
Jim Bell outlined how to design such a market in 1997 here (notice how "digital cash" is in quotes because bitcoin hadn't been invented yet, but was crucial to the design): https://web.archive.org/web/20041209151654/http://jya.com/ap.htm
Iirc the design, many people can contribute to a pool for a given person to be killed, and to win the pool you have to guess the exact day and put down some fairly large amount of money. This encourages the pool to grow, and the down payment helps make it unlikely the assassin will need to share the winnings.
For some entertaining reading about a strange crypto-anarchist, check out Jim Bell's Wikipedia page and the links to primary sources. I first encountered his writing on a cyperpunk mailing list in the late ⁰⁰'s, I think while he was momentarily between prison sentences.
I know it was said as a joke but just to clarify, lower courts are not intended to have the same decisions as appellate courts.
I don't know much about the court system. Can you expand on why this is the case?
IANLAL (I Am No Longer a Lawyer), but I think Daniel's claim is maybe a little too sweeping, but I suspect he means either:
The result of a case can turn on a number of different questions of fact and law and it is typically only questions of law that are subject to appeal. So the higher court is answering a different, narrower subset of the question. But you'd still like the lower court to have come to the same conclusion on the specific legal questions.
And/or
Lower courts are more bound by more precedents than higher courts. So it might consider itself required to reach a conclusion other than the one it would reach from first principles because of an existing decision of another equal or higher court, while a high court might feel free to overrule that decision. This is all true, but a relatively rare species of overruling, AFAIK.
Disclaimer: I am an English lawyer. I imagine that the system is similar in the US, but I'm not sure.
Most of the work done by courts of first instance is determining the facts of cases, for which purpose they hear evidence from witnesses who are cross-examined. Appellate courts do not hear evidence and take the facts as found by the first instance court. This is why Supreme Court hearings are often much shorter than the first instance hearings to which they relate. Somebody needs to make findings of fact in order for the Supreme Court to make a decision, so you can't short-cut the process in the way proposed.
This is all true, Robert, but it's still my understanding that the lower court will also address the specific questions of law which are ultimately appealed in their decision, and optimally their decision ought to match that reached by the appellate court, unless there are issues of differential levels of deference to authority.
Yes, that's right.
Note that, while it's obviously not for money, fantasyscotus.net is quite real and could quite easily be monetized in a different regulatory environment.
For many of these applications, I think one of the biggest challenges for prediction markets will be to beat the crude approximations to prediction markets that we already self-organize on an ad-hoc basis. To take an example from this post where I have some firsthand knowledge, I indeed don't read most of the P=NP proofs, refutations of quantum computing, and other revolutionary developments that cross my field of attention -- but that can be overridden by colleagues I trust staking some of their reputation that, hey, this is interesting, this is worth looking at, etc. Which means that a prediction market wouldn't just have to beat nothing whatsoever, it would have to beat THAT.
In the meantime, though, if it's really important to optimize my use of time, then there's surely lower-hanging fruit to be had, like me not spending hours per day laying in bed with my phone scrolling through social media like now... :-)
Speaking of weird papers, I read an interesting peer reviewed paper once purporting to show that quantum mechanics is a consequence of special relativity. (https://doi.org/10.1088/1367-2630/ab76f7 ) I know "why quantum mechanics" is an interest of yours, so it might not actually be a waste of your time to look at it. ;)
The super short version of the argument:
1) Assume that special relativity is correct.
2).Assume that, in particular, special relativity still makes sense when you allow reference frames to move faster than light with respect to each other.
3) In order for superluminal reference frames to make sense, particles need to act in certain ways consistent with QM but not classical mechanics.
I know enough modern physics to understand what the paper is talking about, but not nearly enough to know if it's clever, trivial, wrong, or not even wrong.
Well, Monday is the day for prediction markets. Can you think of a way to use prediction markets to limit social media use?
Can I ever! https://manifold.markets/NcyRocks/will-the-commitments-community-have
The phrase "the market is wrong" makes no sense to me.
Consider a market created by punters on a horse race. If the longshot going off at 40-1 odds wins, it does not mean the market was wrong. That horse *should* win about 1/40 races. OTOH, if the longshot wins 40 races in a row, that market probably isn't functioning. More important to note, though, is that if the favorite wins 40 races in a row, that also means the market isn't functioning.
If you want to know how well a market is functioning, give the market a Brier score across at least dozens of predictions. The EMH is true by tautology if you have a true market.
