We should be able to see if people really shift all the way to the end if we could see the changes to that graph over time, is that possible in metaculus?
Re. Futuur California fires market – the resolution source says there were 9,639 fires last year, but their description says that "In 2020 there were *8112 fires*, in comparison to there were 5687 cases in 2019." So if you go off the linked statistics it's an obvious 'No', but if you go off their description it's an obvious 'Yes'. I think I'll sit this one out...
It's quite possible I don't fully understand the Futuur betting market, but it seems to me that as regards the people-on-Mars bet, there are plenty of people correcting this mispricing. Out of 6000 transactions on the market, most of them are saying "Nope, no chance", which is correcting the mispricing as much as they feel they can afford (as Scott says, three years is a long time to wait for 17%)
But there are also, predictably, people uncorrecting the corrections - I looked through the bets and there's one Musk fanboy who has wagered 4000 ooms on humans walking on Mars in 2024.
And the weirder thing about this is that it wouldn't change much if it was real money rather than ooms. The 17% would drift down to some Muskman constant and the serious money would stay away because the returns - though as good as guaranteed - are too small and too distant.
When it becomes clear the answer is 0, does the price change and allow one to sell off?
There is a launch window in Q3 of 2022, about 9 months away. (After that, the next launch window is Q4 of 2024, which doesn't allow boots to hit the Martian dirt in calendar year 2024.)
If it's priced like a stock market, there will be no buyers left, the price will go to $0, and everyone can close their positions.
if the "Mantic" model became legal, then anyone with a big enough following and the requisite trust factor could be a William Hill style casino. A 1% "vig" might be really bad for many gambling businesses.
If you think about it, the current 10% vig charged by gambling books are gigantic in this era of communication.
Haha that's a good point! We're not actively looking to disrupt online casinos or sports betting, and are focusing on more niche areas (core question: "what doesn't have a market yet?"). But if it happens by accident, well, I wouldn't complain :)
Totally agreed that their 10% cut (also on e.g. PredictIt) seems high, and we should be able to do better!
I am intrigued by all of Balaji's ideas about how to make speech more free and meaningful. I just wish I understood the terminology better. I too would like some help understanding this Oxford debate thing better. And if anyone can, please also explain his ideas on pseudonymous accounts simply, as well :).
If I am not mistaken, the Soho debates hosted by Reason are hosted using Oxford scoring, so if you want some examples you could play the podcast. Basically:
Relatively long theses for and against, followed by rebuttal, followed by questions. The audience votes both before and after the debate, and the winner is the one that moves the vote in their direction.
I like futuur. I already won my 2 first bets and made some bets that will resolve in about 3 years.
Using virtual ooms only at the moment. I am not touching crypto money out of principle. It is a Ponzi scheme that we should not encourage. If they had option to use better money (US$, euro, pound), I would participate.
Noted, thanks! We do offer USDC, which is pegged to the dollar. Would that allay your concerns? We are considering adding fiat options, but in the gaming space (we operate under a gaming license from Curaçao) fees are very high, and there's a good deal of operational friction, so it might take us a while to roll out fiat options.
Cool to see a futuur person in the comments! I have a few general questions about predictions markets that I'm going to post here in the hopes of getting some answers:
1) Have you analyzed how taxes/fees/opportunity costs from not investing elsewhere impose error bars on your prediction markets? Have you thought about using this info to present error bars around the predictions on your site?
2) less relevant to your project, but whenever I talk about using prediction markets to make decisions, the first criticism I hear goes something like "what if deep-pocketed special interests stood to gain a lot of money from a particular outcome"? I assume economists have thought a lot about this kind of scenario and how to model it. Can you suggest how a non-economist (with decent math literacy) would find an entry-point into that literature?
Interesting questions! Not sure if I fully understood all of them, but will have a go at answering...
1) taxes and fees can and do impact the accuracy of a market, in the sense that if the real probability of an event occurring is say 50%, but the cost of a share is for instance .55, there is a delta of 5 points in which a rational trader will not make a bet, and the probability forecast will not be updated. We intend to address this to the extent possible by offering lower fees, especially as liquidity for a market expands. Regarding opportunity costs, I assume you mean the opportunity to trade other markets? I don't see an easy way to model this but interested to hear ideas. Likewise, not sure how we would present error bars on our markets, but it sounds intriguing.
2) the idea of market manipulation gets raised a lot, and I think there are valid concerns. But: either a) you are trying to make money with your bets, in which case you will be rewarded for accuracy and help to generate better forecasts for everyone, or b) there is some external situation which you hope to influence by betting against your purely financial interests (ie subsidizing wrong forecasts because of the impact it might have). For instance, you could hope to influence media coverage by increasing the probability forecast, and thus influencing the thing being forecast via the forecast itself. The same concern is raised around "assassination markets" and the like. It's an interesting conundrum, but from a practical perspective the amount you would need to invest in a liquid prediction market to influence the outcome would probably need to be much higher than other methods of influence (for instance, hiring a PR team).
1) Re: opportunity costs, you interpreted my question correctly; I assume that by betting in your prediction markets, I lose the opportunity to invest that money elsewhere? This seems like a big problem for long-term predictions, e.g. "Will humans find evidence of life out of earth before 2025?" If I just invest in the stock market, I expect my money to grow by ~30% by 2025, so a rational investor would only prefer to bet on this prediction if they thought the market was more than 30% off. Does that sound right? Maybe this is less of an issue if I'm placing bets using a deflationary currency like Bitcoin?
Yes, makes sense. That's one of the reasons we allow bets in multiple currencies, so if you're bullish on BTC (or DOGE, or LTC, etc), you can make your bets in that currency and benefit from any price rises. But your larger point re opportunity costs is well taken. We are exploring ways to address that longer term, which may involve staking bonuses and the like.
Scott addressed this in an earlier article with the notion of "prediction chaining." Instead of needing to wait until 2025 to close my position, I could bet on a prediction like "The prediction market position for Will humans find evidence of life out of earth before 2025? will be greater than 30% by Dec 31 2022" or some better worded variant of the same.
Intriguing idea! In the case of Futuur, you can sell your position at any time, and there currently is no penalty for doing so (other than the market dynamics of course), so in principle there is an ongoing incentive to bet directionally (ie bet on underpriced outcomes), and no obvious need from an information elicitation perspective to complicate things by creating multiple chained prediction markets. The multiple markets approach would probably reduce liquidity on a per-market basis as well.
I took a quick looks at the post your referred to - https://astralcodexten.substack.com/p/mantic-monday-1115 - and it seems to be discussing a paper by Tetlock et al which takes a non-market based approach along the lines of the Good Judgment Project, where there seems to be a more compelling argument for using these chained forecasts.
I respect the "crypto is a ponzi scheme" position in general, but surely a situation where you really want to use it for a specific practical task is weird time to bring it up?
A permanent work-from-home situation doesn't seem like much of an actual improvement on the previous, just shifting the misery. Some people will like it. Others, like me, find it miserable and lonely. I'm trapped in my apartment all day with nowhere to go. Sometimes I don't leave my apartment for over a week. It sucks. And I don't even have to worry about small children. I'd be thrilled to go back to the office...except that my team moved to open space, which I hate even more than permanent work from home.
It might be coming, though I think a few companies are starting to regret trumpeting the glorious end of the office. If it is, I don't think we're going to see a vast improvement in happiness or productivity. We'll mostly become more lonely and alienated, which will cause a whole host of lousy effects across society. Unfortunately, that's hard to make into a prediction markets question.
Work from home is really "work from anywhere". People interested in this sort of stuff can recreate "offices" on their own; it doesn't work in the opposite direction.
Right, but that probably involves additional coordination and expense on the part of the individual. Finding a situation you can afford that is actually an improvement over your old office might be a lot of work in and of itself. Even if it’s not a financial burden, that’s considerable mental load you didn’t need to deal with before.
Compared to the co-ordination, expense and mental load associated with a) commuting itself and b) being limited to living in areas where you can reasonably commute to work?
I'm in the 'would rather be at the office' camp but the marginal person contributing to the mobility stats is probably thrilled to work from home.
Not sure where you live but where I'm (Czech republic) from co-working spaces are fairly common at least in larger cities (larger cities means everything above 100k people in our country :-) ).
They are used by start-ups or by people like you who prefer company of others when working (I very much prefer working from home or perhaps from a café/bar occasionally) and very affordable.
This requires less effort on your part than renting a flat (you just need to find the co-working space you like and make a reservation there). Of course, running your own co-working space would be a bit more effort (although you could then make it exactly to your liking and perhaps even make some cash on the side by renting it to others).
As the owner of a small manufacturing company, I'm sitting here wondering how coffee makers (for example) are going to be made if everyone is working from home.
> But second, if you go all in on correcting this mispricing, you’ll lock all 10,000 of your Ooms for three years to earn a 17% return.
