(github.com)
They admit no returns.
But it does seem like a fun project and nowhere does it say anything about returns or profits so not scammy imo just funny meme backed code
The bot has zero risk management and I have a strong disclaimer on the github it is essentially a meme.
73% of all polymarkets do resolve to No though.
There's a good dataset on huggingface if you wanted to do some data science
There are actually two theories on insider knowledge. One states that allowing insider trading is beneficial, as it allows prices to better match the underlying reality, the other states that this discourages non-insider trading, which actually makes the prices worse. Stock markets lean heavily towards the second theory, while prediction markets seem to be leaning towards the first.
--
[0] - Or external knowledge, but actual knowledge - thinking of hedge funds stalking CEOs as they fly in private jets, or counting cars in parking lots from satellite photos, to get some probability estimates on factors actually relevant to the performance of a business and possible future events.
Obviously it has come a long way from that, and the markets have become more like gambling. You could probably allow insider trading at this point and the system would continue just fine.
The stock market would not be noticeably less liquid if people had to hold stocks for 24 hours, but volume would drop like a rock.
Prediction markets are doing a bit of that. Some won't take bets on an assassination.
If they get assassinated, those markets will resolve to yes. At least the rules don't specifically exclude that.
This theory is fundamentally not credible, the other side of any trade you make on the stock market is essentially always going to be vastly more sophisticated than you. Insider trading makes zero difference to the end-user.
The credible argument against insider trading is that it's a form of theft. You are making trades based on information which does not belong to you, and which you have an obvious duty to protect. You are essentially stealing from the people you work for.
>These are obviously false.
The purpose of the Ukrainian military is to exhaust the Russian army's materiel and test out our weapons. "Years-long stalemate with Russia? Yes, please." -the US. Seems like an overwhelmingly common Scott Alexander L.
Since when does country A decide what the purpose of country B's military is?
On smaller scale, this is the (in)famous "fire and motion"[0], which works in business as much as it does in military tactics. Make a move, forcing competitors to respond to it. If you're better at it than them (and lucky), you can choose your moves to make their responses go to your advantage.
--
[0] - https://www.joelonsoftware.com/2002/01/06/fire-and-motion/
In any case, the blog is well regarded in these circles.
The original theoretical purpose was to incentivize the creation of new knowledge, not reward insider knowledge that already exists. For example, to fund research that helps answer some unanswered question.
Today, the purpose is obviously gambling. We can see that clearly from the marketing.
There's a ton of scholarly debate about it, and at least most of the early stuff was pretty earnest. But rather than the debate becoming more refined and nuanced over time it seems to have bifurcated along partisan (or partisan adjacent) lines like everything else, similar to the Keynesian/Misesian divide.
The Misesian folks are a lost cause, IMHO. They're hardcore rationalists, self-indulging in circular moral arguments from assumptions that don't apply in the real world.
The question isn't what percentage of bets resolve to no, but whether there is a consistent bias in the prices away from the fair price, which has an expected value of 0, and what direction that bias is in.
I'm sure in the near future, policy decisions or war strategies will be decided by prediction markets' odds, if they are not already being used.
News sources are motivated to get clicks, to appeal to certain audiences, and to retain tribal customers. None of these create incentives for truth. You can seek out smart, well-informed and principled journalists who will prioritize truth-seeking over money-making. There are some. But the fact remains you are relying on character to override incentives. With prediction markets, incentives and truth are naturally aligned. This makes them a powerful and valuable resource imo, even if there is a lot of scumminess that comes along for the ride. The insiders, more than anyone, are contributing to the truth signal.
So the feedback from prediction market turns around, so you can essentially buy events if you put enough money in.
But also motivated to bend the truth to their bet as the journalist in Israel found.
See the loop?
First, you have inside traders. Then, among legitimate bettors, you have smart people using multiple data sources (not just the "news") and doing sophisticated analysis that most journalists cannot do, and are not motivated to do -- again, because their incentives are different.
You can do research to know the US would strike, there's no other point in moving multiple carriers over to somewhere. But exactly WHEN is not researchable. This applies to most other bets. So lets stop pretending there's anything more than 2 cohorts, insiders and degenerate gamblers.
I've been doing it profitably myself for almost 10 years now. I have zero special inside knowledge, and no access to any other non-public information.
> Will the US strike Iran by X date
Last year I did think the market for a strike on Iran was significantly underpriced given the information and conditions within a specific frame of time.
I don't think every smart person can just pop into prediction markets and print money, but I know many smart people who are long-term winners. I also don't try to knock people as degenerate when they have genuine talent.
