The Man Who Beat The Market Algorithm Then Created His Own Samir Varma

TITLE: YyX5mhsiLyA CHANNEL: Unknown DATE: ---TRANSCRIPT--- a PhD rocket scientist, multi-8 figure futures trader, a published author, an inventor with eight US patents to his name. First question you have to ask yourself is, are you George Soros?

Probably not. Probably not. So therefore, you should probably not be changing your position sizes too much. It is true that George Soros said correctly that when you’re right, be a pig. The problem is unless you are George Soros, you don’t necessarily know when you’re right. Only after it’s done. Only after it’s done. Introducing Samir Varma, one of the few people alive who’s been applying genuine scientific rigor to the markets for over three decades. Ranked [music] as the fifth best in the entire United States and still publishing peer-reviewed research on the markets and its algorithms to this day. Volume is the thing that actually tells you something useful that you can use. And all too often I see people completely ignore volume. But volume is the key. And this is one of the reasons by the way that some of these big guys are now starting to trade on these dark pools. They don’t want you to know what the volume is. What is this dark pool? So you have these venues where people large institutions everybody’s going to be doing the same thing at the same time in the same way. And when you do that stuff is going to get arbitrageed away arbitrage arbitrageed away until something really bad happens. Then bang here comes the risk. So that’s the first thing that is very likely to happen with AI. The second thing is your ordinary sources of alpha are going to disappear and you can already see it. What is it 90% of mutual funds that are underperforming the S&P? I mean, it’s some absurd number. At this point, retail traders are going to start to actually gain some power relative to the institutional traders. In this episode, Samir reveals how institutions use dark pools and algorithmic flow to move the markets invisibly. And the biggest edge the retail traders have against the institutions is by utilizing AI right now. And he shares exactly how to do it. I want to get back to the definition we stumbled upon which is in exploiting alpha you need to find a dislocation find the signal for said dislocation and then execute upon it with tested parameters. Therefore I want to understand how do we define signal versus noise and even when there is signal versus noise does the signal have to be causative? That’s a very good question and I can only give you a partial answer to that. So uh the goal seems at least for all retail traders is my goal is to create alpha is to find pockets of alpha and and generate alpha. Is that what is alpha and is that the goal of institutional intents? Okay. So alpha very technically speaking is something that comes out of uh for lack of a better word modern finance theory which is not even that modern it’s about 50 years old and the argument is that if you take the returns the daily returns or some periodic returns of a stock and you plot them on a graph and you plot them versus the returns of the S&P 500 as the market index then and you draw a straight line through the scatter then the amount by which the straight line cuts the y-axis axis however much it’s above or below which would be negative is the alpha that’s the outperformance of the stock and then the variation along that line is called the beta okay now there’s as I said to you offline there’s good evidence not great evidence but there is good evidence to suggest that neither alpha nor beta exist statistically anyway no alpha exists no statistically so we can talk about alpha as a concept but if you talk about trying to measure it may not exist What does that mean? Okay, so here’s the problem. To have a statement that says that alpha exists and therefore beta exists too, they have to. Those relationships need to be stable, right? But they’re not. So my favorite example of this is um I tracked Enron during its bankruptcy while it was collapsing to zero. Its beta never even hit one until right at the end. So according to modern finance theory, it was less risky than the stock market as it was collapsing and going to zero because they’re looking at things in terms of volatility. But volatility is not risk and that’s what drives beta. And then similarly alpha is that there is a there is some fixed relationship between the return of my stock and the return of the market. But who’s to say that such a thing exists? Okay, just to clarify, I thought beta was the benchmark is the S&P. But you’re saying it’s the variance versus the benchmark. Yes, variance versus the benchmark. Okay. So, if if your beta is one, that means that on average um if the S&P moves by 1%, you’ll also move by 1%. If your beta is two, then on average you move twice as much of the market. That kind of thing. But that’s a volatility statement. It’s not really a risk statement. So, uh extending this out, this would imply it’s a perfectly efficient markets if there is no alpha. uh which means every pocket of opportunity has been squeezed out and you’ve had alpha decay in all forms and that’s maybe where the world is headed when we get into the AI topic but it becomes a pendulum no yes and no that’s that’s a fantastic question and now I’m going to confuse everybody and I’ll try not to um you can have an inefficient market that does not have alpha and you can say what does that mean it the the problem is that the if the relationship is unstable and it varies and flops all over the place then you can’t really say whether there is a rate of return or there isn’t excess rate of return or there is an excess rate of return all you can do is make a portfolio watch it over some period of time and say oh look I outperformed so that’s the problem with the definition so it what would be a better definition a better definition is what exactly is it that you want to do with your money and you said it very well a few minutes ago particularly if you’re a retail investor you’re trying to make money you’re trying become a richer person, you’re trying to have a better life, etc. If that’s your goal, then you can find alpha, but you can’t find alpha in the way people tell you to find alpha. And I’ll explain in one minute how you should go about finding alpha. Um, what you’re trying to do there is you’re trying to say, I want to buy some investments that I think in the long term will outperform anything else that I can do. And the anything else that I can do is buy the S&P 500 and sit on it. Right? Now question, how can I do that? And the answer is that you in your profession um particularly if you’re a professional are going to be exposed to all kinds of new services and all kinds of new products and you are going to know something about them that the market doesn’t because you know it in detail. So for example, you’re an opthalmologist. you know about all kinds of eye diseases and you know about all kinds of possible solutions to those eye diseases and you know from the grape vine from talking to friends and so on and so forth that certain solutions might be promising. One way for you to actually really make significant money is realize um that there’s a good chance that one or more of these drugs may turn out to be very profitable and they’re from some small company. Buy it way in advance. Put us some amount of money that you can afford to lose into that company. buy it and forget about it and look at it five years from now or 10 years from now. That sort of bet is something you can make that the market cannot. That is a giant advantage for you and you should be making that. Um the same thing would be true for example years ago, let’s say for taxi drivers. Taxi drivers were the first people to realize that Toyota was producing cars that were way more uh reliable than anybody else’s. So you could have bought Toyota stock years and years and years ago and then ridden the wave of Toyota’s becoming more and more popular. Similarly, you could have been uh somebody that was uh in Formula 1 racing and you would have noticed that Honda was producing amazing engines and then they were thinking about taking these engines and putting them in, you know, regular cars and so on and so forth. Buy Honda stock. It’s just that’s just an example. There’s lots and lots and lots of things like this that you can do that you see in your everyday life. if you just have to keep your eyes open. Hey Titans, let’s take a quick break from the episode to talk about a sponsor and partner of the show that is Ola Prime. Now, a lot of traders have been talking about Ola Prime because they were recently the winner of the fastest payout prop award in the IFX Expo here in Dubai. And something that you don’t see so often is that they are backed by their own brokerage firm, Ola Prime Markets. And a few things that I love about Ola Prime is that they have offers for futures, forex, and crypto traders. And most importantly, they allow you to trade on over eight platforms. And further, they do a 95% profit split. Basically unheard of, which means whatever profit you make, you keep 95% of it. And most importantly, because of their reward, they’re one of the only prop firms that offer a 1-hour payout through a structured 10point 1-hour payout system. Your payouts are practically on demand, which