(Usual disclaimers hedging that I might be misreading Scott's words and/or wrong about the object-level question.)
As far as I can tell, Futuur works like most(?) other prediction markets, where 17% means that you can buy "Yes" shares for (slightly more than) $0.17, and win $1.00 if you're right. That's about a 450% return minus fees if Mars happens, or 150% annualized, plus whatever value you assign to the question resolving earlier than Dec 2024.
Yeah I think you have Scott's sentiment backwards on the Mars landing question. I think a 2024 Mars landing is <5%, probably <1%, and I suspect Scott roughly agrees with me. So you're buying at 0.83 and making $0.17/share (which is technically a bit more than a 17% return since it's 0.17/0.83, but close enough)
Yeah I tend to agree. I'd comfortably call a NASA landing less than 1 in a million. I left my comment fuzzy because I didn't feel like thinking too hard about "how likely is it that Elon says 'fuck it' and does a landing way before we're actually ready". Probably well under 1%, I just haven't really thought it through fully.
I am no expert, but Mars trips with reasonable delta v can only depart at certain times determined by the orbits of Earth and Mars. If this random thing [1] on the internet is to be believed, a mission to Mars landing by 2024 would have to leave *next September*. I think that happening without our having heard of it by now is... more unlikely than the prediction market just stealing your money.
Huh wow, I had a 6 month trip in mind, but hadn't thought about not having launch window options for a 2024 arrival at all. That definitely makes it even less likely
For minimum-energy trajectories, yes, the relevant launch window closes in September/October 2022. There's an opposition-class trajectory window in December 2023 that would arrive before the end of 2024, but require about twice the departure delta-V. So, two years to design, build, and test a high-delta-V propulsion system, and a Mars transfer vehicle, and a Mars lander, and a Mars EVA suit, etc, etc, and oh yes some sort of return capability, and nope nope nope, not going to happen.
So, 17% of the Futuur "money" is in the hands of people who can't or won't do the math, or even the basic literature search.
The video mentions conditional prediction markets (having markets for P(tourism increase|park) and P(tourism increase|no park) ), which seem like a great idea. One problem, and the solution that comes to mind for it, that I'm curious if other people have thought about:
Problem: Having two prediction markets significantly hurts (halves?) your expected returns, because one of the markets will just give back your investment when the condition doesn't happen, so you have to invest ~twice as much capital for the same return. For example, say I think there's a 10% chance of the park being built, a 90% chance of it increasing tourism, the market says 70% chance it increases tourism, and it'll resolve in a year. In a normal market, great, I can get a 30% return ($0.70 investment with a $0.90 expected return, or a $0.20 = 30% profit) in a year, I happily take it. In a conditional market though, there's a 90% chance I break even and a 10% chance I make 30%, so my return is only 3% in a year, which isn't worth betting at all. This will get worse with more than 2 conditional options, or deeper levels of conditioning.
Solution that comes to mind, I'm curious if/where people have discussed this: Have a parent market, "will the park be built", and then instead of that market paying out in dollars, it immediately gives you credits to bet on the relevant conditional market. So betting on "the park will be built" immediately gives me a $1 credit to spend on the "given the park is built" market, which will pay in full if the park is built and gives me nothing back if it isn't.
So in the example above, I can bet $0.10 on "the park will be built" and then put a full dollar of that toward "this will increase tourism". Now I have a 90% chance of losing $0.10, and a 10% chance of [ending up with ~1.43 shares in P(tourism increase|park is built), which have a 90% chance of becoming $1.43 and a 10% chance of becoming 0 so have an expected value of $1.29 or] a profit of $1.19, so my total expected return is $-0.10*0.9 + $1.19*0.1 = $0.03, and we're back to a 30% return on my $0.10 investment.
Question, in addition to just being interested in people's thoughts in general: Is this just recreating some complicated option system or hedge or margin purchase that you can do without the need for the "parent market pays instantly in credits for child markets" system?
Is this better than selling chained probabilities? i.e. make shares pay out on "park built AND tourism increase", and have the $1 bundle split between the 4 outcomes? I guess that many people are bad at the maths for doing probability this way, which might be fixable via UI? It also might not scale well to situations with many more outcomes, like Presidential election chances conditional on a particular candidate winning the nomination.
Hmm good point. On first thought that seems like it's probably exactly identical (and maybe less complicated, so better?).
You just have one market for "park is built and tourism increases", "park is built and tourism decreases", "park is not built and tourism increases", and "park is not built and tourism decreases". Yeah not obvious to me how that'd be any different besides being framed differently. Agreed that people might be bad at the math involved and a smart interface would be necessary, but that's probably fine?
Only concern is that prediction markets I've seen only do steps of 1% (or $0.01), so things with probabilities less than 1% don't really work. That could come up in the chained probability situation in a way that my nested market version would avoid. The 1% steps though are a design choice by the markets, not fundamental at all.
Prediction markets demonstrate worrying dynamics on the Russian issue. The change perfectly reflects my subjective perception shift on this risk in the last two months (i.e. the risk slowly increases).
It’s important to say that a 40% probability within a year still tells us little. As Putin demonstrated in the last 20 years, he never attacks precisely when everyone expects him, but eventually he does.
The last serious escalation between Ukraine and Russia happened 3 years ago (The Kerch Strait incident). It’s a long period, almost a half of the entire war. So it’s reasonable to assume that Putin is willing to fuel the fire soon, as he never received any dividends from this conflict yet. Neither we should underestimate his sacred love for Olympics.
So if we expect an escalation within the next 1-2 years, 40% is a reasonable baseline.
Still we don’t have any idea about the supposed nature of this escalation: whether it will be a full-scale war or something like the aforementioned Kerch Strait incident.
It doesn’t depend on the concentration of Russian troops near Ukrainian border: we see such moves EVERY year since 2014. Whether Russia will start a full-scale invasion depends only on three things: Ukraine’s internal stability, NATO’s willingness to respond, and Putin’s mental health.
But this time the shitshow at the border triggered a few political conflicts inside Ukraine. I’m worried that our politicians and their followers are dumb enough to be blindsided by internal agenda. That’s what made my perception shift recently and that’s where we should pay the most attention now if we want to reason about the Russian threat clearly.
Putin wouldn't invade unless Ukraine decides to force the Donbass issue. Ukraine has long tried to wriggle out of Minsk agreements (which, to be fair, are hilariously unfavorable to it, no idea how it was coaxed into signing theim in the first place), and apparently it finally got the go ahead from the West to ignore them, or is about to get it. So the question is, do they think that risking an invasion is worth it, and in turn, does Putin? Seems like it would mostly be decided by PR considerations from both sides, which isn't reassuring.
Yeah, what a brilliant idea to force the Donbas issue while Russia concentrated forces at the border.
Putin can invade Ukraine regardless and he will definitely portray it as "Ukraine started first", which is an essentially impudent claim: Ukraine has every right to free Donbas as it's already under Russian occupation, it can't "start first" as the war has been already started by Russia in 2014.
Will Ukraine actually take any action first? No. Ukraine is prepared to any scenario, but everyone here understands that Putin will use their actions as an excuse for a further invasion. So, the whole Ukrainian military doctrine is about defense and response, for now.
Of course he can, but Putin isn't random or inscrutable, IMO he pursues a fairly consistent foreign policy, so I wouldn't expect him to disrupt the status quo for no great benefit, the current frozen conflict seems to suit him just fine. In 2014 it was obviously Ukraine that set the chain of events in motion, however much it considered itself justified in doing that.
Compared to what counterfactual? In the 90's Russia tried hard to gain friends "diplomatically", but there were no takers. So he tried an alternative approach, which worked well enough. He continues to be able to sell the stuff that Russia is competitive in (mainly resources and weapons) and undermine Western influence and prestige wherever he can (seems to be going from strength to strength in fact), while remaining pretty popular at home.
Minsk was hilariously, 100% favorable to Ukraine. Ukraine had two goals at the time: stop the advancement of Russian troops and to make Russia diplomatically responsible. Ukraine got both. The troops stopped, and Russia signed the protocol that _it_ was never going to honor, which gave grounds to sanction Russia and to provide aid to Ukraine. How well is current Ukrainian government making use of this diplomatic victory is an entirely different question.
Lol, Russia was designated as responsible the instant that the first bullet was shot, which was always a foregone conclusion. What Ukraine got was legitimized frozen conflict, which I suppose you can argue is the best deal it could've gotten, but still looks pretty bad.
“In Donbass, the war was started by us, and the first shots were shot by us”
— Igor Girkin, translation mine
And the war is won when the other side can’t or won’t wage it anymore. When the other side is the size of Russia, inflicting a military defeat on it just isn’t an option, but 1) freezing the situation, 2) making Russia pay to maintain the situation and 3) making that cost as high as possible is a pretty viable strategy, if not the fastest one.
