The problem with predictions markets is that they incentivize participants to distort the market in order to make their prediction come true (and thus receive a payout). It quickly stops being a place to gauge the likelihood of an event, and turns into a roundabout contract bidding mechanism, the most famous example being "the likelihood that so-and-so famous politician will be dead by next year" is just an anonymized assassination contract. There's a reason that insider trading is illegal; prediction markets are inordinately susceptible to Goodhart's Law.
Goodhart's Law for those who don't remember: any measure used as a performance target ceases to become a good measure.
Or put another way: when people can maximize their profit by manipulating the measurement or market at least some of them will attempt to do so.
Stock markets are more difficult (but far from impossible) to control. To make Target's numbers come in low you'd have to suppress their retail sales or their supply chain. Whereas who wins X Race is a singular event that only requires one person (in the most optimistic case) to participate thereby controlling the outcome. In a sense a lot of what prediction markets make book on amounts to penny stocks and are just as risky. If you want to make such markets widely available you need regulation on what kind of events allow bets and strict rules against insider trading (eg sports participants and their families must be prohibited from participating).
Insider trading is illegal because the non-insiders would stop trading otherwise.
Prediction markets have some problems but this isn't the biggest one. One can even use it for good. If a company gives its associates stocks but prevents them to sell for some time, this is essentially the same mechanism.
For example, we could make prediction markets about upcoming policy and laws and their impact. Then we can tell the policy makers to bet that their proposed changes actually have the good outcomes they promise.
Real estate sort of has insider trading laws, but in the opposite direction. If you know your building has a defect, you have to disclose it to any potential buyers. It's not exactly the same as a ban on insider trading, but it basically prevents the sort of insider trading where you know something bad is going to happen and so you dump the stock. Also there's a technical point: "insider" has a specific meaning (ie. an employee having material non public information). For real estate, commodities or crypto, it basically doesn't exist.
Not so, I would just buy whatever the insiders are buying a bit later. The argument against insider trading is a bit weak because it seems impossible to enforce effectively and I don't see why anyone would assume insiders are anything but active in the market.
Picking on stocks as an example, there is a background fair market rate that everything should theoretically achieve. Insiders way outperform that - but most people are in the market to target the background rate, not the outlier rate insider traders can achieve.
My favorite part of insider trading is that it only punishes those who take action. You buy or sell. If you have insider information that prevents you from buying or selling, that’s just fine (at least that is my understanding).
>The problem with predictions markets is that they incentivize participants to distort the market in order to make their prediction come true (and thus receive a payout)
As opposed to the stock market?
>There's a reason that insider trading is illegal; prediction markets are inordinately susceptible to Goodhart's Law.
Insider trading is only illegal because you're abusing your position of trust as an employee. You having an incentive to make the stock go higher is totally fine. In fact that's how activist investing works. You buy shares in a company, use that to get control and turn the company around, and sell your shares in the now more valuable company.
The reason governments finally have gone hardball on crypto was the realization that it created an incentive for people to bet against and ultimately attack the shared economic system.
Silicon Valley bank run triggered by investors with well known crypto exposure was only the final straw
yes, as opposed to the stock market; if i go buy AAPL calls, i can’t expect to manipulate apple’s share price by, say, buying a bunch of iphones.
conversely, i can’t buy puts, hoping that my stock sell off will effect them being in the money at expiry.
something something market forces and what not, but the long and short of it (no pun intended) is that.. if it were profitable, people would be doing it.
insider trading is illegal for numerous reasons, “abuse of trust” isn’t wrong, it’s just wholly incomplete.
In theory the same thing is true with horse race betting; you can just assassinate the horse. Or the jockey. Or if you want to be more subtle about it, just slip something into the horse feed. But this doesn't really happen in practice.
Match fixing happens regularly, so regularly that we have the term "match fixing" with a Wikipedia article dedicated to it. The presumption that match fixing doesn't happen in practice is unfounded.
It's a question of scale. A few horses aren't a big deal, and both sides (the horse racing, and the bet) are heavily constrained in scope. Both sides become far less constrained with prediction markets.
Maybe if we were objective, horse betting should be shut down -- but we kinda don't care.
Which comes back around to the original comment: the existence of a predictions market influences the outcome of the event being predicted, and in the worst case this effect is so large that it completely invalidates the purpose of attempting to predict the outcome in the first place; the tail wags the dog.
If prediction markets could tell the future one would imagine the first person to harness them would now be the most wealthy and powerful person to have ever lived.
>If prediction markets could tell the future one would imagine the first person to harness them would now be the most wealthy and powerful person to have ever lived.
1. It's unclear how you or any person in particular would be "the most wealthy and powerful person to have ever lived", because other people would also be able to exploit it
2. Many predictions are hard to act upon, even if you know they're accurate.
Despite the pearl clutching on HN about assassinations (killing a CEO after shorting stock is a very similar profit outcome, yet no one does this because that’s totally insane), prediction markets have been fairly popular for a few years as blockchain tech has risen. And the sky has not fallen!