means you can spend more time on the chart trading, withdraw your profits, and go back to the markets. With all these steps, measures, and awards in place, they are truly redefining transparency and trust in the prof space. So, if you want to work with a prof that you can trust and a partner of the show, click the link in the description or use the code toot for Titans of Tomorrow to get the best prices and discounts that I’ve personally negotiated for you guys, our Titans of Tomorrow audience. With that being said, let’s get back into today’s episode. So, the the end results I completely get. I wish I invested in Uber and Airbnb and so forth, but the idea is this is a maybe a form of survivorship bias because we’re looking at the best, but how many bets could I take with that same mentality that this this could be something big and you’re looking for a unicorn, but then you you burn a lot of capital or you have dead weights because you’re tied up and illquid. Yes, that’s why you don’t want to do a massive amount on it until it proves itself out. But I mean, I’ll give you an example. um a friend of mine um bought a very significant quantity of Apple um because he was a Windows guy and over the years I kept telling him in the early 2000s I said look a Windows is not a very good operating system B you have a degree in CS and so therefore C why don’t you look at this new thing that’s on the pike I’m telling you this is really good technology so after months of browbeating him he actually decided to do his thing and he loaded the boat up on Apple sold it 8 years later for some absurd quantity of money. That’s the sort of thing you can do because he had a CS degree. He understands computer science. He’s able to understand what’s going on. That’s where your advantage is. Your advantage is not in trying to beat the institutions at their own game, which by the way isn’t a great game anyway. Why can’t the institutions play the same game? And why does it become the resour? Why? because then they would have to run a concentrated long portfolio and then you’re put into the you’re a long only manager bucket and I don’t really care what your rate of return is you you know you’re only worth a few basis points a year in management fees so people don’t do it now there are people that can get around it there are funds that have managed that have like tiger cup parentage and all this other stuff and they’re able to sort of get around it but they’ve got branding they’ve got marketing they’ve got sales they’ve got all that stuff but it’s very difficult if you’re quote long only end quote I want to get back to the hypothesis of maybe alpha doesn’t exist. So pondering upon it um the crux of the problem is that alpha maybe is unreliable based on the definitions uh that we that we put upon it but unreliable means it’s harder to predict but it it would exist at times. So would would our job therefore be understand a dislocation find the signal for said dislocation and whenever it appears capitalize upon it. Yes, that there’s nothing else that needs to be said. Yes is the answer. Nice. Um the uh what I can tell you is that if you treat alpha as a concept, not as the way the finance industry treats it, it’s actually quite useful because then you’re asking a different question. You’re asking given this set of alternatives, what should I do? Which is a very good question. That is the correct invest question when you’re trying to invest. But that is not quite what people are asking. They’re always talking about these incredible I call it the tyranny of measurement. They’re always trying to talk about these incredibly precise things. But the market is not precise. And the error that a lot of people make is that they would like to be precisely wrong rather than approximately right. I like that. As a physicist, scientist everything is about data, evidence, conclusions, theory, statistical significance and the entire process. How do you reflect upon this idea that you kind of have precision in the markets? Good question. So the answer is that whatever it is that you decide to do in the market, it needs to be robust. Um so I actually as tongue and cheek about six weeks, eight weeks ago I wrote trading kans for uh you know traders. um basically the idea being sort of tongue and cheek make people laugh but also tell them something useful and so one of them I’m paraphrasing myself because I can’t quote myself I don’t remember um is that um you know there’s a master trader sitting at a desk and um some young pupil comes in and he says oh master trader here I’ve created this model it has seven variables it fits the data beautifully you know it looks like it’ll have this amazing back test and whatever whatever and [snorts] the master trader looks at him and says Okay. So, um uh what happens when it sees something that has never happened before in the market? And the guy says, “I don’t know.” And he says, “Well, so the answer is you should have something simple here. I’ll show you.” And he puts a 200 day moving average. And basically says, “When things aren’t working, that’s going to work.” Why? Because that’s robust. That’s the point. Um you need robustness in your rules. And that means you need to make them approximate. It means you need to stop shooting for precision. It means that you need to start thinking in terms of algorithms and decision systems, not in terms of linear regressions and mathematical differential equations and things like that that it’s the wrong arena for that. Does robustness or adhering to robustness imply you need to have an element of discretion? And as a systematic trader, how do you uh put systems around approximations? No is the answer. You should not need uh to have discretion. Uh I think last time I told you that you need to have a system that fits your personality. And to make a system fit your personality requires firstly losing lots of money and secondly um testing things ad nauseium until you’re literally actually adnauseium. ready to throw up and then you test them some more. Why? Because you’re trying to figure out when this system is not going to work and you need to either be able to willing to live with when it doesn’t work or you need to figure out a way of getting it to work in advance. The biggest problem with every systematic trader is trying to make decisions by the seat of their pants. When something happens that they don’t like and they’re like, “Oh my god, what do I do now?” Um it’s almost always the case that a decision made like that under fire is going to be the wrong decision because of the emotional elements because of the emotional element. Okay. And it also means that you did not find a system that fit your personality uh or you made some uh error in your back test by not looking at certain kinds of periods that may freak you out or whatever it is. And by the way, that’s again part of your personality. You need to know what’s going to freak you out and then you need to know what you’re going to do in that situation if you’re going to freak out and that needs to be in your system. That makes sense from from the medical background that I have is when you’re having a patient who’s having a seizure is not panic and try and figure out a protocol. There’s predefined steps. It’s it’s approaching with this lens. Exactly. The the the push back I want to give is when we look at all of the legendary names in finance and the biggest and best traders that we see, they are usually discretionary traders and would you argue therefore that’s a survivorship bias or they are anomalies or it has advantages that we should all look towards. It has great advantages if you can do it. So this is the problem with that is that um it’s not that it can’t be done just like it’s it it somebody has to be world number one at at tennis but the number of people that are going to be world number one at tennis is not that high. It’s one it’s one exactly that’s the problem. So, can you have a George Soros that’s absolutely singular who says that his back hurts every time he’s taking risks that he shouldn’t? If you’re George Soros, of course you can. But to get yourself to that level of sophistication requires a great deal of training and study and losses, frankly, you just you can’t learn any other way other than by losing money. And it’s a very expensive bit of tuition. And so, the cheaper way of going about it is to figure out in advance what you’re going to do. So the last time we spoke the opening conversation was around this find alpha or find a strategy and see if it’s congruent with your personality so that it holds in stressful times. I always at the time specifically held psychology as a topic that is it’s a scapegoat because people don’t have an edge and therefore it’s easier to blame by mindsets and morning routine rather than go and find a system and at nauseium test it. But then after speaking to a lot of authors and psychologist specialists it started to make sense to me. So I want to I want to throw this at you which is the idea of okay dopamine dopamine dopamine is a neurotransmitter that it’s it’s to push you through uh hard times to to then have the delayed grati gratification. So then dopamine can be used as a tool. Then I also learned about let’s say when you’re in a fight orflight situation when you’re feeling anger the certain neurotransmitters that are released in your brain