I think we disagree about the goals of Russia's involvement. Mainly, its interest was to protect the Russian population in the regions with significant presence thereof, which was easily achieved. If the goal was to utterly crush Ukraine, no matter the cost, that would be easy too, but that never was the goal, and still isn't.
That’s not interest, that’s a pretext. How exactly are Russia’s interests served by protecting 1M+ Russian-speaking people from their homes and 10k+ Russian-speaking people from being alive, all at a high monetary and diplomatic cost?
>So it’s reasonable to assume that Putin is willing to fuel the fire soon, as he never received any dividends from this conflict yet.
From Putin's perspective, he has. His foreign policy for the last 15 years has boiled down to "make the West look limp-dicked and Russia look like it gets what it wants". And the Ukraine situation has done that.
His goals are domestic popularity and international respect, not direct material gain. You see it in Ukraine, in Syria, in the assassinations, in putting up Snowden, and in the interference with US elections.
I'm not saying Putin *won't* make some more moves in Ukraine, of course, but if you think he hasn't gotten anything already you're not thinking in his terms.
That's true that Putin wants to make the West look limp-dicked, but that's just a mean, not the end. His will for influence and respect goes as far as restoring Russia's status as a world power, like the status it had during the Cold War. He wants to split the world into "spheres of influence" and assert his control over what he is calling "the Russian world".
Ukraine is the most important piece in his puzzle, both metaphysically and geopolitically, and he had much more control over it back in 2013 than he has now. Aside "taking Crimea home", which is mostly just a domestic victory, he is very far from achieving his goal of controlling the rest of the country. He also haven't got any respect for "Russian interests" in Europe from the US and NATO countries.
So I don't think he is interested in freezing up the existing conflict as long as those goals aren't achieved.
Thanks, Scott, for the coverage of what we’re working on at Futuur! We’re fans of ACX and its precursors so it’s great to see Futuur mentioned here :)
As indicated, we’ve recently launched our real-money markets, with trading available in various cryptocurrencies (including the USDC stablecoin for those concerned about crypto volatility). We offer hundreds of markets across all major topics, facilitated by an automated market maker. We are just getting started, so there are certainly opportunities for savvy traders as we build liquidity!
Regarding the point about our play-money markets, it is true that they can generate some odd forecasts, and we are exploring ways to address this (for instance, phasing out bets that came from inactive accounts on long-term forecasts). Overall however our play-money markets have performed reasonably well; we’ve posted a couple analyses on Medium, here: https://medium.com/futuur
And of course, we expect our real money markets to perform even better, although it’s not guaranteed. We’ve created a market on precisely this question here:
Very interested to hear everyone’s feedback — positive, negative or neutral — and happy to respond to questions or suggestions folks might have, including suggestions for new markets, feature requests, etc!
> It seems like the general rule is that every month, the date when people expect normality to return moves about a month and a half further out.
Interestingly, there's a very simple Bayesian model which results in this behavior. Namely, if the waiting time for some event is exponentially distributed, and you have a gamma prior on the rate, then the additional expected waiting time after waiting for time T will scale linearly with the time waited so far. It's kinda neat, since you actually have less than one observation in some sense, but you can still compute a posterior (just using P(X > T), instead of P(X = t)).
I originally thought about this in the context of nuclear war (of course neither nuclear war or pandemics are memoryless processes, but all models are wrong etc.). If we haven't observed a single nuclear war so far, how can we estimate how likely they are? You can't, but if you have a prior it also tells you how much longer each year without a war means you should expect to wait.
Calculating the rate of nuclear war from observational evidence gets weird and subjective, because you have to correct for your own personal anthropic bias. Any period during which you were guaranteed to die in the event of Global Thermonuclear War (e.g. via living in Washington DC) is zero evidence *for you* against GTW because if there had been GTW then you wouldn't be asking the question now, and any period during which you were significantly likely to die in the event of GTW (e.g. not in a nuke target but no stockpiles against supply-chain disruption) should be weighted less.
And yes, that means that different people get different correct answers.
One way to make a Mars mission profitable would be to make your preparations in total secret, then bet against everybody on prediction markets that someone would do it by a given date.
Of course it would be too hard to completely hide the existence of a giant, well-funded rocket company, so you'd just need to appear to be a giant well-funded rocket company that was making suspiciously slow progress, like Blue Origin. Hey, wait a minute....
They'll definitely get the last laugh when their $100 billion effort nets them the $20,000 or so that's currently tied up in prediction markets on this question.
I'm not sure I know Balaji's idea but you could try to host a debate about a topic and see which way the market moves. You could make people bet on the resolution and hold their position.
My problem with this is that prediction markets are being touted as 'ways to find the Real Truth of whatever thing is going on'. But people don't play the stock market to find the 'Real Truth of is Wiggins' Widgets correctly priced', they do it to make money. If that means over- or under-valuing the share price for Wiggins' Widgets, that is what will happen. This is how we get bubbles; everyone thinks they have secret inside information that will net them a fortune. Then everyone else sees the price of Wiggins' Widgets soaring, and they want in on this sure-fire gravy train.
So I don't think using prediction markets to settle debates or whatever will work. If I've got enough money to make it worth my while to bet seriously in a prediction market, rather than just play at it where losing won't impact me that much since I haven't wagered that much, I will bet on "what is the result that gets me the most money?" not "do I think Penny Stampe had the better argument on the topic 'Do butterflies dream of fountains?'"
If my estimation is "Penny is hot and her opponent, Dundreary Sam, is not, so people are going to vote for whatever Penny argues never mind if it's batshit insane", then I'll spend my money on "Yes" to the butterfly question. Even if I think it's batshit insane to even ask if butterflies dream. Because that will get me money.
Or conversely, I will vote on "No" because I expect Sam to lose (since he's an uggo) even if I firmly believe in butterfly sentience. Because that will get me money.
Can Americans simply lie about their nationality and use Mantic anyway? If it uses crypto it sounds like you shouldn't need to give them bank account details or anything tying you to a location.
I'm constantly amazed that despite seeing dark net market after dark net market busted, and tons of evidence that the US federal government works on chain analysis and deanonymization, yet people still think that it's fairly straightforward to hide your crypto from the government. Sure, they won't burn the methods they used publicly, but they certainly have lots of other ways to claim they found you after you're charged. (Of course, you're probably not doing anything illegal enough for the NSA to care, so you'll probably be fine.)
If *all* you do is make a technically-illegal $10,000 or so, and you're careful to maintain broad and thorough operational security (which takes a lot more than just "not being stupid"), then there's a good chance that the Feds will ignore you rather than risk revealing their capabilities. But that's a lot of risk for a little gain.
I believe this is what some existing crypto markets like Polymarket do: warn Americans against using it but not enforce it in some way.
Honestly, we're still figuring out how the crypto/real-money side of things could work from a legal perspective! Our expertise is in the software side, which is why we've built out the web prototype using fake money. My expectation is that we'll keep operating that for a while, and possibly keep using play money forever (as a second option)
Balaji seems to be using prediction markets as a buzzword the same way tech companies keep putting things on the blockchain that do not need to be on the blockchain. Oxford debate has people spend some time arguing in a particular structured format then has the audience vote on who they think had a better argument. There's no window for usefully predicting the outcome (are you gonna do it halfway through the debate and predict 30 minutes in the future, or before the debate has even started when all you know is the topic and who's involved?) and also debates are fundamentally entertainment, this is like proposing a prediction market on who will die in a TV show that barely anyone watches.
"This is good because it maintains everyone’s faith in the objective process, but bad because what people actually care about is whether there will be something that feels like a crisis situation, which is hard to instrumentalize in hospital numbers. Mantic wants to lean into the subjective side of prediction markets and see what happens."
I had no idea that what I needed to explain prediction markets to me was an animated dog in a wizard hat, but there you go! 😀
However, it just looks more and more like betting on horse races to me ("I am really sure that Silver Shoes will win the 3:30 at Kempton because of the going, his past performances in these conditions, and my uncle's best friend's nephew is head lad in the stables and they swear Silver Shoes will run a blinder" - now just slap a probability estimate on how you think Silver Shoes will do, and it's the same principle as "Joe Doggen will legalise black magic").
I think this is the whole point. I don't know anything about horse betting, but sports markets in general does tend to be remarkably good at predicting outcomes, as far as I can tell much better than individuals are generally able to.
As I understand it, prediction markets are only profitable if there are enough people making bad predictions to lose money to the people who make good predictions, and prediction markets might tend to drive unskilled people out.
Presumably, prediction markets need to be enough fun to keep the fish playing. I have some evil ideas for that, probably combining the prediction market with a lottery.
If prediction markets have an overlap with gambling ones, I don't think there will be a shortage of unskilled participants. Most of the gamblers I know have been on a losing streak since the day they first laid a bet and plough on undeterred.
So the most perfect prediction market would be one where there's no point participating, as the outcomes are all perfectly in line with the predictions, so nobody makes any money?