You though, are claiming that the market is perfectly priced, or should be, such that this strategy won’t work. It’s pretty hard to balance the odds of an animal seeing their shadow vs the expected strike price of the nasdaq. It’s clear you’re not familiar with betting markets, which is in your best interest most likely, but that’s not how this works.
You’re arguing against yourself… against a point nobody made but you.
I bet the average price for a no bet across these markets is 73 cents.
Whether there is enough of a predictable bias there to snag enough low return high probability bets to beat the vig and not shift the markets I have not looked into in any way,but it is a known bias with them.
The real money to be made in prediction markets is being the ones with the actual knowledge which is arguably why they are useful and why for some topics, people find them abhorrent.
If the market isn't resolving soon, the small return might not be worth it.
From their docs, it looks like they charge fees to bet "takers" (as opposed to makers), but exclude the geopolitical and world-events markets where they don't charge fees.
I have to imagine that may be related to some of the blow-back towards prediction markets about profiting on topics like war & their potential for manipulation.
Given it sounds like the bot bets everywhere other than sports, many of those categories would likely have fees in this case.
Or something like that.
Edit: conversely, if the average no costs _more_ than 73 cents, but the 73% of all polymarkets resolve to No, that would imply that an everything-always-happens strategy is profitable (neglecting slippage)
Or just the bid-ask spread; price no at 73.25 and yes at 27.5 and you have a profitable but theoretical mid-market price.
Behavioral economics has already answered the question of whether humans are, on average, perfectly rational economic actors. They are not.
To the contrary, there is substantial evidence indicating a meaningful number of humans will mis-estimate the likelihood of uncommon future events.
The alternative would be that there's a bunch of free money sitting there waiting for someone to decide to pick it up, and nobody is, not even you.
The whole premise of gambling is that they don't
> [...] if prices perfectly reflected available information, there is no profit to gathering information, in which case there would be little reason to trade and markets would eventually collapse.[2]
That's a stupid way to formulate this. Markets wouldn't "collapse". They would get slightly less efficient until equilibrium is restored to where arbitragers can make enough money to keep prices at that level of efficiency.
Meanwhile Two Sigma is hiring alpha quants to be AI research scientists at $250k starting salary + bonuses.
Even if we're just talking about the HFT/sell-side, there clearly exist various anomalous inefficiencies that can be exploited.
Fama's guy doesn't agree either [1]
https://www.ft.com/content/813b3d76-6ef1-427d-a2e0-76540f58a...
You don't believe in the existence of residual return orthogonal to priced cross sectional risk factors (alpha)? E.g. Trends, momentum, volatility clustering, etc. many easily demonstrable inefficiencies. VPIN and order flow toxicity are highly predictive features. Most HFT MM especially in crypto involves hybrid alpha in addition to the (visible) bid-ask spread, which it itself an "inefficiency" to compensate market makers like Jane Street and other successful firms that operate on the assumption that weak form EMH is not accurate.
* https://www.kaggle.com/competitions/jane-street-real-time-ma...
Successful project imo.
I wonder what it means exactly. Typical Polymarket looks like this:
X happens before May. [Yes][No]
X happens before June. [Yes][No]
X happens before July. [Yes][No]
...
So even if X ended up happens in December, it's still 12.5% Yes and 87.5% No?
So it's not a useful trading strategy. Good to know.
It might have worked out that the human tendency towards optimism biased the Yes side, but Polymarket is watched closely enough by traders that the pricing is apparently realistic.
Now if you could bet against minor crypto coins, which almost always go down... But if you could, there would be traders pricing them realistically. Everybody has analytics now, and mispriced markets are detected and exploited quickly.
I laughed. That's inspired. Quite the nerd-snipe as well, based on the rapidly accumulating threads on effectiveness, probabilities and markets.
Like with this bot, I have no idea if that will still lead to actual positive returns. This might just be a remnant from a time when these betting lines were set less intelligently. But all things being equal, it seems logical that "boring" bets would have a better return in the long run than "exciting" bets as long as some betters are at least partially motivated by entertainment.
There's probably a lot of knowledge like this that sports betters have built up over decades that could apply to these new forms of non-sports gambling.
Here's how the mechanism works: I find that something is statistically worth $0.70 but I am able to buy it for $0.60 and statistically sell it for $0.70 (in the average). I make $0.10 each trade on average. Until you come along, copy my strategy and change $0.60 to $0.61 to frontrun my trades. Then someone else does it for $0.62. Until the market finally reprices to $0.70 where it should be. The guy who tries $0.71 loses money and stops, and then it goes back to $0.70. It's a stable feedback loop.