can up to red can reduce up to 50% of your free prefrontal cortex activity or if you are on low sleep below 4 hours your preffrontal cortex activity drops significantly and that is the part of the brain that is responsible for cognition and logic logical thoughts. So then we are fighting neurochemistry here. Uh when your amydala which is the fight orflight part of your brain overpowers your critical thinking part. How do you build frameworks around neurochemistry? So did you ever watch Greg Norman play golf? No. So Greg Norman was a big name in the 1990s and he was by far in terms of talent the best golfer in the world. He was so dominant in terms of his talent that he basically was Tiger Woods. And he did some things Tiger Woods has never done. But he didn’t win very many tournaments. The only thing he ever won with that massive talent stack was two British Opens. And in both of those he collapsed. He just didn’t collapse enough that he lost the tournament. He had the third round lead in in in major tournaments in 1986. In all three major tournaments, he had the third round lead. He won one cuz he collapsed the other three times. What am I saying? What I’m saying is that I’m agreeing with you. The the the exact problem is that you start to freak out. You can’t make rational decisions. You do something rash or foolish or stupid or you’re on tilt or whatever it is that you know happens to you on say a poker table and you end up losing lots of money and it’s incredibly difficult to recover from it. The way around it is either you train your brain like an athlete. Yanekx sinner is Yanek sinner because he’s trained his brain not to freak out. That’s one thing you can do which is the same thing George Soros has done. Or you can figure it out in advance and have a set of rules that you’re going to follow plus some fail safe. You should always have a fail safe. I don’t mean a stop-loss. I mean a fail safe. The two are not quite the same thing. Um that’s just simply going to stop you doing too much damage. That’s how you protect yourself. There’s no other real really I I don’t know of any other way of doing it. So let let me ask you because the idea of I know the the early signs of when I’m about to go on tilt. Let’s say I’ve taken a loss. I’ve taken two losses now and then in the past has led to five losses, 10 losses in a row. But the more you go through the stimulus, the more you become self-aware, the more likely you are to catch yourself in the moment. But the key portion becomes is I can have frameworks and and systems in place and fall backs. But it still relies on one thing which is me to adhere to it. Yes. And the adherence part is the psychology part and that’s the testing. So, how testing? What do you mean? Testing. That’s what I mean by test and test and test and test yourself. You’re testing yourself as you’re testing the system because you have to imagine what you’re going to feel like when the system does what it does. And you have to be willing to live with whatever the back test showed you because that could happen. You’ve already seen the back test showed you it could happen. How do you simulate feelings? You Well, you can’t very well, but you can try. Um there’s there’s nothing else you can do. And you’re right, you have to be exposed to the situation anyway. But that’s why you always want to fail safe. So the two answers I’ve arrived to is the first one I agree with you. You just have to gain enough market experience. You you got to acquire the scars and then uh eventually you build the resilience. Uh so there’s no shortcuts and you can have uh systems in place and with time you adhere to it and and then eventually you get it. But then the other one that I I think is uh the first thing that you said last time to me which is rather than trying to force yourself to fight your temperament in the heated moments of a losing period uh just change your approach to the markets. I don’t like sitting in trades that I’m sitting around my entry for 2 weeks. I’m not a swing trader anymore. Um to help me uh build out a framework for someone watching specifically who’s learned uh things online. They’ve learned strategies online but they are not finding success but they find it in back testing but in live markets they don’t. it can only be a psychological gap. How can they go about fixing it now and say this is how I find a more congruent strategy to my temperament? So the first question you have to ask yourself is are your back tests valid? Um because frequently your back test is not valid even if it shows a very large rate of return. So I’ll give you the classic example of this which many people still do which is surprising. So people will say I want to trade a moving average crossover strategy and then they’ll find you know some asset that they want to trade and then they’ll find that exact combination of two moving averages that produce the highest back test and then they’ll say well I need to trade this. The chances that that’s going to work in the future are very low. Why? because you’ve almost certainly hit upon some peak in the back test and that just happened to be at random at that specific spot and the future returns are going to be substantially lower if not negative. What you really need to do if you’re going to do that is you want to find a stable set of parameters. You’re not looking for the highest rate of return. You’re looking for the most stable rate of return. Did you find a plateau? Can you vary your parameters and still get basically the same rate of return? And as I told you last time, can you add noise to your to the prices that you’re using? And by adding noise, your return slowly degrades. Is it doing that? That’s that’s a good sign too that you’ve found a good back test. So that’s the first thing you got to do. You got to get a good back test, right? How do you do it? A good back test. Yeah, cuz I mean myself and most we we go on Trading View and we rewind and we play price action and we see what we would do. So manually, you have to first Good, it’s a good question. You have to first ask yourself, you don’t have supercomputers. You don’t have 100 PhDs working for you. So, you can’t play the renaissance game of let’s find every patent and trade them all. So, you have to do the opposite. You have to really drill down. You have to ask yourself, what is my thesis? What am I saying? What is it that I know or understand that is allowing me to make this excess money in the market? Once you can define that and you can actually state it precisely and clearly I think my thesis is the following then you can put together a back test that’s of some use. So I’ll give you an example. Um uh recently I I started back testing a couple of things just basically for fun although I haven’t started using them yet. Um, one thing that I read uh from a bunch of commentators is that we the US has lots of debt and its debt keeps going up and so therefore the dollar is going to collapse and if the dollar collapses that’s going to be terrible for the market. So I said okay that seems reasonable to me. I don’t know if it’s true or it’s false. Why don’t I look at DXY and ask myself do extreme moves in the DXY which is the dollar index in the past have they ever told me anything useful about the market? The answer is no. So maybe the dollar will collapse, maybe it’ll be bad for the market, but is there any evidence for this assertion? None that I can find. So can you trade that? I wouldn’t. On the other hand, um you can look at for example the high yield option adjusted spread, right? uh for high yield bonds. There is some evidence to suggest that when there’s stress in the high yield bond market that the stock market is about to have significant risk correlation or causation. Good question, but it is fairly consistent which suggests that it’s causation or at least an early warning canary if you want to call it that. So in other words, there is stress in the system. the stock market hasn’t seen it yet because it hasn’t quite affected stocks and so uh you could treat the high yield option adjusted spread um however you want to do it as a leading indicator of something and then you have to think about it but so that’s the dichotomy so you’re always trying to find a reason a good sensible reason for why this might be true you know the old joke is that you know butter production in uh in Bangladesh is uh heavily correlated to stocks that begin with A or something like that. I forget. I forget the exact thing. Something like that. Okay. I mean, I’m giving you a ridiculous example because it is ridiculous. But you have to be very careful with what it is that you’re actually back testing because there’s no other way of finding a signal. For the last 2 years, a proud sponsor of the show is a top ranked leading prop firm, Alpha Capital. And for the years that I’ve been working with them and the thousands and thousands of viewers, you guys that have been working with them through the discount codes of Titans of Tomorrow, it’s clear for me to see why they are top ranked prop firm in the industry. They have also reached a monumental milestone of $100 million in payouts. And with the multiplestep plans and the multiple package types they have, there’s going to be an option catered specifically for what you’re looking for. So, you can buy an evaluation account catered to your needs at the most