"OTOH, if the longshot wins 40 races in a row, that market probably isn't functioning."
Or you should check to see if the horse is a ringer 😀
This sentence has a minor error where the order of 2 words is reversed, but it happens to be an error that tricked my internal parser into going down a garden path that actually took me a nontrivial amount of time to back out of:
"I throw these out all unread the time when I get them in the mail"
I considered doing the fix Valinor's backyard myself thing, but (a) it was unclear to me that they were actually angling for that and I thought it might be intrusive if I just showed up and started touching their stuff, and (b) if I'm going to do manual labor I want to be paid in real actual money and not Monopoly money ¯\_(ツ)_/¯
Cytomegalovirus should definitely feature in the biodeterminist's guide to parenting. I don't see it anywhere in that list.
After reading your blog, this is hard agree, and as a future parent I'd like you to know that thanks to your blog we both tested :)
On why we don't have more geniuses: "His summer following his father around learning medicine was probably good for him, but not outside the bounds of what still happens today (I followed my father around learning medicine)."
On whether small prizes are great for thoughtful, complex: "I put in 20 hours of research and won. If I’m typical, this is proof of concept that [...]"
This is maybe not the most effective argument that parental tutoring is less than very effective, or that medical students can be cheaply incentivized to do good lit reviews.
Don't wait too long to do your "deep dive" on pregnancy interventions, so that my wife and I can benefit from it in the (somewhat) near future! :)
Same! I'm excited to read it so I can add to the list of things I nag my wife about! She can't stand the word "pthalates" anymore.
Thirded - I am about to start making children, and need that guy stat! I did read the old one and adjusted accordingly, but I assume new facts will come to light since several years ago that the old one got made.
> People don’t assassinate Elon Musk because then they’d be investigated for murder.
Well that's an interesting quote.
In regard to the historical prediction market, I wonder how you could use them to judge the consensus on past events. There are many things that happened in the past that are still unclear, with papers being published explaining that this and this happened and two months later a counter-paper that says it totally did not. It is often hard for non-experts to
1/ judge what is the most plausible explanation of what happened,
2/ judge how strong the consensus is.
I am starting this Manifold market with the first disputed event that came to my mind: did Joseph Stalin died of natural cause? (https://manifold.markets/TaoSumer/did-stalin-die-of-natural-cause)
It will be probably be too obscure, too niche, too complex and too irrelevant for it to gain traction (let me know if you got better ideas for a disputed event, which should be easy to find). I thought of using the same technique in my previous experimental market, as in "Convince me that XXX is true and I'll resolve the market this way" (https://manifold.markets/TaoSumer/are-there-more-doors-than-wheels-in), but I feel like such a world-changing event is too complex to ask people to debate constructively in the comments. I guess the best way to make it interesting would be to invite historians who wrote papers on Stalin deaths to bet money and debate/argue, and then get the crowd to decide who got the best arguments - tho rethoric would probably come in the way, with people voting for the best speaker/writer.
Do you have any ideas on how we can make such markets better? I can implement them for the Stalin market - or create a more interesting one.
> let me know if you got better ideas for a disputed event
I love this comment! I guess a market on "did Jesus physically resurrect from the dead?" would be a prime example. The question has a lot of significance (historically, and still today for millions of people), has been debated at length, yet each side has people who fervently argue for it.
I would be very skeptical though that this marked could be resolved... much more so than for the Stalin case. Maybe one day when we get wormholes and space/time loops ;-)
Even just a "did Jesus exist as a historical figure" would make for an interesting market.
I thought that question was rather settled in the positive?
https://en.wikipedia.org/wiki/Historical_Jesus claims that "Virtually all scholars of antiquity accept that Jesus was a historical figure, and attempts to deny his historicity have been consistently rejected by the scholarly consensus as a fringe theory."
I agree that it would nevertheless make for an interesting market, and would definitely be easier to resolve than the resurrection.
https://rationalwiki.org/wiki/Evidence_for_the_historical_existence_of_Jesus_Christ
See here for counterarguments ;)
> Do you have any ideas on how we can make such markets better?
Idea: reverse the role of market-maker and participants. Participants are *only* allowed to sell shares, and the automated market-maker (or the creator of the market) only buys shares. Perhaps implement by mandating that a given user can have open sell orders only on one side of the market at a time, and submitting a buy order on one outcome requires submitting a buy order on the other outcome simultaneously?