There are lots of positive EV strategies lying around in these inefficient markets that Citadel hasn't (yet) descended upon. The best advice I can give is if you find one, trade the hell out of it and don't open source it or tell anyone about it, because as soon as more people run it, it will cease to be positive EV and then after that it becomes an infrastructure game.
If it's popular on Github it probably doesn't work.
If you found something that works and is paying your rent, don't put it on Github. My 2 cents.
Your strategy doesn't work on sportbooks to begin with, because bookmakers don't move the odds with the action.
That is, there is no such phenomenon as "the over is exciting therefore overpriced". Bookmakers price purely based on facts and statistics. Their pricing isn't affected by excitement nor by how many people are betting a certain way.
A bookmaker is a market maker, and they ideally want to end up with no net interest in a position. They then take guaranteed profit in the bid-ask spread, which in sportsbooks is the 'vig'. Bookmakers who adjust their odds in real-time don't have to be particularly clever about the fundamentals, just responsive to the competing demands on either side.
A bookmaker who intentionally takes a position on a game is the equivalent of a proprietary trader or hedge fund. It's potentially more profitable, but it's also adversarial against 'sharp' traders.
Bookmakers who set odds at the beginning and don't move with the action must set larger bid-ask spreads to compensate.
If this were true, lines would never move unless there was breaking news, but we see lines move all the time without there being any material change to those "facts and statistics".
Without there being any material change you can see. If you had access to all the tips and data and insider information that sportbooks operate with, you could be a bookmaker too.
Can you give an example of what you're talking about here? Because it sounds like you're accusing these large publicly trade companies of participating in organized crime. There is regulation when it comes to sports betting that doesn't exist with general prediction markets. An athlete can't just feed a sportsbook "insider information" in the way you're suggesting. The only private info that they are supposed to have is better behavior details that you claim doesn't factor into their decisions.
Where did that entity get that information, and how are they right so often? Your guess is as good as mine. I'm not accusing anyone of anything.
Well that's the source of our confusion then. I agree with this, but it conflicts with what you said a few comments up:
>because bookmakers don't move the odds with the action.
Different from a prediction market like Polymarket or Kalshi whose income probably comes mostly from transaction fees rather than house edge. Otherwise these platforms wouldn't welcome bots so much. Bots => efficient pricing + transaction volume => profit for them
Yes, you can find positive EV trades on Wall Street as well. I've been diving into this a lot lately, all I can say is it's 10X harder to find strategies that work on Wall Street than prediction markets.
With one exception.
The one easy long-lasting anomaly to exploit on Wall Street which actually does NOT exist on prediction markets: "American stocks go up most of the time". This is actually a massively exploitable structural anomaly (you just buy and hold forever and effectively reap the rewards of a biased coin) and the source of the anomaly is mostly US monetary policy, US foreign policy, and US tech expertise put together. However, it still is an anomaly. The thesis that SPY will continue going up forever is also predicated on these things continuing to work the way they have in the past.
Also requires a lot of volume to be "predictable" obviously, since 1 loss sets you back 10-20 wins. It's surprisingly hard to find reasonable-liquidity markets after all your filtering. Many have huge spreads or thin books. Scare quotes around "predictable" because you never know if others will use this strat or a lot of unlikely events will happen due to insiders.
Another thing, just like the author, I was excluding sports in all the above. Yes Polymarket is famous for letting people bet on world events etc, but turns out it's still more about sports. Betting on the overdog in sports markets seems more appealing because there are plenty of those events with large volume, they're kinda homogenous, you know exactly when they resolve, and they're harder to rig. I simply never got around to putting real time or money into the overdog strat.
didn't look at the numbers, but this one sentence reminds me of selling options for 'passive income' (don't do that)
Polymarket is also holding onto the money in the meantime. Idk what they do with it, but it's not like some other platforms where they at least work with a bank to earn you some tiny interest on it.
Good old eat like a bird, poop like an elephant.
I think timing is the missing piece of this. Just randomly betting no on everything likely isn't going to give good results, but if you tied in a news API and just bet no on anything related to a major story right after the news starts picking it up, I would expect you could make a solid return.
I've had success playing the markets in these specific cases. I did fritter away a lot of my gains from the financial crisis thinking I was a genius market timer. But I learned my lesson and didn't waver once I jumped back in after covid.