competitive prices. And with our discount code TOT for Titans of Tomorrow, you’re able to get the most unbeatable unmatched prices in the industry with a leading trusted prop firm. And with that being said, let’s get back to the episode. This begs the question, well, two questions. Number one, most people watching, especially new traders, uh the arena when we see how can we estimate price, it’s fundamentals, it’s sentiments, it’s uh trying to understand people’s behavior, psychological, and then there’s also technicals. price. Most people end up at price because it’s easy to open up a chart and and draw lines. Uh what would you argue in the case of can technicals alone be a way to model and trade price? If you use technicals because you’re trying to control your risk, there’s a decent chance it will work. If you’re using technicals with some exceptions, we can get into that to try to pick assets or um stocks or whatever it is that are going to go up, say, there’s a decent chance with some exceptions that that’s wrong. That’s the best thing that I can I can tell you be because it’s a question of base rates. I mean, you know, you you know medicine as well as I do. The base rates are in your favor if you’re talking about risk. The base rates are against you if you’re talking about return with one exception. Okay. The reason I was asking about technicals is because the the premise has to be to them believe it or not is can past price predict future price. Yes, to some extent. But the question is can does it predict it enough that you can make money. Is there a variance on time horizons? Shorter versus longer time horizons. Yes, there is. So if you are a small investor and you decide that you want to day trade, you actually have a chance of making some money with small positions. And I can even explain how um if you are a larger than small investor, then it’s quite excuse me difficult to deploy enough capital on a very short-term time horizon as an individual investor. You don’t have the nancond execution. Um the the exception that I was mentioning is that um the one thing that seems to work in almost every market and almost not quite every time frame is momentum. So you can certainly trade for example breakout systems if you’re willing to live with the risk. So essentially you know you have a channel draw it however you want doesn’t matter Ballinger bands lines it makes no difference and when the market breaks above you trade one way market breaks below another way you you have to have strict risk control you have to worry about um stop losses you have to worry about all kinds of things but actually you have a chance of trying to make some money there’s a limited amount of capital you can deploy but you can do it so momentum is the one exception to that statement I I I assume the reason you say limited capital can be put through that needle of alpha yes is because uh the market uh what? No, you’ll move the market because you’ve only got so much liquidity at that point and you need to enter at that point because if you don’t enter at that point is no longer profitable. Mhm. I I remember last time you told me that there’s a lot of liquidity in the market but at at an instant it’s it’s very uh finite or limited. Yes. Uh even in something like the S&P which I found pretty cool. Exactly. Um okay. Okay. So, let’s say my my goal is now to find uh something like this, a technical pattern like a breakout and I’m I’m trading momentum and then there is a certain limits that I could uh put into it before I’ve moved the market and I missed the opportunity. Is that something a retail trader should be concerned about or is it very big limits relatively speaking? It’s pretty big limits for a retail trader you don’t have to worry about it too much. And then again for a retail trader there are some other very clever things that you can do. Um one of them is uh there are three kinds of momentum. And so as a retail trader, you should look at all three kinds of momentum and see if any of them are to your liking, fit your personality and you want to trade them. Um, and those three kinds of momentum are uh time series momentum. So that is to say the momentum of the thing itself, right? So like if you decide to trade uh let’s say you take a moving average and you buy it when it’s above and sell it when it’s below, that’s time series momentum. You time series momentum is is effectively how much it moved in a set amount of time. Um well the reason it’s called time series momentum is that it’s just calculated from the time series of the thing that you’re trading. So you just a moving average is the last n days average together. Right? That’s why it’s called time series momentum. Got it? So that’s one kind of momentum. The second kind of momentum is called cross-sectional momentum. So in cross-sectional momentum you you don’t look at how the asset did by itself. You put the asset in a basket of relatively similar assets. Let’s say all stocks or something or the other. and then you say, “Okay, I’m going to buy the top x% of the performers in this basket.” That’s that’s cross-sectional momentum. The reason it’s cross-sectional is that you have more than one asset in that basket. And peculiarly, you can even take unrelated assets and throw them in there and do the same thing and it’ll work. Don’t ask me why, but it does work. Okay. Um, so that’s two kinds of momentum. But there’s a third kind of momentum, which is uh a way of taking advantage of the stupidity of academic finance. Okay. So, you remember I was telling you earlier that alpha may not be defined. It might be, it might not be. I just wish people would realize that it’s not as clear-cut as people claim it is. So, because of because it’s unstable, there’s something clever that you can do. You can take any given stock or even portfolio like an ETF or whatever and you can quote decompose it into factors. So you can ask how much of its return comes from uh size and how much comes from value and how much comes from growth and how much that’s called those are factors. Now the fact is my pun that factors are themselves unstable. So this entire statement that you can decompose a set of returns of a portfolio into factors is to put it politely the technical term is crap. But never mind. Um that gives you something that you can use and the something that you can use is use factor momentum. So let’s say that you’ve got a um you’re looking at um oh I don’t know some collection of stocks that you decide to break out in a certain way. You can ask which factors are working now and then you can invest in the factors that are working now and because they are unstable for a short period of time at least that factor is likely to outperform the other factors. So you can make a bit of extra return from being in those factors and not in the other factors. Now you have to keep watching this and so when that changes you have to change to the other factors and you just keep hopping from factor to factor to factor to factor. This works because of the instability. would all three of these uh types of momentum especially the factors one could be condensed down into just reading price it would reflect on a chart. No, because two of them are relative. So you’d actually have to put a few hundred stocks on the same chart say and then try to figure out you know which are the strongest and which are the weakest. So then you’re better off not doing it that way. Mhm. I want to get back to the definition we stumbled upon which is alpha is or in exploiting alpha you need to find a dislocation find the signal for set dislocation and then execute upon it with uh tested parameters therefore I want to understand how do we define signal versus noise uh especially with chaos theory let’s say and even when there is signal versus noise does the signal have to be causative it caused uh the reason that price is price is moving is causative and and you’re reacting to a causation and would correlation therefore be functional? Correlations in my head are moving averages or even patterns. Price patterns seem to be a not a causation. Seems like a correlation. So I want I want to clear up that definition. That’s a very good question and I can only give you a partial answer to that. So uh two three years ago I think Hindenburg research put out this research report about um Adani Gautam Adani who runs a whole bunch of he’s the richest man in India. Yes. And he said and they said that uh there’s a bunch of fraud in his companies and this is all wrong and all this other stuff. Um now it turned out and the stock collapsed of course it turned out that um a lot of the accusations it seems were overblown. There was some truth to them but they were massively overblown. That’s a dislocation. Uh and there’s a good chance that if you do and you have time because this doesn’t reverse right away. If you do some serious digging, you might realize that in fact this is overblown. And you might then say, I think I’ll buy this stock. And guess what? If you done that, you’d have been very happy. So um you’re right, dislocations are the place to look at. And in fact, dislocations frequently are large enough that even in um institutional investors can get into them. Um you as a uh uh retail investor have a big advantage that you can do whatever research you want which frequently the big guys actually cannot do uh because they