Hmmm - is there perchance a flaw in the reasoning behind them? 🤔
Probably they'll work because very few people are always 100% right about everything, but I don't know if I would trust them for making actual policy decisions: Mayor Piecrust makes the decision to go ahead and build that park in order to boost tourism, based on the prediction market forecasts, but the reason they need to boost tourism in the first place is that nobody wants to visit the town because it's in the back-end of nowhere, there is nothing to do there, the weather isn't good enough for sun holidays/winter skiing holidays, it's just 'average nowhere town' and a park isn't going to be enough to entice visitors.
However, since everyone participating in the prediction market are the locals, who are the only ones left in the dying town, and they're all desperate for something, anything, to boost the fortunes of the town, they allow optimism to overcome good sense ("a park helped Greenville get more tourism!") and so the prediction market result is skewed to "build a new park!"
Only locals are going to be interested in making predictions about "Will a new park boost tourism to Deadend?", and if you are hoping for outsiders to help make it more aligned with reality, the only outsiders who are going to invest will be those who are hoping to make money off prediction markets, so they look for local interest markets like this, and evaluate it that "the locals hope the answer will be yes but realistically the prospects are no; since the locals are all going to vote yes, I can beat this by buying up all the no predictions".
I wasn't assuming that the predictions were perfectly correct, just that they would be close enough to correct that it wasn't worth the work to do better.
In the long run I expect that hype will drive dumb money into prediction markets in much the same way that it has for *checks notes* bad furry pixel art.
Before that, there's a lot of talk about subsidising new prediction markets with some kind of crappy algorithmic trader to get people trading on them.
2. There are kinds that attempt to predict "if we took up a collection to buy a rare-book NFT of the Constitution, could we raise $40 million".
3. There are kinds that attempt to predict "if we took up a collection to buy a rare-book NFT of the Constitution, would $40 million be enough".
7. Attempting to predict “what percentage of the US population supports <insert hot-button political topic>”, something which would be best measured by a statistical survey rather than a financially motivated system.
The Ukraine one makes sense to me. Biden isn't threatening to go to war if Russia invades Ukraine, he's threatening harsh economic sanctions. I don't think Russia invading Ukraine this year triggers an actual war between the US and Russia right away, but it does increase the long term risks of war.
I think there is a reasonably chance that every intelligence agency of reasonable budget/power that decides having the general public (and their adversaries) having access to good prediction data is against their national interest, will immediately move to shut down/discredit/hack prediction markets until the markets are unworkable.
It seems to me that it would be very cheap and easy to do.
A common problem with prediction markets is that it's inefficient to act to correct to prices far in advance of the resolution date, because of the opportunity cost of the returns that you could otherwise be earning on the funds locked up in the market. The obvious solution is to invest the market's funds to realize those returns.
It's less obvious how the details should work:
- Should the amount of investment returns payed to each participant be proportional to their purchase cost, their winnings, or the average value of their position over the period it was held?
- How should the funds be invested? Is there a way to accommodate market participants with different risk/return profiles?
- What happens when there's covariance between the prediction and the underlying investment? Doesn't that mess up the nice property of directly treating prices as probabilities? Can you correct for it?
- While we're investing the underlying funds, why not invest them in *other* prediction markets? To frame it a different way—a position in the prediction market (usually) has non-zero value and is fairly liquid, so you should be able to borrow against it to make other bets. Or, you could reinvest in the same bet and get leverage. Would available leverage improve or detract from the accuracy of a prediction? What special considerations are there for a lender against this sort of collateral?
Yes, the problem of opportunity costs gets brought up a lot. There are some potential solutions:
1. I'm sure there are a bunch of ways it could be set up, but the natural way IMHO is to make a "yes" wager $X, a "no" wager $1-X, and then take their combined dollar and invest it. The winner gets the investment.
2. From a mathematical perspective, the natural investment would be a (mostly) risk-free instrument, so that the bets are on the prediction and not on the investment. However, risk-free instruments don't have very good returns and might not be enough to convince people to take really long-term bets. In this case, I could imagine using a diversified market portfolio, or allowing the "yes" and "no" sides to agree on an investment at the time they place their wagers (though this has the drawback that bets aren't comparable if they use different investments, due to the covariance issue you mention).
3. Even with uncorrelated investments, prices don't imply probabilities unless you assume risk-neutral investors. However, the average investor is risk-averse, so prediction markets will tend to overpredict bad things happening.* You can build a model of correlations and risk premia and try to estimate probabilities from market prices, but there's no theoretically guaranteed way to convert prices to probabilities. Ideally, people should only speak of prices, not of probabilities.
*I should add a caveat: this might not be true in the early days of prediction markets, where average investors aren't participating. For example, there might be some selection bias where risk-seekers are more common.
4. You have the right idea. Once something is liquid enough, you can potentially borrow against it and make other investments. This will generally make markets more efficient as investors have more ability to get exposure to the risks they want. However, a drawback to prediction markets is that they are (usually) binary outcomes: you either get paid or get nothing. That makes them risky collateral for a loan, so I wouldn't be too optimistic about ever being able to borrow much against prediction market investments.
My thoughts here are based on my experience working with financial derivatives, which have many of the same issues. If you wanted to do some reading, #3 is something covered in quant finance curriculum, though I'm afraid I can't point to a specific source.
I was wondering if someone could explain something to me.
How does a question on a prediction market generally get started? As in, how does the exchange itself sell the initial "futures" to predictors, like an IPO? They need to sell them for a high enough price to pay out, on average, to the holders of the futures if and when the event takes place. How do they ensure they achieve this with enough of a safety margin?
Is there some clever trick pairing off the "yes" and "no" markets together? Do they act as a normal market participant (would this not require the exchange themselves to partake in prediction and need to do research)? I've thought about it for ten minutes and haven't come up with a low-risk mechanism that the exchange can use to get a market started, but I'm sure I'm missing something.
Assuming that: (A) large numbers of people will soon be displaced by automation, thus making something like UBI necessary to avoid pesky bread riots and whatnot, and (B) if large and liquid markets are generally the best available (albeit imperfect) predictors of future events, even taking "dumb" money into account, I am trying to think of a way to tie prediction markets to UBI.
Each participant gets a certain amount of money at the beginning of each month, which they are required to use to make prediction bets on one or more matters of social interest. The amount the participant receives at the end of the month, above a certain minimal threshold, is tied to their success in making such bets. For bets which don't or cannot resolve in the next thirty days, the participant can either let the bet ride for the next thirty days or until resolution, whichever happens first, or the participant can cash out at the current bid-ask price.
The participants get UBI and something potentially socially useful to do (and that has the potential of an additional payday), and the mechanism may be more effective than paying consultants, futurologists and analysts. Also, participants which exhibit consistently successful track records may be selected to be groomed for Bigger And Better Things.
I just thought this up because I don't feel like working ATM, so tell me what is wrong with this idea?
regarding the VA workplace activity question, the prediction probability density graph seems to agree with your conclusion as there is a high spike at the end of the graph (img: https://imgur.com/a/ZduqJaK, question: https://old.metaculus.com/questions/7137/va-workplace-activity-reach-baseline-lvl/)
We should be able to see if people really shift all the way to the end if we could see the changes to that graph over time, is that possible in metaculus?
Re. Futuur California fires market – the resolution source says there were 9,639 fires last year, but their description says that "In 2020 there were *8112 fires*, in comparison to there were 5687 cases in 2019." So if you go off the linked statistics it's an obvious 'No', but if you go off their description it's an obvious 'Yes'. I think I'll sit this one out...
It's quite possible I don't fully understand the Futuur betting market, but it seems to me that as regards the people-on-Mars bet, there are plenty of people correcting this mispricing. Out of 6000 transactions on the market, most of them are saying "Nope, no chance", which is correcting the mispricing as much as they feel they can afford (as Scott says, three years is a long time to wait for 17%)
But there are also, predictably, people uncorrecting the corrections - I looked through the bets and there's one Musk fanboy who has wagered 4000 ooms on humans walking on Mars in 2024.
And the weirder thing about this is that it wouldn't change much if it was real money rather than ooms. The 17% would drift down to some Muskman constant and the serious money would stay away because the returns - though as good as guaranteed - are too small and too distant.
When it becomes clear the answer is 0, does the price change and allow one to sell off?
There is a launch window in Q3 of 2022, about 9 months away. (After that, the next launch window is Q4 of 2024, which doesn't allow boots to hit the Martian dirt in calendar year 2024.)
If it's priced like a stock market, there will be no buyers left, the price will go to $0, and everyone can close their positions.
if the "Mantic" model became legal, then anyone with a big enough following and the requisite trust factor could be a William Hill style casino. A 1% "vig" might be really bad for many gambling businesses.
If you think about it, the current 10% vig charged by gambling books are gigantic in this era of communication.