In both cases I got out before a bulk of the crash and timed the bottom almost to the day. Lucky I know, but I had reasons for both. For the financial crisis it was when Bill Fleckenstein closed his bear fund and put it all in MSFT. For covid it was when it looked like the lockdown was working and NYC hospitals weren't going to completely fall over like Northern Italy or Wuhan.
For any non black-swan scenarios, I assume I'll never get one up on the masters of the universe and just leave everything in blended age-appropriate funds.
I'm very concerned about an AI crash and the future of white collar work in general. But it feels more like a slow death to me than a black swan. So I'm just hedging with bonds and cash and stocks that hopefully don't crash as hard in a recession.
One of my go-tos on this is the Fukushima nuclear accident. IIUC there were plenty of folks in Japan who knew of the high risk. Perhaps many interested in nuclear energy outside of Japan, too.
But the average adult if asked about the prospect of a major nuclear incident occurring say, "tomorrow," would narrow their eyes in skepticism. There's almost an instinctual level seeding of doubt.
This can be a good thing. LK-99 was an excellent test of the dissonance from dramatic changes in reality and costs of inaccuracy.
The greatest VCs I have known are exceptional at suspending disbelief to test their ability to basically shape world building.
Consider this bot running on us military outcomes or something.
It's really no different than a casino: if you ever find yourself with more money than you walked in with, cash out and leave.
Best strategy for most people though is to simply not participate and you'll break even.
It gives us normies a way to see what the powerful are thinking.
And, I'm not even contemplating gambling addiction. There's a huge market of people that just go to Vegas once or twice a year and come home thousands of dollars poorer. But they don't need it, they may not gamble outside of Vegas, or nothing that would signal an addiction.
If Polymarket were regulated like a casino, I’d actually have no problem with it.
Weird way to validate polymarket.
Again, no idea if anyone sees this as a true substitute or not. My guess is not as Polymarket bets don't feel entertaining at all (IMO). So it's not filling that void for anyone, but it hypothetically could.
You can make money off of all sorts of stuff. You can "sell" the bets, so there's lots of live pump and dump.
We've gone full circle. The bookie with no neck that smelled like onions was more honest than these platforms.
But isn’t weird the betting platform is sending an app notification saying “hey bet on this dude to win $X”?
Joe Dart elected president Y/N
Cory Wong elected president Y/N
A no bet on Joe Dart is not a yes bet on Cory Wong.
It is trivial. Saying NO to a candidate means you're saying YES to ALL other candidates with varying probabilities that would sum to the neg risk of that NO bet.
Perhaps this is pedantic, but this equivalence is ignoring fees, spread, and slippage, right?
If you don't think it's true, then go ahead and arb polymarket for all the incorrect pricing.
> yes this has to buy below 0.73 long term, the bot has a configurable ceiling set at 0.65 and checks for new markets buying closer to .5
What other question would you like to be backtested? This one is fairly easy
For example, for markets that are between 60 and 70, is it the case that around 65% of them resolve to yes?
I guess you want to take a certain time before out finishes, so focus on sports.
If you put all your money on no, you get 4% if you win, and if Jesus comes back and you lose, money won't matter.
This locks up your money in the meantime, right? If so, considering the fed funds rate is 3.64% (and you can probably get higher rates on stablecoins), a huge chunk of those "winnings" is going to be eaten up by the opportunity cost of the money.
For example, recent events show that any bet can be selectively disputed by arbitrary reason ("we found insiders", "we found this immoral/illegal", etc.).
And for perpetual events - there is not a single week without a hack (https://www.web3isgoinggreat.com/)
Who does "you" refer to in this sentence? Polymarket itself?
I'm pretty sure if Polymarket itself decides it wants to screw you, you're gonna lose no natter what your strategy is.
Has anybody looked into the repo in more detail? I imagine it's useful for infrastructure inspiration to build your own bot pursuing more differentiated trading strategies.
> Why predict the future when 73.4% of all Polymarkets resolve as No?
(Manifold doesn't use real money, so there's more "free money" lying around waiting to be picked up than on most real-money markets)
The author [page](https://github.com/sterlingcrispin) is there on github, but I can't even find his full list of his repos to confirm it's still there (I also get a 504 on that).
Say 70% of the time it resolves to ‘no’, you still don’t make money by blindly choosing ‘no’.
Guess why?
Hint: This strategy is also described with the macabre analogy: picking up pennies in front of a steamroller.
Do you want to pick up pennies in front of a steamroller?
This or any other statistical play is only 'in front of a steamroller' if you do it with leverage, especially if notionally uncorollated bets suddenly move together. Bets on Polymarket have limited downside by design, and bets in different categories are obviously unrelated to each other.