have wall crossing problems and they can’t ask this question and they can’t talk to that guy and they can’t do this and they can’t do that whatever there’s lots of restrictions you don’t have them so you can do a lot of soouththing and digging and when you find these dislocations find the occasional gems and invest in them. Uh the other one is if a stock goes into bankruptcy, the statement is more or less never should you buy it because it’s a disaster always. So if a stock goes into bankruptcy, yeah, except there’s one giant exception to that, Hertz. Hertz was a huge win for retail investors who are the ones that were piling into it while it was in bankruptcy because they had noticed that um it actually probably had more assets than liabilities and it was pretty silly to put it into bankruptcy in the first place. So that’s another example. These are the places where you as a retail investor can actually make significant money. So I’m getting two sense of alpha that we’re describing. One is this opportunistic it’s hunt for the gems of areas of dislocation where something isn’t making sense and you capitalize upon it or something is undervalued and it’s about to go. Yes, that’s one type of alpha. The other type of alpha that I would imagine is most commonly uh exploited or attempted to be exploited is what I would want to say let’s you’re data mining. You’re finding pockets of opportunity in price in technicals on the same thing. I always trade gold. I always trade uh the S&P. Um what about finding alpha as a day trader in that arena in that style? So what you’re doing then is the human brain is an amazing pattern recognition engine. What you’re doing then is you’re effectively seeing if there are patterns in those prices that are being caused by the very large players and you’re surfing those patterns is what you’re trying to Okay, I like this a lot because what I what I see is when I go on and learn anything about technicals, uh I see okay head and shoulder, I see Fibonacci levels, I see EMAs, I see support and resistance and all of these engulfing type of candles. Okay, they they’re all nice. And I always thought when I speak to these big hedge fund managers I’ve spoke to billion dollar hedge fund managers six even a $600 million hedge fund manager recently and I would have thought okay they must have stellar technicals beyond what us retail can access and I’m surprised to say I would imagine the guests I’ve had on the show that have retailed that have done well have infinitely superior technicals than a hedge fund manager who is at times just finding a range waiting for a breakout and using an EMA to protect themselves. And when I try and dig deeper, they’re shocked that I’m asking because they’re like, Technos is just that. So, this leads me into a confusing web. But I love that you said our job is to surf on their wave because at least in my style, I I did I wanted to find something that was causative. Yes. And I don’t know how to find something that’s causative, but I I could sense that a head and shoulder pattern can’t be causative. It’s a nice probably a correlation. Same for a Fibonacci level. I would assume the same. But causative I would assume is how can I find a digital footprint of where large participation happened and then I try and find areas of max pain and where there’s max confusion and then I try and find a winner of that tugof-war between buyers and sellers and then capitalize upon that. I’m not going to tell you how I do it just yet, but I want to hear your thought process on how I one could achieve that. So, um, I agree with everything you just said and I would add to it that the way to do that is by looking at volume patterns because that’s what’s actually confirming what you’re trying to figure out. So, what you’re trying to do is you’re trying to surf the wave. You’re trying to surf the wave of some big guy. You’re trying to say if there’s an iceberg, we spoke about this last time, if there’s some iceberg where there’s a tip here and there’s a massive thing below the surface, you’re trying to figure out when that exists. The the way to figure that out is by looking at volume patterns. So, you look at volume in relation to price and you’re asking yourself things like um when this stock goes up, does volume increase as it goes up and decrease as it goes down? That suggests that there’s some upside left to the stock. Um or the other possibility is um the stock is falling and then all of a sudden there’s like a 20 times volume day. Okay, that sounds like somebody puked it out. It may be time to buy it. it the volume is the thing that actually tells you something useful that you can use and [snorts] all too often I see people completely ignore volume but volume is the key and this is one of the reasons by the way that some of these big guys are now starting to trade on these dark pools they don’t want you to know what the volume is what what is this dark pool so you have these venues where people large institutions can only participate by invitation and um they trade with each other basically and [snorts] that stuff is never even shown on the exchange but it would reflect in price would reflect eventually in price once the trade has taken place but that but the but the activity takes place of exchange and therefore I mean this is just a gap in price that you can’t really there’s not not much you can do about it at the moment it is still true that the volume action uh for relatively liquid stocks in the US market is still reflecting what’s actually going on. If um trading keeps moving to dark pools, that may not be true anymore, but it is still true. So you as an individual investor can still do something about it. You have to learn to read volume patterns. Would therefore a conclusion of what we just discussed be if I want to be a intraday or scalper and I want to be using technicals predominantly and I want to surf the wave of institutional movements then a necessity becomes look at the level two look at the order flow 100% and therefore don’t trade gold don’t trade forex don’t trade these things where you cannot access the level two yep because you’re at a significant disadvantage without this level two and also also the even bigger disadvantage you have when you trade say forex is you have no idea what the volume is, whereas you do in the stock market. So that’s the other problem. So here’s the other thing. So this is just an FYI, funny story. I won’t name names. Um, but more than one master of the universe has spun out of places like Goldman Sachs and raised some gigantic sum of money with all kinds of people saying, “This guy’s the next trading genius.” and then 3 years later they shut their shop up very quietly and basically go home with a tail between their legs because they couldn’t trade. What they don’t realize when they’re masters of the universe is that the reason they made money in the first place is because they were seeing the flow. And when they left Goldman say they were no longer seeing the flow and it’s absolutely apparent to anyone with a functioning gray cell that that’s what’s going to happen. But somehow if you come out of Goldman and you have the right pedigree, people’s gray cells seem to stop functioning. The level two, the the flow of orders, you can now read the story which is buyers versus sellers. And the analogy I I always visualize like this because if we have a team of buyers and team of sellers and they’re doing a tugof-war and a rope, the only way the buying team can win is you have more buyers or stronger buyers or less sellers. That’s the only way price is going to go this way and vice versa. Yes. So obviously if you could read the that in in the order flow then you’ve understood what price is likely to do. Yes. Is there a way to do this not in in live action which is on the pulse of all the order flows and executing upon but rather you wait for the resultant. You wait for which team won. You you wait for the price to move in that direction. Now you’ve got a confirmation and now you enter upon and ride the wave. Would that be would that be? Yes. With a giant caveat. You can’t do much money that way, but you can do it. So, I’ll give you a a very good example. Uh I haven’t done this in a while, but in the old days, well, not that old, but in a while ago when I was doing this stuff, um uh we found that you could take a a simple thing like a uh opening 15-minute bar and you could trade the breakout from that bar and more or less for the rest of the day, the price was going to go in the direction of the breakout. M. So that’s something that you could trade and it existed. I’m not sure if it still does or not, but it did. And that’s exactly what you’re talking about actually is because what’s happening there is that that somebody’s trying to buy a very large position and the breakout is going to continue to drive in that direction because they’re buying a large position. They have no choice. They got to buy it. So you’re basically riding that and what you’re identifying with the breakout is that. And then you can do things with volume and so on to try to confirm it that this is real. If if a good trader’s job is to react, not predict, the best thing we could react to as we’re concluding is the order flow specifically for shorter time horizons, intraday trading. If that therefore becomes the formula, how does someone build a