Transcription
In 1987, a desk at Morgan Stanley reportedly made $50 million in a single year, which was huge for the time. And that was because a Columbia University computer science student figured out pairs trading for the first time. This was super lucrative for Morgan Stanley for a short stretch of time when they were the only people who knew about it and the only people capable of doing it with their computer infrastructure.
But then just a couple years later, the edge completely tapered off and evaporated. This is because traders at Morgan Stanley left to create dehaw and two sigma and some other giants. And as more capital flooded into this trade, it got crowded and the edge evaporated.
This phenomenon of an edge evaporating after it's publicized is not a one-off at all. And actually, that's the way that markets are supposed to work. You're never supposed to have an edge. If market prices perfectly represented the value of an asset at every single moment in time, there would be no way to get an edge. That would be an efficient market. And assuming that you're trading a real edge, every single trade you take is bringing the market closer to efficiency. If you have an edge that accurately tells you the price of an asset is too high, you short it and that brings its price a little bit closer to what it should be. If you have an edge that tells you an asset is too inexpensive, you buy it and that brings the price up a little bit closer to fair value.
As more people find out about an edge, they're going to crowd into the same trades until the price of the asset converges to the perceived fair value. This happens with edge after edge after edge after edge after edge after edge after edge after edge after edge after edge. And the mispricings get smaller and smaller and smaller and smaller and smaller. This will happen until the edge is completely gone and the strategy no longer makes money.
And this brings me to an interesting point. There are lots of edges that you can find today if you analyze them in a really, really naive way. You could look at a super basic and antiquated pairs trade and analyze it as if you were trading a strategy that had no slippage and no transaction costs or something like that. That's not realistic at all and think that you could find an edge there. But the truth is, of course, that with any well-known trade like that, the moat is latency. If you're analyzing some public and obvious strategy and you think you have an edge, most likely you're just analyzing it in a way that's not comparable to the real world, for example, no slippage, no transaction fees, or you're seeing instantaneous mispricings that are arbed out by high frequency trading firms.
This brings me over here to the idea that an anomaly or an edge will not actually attract arbitrage capital to arbit it out if it's not really profitable to trade. As the authors of this paper point out, you see this a lot in academia where authors of academic papers will say that they've found an edge but neglect to consider transaction costs or slippage or things that you encounter in the real world when you actually try to capitalize on that edge. So if we estimate the effect of statistical bias to be about 10% in research papers that claim to have found edges in financial markets, then this paper which analyzes the effects of publication on the returns of edges which were published estimates that they decline by about 25% after publication. Meaning a private edge returning 100% per period would lose 25% of those returns after publication with this 25% being due to other traders who started trading the published strategy. And any remaining returns of these strategies are likely just artifacts of the things that we talked about before.
This is why real profitable proprietary trading strategies are hidden from the public by all sorts of legal protections. And breaking these legal protections to trade with or publicize a private trading strategy is punished very harshly. This was unsealed in January of 2025. A research developer and quantitative trader at a quantitative trading firm ran away with the source code from that firm. He tried to trade on his own using one or more of their proprietary trading strategies and the FBI came after him and got him.
So, if keeping your edge private is so critically important to it not decaying, then why are all these YouTubers posting their strategy for the entire world to see? Why does TJR post a new video every couple days sharing his entire strategy in depth? any edge he has would surely decay almost immediately. And the common retail rebuttal to that is, "Oh, quant firms wouldn't want to trade these strategies. Their position sizes are too large, and they wouldn't be lucrative for larger firms." That's not true. TJR and these other traders trade some of the most liquid securities in the entire world, sometimes with more than a trillion dollars traded in a single day. The strategies do not require sophisticated HFT infrastructure and could be fully automated. I am fully aware that these strategies aren't mechanical, but that does not mean they can't be automated with neural networks. it would be very easy.
But let's actually drop the idea of institutions trading these strategies at all for one second. How have TJR's 1.2 million followers not armed this out yet? If they do have a positive edge here, they're executing, which means they're pushing price closer to what should be its true value, right? And people can pile into these trades and scale and scale and scale until price converges to its perceived fair value. So, if the strategy is working, why haven't they? Well, maybe they are fully extracting some alpha and then the battle becomes latency like we talked about before. So assuming TJR strategy works, sharing it would make it a battle of latency like we saw in those dozen examples that became dominated by HFT firms. HFT firms with colllocated servers and specialized fiber optic cables for faster connection to the exchanges. But TJR trades from his bedroom in Bickl Miami on his phone. So what's going on? This isn't just another day trading course. This is the day trading blueprint that helped me go from a literally the entire trading industry. Maybe it has something to do with his high ticket course that you need to book a call with a closer to purchase.
The point of the video is this. If you don't create the edge yourself or have to sign an NDA or some legally binding document to keep it secret, it is [ __ ] If the trading strategy is public or for sale, it is a scam. TJR says he sells his strategy to make some extra money on the side. If he had a strategy with an edge on some of the most liquid assets in the world, the most scalable assets in the entire world, why wouldn't he scale it to tens of millions or hundreds of millions of dollars? Why would he still be trading on his phone alone? Why are you not safeguarding this edge with your life and out raising money from investors to build a fund? Why are you not hiring quantitative developers on NDAs to build out a programmatic implementation of your strategy? These are the steps you would take if you had the edge that you say you have. Nobody would have any idea what your strategy is.
The entire way that retail has been made to understand trading is fundamentally wrong from the very first step. Everybody is starting from step two, choosing an influencer whose strategy they want to copy and learn. If the strategy is public, it doesn't work in the first place. Step one is building the strategy from scratch. Discovering unique and novel relationships between features that you engineer. noticing something that other people don't notice. And if and when you find something, then from the very beginning, you're silent. Any narrative other than that is sickeningly incorrect. But nobody wants you to know that.
Under videos that I've posted exposing ICT or TJR, I coded TJR strategy a while back and another guy named Justin Worlin. And I get these comments under them. If ICT concepts don't work, then what does? Something proprietary. Nothing you're going to find online. If you can find it published, it doesn't work. If somebody's selling it to you, it doesn't work. you have to build it yourself. Since this information is somehow almost impossible to find on the internet because it's just flooded with videos of gurus, I'm going to go step by step from step zero to step done. From zero, not one buying a guru's course, but step zero, obtaining data, pre-processing it, engineering features, everything all the way through properly testing and deploying. Nobody should ever buy a course again after this. They should create their own strategies. And nobody should ever trade a strategy that does not truly have an edge because they'll know how to test it properly.