Haha that's a good point! We're not actively looking to disrupt online casinos or sports betting, and are focusing on more niche areas (core question: "what doesn't have a market yet?"). But if it happens by accident, well, I wouldn't complain :)
Totally agreed that their 10% cut (also on e.g. PredictIt) seems high, and we should be able to do better!
Tsk tsk, Scott. I had to look up CFTC to find out it means Commodity Futures Trading Commission.
It isn't ChloroFluoro-Trilateral Commission?
I am intrigued by all of Balaji's ideas about how to make speech more free and meaningful. I just wish I understood the terminology better. I too would like some help understanding this Oxford debate thing better. And if anyone can, please also explain his ideas on pseudonymous accounts simply, as well :).
If I am not mistaken, the Soho debates hosted by Reason are hosted using Oxford scoring, so if you want some examples you could play the podcast. Basically:
Relatively long theses for and against, followed by rebuttal, followed by questions. The audience votes both before and after the debate, and the winner is the one that moves the vote in their direction.
An example involving intellectual property rights is at https://reason.com/video/2021/11/24/abolish-copyrights-and-patents-a-soho-forum-debate/
I like futuur. I already won my 2 first bets and made some bets that will resolve in about 3 years.
Using virtual ooms only at the moment. I am not touching crypto money out of principle. It is a Ponzi scheme that we should not encourage. If they had option to use better money (US$, euro, pound), I would participate.
Noted, thanks! We do offer USDC, which is pegged to the dollar. Would that allay your concerns? We are considering adding fiat options, but in the gaming space (we operate under a gaming license from Curaçao) fees are very high, and there's a good deal of operational friction, so it might take us a while to roll out fiat options.
Cool to see a futuur person in the comments! I have a few general questions about predictions markets that I'm going to post here in the hopes of getting some answers:
1) Have you analyzed how taxes/fees/opportunity costs from not investing elsewhere impose error bars on your prediction markets? Have you thought about using this info to present error bars around the predictions on your site?
2) less relevant to your project, but whenever I talk about using prediction markets to make decisions, the first criticism I hear goes something like "what if deep-pocketed special interests stood to gain a lot of money from a particular outcome"? I assume economists have thought a lot about this kind of scenario and how to model it. Can you suggest how a non-economist (with decent math literacy) would find an entry-point into that literature?
Interesting questions! Not sure if I fully understood all of them, but will have a go at answering...
1) taxes and fees can and do impact the accuracy of a market, in the sense that if the real probability of an event occurring is say 50%, but the cost of a share is for instance .55, there is a delta of 5 points in which a rational trader will not make a bet, and the probability forecast will not be updated. We intend to address this to the extent possible by offering lower fees, especially as liquidity for a market expands. Regarding opportunity costs, I assume you mean the opportunity to trade other markets? I don't see an easy way to model this but interested to hear ideas. Likewise, not sure how we would present error bars on our markets, but it sounds intriguing.
2) the idea of market manipulation gets raised a lot, and I think there are valid concerns. But: either a) you are trying to make money with your bets, in which case you will be rewarded for accuracy and help to generate better forecasts for everyone, or b) there is some external situation which you hope to influence by betting against your purely financial interests (ie subsidizing wrong forecasts because of the impact it might have). For instance, you could hope to influence media coverage by increasing the probability forecast, and thus influencing the thing being forecast via the forecast itself. The same concern is raised around "assassination markets" and the like. It's an interesting conundrum, but from a practical perspective the amount you would need to invest in a liquid prediction market to influence the outcome would probably need to be much higher than other methods of influence (for instance, hiring a PR team).
Regarding literature around prediction markets, the writings of Robin Hanson are probably a good jumping off point: https://mason.gmu.edu/~rhanson/ifpubs.html#Hanson
https://www.overcomingbias.com/tag/prediction-markets
Beyond that, there is an expanding body of work accessible on Google Scholar here: https://scholar.google.com/scholar?hl=en&as_sdt=0%2C5&q=%22prediction+markets%22&btnG=
Hth! :)
Thanks for your answers!
1) Re: opportunity costs, you interpreted my question correctly; I assume that by betting in your prediction markets, I lose the opportunity to invest that money elsewhere? This seems like a big problem for long-term predictions, e.g. "Will humans find evidence of life out of earth before 2025?" If I just invest in the stock market, I expect my money to grow by ~30% by 2025, so a rational investor would only prefer to bet on this prediction if they thought the market was more than 30% off. Does that sound right? Maybe this is less of an issue if I'm placing bets using a deflationary currency like Bitcoin?
Yes, makes sense. That's one of the reasons we allow bets in multiple currencies, so if you're bullish on BTC (or DOGE, or LTC, etc), you can make your bets in that currency and benefit from any price rises. But your larger point re opportunity costs is well taken. We are exploring ways to address that longer term, which may involve staking bonuses and the like.
Scott addressed this in an earlier article with the notion of "prediction chaining." Instead of needing to wait until 2025 to close my position, I could bet on a prediction like "The prediction market position for Will humans find evidence of life out of earth before 2025? will be greater than 30% by Dec 31 2022" or some better worded variant of the same.
Intriguing idea! In the case of Futuur, you can sell your position at any time, and there currently is no penalty for doing so (other than the market dynamics of course), so in principle there is an ongoing incentive to bet directionally (ie bet on underpriced outcomes), and no obvious need from an information elicitation perspective to complicate things by creating multiple chained prediction markets. The multiple markets approach would probably reduce liquidity on a per-market basis as well.
I took a quick looks at the post your referred to - https://astralcodexten.substack.com/p/mantic-monday-1115 - and it seems to be discussing a paper by Tetlock et al which takes a non-market based approach along the lines of the Good Judgment Project, where there seems to be a more compelling argument for using these chained forecasts.
I respect the "crypto is a ponzi scheme" position in general, but surely a situation where you really want to use it for a specific practical task is weird time to bring it up?
A permanent work-from-home situation doesn't seem like much of an actual improvement on the previous, just shifting the misery. Some people will like it. Others, like me, find it miserable and lonely. I'm trapped in my apartment all day with nowhere to go. Sometimes I don't leave my apartment for over a week. It sucks. And I don't even have to worry about small children. I'd be thrilled to go back to the office...except that my team moved to open space, which I hate even more than permanent work from home.
It might be coming, though I think a few companies are starting to regret trumpeting the glorious end of the office. If it is, I don't think we're going to see a vast improvement in happiness or productivity. We'll mostly become more lonely and alienated, which will cause a whole host of lousy effects across society. Unfortunately, that's hard to make into a prediction markets question.
Work from home is really "work from anywhere". People interested in this sort of stuff can recreate "offices" on their own; it doesn't work in the opposite direction.
Right, but that probably involves additional coordination and expense on the part of the individual. Finding a situation you can afford that is actually an improvement over your old office might be a lot of work in and of itself. Even if it’s not a financial burden, that’s considerable mental load you didn’t need to deal with before.
Compared to the co-ordination, expense and mental load associated with a) commuting itself and b) being limited to living in areas where you can reasonably commute to work?
I'm in the 'would rather be at the office' camp but the marginal person contributing to the mobility stats is probably thrilled to work from home.
Not sure where you live but where I'm (Czech republic) from co-working spaces are fairly common at least in larger cities (larger cities means everything above 100k people in our country :-) ).
They are used by start-ups or by people like you who prefer company of others when working (I very much prefer working from home or perhaps from a café/bar occasionally) and very affordable.
This requires less effort on your part than renting a flat (you just need to find the co-working space you like and make a reservation there). Of course, running your own co-working space would be a bit more effort (although you could then make it exactly to your liking and perhaps even make some cash on the side by renting it to others).
As the owner of a small manufacturing company, I'm sitting here wondering how coffee makers (for example) are going to be made if everyone is working from home.
In China? :/
> But second, if you go all in on correcting this mispricing, you’ll lock all 10,000 of your Ooms for three years to earn a 17% return.
(Usual disclaimers hedging that I might be misreading Scott's words and/or wrong about the object-level question.)
As far as I can tell, Futuur works like most(?) other prediction markets, where 17% means that you can buy "Yes" shares for (slightly more than) $0.17, and win $1.00 if you're right. That's about a 450% return minus fees if Mars happens, or 150% annualized, plus whatever value you assign to the question resolving earlier than Dec 2024.
Yeah I think you have Scott's sentiment backwards on the Mars landing question. I think a 2024 Mars landing is <5%, probably <1%, and I suspect Scott roughly agrees with me. So you're buying at 0.83 and making $0.17/share (which is technically a bit more than a 17% return since it's 0.17/0.83, but close enough)
I'd say there isn't a snowballs hope in hell of humans landing on Mars in 2024, so I guess that would equate to less than 1%..