system around that hypothesis for now? That’s difficult. And the reason it’s difficult is that it’s you if you’re going to do that, you have to simulate with past tick data. Firstly, you need tick data. Secondly, you need level two data. And thirdly, you need an execution model. Even forgetting whether you have a signal or not, right? So you you have to then specify if you’ve got a signal, how you going to get into the market, how you going to get out of the market. And that in itself has issues with commissions, which now zero, but never mind. slippage, bid ask spread, market impact, and so on. And you got to model all of those, and you got to get them right. So, if you’re going to write a system for intraday trading, those are all the things you need to think about, and you need to get them right. Otherwise, you can’t write a system because it’s actually not that difficult to find a system where you say all the costs are zero and you basically trade intraday momentum. That’s actually pretty easy. You can do that in a spreadsheet in no time at all. But none of that or not none of that almost none of that is tradable. So now you got to find the stuff that is tradable and the only way to do that is to make sure that you’ve got realistic costs and realistic executions and so on embedded into your system and then test that. So that by the way that’s the other tip for people creating systems. Do not treat your um uh entry and exit points as um uh just numbers. You have to model your entry and exit points because that modeling could take away the return that you think you have and that’s critical. So in other words, your system is not just your signal. It’s your signal plus your entry plus your exit. And you have to define all three. What is an example of someone thinking they’re doing something right, but they’re not regarding the execution criteria that could erode the edge momentum. So frequently you will find in academic finance um uh very well- constructed articles that say if you have this kind of momentum then you’ll outperform the market by some number and they’ll tell you what the number is. But then when you dig through what they did to get that number you’ll find that the assumptions they made for how they were getting in or getting out are not realistic. And you would just if you naively looked at that and said oh wow that makes an extra 31% per month. Wow. I think I’m going to do this. You haven’t necessarily actually worked it out. And then if you’re going to short, you have another problem. In fact, you have two problems. The first is, do you have historical hard to borrow data? And second, do you have a historical cost to borrow data? Because frequently a shorting strategy will look profitable until you put those two costs in and they won’t be profitable anymore. I finally have a special offer to share with all of you from the US or my futures traders which is over 20% of the listeners of the show and that is Alpha Futures a leading futures prop firm that is working with trade of eight and ninja trader that are compliant with CME regulations with the largest end of day balance draw down in the industry and 90% profit split and same day payouts and with the most competitive pricing in the industry with accounts starting at just $79. On top of that, just by being a viewer of the show, you get up to 40% off all evaluations. So, why not get started with an evaluation right away? Trading $50,000, $100,000, and you already know the power of prop firms and larger capital. So, go ahead and use the link in the description or code toot for the best prices in the industry plus the best discounts in the industry to make this a home run offer if you are a futures trader. Just to let’s say wrap up the point that I was that was in my head. Would you say therefore the best conclusion we can come to is if I want to be an intraday or intracession or scalper that is technically driven. I want to capitalize upon momentum that the all the causations that could exist whether it’s fundamentals different type of participation XY Z it is all upstream of orderflow and then orderflow can be how we synthesize all of that participation and intent. And then you don’t need to look at all of the other things because it’s reflected in the level two. As long as you have robust frameworks around that. I’m trying to decide if you’re 90% right or 100% right. Um what is that 10% that’s holding you? Um if you look at investment banks, it’s very clear that a huge amount of money that the desks make is just from the flow. Matching flow, bid asks, spreads, blah blah blah. you know it’s all it’s all just flow that suggests that your statement is 100% true the only thing that gives me pause I don’t know the answer is what if some of that trading is in some sense fundamentally driven so that is to say somebody did some deep research and they said okay this stock is massively undervalued or whatever it is and I’m going to buy it um does that count as flow or does that count as something else you could say it counts as flow because they’ve done the research and now they want to buy it. But it could also not necessarily be just pure flow because frequently pure flow is just portfolio rebalancing or pure flow is some statistical arbitrage or whatever. Right? So that’s not quite the same. So I don’t know if it’s 100% right or 90% right but it’s somewhere in that in that range but yes it it is flow. The the other uh confusion I see people have is they may have an edge and that’s a bold assumption, but if you found an edge and then you have the execution criteria that you were mentioning and you’ve tested it all, but then you mess it all up and you actually make no gains because of how you varied your position sizing. Yes. And the idea, I mean, you have one camp which is just standardized risk to even it out. But then the other camp is if you really want to capitalize that the market doesn’t always provide equal opportunity. There are pockets of better opportunity at certain times. Even people have argued to me that 80% of the returns come from 20% of the year to that effect. So you need to know when to have your A+ setups and size big. But that variation in sizing done wrong can mess everything up and then obviously ignited or amplified by human psychology as well and the wrong behaviors that can be downstream of that. What is your thoughts on position sizing to capitalize gains but also longevity and wide? Superb question. The first question you have to ask yourself is are you George Soros? Okay. Probably not. Probably not. Okay. So therefore, you should probably not be changing your position sizes too much. It is true that George Soros said correctly that when you’re right, be a pig. The problem is unless you are George Soros, you don’t necessarily know when you’re right except after it’s done. Only after it’s done, right? The second thing is that um there is a way of doing this that will get you 90% of the way there with 0% of the FS. And that is what I call the Kelly criterion. Yes. Right. I mentioned this before. Now, no one has the stomach to trade Kelly. No one. Literally nobody. But what everybody has a stomach for should is to trade a fractional Kelly. So what you do is you you you figure out what the Kelly size bet is and then you trade some small fraction of it and you keep doing that over and over again. And what that says is you should always be trading some fixed fraction of your total portfolio value. So as you lose money, you trade smaller. As you make money, you automatically trade bigger because it’s always a fixed fraction of that. That is and and I’m going to say a second thing in a minute. That is the most robust way I know of of position sizing. Now there’s a second aspect to position sizing. This is the thing I was saying earlier. The second aspect of position sizing is okay, I’ve got a bunch of positions. How do I size each of these positions? So many people will tell you size them by conviction. So others will tell you size them by risks. Whatever it is. Well, maybe. Are you George Soros? No. Okay. So the correct answer always the most robust answer because that’s what I always worry about is make them equally sized. And it sounds silly but it works. And not only does it work it even works in cases where it has no business working. So for example let’s say you’ve got a portfolio where there’s a bond ETF and a few stocks. So the bond ETF is this heavily diversified thing and you know you you you would think that it has lower risk and these stocks have higher risk or whatever it is. So you have no business treating them all as if they were the same thing. Treat them all as if they were the same thing. Any position you have in your portfolio, more or less equalize it. And you will find in the long run that’s robust. It stops you from going broke. And because it stops you from going broke, it leads to the largest compound rate of rate of return in the future. That’s the simplest way of doing it unless you’re George Soros. Can you walk me through briefly uh the Kelly criteria? I only recently came across it to be honest. So the Kelly criterion basically says that you should be sizing your bets based on whatever your largest expected loss is. And then he has a formula which I don’t even remember but essentially you you you size your bets based on the largest expected loss and that tells you how much