Theory:
The "obvious" prediction market resolution condition for, e.g., "did Joseph Stalin die of natural causes?" is something like "does the market reach $0.95 for yes within a month?". But by the prediction-market-analogue of Lob's theorem, this is equivalent to a market betting on the question "will this market reach $0.95 for yes within a month?" without any reference to Stalin. The problem, I think, is that it's so much cheaper to shift the market than to make an accurate prediction of the outcome of the Keynesian beauty contest that the market just becomes an action market and the market maker is incentivizing shifting the equilibrium in whichever direction is easier to shift, so you don't actually get information out.
Alternate analysis: Buying "yes" shifts the probability of "yes" higher, so all you have to do to win is flood the market with enough money that the price shifts in your favor.
But if you flip the roles of participant and market maker, then the market-maker is actually disincentivizing action that shifts the market.
What's the incentive, here?
In this scenario, you only make money when you're wrong. When you're right, you have to pay out money (in proportion to how uncertain you were that you'd be right). So your incentive (at least if you're risk-averse) is to find the markets where you're most confident in your prediction (high probability), but also where you are the least meta-confident in your confidence. (Huh, this is interesting. It rhymes with the thing that I think brains do in predictive processing, which is attempt to maximize the amount of surprise they can eliminate/explain/understand/resolve.)
This is also cute: in normal prediction markets, you make money by being *more right* (buying the correct outcome, with more accurate predictions than the market), while in these reversed prediction markets, you make money by being *less wrong* (selling the incorrect outcome, but at a price which indicates that you think it's less likely to happen than the market does).
Coming back to the point, as long as there are no secondary markets on these "reversed prediction markets", I think there's no way to make money by influencing the outcome within the market. (Someone please do a better analysis than me, though!)
> let me know if you got better ideas for a disputed event
The US government declassifies some of its classified material on a certain schedule. I recommend looking up e.g. the FBI's automatic declassification schedule, then making a market set to resolve after a relevant FBI document release occurs.
By the way, I made a prediction in your Stalin market. Let me know whether the comment I attached to the prediction takes a reasonable approach.
>> Will Valinor unfuck its backyard eating area by April 1st.
Man I thought Jeopardy rules applied. Or it's a trick on two levels.
Oh man, you wrote the blog post I wanted to write! (But you got some things wrong.)
> But it’s strictly worse than a regular bounty system.
Nope. In a normal bounty system, you set a bounty and if anyone is willing to do it for that price, they do the job and get the bounty. If they were willing to do the job for less than the bounty you set, they get the excess as profit.
Action markets have one advantage and one difference (which might be an advantage).
The advantage is that they allow decentralized claiming of the bounty:
Someone can buy up all the "yes" shares, clean half the backyard, then offer to sell them at the the halfway mark between the price they bought and 1, and thereby claim half the bounty. Moreover, in a sufficiently liquid market, the price that they get for their shares will reflect the market consensus on how much their action contributed to the goal (so you'd get less than half the bounty for cleaning the easier half).
The difference is that in normal bounties, the excess bounty (over what people are willing to do the job for) is paid to the person taking the bounty. In action markets, that excess is paid to people who are making actuate predictions and giving information to the market. I actually think this is pretty cool (but maybe you'd prefer to tip the people who perform your bounty rather than paying that money out to the people who are teaching you that you're wrong about how much it costs to get something to happen in the world?)
> For one thing, if I suspect someone else will take care of the backyard, then just by registering that prediction I can share the winnings with them, even though I didn’t do any work
You can, but only if you believe that sharing winnings with them will not push the bounty below what they'd be willing to take. And if you do that, you're providing information to the market.
For example, say there's 1000 shares of yes that can be bought for $0.70. This means there's a $300 bounty on cleaning the backyard. If you think someone is about to clean the backyard and try to preempt then by buying half the shares, you decrease the bounty to $150. If that's still enough for them to clean the backyard, that means the bounty-setters were miscalibrated on the bounty required to clean the backyard, and you're claiming the excess. If it's not enough, then you're betting incorrectly, and the price should not rise.
> (since this is at 70% (rather than 100%), someone must be doing something like this).