Without having looked into this in detail, however, I suspect the problem would be limited capacity; markets that are both deep and so trivially irrational are probably fairly infrequent. You might pick up pennies but only pennies.
It’s the 90% chance of making 1$ vs 10% chance of losing 100$.
The exact numbers vary, the expectations even out with high volume stocks but prediction markets do not because of rounding that favors the house.
My point is equally valid for "markets consistently value 'yes' too highly".
The occasional 'no' will wipe you out on average.
But your statement seems stronger, i.e. that such a strategy is somehow fundamentally and inherently impossible, so I think it's on you to explain why that is supposedly the case.
For example, assuming a hypothetical consistent "yes" bias of 10%, would you still say it's impossible? Why? Basically, are you saying it's impossible because of the actual observed "yes" bias being too weak or for some other reason?
Though I agree it's bad math, even if 70% resolve to no, there's a high variance among all of them, and to know whether it's a good bet or not... you have to do your DD on that particular market. Even if you follow the Kelly criterion, randomly choosing bets will probably tank your bankroll sooner or later.
No, all these variables cancel out.
If you were picking and choosing, yes. But this approach is basically betting no on all the markets.
The textbook explanation of this is the central limit theorem, proving this mathematically is a bit more involved for power-law systems like this but it’s empirically valid.
nice to see heroku still alive...
https://huggingface.co/datasets/SII-WANGZJ/Polymarket_data
"A comprehensive dataset of 1.9 billion trading records from Polymarket, processed into multiple analysis-ready formats. Features cleaned data, unified token perspectives, and user-level transformations — ready for market research, behavioral studies, and quantitative analysis."
And the chance of losing at least once in a 99% sure bet after 100 rounds is around 60%. Even if you reduce to 30 rounds it still is around 30%.
This may seem smart at first glance, but the math doesn't really checks out.
But they aren't independent there are a lot of correlations. Global geopolitics for example.
The way the math works out, 73% of markets resolve to No, If you buy No at 0.73 each time you would break even.
Not financial advice of course
The stopped clock is right twice a day, but it reads noon and we're at half past three
I have been making money with a bot off a statistical anomaly in prediction markets lately. There is no way in hell I will open source it or tell you what that anomaly is because I have capacity back-tested it and there are so many players in the market; if all of HN and Github start downloading and use my code it WILL cease to work.
Put another way, your orders are helping move the market and price the market more efficiently; that's the market compensating you for pricing things better. If a thousand people run your strategy, prices will get moved to exactly the point where your strategy stops working. You effectively split that pie with a thousand people.
you wouldn't get it
Nevertheless, Polymarket is a very interesting marketplace of sentiments and information, and it can be a very strong leading indicator of huge price movements in "real" markets like the NYSE, in part because it directly measures one factor of sentiment, i.e. whatever the prediction is about. Market sentiment determines market prices on very large and deep markets, too.
In the run-up to the election, when Trump was running against Biden, a betting market was a leading predictor of NASDAQ (a very deep, very liquid index of stocks). I wrote up the findings here: https://medium.com/@rviragh/does-the-stock-market-react-posi...
This indicator was the best one anyone has ever shown for NASDAQ for any signal, period. The signal was so strong it trumped all other signals and variances of any kind. Traders trading with just this signal and no other signal of any kind could have made practically an unlimited amount of money as long as the signal was intact. (Basically, until Biden dropped out.)
I myself didn't place any bet due to my role as a political advisor at the time, but the size of the correlation is still the biggest and most surprising one I've ever seen.
Same way Binance did [1]. Assuming they wouldn’t get charged.
> Restricted Jurisdictions. You acknowledge and agree that you are not permitted to access, use or trade with the Contracts on the Platform if you are residing in, a citizen of, organized in or located in the following jurisdictions (collectively, the “Restricted Jurisdictions”): Australia, Belgium, France, Germany, Italy, the Netherlands, Ontario, Poland, Quebec, Russia, Singapore, Taiwan, Thailand, the United Kingdom, the United States; or a jurisdiction or territory that is the subject of comprehensive country-wide, territory-wide, or regional economic sanctions by the United States, including but not limited to Iran, Syria, Cuba, North Korea, and the Crimea, Dontesk and Luhansk regions of Ukraine.
Maybe I'm supposed to ignore that?
But also it’s not illegal for a US citizen even before it was simply not legal for them to do business with US citizens because of lack of kyc
[1] https://en.wikipedia.org/wiki/Patrick_Crusius#In_popular_cul...