leverage you should be using. Um and the the upshot of it is you don’t need to even do it that way. Just do fixed fraction of your portfolio betting. The other lever that I can think people would pull to get outsiz gains with the edge that is your northstar, you can’t really move away from it. You’ll always end up there, but you can temporarily dislocate is either bigger bet sizes at times through higher conviction plays which uh could end up in the same place or end up worse if you cannot accurately decide what it was was a A+ setup. Um the other one is how you scale out of a position. Yes. And this is uh something I spent a lot of time on a couple years back and I’ve stayed on it ever since. So I wonder if I should revamp it which is why I’m asking because I’m an intraday intraession kind of trader. I started to do price action based targets and it could be you know previous day high it could be previous session high it could also be market structure trend. There was so many lines on my screen that price could go towards and it became discretionary and more importantly based on my hope versus greed versus optimism, pessimism that navigated how far or short my targets would actually be. And then I scrapped it all and tried to do it in a datadriven way which was uh I did a whole study uh and by by chance it ended up being I take my first partial at the average session volatility. What historically the session will do take uh take one chunk off and the remaining at the average daily volatility and correlated to a fixed risk reward because my stop loss size is in a tight parameter and therefore my risk reward can be in tight parameters around these uh ranges. Would this be the optimal way to do things on based on how I described it or would there something be better? So I think the first question you have to ask yourself is do you have a trade record of the time when you were doing your discretionary trading? I I looked back around 2 years. Okay. So, if you’ve got that, then one advantage you have is you can go back through your trades and ask yourself with those trades, what were the optimal entry points and the optimal exit points? Uh, better than what I did. Better than what you did. With the benefit of hardness. Okay. The reason that’s important is that you’re asking, is there a pattern that I missed or didn’t think of or whatever it is in those traits? Um, that’s the first thing you want to ask yourself because if there is a pattern and it’s discernible, then that you’ve already you’re way ahead of the game already. Now you know something that’s useful. Um, the second thing is that every time you’re doing partial trades, you’re actually saying that this is a new trade. So you should really be thinking of that as a portfolio of trades. So every entry is a new trade. Every exit is is a exit of a previous trade. Right? This is exactly the conversation I had with myself. correct me if if this is where you’re headed, but I thought uh as I go to higher riskreward levels, I’m no longer thinking, okay, what is the total big number I can get is also the riskreward of it going back to break even. So, it’s the I’m already up three R uh riskreward and I’m targeting a a 4R. Well, I’m potentially going up three to gain one. So, as you hold it, you end up inverting and having a negative risk reward to hold it. Is is this where you’re headed? Yes, that’s where I’m headed. But the thing is, you’ve got to test it because you may also be giving up return that you think you have. And the problem is you can grow broke. You can go broke taking a profit. And the reason is that if your profits aren’t outweighing your losses because you took them too soon, then despite having good setups, you may actually be losing money. So you got to test it. But but yes, that is where I’m going. So there’s there’s a delicate balance. It’s a very delicate balance that you have to strike. So yeah, there are three parts of the question. And if you can just clarify for me, one was let’s say hypothetical price action based targets and there can be many. The other one was based on average movements or volatility based on a session based on a day. And then the other one would be fixed R multiples based on a basket of trades. What was the optimal? So you’re not leaving money on the table but not not shooting too far. Which one of those three would be the best? [sighs] It depends on the asset you’ve got in the in the basket. Is it all stocks for example? Yes. So if it’s all stocks then what you what you want to do is to set up your rules such that you have the cleanest possible entries and the cleanest possible exits. However you want to do that. Okay. And that the the cleanliness of the entrances and the exits is what is going to make the system robust for you in the long run. um the more you have to uh fiddle with your entries and exits to make it profitable, the higher the chances that you might lose money in the future. So with the benefit yeah with the benefit of hindsight I can always find better entries uh because I can see what happened and what I could have done of course but the challenge for me becomes is I I I’ve there’s many entry types I could do but then how do I know in the moment which one it will be? So you’re that’s not what you’re looking for. What you’re [clears throat] looking for is what is the pattern if any that would have given me better entries on the routine basis or better exits on the routine basis. If you can find a pattern then you’re ahead of the game. If you can’t find a pattern then it might just be random for all you know. So let me present you three scenarios that I often encounter which is a zone that I want price to reject from and I’m looking how to get in and I’ve got the time of day. I’ve got the momentum XY Z. So I’ve got a trade ideation and I want to convert it to an execution. Now there’s three camps I could put it into. One is put a limit somewhere on there. So as it approaches I’m already tapped in. The second one would be as it approaches wait for a confirmed rejection. Could be a break of structure. It could be many types. That confirmed rejection is price action based and then I execute cover the high. The last one would be uh a s a sort of market execution or a limit based on momentum in the moment engulfing. But then your entry is going to be a bit after the engulfing. So you’re you got a delayed entry, larger stop loss relatively speaking and then price could always come back and give me the entry that I was looking for. So then the paradox for me becomes is I could take the market execution trade but then give it a couple more minutes it might come back for my confirmed entry. So then I kind of don’t do the market execution one but then if the market execution one presented and didn’t come back for my limit entry I missed the trade. Yes. So then I don’t know how to navigate these three. So then just a rule I made for myself is do the do the confirm one and accept I miss the rest. So, okay, there is an answer to your question, but it’s a little bit complicated. The answer to your question is that you’re trying to figure out if in the time frame of the execution the asset in question is mean reverting or trend following. It really depends of what time frame I’m defining. Yes, correct. No, no, no. This is the interesting part. The thing that confuses everybody about the stock market is that depending on what time frame you pick, it can go from mean reverting to trend following to mean reverting to trend following to mean reverting to trend following. Right. Which is why top down analysis is confusing. Confusing. Yes. So what you need to do is you have two things there that are mixed together. The first is your signal and the second is the execution of the signal. Okay. Yes. Right. The execution of the signal [clears throat] is at a much shorter time frame than the signal. Of course. Yes, it has to be right. So ask yourself what is that execution time frame? I I’ll pick a number for you. 1 minute. Okay. Ask yourself if the asset or assets in question are mean reverting over 1 minute or trend following over 1 minute. Oh, so I’m looking at one minute trends to define if I’m mean reverting or not. Yes. Not your signal. Your signal will be automatically over a longer time frame than the trade. Let me put some numbers so that we can clarify it and better visualize it. I I used to do the top down analysis and as you said daily could be bullish 4hour bearish 1 hour bullish M30 it becomes zigzag all over the place. What is the trend? I don’t know. So what I kind of did is package time frames. The daily time frame I just look at to see am I anywhere relevant? I mean an area of interest. Is anything happening? If not most of the time you’re in no man’s land. So I disregard and then my home becomes the 15-minut time frame which helps me assess intraday trends. You know what happened the last couple of days? And that helps me assess my real relevant trends, direction, and points of interest. And then with that, I’ll use the one minute for execution as you were describing. So then I’ll try and be uh trend on 15 minutes, the intraday trend. And then the the one minute time frame I’m looking for rejection, which I guess would be me reversing. Yes. Would that be a healthy balance? It could be. That’s where you have to test it. So for example, let’s say that uh in this case, your your