Yeah I tend to agree. I'd comfortably call a NASA landing less than 1 in a million. I left my comment fuzzy because I didn't feel like thinking too hard about "how likely is it that Elon says 'fuck it' and does a landing way before we're actually ready". Probably well under 1%, I just haven't really thought it through fully.
I am no expert, but Mars trips with reasonable delta v can only depart at certain times determined by the orbits of Earth and Mars. If this random thing [1] on the internet is to be believed, a mission to Mars landing by 2024 would have to leave *next September*. I think that happening without our having heard of it by now is... more unlikely than the prediction market just stealing your money.
[1] https://www.reddit.com/r/SpaceXLounge/comments/dtm5bc/mars_launch_windows_20202030/
Huh wow, I had a 6 month trip in mind, but hadn't thought about not having launch window options for a 2024 arrival at all. That definitely makes it even less likely
For minimum-energy trajectories, yes, the relevant launch window closes in September/October 2022. There's an opposition-class trajectory window in December 2023 that would arrive before the end of 2024, but require about twice the departure delta-V. So, two years to design, build, and test a high-delta-V propulsion system, and a Mars transfer vehicle, and a Mars lander, and a Mars EVA suit, etc, etc, and oh yes some sort of return capability, and nope nope nope, not going to happen.
So, 17% of the Futuur "money" is in the hands of people who can't or won't do the math, or even the basic literature search.
What is the probability that Hell is frozen rather than fiery?
Not sure, but I hear they have terrible problems with demonic warming down there..
According to Dante's fine travel documentary works, it's both!
Pretty good, ever since Duke Nukem Forever came out.
The video mentions conditional prediction markets (having markets for P(tourism increase|park) and P(tourism increase|no park) ), which seem like a great idea. One problem, and the solution that comes to mind for it, that I'm curious if other people have thought about:
Problem: Having two prediction markets significantly hurts (halves?) your expected returns, because one of the markets will just give back your investment when the condition doesn't happen, so you have to invest ~twice as much capital for the same return. For example, say I think there's a 10% chance of the park being built, a 90% chance of it increasing tourism, the market says 70% chance it increases tourism, and it'll resolve in a year. In a normal market, great, I can get a 30% return ($0.70 investment with a $0.90 expected return, or a $0.20 = 30% profit) in a year, I happily take it. In a conditional market though, there's a 90% chance I break even and a 10% chance I make 30%, so my return is only 3% in a year, which isn't worth betting at all. This will get worse with more than 2 conditional options, or deeper levels of conditioning.
Solution that comes to mind, I'm curious if/where people have discussed this: Have a parent market, "will the park be built", and then instead of that market paying out in dollars, it immediately gives you credits to bet on the relevant conditional market. So betting on "the park will be built" immediately gives me a $1 credit to spend on the "given the park is built" market, which will pay in full if the park is built and gives me nothing back if it isn't.
So in the example above, I can bet $0.10 on "the park will be built" and then put a full dollar of that toward "this will increase tourism". Now I have a 90% chance of losing $0.10, and a 10% chance of [ending up with ~1.43 shares in P(tourism increase|park is built), which have a 90% chance of becoming $1.43 and a 10% chance of becoming 0 so have an expected value of $1.29 or] a profit of $1.19, so my total expected return is $-0.10*0.9 + $1.19*0.1 = $0.03, and we're back to a 30% return on my $0.10 investment.
Question, in addition to just being interested in people's thoughts in general: Is this just recreating some complicated option system or hedge or margin purchase that you can do without the need for the "parent market pays instantly in credits for child markets" system?
[Edited to clarify math and language a bit]
Is this better than selling chained probabilities? i.e. make shares pay out on "park built AND tourism increase", and have the $1 bundle split between the 4 outcomes? I guess that many people are bad at the maths for doing probability this way, which might be fixable via UI? It also might not scale well to situations with many more outcomes, like Presidential election chances conditional on a particular candidate winning the nomination.
Hmm good point. On first thought that seems like it's probably exactly identical (and maybe less complicated, so better?).
You just have one market for "park is built and tourism increases", "park is built and tourism decreases", "park is not built and tourism increases", and "park is not built and tourism decreases". Yeah not obvious to me how that'd be any different besides being framed differently. Agreed that people might be bad at the math involved and a smart interface would be necessary, but that's probably fine?
Only concern is that prediction markets I've seen only do steps of 1% (or $0.01), so things with probabilities less than 1% don't really work. That could come up in the chained probability situation in a way that my nested market version would avoid. The 1% steps though are a design choice by the markets, not fundamental at all.
Prediction markets demonstrate worrying dynamics on the Russian issue. The change perfectly reflects my subjective perception shift on this risk in the last two months (i.e. the risk slowly increases).
It’s important to say that a 40% probability within a year still tells us little. As Putin demonstrated in the last 20 years, he never attacks precisely when everyone expects him, but eventually he does.
The last serious escalation between Ukraine and Russia happened 3 years ago (The Kerch Strait incident). It’s a long period, almost a half of the entire war. So it’s reasonable to assume that Putin is willing to fuel the fire soon, as he never received any dividends from this conflict yet. Neither we should underestimate his sacred love for Olympics.
So if we expect an escalation within the next 1-2 years, 40% is a reasonable baseline.
Still we don’t have any idea about the supposed nature of this escalation: whether it will be a full-scale war or something like the aforementioned Kerch Strait incident.
It doesn’t depend on the concentration of Russian troops near Ukrainian border: we see such moves EVERY year since 2014. Whether Russia will start a full-scale invasion depends only on three things: Ukraine’s internal stability, NATO’s willingness to respond, and Putin’s mental health.
But this time the shitshow at the border triggered a few political conflicts inside Ukraine. I’m worried that our politicians and their followers are dumb enough to be blindsided by internal agenda. That’s what made my perception shift recently and that’s where we should pay the most attention now if we want to reason about the Russian threat clearly.
Putin wouldn't invade unless Ukraine decides to force the Donbass issue. Ukraine has long tried to wriggle out of Minsk agreements (which, to be fair, are hilariously unfavorable to it, no idea how it was coaxed into signing theim in the first place), and apparently it finally got the go ahead from the West to ignore them, or is about to get it. So the question is, do they think that risking an invasion is worth it, and in turn, does Putin? Seems like it would mostly be decided by PR considerations from both sides, which isn't reassuring.
Yeah, what a brilliant idea to force the Donbas issue while Russia concentrated forces at the border.
Putin can invade Ukraine regardless and he will definitely portray it as "Ukraine started first", which is an essentially impudent claim: Ukraine has every right to free Donbas as it's already under Russian occupation, it can't "start first" as the war has been already started by Russia in 2014.
Will Ukraine actually take any action first? No. Ukraine is prepared to any scenario, but everyone here understands that Putin will use their actions as an excuse for a further invasion. So, the whole Ukrainian military doctrine is about defense and response, for now.
Of course he can, but Putin isn't random or inscrutable, IMO he pursues a fairly consistent foreign policy, so I wouldn't expect him to disrupt the status quo for no great benefit, the current frozen conflict seems to suit him just fine. In 2014 it was obviously Ukraine that set the chain of events in motion, however much it considered itself justified in doing that.
Whatever you believe in.
Disrupting status quo for no great benefit is the definition of Putin’s foreign policy. Has he ever gotten any kind of diplomatic benefit?
Compared to what counterfactual? In the 90's Russia tried hard to gain friends "diplomatically", but there were no takers. So he tried an alternative approach, which worked well enough. He continues to be able to sell the stuff that Russia is competitive in (mainly resources and weapons) and undermine Western influence and prestige wherever he can (seems to be going from strength to strength in fact), while remaining pretty popular at home.
Define “worked well enough”? Other than lining his own pockets, what came of it?
How can you free a people that don't want to be free? The people of Donbass want to be part of Russia just like the Crimeans did.
Says who?
Minsk was hilariously, 100% favorable to Ukraine. Ukraine had two goals at the time: stop the advancement of Russian troops and to make Russia diplomatically responsible. Ukraine got both. The troops stopped, and Russia signed the protocol that _it_ was never going to honor, which gave grounds to sanction Russia and to provide aid to Ukraine. How well is current Ukrainian government making use of this diplomatic victory is an entirely different question.
Lol, Russia was designated as responsible the instant that the first bullet was shot, which was always a foregone conclusion. What Ukraine got was legitimized frozen conflict, which I suppose you can argue is the best deal it could've gotten, but still looks pretty bad.
Exactly true on both points.
“In Donbass, the war was started by us, and the first shots were shot by us”
— Igor Girkin, translation mine
And the war is won when the other side can’t or won’t wage it anymore. When the other side is the size of Russia, inflicting a military defeat on it just isn’t an option, but 1) freezing the situation, 2) making Russia pay to maintain the situation and 3) making that cost as high as possible is a pretty viable strategy, if not the fastest one.