I would like to test whether prediction markets are worse than they were during zero-interest land. The incentive to create sophisticated models to get low percentage points of edge is much lower today than it was before. Why risk time and effort trying to predict the 2024 election for a few pp of edge when you can… just buy a 1y tbill instead? Sure there are probably better ways to express an opinion on these types of things too. Options that favour a team red or team blue win come to mind, but no matter what you’re still competing with 5% yield.
I think this is solvable by having some kind of betting token that itself accrues interest over time to reduce the opportunity cost but if regulators already don’t like prediction markets boy they especially wouldn’t like that.
This is indeed a problem for long-term markets but long-term is more about 10 years and longer here. Something like "Will China invade Taiwan before 2040?" is a long-term example.
Prediction markets are nearly 50-50 for democrat vs republican president currently. That is a 100% yield if you are correct. It isn't about the yield. It is about much more accurate you are. If your oracle gives you 60-40, you will still lose lots of bets but you will easily beat the 5% yield of treasure bills.
While the game-ability of prediction markets is a major concern, I believe that the potential of prediction markets to further exacerbate inequalities may be a greater threat. It is important to keep in mind that while markets optimize for total value, it doesn't guarantee that it will optimize for things we as individuals want.
Free markets tend to lead to a positive feedback loop. Someone who has a lot of money will be able to predict the market better than someone who has less. Their advantage generates more wealth which helps them to predict the market better. Wealth generates more wealth.
In theory, this is offset by the notion that the market is not a zero sum game. The person may be generating new, previously unavailable, wealth. But while stock performance is tied to the success of a company (which can generate new wealth), prediction markets are tied to transient questions (where intrinsic value diminishes the closer you get to expiry). In this sense, how "predictions" might generate "new value" is unclear.
I'm no expert, but I assume the difference is the underlying asset that you're betting on. Stock vs commodity price at given time vs "event coming to pass or not".
They work a lot like binary options, which can fool a lot of unsophisticated investors. The line is a lot thinner between prediction markets and gambling.
Prediction markets and bookmakers really strongly expected a Clinton win in 2016. Irish bookies were so sure Trump would lose that one firm began paying out to those who bet Clinton even before the election.
And we all know how that turned out.
Nov. 7, 2016:
"Hillary Clinton’s odds of winning the presidency rose from 78% last week to 91% Monday before Election Day, according to CNN’s Political Prediction Market."
You can go look at the old Predictit market if you like - it's still around[1]. You can also check election betting odds[2]. Clinton hovered around 60-80 in the week or two before the election.
Even if it were 90%, I'm not sure that's an indictment of prediction markets. A probability of 90% means that 10% of the time, the other thing happens. It's a bit like saying "wow, it's cold today; I guess global warming isn't real." You'd want to to an analysis across many many prediction markets in order to see if they're generally accurate. That already exists, it's here if you'd like to look at it[3].
We have no way of exploring the multiverse or running the same election more than once, so there isn't any evidence that the payout from a prediction market matches the probability of something happening. It's just collecting the predictions of people who are willing to bet money, weighted by how much they spend, and calling that the prediction. I don't think rich people are necessarily better than not-so-rich people in predicting something not related to their expertise, but prediction markets are weighted according to how much was bet.
Is this a response to the metaculus track record, which is an aggregate of the accurate of all predictions ever made on the platform, and shows that predictions fall close to accurate probabilities? Which part of the track record do you find fault with?
Their FAQ says that they aren't a prediction market, and they operate differently. People aren't placing bets, so it isn't weighted in favor of the wealthy. They weight people based on past prediction success.
#3 makes it look like things predicted as "very likely not" (10-20%) and "very likely so" (80-90%) tend to be biased towards the contrarian view. Does Metaculus have limits that make it unprofitable to make bets things that would push them outside the window? Anecdotally I've seen quotes from prediction marktets that show a surprisingly high chance for things that are incredibly improbable (to the point that I would call them "impossible" in casual conversation).
I've been snagged by a few markets that were in the high 90s or low 10s and looked like free money, but ended up resolving the other way. It's usually either 1) there was some subtlety in how the market was written that causes it to resolve unexpectedly or 2) a true inversion - it just seems like human nature to under-predict how often they actually happen.
In this case the markets are overpredicting inversions; there's at most a very small difference between how often markets predicted at 5% and markets predicted at 20% resolve as "no." Same for the 75%-95% range.
Same response I gave the other guy - you can go look up the market here rather than conjecturing [1]. It was about 25% the day before. In any case, it seems very silly to discount prediction markets because the probability didn't work the way you expected it to in a single event. The whole point of probability is that it's probabilistic!
If that were true then the results of prediction markets wouldn't closely track the actual results that happen, as the link I provided shows that they do. Also, you could be making tons of money right now! How's that going?
It is something of both. Like playing Poker can be gambling, yet one can also play it seriously and systematically earn money.
The difference is that a Poker tournament is just entertainment for the audience. With prediction markets there is the very useful side effect of accurate forecasts.