This is not true. If the probability is at 100% there's no bounty to be had. (Maybe you meant "rather than 0%"?). Additionally, *you can't decouple the action from the prediction*! You're supposed to set up an action market at a price $p and volume N shares such that you believe $(1-p)N is the minimum bounty required to make the action happen with probability p. If you differ from this constraint at all you will lose money (in a sufficiently liquid action market) to people who are better than you at predicting.
> For another thing, it incentivizes people to bet “no” and then sabotage cleanup projects.
Only if you set up your market wrong. If you're trying to make "yes" happen, you're not supposed to offer any shares of "no".
Importantly, none of the money that you are bountying towards "yes" can be extracted by the people who are trying to sabotage you. If the the yard gets cleaned, you pay out the bounty to the bounty-taker and the people who predicted better than you. If the yard doesn't get cleaned, you get compensated.
So insofar as there is an incentive to "bet no" and sabotage the yard cleaning, it must come from something other than your bounty (maybe it comes from the people playing the prediction game on your bounty market? maybe it comes from making the outcome and the current prediction more legible?)
> Banned because Wall Street developed a financial instrument that let them short ConTrackeds, then tried really hard to prevent bridges from being built
Ha, but this isn't an argument against action markets, it's an argument against inadequate equilibria. Why couldn't Wall Street come up with a financial instrument that lets them short bridge building directly and then go interfere in bridge building? If the shorting method involved literal borrowing of ConTrackeds, then this is an argument for the stupidity of the lenders, who should have realized that lending this token incentivizes the lendee to sabotage the project. At worst, you're not supposed to ban the action market, you're supposed to ban the derivative markets that incentivize antisocial behavior.
> Commenters brought up that you get weird incentives as soon as more than one person owns ConTrackeds
Link? The incentives, as I understand it, are:
1. If you bought the token to claim the bounty, you have an incentive to make partial progress on building the bridge, and to sell as soon as the market adjusts to account for the partial progress you believe you will make.
2. If you bought the token because you thought the market was wrong and you know better, you have an incentive to sell as soon as either the market adjusts or you learn otherwise.
3. You are incentivized to buy only if you either think you know better than the market or you are planning to make progress on the bridge building.
> I think you could solve this problem by subsidizing a prediction market at 25% chance of recovery, and letting doctors bet on other odds.
Yes, you can. This is what my formula above says in general.
> This would still have some problems - if someone saw a famous doctor betting, they could bet too, and “steal” some of the “winnings” from the doctor - but it seems to be kind of the right idea.
No, this means that either you were dumb and set a higher bounty than necessary (the doctor still buys enough shares to make it worth it to treat you) or that these people are dumb and will lose money to the market (there isn't enough bounty left to incentivize the doctor).
> Different prediction markets “leak” into being action markets at a different rate: one about when a distant star will supernova is 100% prediction; one about when a President will die is 99% prediction; one about when the forecasters’ backyard will be cleaned is maybe 80% action.
Important question: can you read off the leakage rate for a given market from market data? I think solving this problem is required for solving AI interpretability, based on the connection you mentioned.
Important follow-up: If you can't read it from market data but you can read it from brains / predictive processing / action-inference, what does this say about the design of brains, and can we port this design back over into the marketplace to make better prediction-action markets?
And finally, I want to note that this issue is already present in existing AI and prediction markets. The book Surveillance Capitalism talks about "behavioral futures markets" that Google and all the other big Internet companies run (I'm not sure how much this is a literal thing and how much it's a model), which are used for advertising. The book also talks about how they want to control our actions to make them predictable. (I think this book is secretly about misaligned super-human AI that is actually already here, and how we really better get our act together and align it before it's too late (unless it already is), except the author doesn't quite realize this. (I want to write a book review on it, or collaborate with someone on writing a book review on it.))
Final thoughts: there are five directions of incentives that action markets will incentivize:
1. They incentive better predictions. This is the standard incentive from prediction markets.
2. When there's much more money on selling "yes" (/ buying "no"), they incentivize "yes"
3. When there's much more money on selling "no" (/ buying "yes"), they incentivize "no".
4. When there is the same amount of money on both sides ("yes" and "no"), they incentivize whichever direction is easier to influence, i.e., they incentivize making behavior more predictable. This is the issue talked about in Surveillance Capitalism.
5. When the amount of money on each side, after being adjusted for the marginal cost of influencing the outcome in that direction, is the same, I *think* the market disincentivizes any action that shifts the likelihood away from the current prediction.