I think we disagree about the goals of Russia's involvement. Mainly, its interest was to protect the Russian population in the regions with significant presence thereof, which was easily achieved. If the goal was to utterly crush Ukraine, no matter the cost, that would be easy too, but that never was the goal, and still isn't.
That’s not interest, that’s a pretext. How exactly are Russia’s interests served by protecting 1M+ Russian-speaking people from their homes and 10k+ Russian-speaking people from being alive, all at a high monetary and diplomatic cost?
>So it’s reasonable to assume that Putin is willing to fuel the fire soon, as he never received any dividends from this conflict yet.
From Putin's perspective, he has. His foreign policy for the last 15 years has boiled down to "make the West look limp-dicked and Russia look like it gets what it wants". And the Ukraine situation has done that.
His goals are domestic popularity and international respect, not direct material gain. You see it in Ukraine, in Syria, in the assassinations, in putting up Snowden, and in the interference with US elections.
I'm not saying Putin *won't* make some more moves in Ukraine, of course, but if you think he hasn't gotten anything already you're not thinking in his terms.
That's true that Putin wants to make the West look limp-dicked, but that's just a mean, not the end. His will for influence and respect goes as far as restoring Russia's status as a world power, like the status it had during the Cold War. He wants to split the world into "spheres of influence" and assert his control over what he is calling "the Russian world".
Ukraine is the most important piece in his puzzle, both metaphysically and geopolitically, and he had much more control over it back in 2013 than he has now. Aside "taking Crimea home", which is mostly just a domestic victory, he is very far from achieving his goal of controlling the rest of the country. He also haven't got any respect for "Russian interests" in Europe from the US and NATO countries.
So I don't think he is interested in freezing up the existing conflict as long as those goals aren't achieved.
Thanks, Scott, for the coverage of what we’re working on at Futuur! We’re fans of ACX and its precursors so it’s great to see Futuur mentioned here :)
As indicated, we’ve recently launched our real-money markets, with trading available in various cryptocurrencies (including the USDC stablecoin for those concerned about crypto volatility). We offer hundreds of markets across all major topics, facilitated by an automated market maker. We are just getting started, so there are certainly opportunities for savvy traders as we build liquidity!
Regarding the point about our play-money markets, it is true that they can generate some odd forecasts, and we are exploring ways to address this (for instance, phasing out bets that came from inactive accounts on long-term forecasts). Overall however our play-money markets have performed reasonably well; we’ve posted a couple analyses on Medium, here: https://medium.com/futuur
And of course, we expect our real money markets to perform even better, although it’s not guaranteed. We’ve created a market on precisely this question here:
https://futuur.com/q/141989/real-money-or-play-money-markets-which-will-be-more-accurate-through-the-end-of-2021-on-futuur
Very interested to hear everyone’s feedback — positive, negative or neutral — and happy to respond to questions or suggestions folks might have, including suggestions for new markets, feature requests, etc!
> It seems like the general rule is that every month, the date when people expect normality to return moves about a month and a half further out.
Interestingly, there's a very simple Bayesian model which results in this behavior. Namely, if the waiting time for some event is exponentially distributed, and you have a gamma prior on the rate, then the additional expected waiting time after waiting for time T will scale linearly with the time waited so far. It's kinda neat, since you actually have less than one observation in some sense, but you can still compute a posterior (just using P(X > T), instead of P(X = t)).
I originally thought about this in the context of nuclear war (of course neither nuclear war or pandemics are memoryless processes, but all models are wrong etc.). If we haven't observed a single nuclear war so far, how can we estimate how likely they are? You can't, but if you have a prior it also tells you how much longer each year without a war means you should expect to wait.
(Original stackexchange question and answer I wrote: https://math.stackexchange.com/questions/4113942/growth-rate-of-expectation-of-exponential-random-variable-given-no-arrival-so-fa)
Calculating the rate of nuclear war from observational evidence gets weird and subjective, because you have to correct for your own personal anthropic bias. Any period during which you were guaranteed to die in the event of Global Thermonuclear War (e.g. via living in Washington DC) is zero evidence *for you* against GTW because if there had been GTW then you wouldn't be asking the question now, and any period during which you were significantly likely to die in the event of GTW (e.g. not in a nuke target but no stockpiles against supply-chain disruption) should be weighted less.
And yes, that means that different people get different correct answers.
One way to make a Mars mission profitable would be to make your preparations in total secret, then bet against everybody on prediction markets that someone would do it by a given date.
Of course it would be too hard to completely hide the existence of a giant, well-funded rocket company, so you'd just need to appear to be a giant well-funded rocket company that was making suspiciously slow progress, like Blue Origin. Hey, wait a minute....
They'll definitely get the last laugh when their $100 billion effort nets them the $20,000 or so that's currently tied up in prediction markets on this question.
I'm not sure I know Balaji's idea but you could try to host a debate about a topic and see which way the market moves. You could make people bet on the resolution and hold their position.
My problem with this is that prediction markets are being touted as 'ways to find the Real Truth of whatever thing is going on'. But people don't play the stock market to find the 'Real Truth of is Wiggins' Widgets correctly priced', they do it to make money. If that means over- or under-valuing the share price for Wiggins' Widgets, that is what will happen. This is how we get bubbles; everyone thinks they have secret inside information that will net them a fortune. Then everyone else sees the price of Wiggins' Widgets soaring, and they want in on this sure-fire gravy train.
So I don't think using prediction markets to settle debates or whatever will work. If I've got enough money to make it worth my while to bet seriously in a prediction market, rather than just play at it where losing won't impact me that much since I haven't wagered that much, I will bet on "what is the result that gets me the most money?" not "do I think Penny Stampe had the better argument on the topic 'Do butterflies dream of fountains?'"
If my estimation is "Penny is hot and her opponent, Dundreary Sam, is not, so people are going to vote for whatever Penny argues never mind if it's batshit insane", then I'll spend my money on "Yes" to the butterfly question. Even if I think it's batshit insane to even ask if butterflies dream. Because that will get me money.
Or conversely, I will vote on "No" because I expect Sam to lose (since he's an uggo) even if I firmly believe in butterfly sentience. Because that will get me money.
Someone has nefarious plans with that fires bet.
Can Americans simply lie about their nationality and use Mantic anyway? If it uses crypto it sounds like you shouldn't need to give them bank account details or anything tying you to a location.
Probably - but if they later withdraw the money to the US, they've now violated lots of laws.
Probably not - Mantic would be liable for violating the law. They are probably doing IP checks - but a VPN would get you around that.
The cool thing about crypto is that if you're not stupid about it, the government can't prove where it came from.
I'm constantly amazed that despite seeing dark net market after dark net market busted, and tons of evidence that the US federal government works on chain analysis and deanonymization, yet people still think that it's fairly straightforward to hide your crypto from the government. Sure, they won't burn the methods they used publicly, but they certainly have lots of other ways to claim they found you after you're charged. (Of course, you're probably not doing anything illegal enough for the NSA to care, so you'll probably be fine.)
If *all* you do is make a technically-illegal $10,000 or so, and you're careful to maintain broad and thorough operational security (which takes a lot more than just "not being stupid"), then there's a good chance that the Feds will ignore you rather than risk revealing their capabilities. But that's a lot of risk for a little gain.
I believe this is what some existing crypto markets like Polymarket do: warn Americans against using it but not enforce it in some way.
Honestly, we're still figuring out how the crypto/real-money side of things could work from a legal perspective! Our expertise is in the software side, which is why we've built out the web prototype using fake money. My expectation is that we'll keep operating that for a while, and possibly keep using play money forever (as a second option)
Balaji seems to be using prediction markets as a buzzword the same way tech companies keep putting things on the blockchain that do not need to be on the blockchain. Oxford debate has people spend some time arguing in a particular structured format then has the audience vote on who they think had a better argument. There's no window for usefully predicting the outcome (are you gonna do it halfway through the debate and predict 30 minutes in the future, or before the debate has even started when all you know is the topic and who's involved?) and also debates are fundamentally entertainment, this is like proposing a prediction market on who will die in a TV show that barely anyone watches.
"This is good because it maintains everyone’s faith in the objective process, but bad because what people actually care about is whether there will be something that feels like a crisis situation, which is hard to instrumentalize in hospital numbers. Mantic wants to lean into the subjective side of prediction markets and see what happens."
Yeah - I discussed this tradeoff more expansively here - https://www.lesswrong.com/posts/JnDEAmNhSpBRpjD8L/resolutions-to-the-challenge-of-resolving-forecasts and here - https://www.lesswrong.com/posts/KrvXNGLxhkifQCch9/systematizing-epistemics-principles-for-resolving-forecasts
I had no idea that what I needed to explain prediction markets to me was an animated dog in a wizard hat, but there you go! 😀
However, it just looks more and more like betting on horse races to me ("I am really sure that Silver Shoes will win the 3:30 at Kempton because of the going, his past performances in these conditions, and my uncle's best friend's nephew is head lad in the stables and they swear Silver Shoes will run a blinder" - now just slap a probability estimate on how you think Silver Shoes will do, and it's the same principle as "Joe Doggen will legalise black magic").
Interesting summary of an experiment showing crowd sourcing bets on horse races performed better than 'expert' bettors:
https://pdfs.semanticscholar.org/d579/d6db9d2774d2655a04cca870fc3c9fdc8186.pdf
I think this is the whole point. I don't know anything about horse betting, but sports markets in general does tend to be remarkably good at predicting outcomes, as far as I can tell much better than individuals are generally able to.
As I understand it, prediction markets are only profitable if there are enough people making bad predictions to lose money to the people who make good predictions, and prediction markets might tend to drive unskilled people out.
Presumably, prediction markets need to be enough fun to keep the fish playing. I have some evil ideas for that, probably combining the prediction market with a lottery.
Am I missing something?
If prediction markets have an overlap with gambling ones, I don't think there will be a shortage of unskilled participants. Most of the gamblers I know have been on a losing streak since the day they first laid a bet and plough on undeterred.
So the most perfect prediction market would be one where there's no point participating, as the outcomes are all perfectly in line with the predictions, so nobody makes any money?
Hmmm - is there perchance a flaw in the reasoning behind them? 🤔
Probably they'll work because very few people are always 100% right about everything, but I don't know if I would trust them for making actual policy decisions: Mayor Piecrust makes the decision to go ahead and build that park in order to boost tourism, based on the prediction market forecasts, but the reason they need to boost tourism in the first place is that nobody wants to visit the town because it's in the back-end of nowhere, there is nothing to do there, the weather isn't good enough for sun holidays/winter skiing holidays, it's just 'average nowhere town' and a park isn't going to be enough to entice visitors.
However, since everyone participating in the prediction market are the locals, who are the only ones left in the dying town, and they're all desperate for something, anything, to boost the fortunes of the town, they allow optimism to overcome good sense ("a park helped Greenville get more tourism!") and so the prediction market result is skewed to "build a new park!"
Only locals are going to be interested in making predictions about "Will a new park boost tourism to Deadend?", and if you are hoping for outsiders to help make it more aligned with reality, the only outsiders who are going to invest will be those who are hoping to make money off prediction markets, so they look for local interest markets like this, and evaluate it that "the locals hope the answer will be yes but realistically the prospects are no; since the locals are all going to vote yes, I can beat this by buying up all the no predictions".
I wasn't assuming that the predictions were perfectly correct, just that they would be close enough to correct that it wasn't worth the work to do better.
In the long run I expect that hype will drive dumb money into prediction markets in much the same way that it has for *checks notes* bad furry pixel art.
Before that, there's a lot of talk about subsidising new prediction markets with some kind of crappy algorithmic trader to get people trading on them.
"seems like the general rule is that every month, the date when people expect normality to return moves about a month and a half further out."
The manana law:
https://books.google.com/books?id=XAc_AwAAQBAJ&pg=PA157&lpg=PA157&dq=power+laws+manana+law&source=bl&ots=pdvm6M8Vaj&sig=ACfU3U0wr7f0D_xP53jriB6mpHAa30tA1A&hl=en&sa=X&ved=2ahUKEwiF8Z6HkYf1AhUtn-AKHYT6AxsQ6AF6BAgiEAI#v=onepage&q=power%20laws%20manana%20law&f=false
Neat!
If I write a really good question, what stops a bigger fish from just copying it?
It's not Monday, but I'm feeling man(t)ic.
A taxonomy of prediction markets.
2. There are kinds that attempt to predict "if we took up a collection to buy a rare-book NFT of the Constitution, could we raise $40 million".
3. There are kinds that attempt to predict "if we took up a collection to buy a rare-book NFT of the Constitution, would $40 million be enough".
7. Attempting to predict “what percentage of the US population supports <insert hot-button political topic>”, something which would be best measured by a statistical survey rather than a financially motivated system.
The Ukraine one makes sense to me. Biden isn't threatening to go to war if Russia invades Ukraine, he's threatening harsh economic sanctions. I don't think Russia invading Ukraine this year triggers an actual war between the US and Russia right away, but it does increase the long term risks of war.
I think there is a reasonably chance that every intelligence agency of reasonable budget/power that decides having the general public (and their adversaries) having access to good prediction data is against their national interest, will immediately move to shut down/discredit/hack prediction markets until the markets are unworkable.
It seems to me that it would be very cheap and easy to do.
A common problem with prediction markets is that it's inefficient to act to correct to prices far in advance of the resolution date, because of the opportunity cost of the returns that you could otherwise be earning on the funds locked up in the market. The obvious solution is to invest the market's funds to realize those returns.
It's less obvious how the details should work:
- Should the amount of investment returns payed to each participant be proportional to their purchase cost, their winnings, or the average value of their position over the period it was held?
- How should the funds be invested? Is there a way to accommodate market participants with different risk/return profiles?
- What happens when there's covariance between the prediction and the underlying investment? Doesn't that mess up the nice property of directly treating prices as probabilities? Can you correct for it?
- While we're investing the underlying funds, why not invest them in *other* prediction markets? To frame it a different way—a position in the prediction market (usually) has non-zero value and is fairly liquid, so you should be able to borrow against it to make other bets. Or, you could reinvest in the same bet and get leverage. Would available leverage improve or detract from the accuracy of a prediction? What special considerations are there for a lender against this sort of collateral?
I'd appreciate any pointers to prior work.
Yes, the problem of opportunity costs gets brought up a lot. There are some potential solutions:
1. I'm sure there are a bunch of ways it could be set up, but the natural way IMHO is to make a "yes" wager $X, a "no" wager $1-X, and then take their combined dollar and invest it. The winner gets the investment.
2. From a mathematical perspective, the natural investment would be a (mostly) risk-free instrument, so that the bets are on the prediction and not on the investment. However, risk-free instruments don't have very good returns and might not be enough to convince people to take really long-term bets. In this case, I could imagine using a diversified market portfolio, or allowing the "yes" and "no" sides to agree on an investment at the time they place their wagers (though this has the drawback that bets aren't comparable if they use different investments, due to the covariance issue you mention).
3. Even with uncorrelated investments, prices don't imply probabilities unless you assume risk-neutral investors. However, the average investor is risk-averse, so prediction markets will tend to overpredict bad things happening.* You can build a model of correlations and risk premia and try to estimate probabilities from market prices, but there's no theoretically guaranteed way to convert prices to probabilities. Ideally, people should only speak of prices, not of probabilities.
*I should add a caveat: this might not be true in the early days of prediction markets, where average investors aren't participating. For example, there might be some selection bias where risk-seekers are more common.
4. You have the right idea. Once something is liquid enough, you can potentially borrow against it and make other investments. This will generally make markets more efficient as investors have more ability to get exposure to the risks they want. However, a drawback to prediction markets is that they are (usually) binary outcomes: you either get paid or get nothing. That makes them risky collateral for a loan, so I wouldn't be too optimistic about ever being able to borrow much against prediction market investments.
My thoughts here are based on my experience working with financial derivatives, which have many of the same issues. If you wanted to do some reading, #3 is something covered in quant finance curriculum, though I'm afraid I can't point to a specific source.
I was wondering if someone could explain something to me.
How does a question on a prediction market generally get started? As in, how does the exchange itself sell the initial "futures" to predictors, like an IPO? They need to sell them for a high enough price to pay out, on average, to the holders of the futures if and when the event takes place. How do they ensure they achieve this with enough of a safety margin?
Is there some clever trick pairing off the "yes" and "no" markets together? Do they act as a normal market participant (would this not require the exchange themselves to partake in prediction and need to do research)? I've thought about it for ten minutes and haven't come up with a low-risk mechanism that the exchange can use to get a market started, but I'm sure I'm missing something.
Assuming that: (A) large numbers of people will soon be displaced by automation, thus making something like UBI necessary to avoid pesky bread riots and whatnot, and (B) if large and liquid markets are generally the best available (albeit imperfect) predictors of future events, even taking "dumb" money into account, I am trying to think of a way to tie prediction markets to UBI.
Each participant gets a certain amount of money at the beginning of each month, which they are required to use to make prediction bets on one or more matters of social interest. The amount the participant receives at the end of the month, above a certain minimal threshold, is tied to their success in making such bets. For bets which don't or cannot resolve in the next thirty days, the participant can either let the bet ride for the next thirty days or until resolution, whichever happens first, or the participant can cash out at the current bid-ask price.
The participants get UBI and something potentially socially useful to do (and that has the potential of an additional payday), and the mechanism may be more effective than paying consultants, futurologists and analysts. Also, participants which exhibit consistently successful track records may be selected to be groomed for Bigger And Better Things.
I just thought this up because I don't feel like working ATM, so tell me what is wrong with this idea?