So beware that playing market maker and adding liquidity to prediction-action markets is not morally neutral. You are incentivizing whichever outcome is easier to cause. Since destruction is generally easier than creation, this says that we should tax market makers of action markets at the rate of destruction their market-making is liable to induce. (This is a great use case of being able to read the leakage rate off of market data! If we can read off the leakage rate, then we can accurately tax market makers!)
While I'm here, I want to share an interesting tangent about how action markets are cool that didn't make it into the above comment.
I want to investigate an interesting incentive / financial product around ConTrackeds. If you buy ConTrackeds intending to claim the bounty, and then a natural disaster hits making bridge building much more expensive, you're out a lot of money. The solution to this, I think, is *more prediction-action markets*. If you set up a market in "I will invest at least $X into building this bridge", you should be able to derisk your ConTrackeds by transferring your investment to this market. Namely, you buy a bunch of ConTrackeds, declare your intent to build the bridge, and then offer to buy up shares of "yes" in this market (equivalently, sell shares of "no") using the proceeds of selling off the ConTrackeds. You should pick actions that are sufficiently hard to interfere with yet sufficiently correlated with actually building the bridge that the bounty you're placing on sabotage isn't actually significant. For example, "I will place an order for this amount of materials with this provider at this time, and earmark them for the bridge building project, provided the price is at most this amount" is better than "I will have this amount of materials by this date" because it's much harder to sabotage "placing an order conditional on price" than "receiving materials".
If there's enough liquidity in the market and you set it up right, I think you can derisk pretty quickly. I think this is pretty cool!
(It does show that we're going to need something like contract lawyers and bug finders to get action market incentives set up right.)
> Or by going long on train companies, and bombing a plane?
I assume you mean train, not plane here.
(I don't know anything about finance, but) maybe he meant that bombing a plane might temporarily reduce public confidence in planes for mid-distance travel so demand for train travel as an alternative increases, boosting train company stocks? Or something like that?
Ah, indeed, I misread "long" as "short".
“The agreed-on solution for this is lots of conditional prediction markets.” - a downside here is that each individual market ends up being lower confidence, as bettors have to split up their time across many questions. But if you think that there are diminishing returns to time spent on a question, this is probably am overall win?
For the pregnancy market, you are assuming women are already taking prenatal vitamins, right? (Otherwise folic acid or iodine or something will be best).
I think those are the low-hanging fruit, certainly if you are planning to get pregnant then every resource including your doctor tells you to supplement with folic acid, stop smoking, cut back on drinking, eat healthily, get enough sleep, etc.
That's why I'm punting on the "natural birth versus Caesarian". Caesarians are a procedure that are becoming increasingly common and there's a lot of pro- and anti- debate around them. There's the "too posh to push" bit, where the claim is that when it is not medically necessary, women are asking for Caesarians in order to cut down time in labour. There's the "vaginal delivery puts stress on the newborn, including physical risks" claim, where Caesarians are said to be better for both mother and baby.
Like any medical procedure it has risks, including if you're intending to become pregnant again:
https://www.hse.ie/eng/services/publications/clinical-strategy-and-programmes/delivery-after-previous-caesarean-section.pdf
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3765853/
There should be enough data now, since it is becoming more common, to see what the effects (if any) on newborns of natural versus surgical delivery are, as well as fertility effects (see paper linked above).
Sure, there are papers like this that show caesarian-born children are at higher risk of asthma, obesity, and respiratory tract infections: https://www.mdpi.com/1660-4601/17/21/8031
However, I don't know of any randomized controlled trials, and the confounding factors seem pretty significant here (at-risk births are more likely to end up as c-sections). I don't even think the above link restricted the analysis to elective c-sections (it's not mentioned in the abstract; I didn't read the paper).
> I don’t think there’s any site as convenient as Mantic
Should be Manifold I think
> when I was a poor medical student, Zvi Mowshowitz’s company ran a contest for the best literature review of mineral supplementation with a $5000 prize. I put in 20 hours of research and won. If I’m typical, this is proof of concept that big enough bounties can incentivize poor medical students to do good lit reviews
Is there a website which aggregates links to these sorts of contests? (It could attract entrants who are indifferent to anything beyond the semblance of accuracy, so I wouldn't be surprised if biomedical companies wouldn't actually use it.)
Had you expected a large number of entrants, would you still have entered the contest? If so, you were probably motivated by more than just the money.
Start-up idea: a repository for unusual bounties of all-kind / a prediction market platform adapted for conditional markets (for example with the ability to visualize conditional markets in a tree-like structure of some kind)
About the idea with organising market to decide which papers should be sent to e.g. Scott Aaronson: isn't it just a non-expert version of peer-review process?
One quibble: do you think the comment you made about the pregnancy interventions *won't* interfere with the betting?
Perhaps it will have the same effect as a company announcing quarterly results and the stock market adjusting accordingly?
What's your point? Not as a snarky retort, I just genuinely don't understand.
The question about the dead drummer is not a question about past events it is a question about a future event -- i.e. the contents of the autopsy report. If the report is not made public, it may actually be about what the ME says about the report.
Yeah, the natural assumption is that drugs were involved, but it could be due to a completely unrelated health problem. We won't know for sure until the report.
There are a lot more points of failure and assumptions than that:
1. The autopsy could fail for reasons beyond anyone's control. E.g. the body may have been contaminated or too decayed, or the cause of death may be beyond the ability of curreent technology to detect it.
2. The pathologist could be incompetent or dishonest.
3. The report could be lost, garbled, or miscommunicated to the coroner.
4. The coroner may alter the report or hold it back from publication to save the family embarrassment. Ever heard of auto-erotic strangulation?
5. The press may misunderstand, misreport, or lie about the corner's statement.
We assume that the autopsy will produce reliable results which will be correctly and competently reported to public officials who will accurately disclose their contents to the communications media. who will report on them accurately.
In the ordinary course these assumptions are true, but they are assumptions and none of this has happened yet. https://www.youtube.com/watch?v=zhyu4Rh_w5Q
in 2017 someone bombed bus of top German soccer team, having “allegedly bought options to short sell 15,000 shares of Borussia Dortmund stock for 78,000 euros [£65,000]".
https://www.theweek.co.uk/83549/borussia-dortmund-attack-stock-market-trader-arrested
Wow, that was a *serious* Schalke fan! 😁
(I shouldn't laugh, that was potentially very bad outcome)
It seems like some of the moral hazard issues with prediction markets would be well addressed by limiting the players and the potential stake of any players. Even if we assume that someone might be willing to do antisocial things for money, the effort involved should at least make low-stakes bidding possible. ... until you get multiple, redundant markets, that is.
Can market fidelity be a thing?
I think the bigger issues are:
1. Market-making in a prediction-action market is not morally neutral. It incentivizes whichever outcome is easier to make happen.
2. As far as I can tell, no one yet knows how to determine the point at which prediction markets leak into action markets. (Unlike what Scott implies, it's not a fixed rate, but a function of how much money is flowing through the market in each direction.)
2. As far as I can tell, people don't currently understand the moral implications of the four ways of participating in a prediction-action market. If cleaning the backyard is prosocial, then selling "clean" and buying "dirty" are both prosocial (the former increases the bounty on "clean" and the later decreases the bounty on "dirty"), buying "clean" is conditionally antisocial (you are decreasing the bounty on "clean", but if you are planning to clean it yourself that's fine and if you think the bounty is higher than it needs to be than you are claiming the part of the bounty that's actually for getting information about what the minimum bounty is), and selling "dirty" is actively antisocial (you are increasing the bounty on "dirty")
What is market fidelity?
The 'starting a prediction market for whether the yard will be cleaned up' strikes me as the (far less dysfunctional) rationalist analogue of this: https://www.theonion.com/marxists-apartment-a-microcosm-of-why-marxism-doesnt-wo-1819566655
How about an attention market where the currency is attention of every participant?
Every person upvotes content that they think deserves your attention and downvotes content that will waste your attention. They have some starting balance of your attention tokens and with every upvote/downvote they place a bet. If you upvote some content - you reward everyone who upvoted it by giving them 100 tokens of your attention (divided equally). And you penalize those who downvoted it by taking away some percentage of your attention tokens they hold.
You do the opposite when you downvote content - penalize those who upvoted and reward those you downvoted.
Then in such system you can rank content based on the amount of your attention token bets and the content that is ranked at the top naturally gets your attention.
This creates a feedback loop where you give your attention to those who have earned your attention in the past.
I am building such a system at linklonk.com and would appreciate any feedback.
Note on crackpots: you do not, in fact, need to be Scott Aaronson. In my experience, any cosmology PhD student who's put anything on arXiv will receive a good amount of "refutations" of all of modern physics!