setup comes and you want to trade a breakout. Let’s say, okay, what you then want to see is in the one minute that I’m going to be putting the trade in, is the stock a mean reverting stock or is it a trend following stock? If it’s a mean-reverting stock, then you wait and you execute towards the end of the bar or you execute when the market falls a little bit or something like that. So, you get a better execution. On the other hand, that minute may be a trend following minute, right? On a second to second basis, say that means you have to actually execute immediately. That’s the right way to look at it. So break your time frame up into a signal time frame which you’ve already done and your execution time frame which is by definition much shorter than the signal time frame and you’re asking whether what the statistical characteristic is in that tiny time frame. What was what is the best way on that one minute time horizon to assess uh trend versus mean reversion? Would it be simply market structure? Oh no. Well, you need to get the tick data and you have to say, okay, at instant x uh there was a signal. Then you have to ask yourself uh actually that’s too complicated. There’s even a simpler way. Just basically take um uh one minute bars uh and and take and break each one minute bar into uh let’s say 1 second intervals. Oh, I thought you were collecting a series of one minute treads. It’s actually breaking down the one minute. breaking down the 1 minute into 1second intervals and then ask on a second to second basis. Does the price in this bar tend to go up? So, he’s waiting for an upside breakout or does it tend to go down? Because it could do either depending on the time frame and that’ll be consistent. By the way, this is the interesting thing. And then you know this is over the minute it’s mean reverting but over 15 minutes is trend following. That means you wait to trade. Um if on the other hand it’s trend following over the minute and trend following over 15 minutes then you got to buy it immediately. So this sounds great in theory but I’m visualizing how it executes upon this and um I mean you you you’re making a decision in seconds and you’re looking for confirmations in seconds which could be eroded in seconds. No, you know that already. You’ve already done the work. Okay. You don’t do it at at the time. I should have clarified. You’re just trying to ask statistically speaking. Got it. Is it the case that one minute bars are mean reverting or one minute bars are trend following by looking at lots of one minute bars and you’re asking what is the average behavior of the one of one minute bars and I’m assuming for sensible stop-loss placements you’re you’re not protecting yourself on the one minute uh trend correct it’s on the 15 minutes on the intraday correct okay otherwise you’re going to have a 1 to100 risk rule you know exactly cool exactly and then the one superpower I feel I have in my system is I I’m entering not that extremely but an intense version uh for most which is on the one a series of one minute candles. I’m entering on the one minute time frame and the reason I do that is yes I miss opportunities all the time and my temperament inspired by what you said I’m okay with that. I’m happy to miss opportunities. I don’t really feel FOMO that bad. But what I do like is if I have a confirmed entry I have a better win rates because I’ve got confirmation. I have a healthy risk-to-reward and also I’m able because of the price rejection points uh is pretty precise. I’m able to break even very fast uh just on a one minute shift in my direction. I protect myself. Yes, I get kicked out of trades all the time. Uh but when I it means I also avoid of a lot of losses where I didn’t understand the trend correctly or something changed but at least I was in I could take a partial at the session movements which isn’t that much and then the price may collapse on me because I got the trend wrong but I was able to catch a pocket. So a lot of my returns actually become wrong trades, wrong direction, but I go I grab the pockets. So the break even becomes very important for me. I know a lot of people don’t really use break even or believe in it. So I’m curious on your thoughts on how we use break even a trade as a tool. It’s completely dependent on the statistical properties of the thing you’re trading. Literally there’s nothing else to it. And so you’re just asking in this case um if it’s likelier than not to be breaking down and and I mean over you know a lot of data and over a fair number of of securities too not just this one particular thing at one particular time and so if the statist statistical properties are one way then you should be doing one thing and if the statistical properties the other way you should bring it the other way and you can just do that by getting tick data and figuring it out. It’s it’s not a difficult thing to do because all you want is averages. Next part I want to talk about is just scaling as an individual. So there is many ways one could scale but I’m curious at the path you chose which is uh hedge fund other people’s capital and then that becomes also a rat’s race in a form where you’re just chasing more under management the management fee becomes now something lucrative that could distract you. So where is the sweet spot you found in your career of u how to get the most out of markets and also keep it in its purest form without ending up corrupting yourself in in the areas you were describing earlier? Ah that’s a great question. The answer is by risking my own capital. So I the only thing I invest in more or less with some small exceptions is in my own fund. So my investors know that if they’re feeling pain, I’m feeling twice as much pain because this is all I’ve got. Um the nice thing about it is um was it Benjamin Franklin? I forget who said it that you know if we’re going to be hanged we should all be hanged together, you know, something like that. Okay. And that’s really true is that you need to be your portfolio. Skin in the game. Skin in the game. And that that’s that’s what solved it for me. Um the moment it was my skin in the game, it was much easier for me to think than otherwise because then I don’t have to worry about anything. I just do what I think is the right thing to do. How many years have you been participating in the market? Since I’ve been a money manager since

Oh, that’s older than me. I’m 30. Wow. Okay. And I’ve been I’ve been running the current fund that I have for 10 and a half years. The reason I ask is because I recently spoke to a guest on the show and he argued especially on intraday trading styles the market evolves every quarter meaning there’s there’s a high rate of you got to adjust your strategy in the decades that you’ve been involved in the markets how often have you had to reinvent your system subject to alpha degay potentially. Yeah. So that’s a good question. I um when I started off I was like everybody else. I tried to predict and that worked for a while. Worked very well actually. Then I decided that well you know I should be in the business of alpha. So I started doing alpha trading which is to say you know long this and short that. But then I realized 15 17 years ago at this point that the moment we start talking about light speed limits in trading this is a style of trading I’m not that interested in we need ncond execution nancond execution because that I mean look if you can make money from it good for you for me that’s boring I I can’t possibly be interested so I started wondering what I should be doing that doesn’t have this alpha decay problem and I came up with the radical idea of just don’t bother with alpha altogether and And I don’t I trade indexes and I trade indexes by taking a risk view. And what I’m trying to do is I’m trying to write models such that they can detect when the chance of a large draw down in the market is high. And if it’s if the chance is high, I’m going to be out of the market or have a partial position or something. Um but the rest of the time I wish to be long and levered. [snorts] And so I’m a combination of long and levered or cash and that’s it. It’s simple and it’s done on purpose so I don’t have to worry about alpha. So yeah, the last episode we spoke about your beliefs and your own trading systems, but I guess it hasn’t evolved in the last year. You’ve you’ve stayed true to what it was. I’ve been thinking about how to improve it, which is becoming harder and harder because it actually is pretty darn good. Um, and we made some small tweaks to improve it, which is great, but they’re not massive tweaks. No. Um, and that’s by design. Yeah. The the reason I ask is because when somebody goes through a losing period, the questions can become I’m the problem and my external or internal factors, the edge is the problem for decay or or it’s uh oh yeah, it’s just performance. It’s just performance or markets actually. It would be it would be or adherence these three. Yep. I guess the uh performance side can fluctuate a lot. The adherence side can fluctuate a lot. The alpha decay is probably slow moving. It’s not on a monthly, quarterly, yearly basis. That’s true. Um uh but the problem is that if you’re in the alpha game, you it’s always going to decay. And so you need a very large research staff to try to keep finding more and more sources of alpha. And with the advent of AI, ordinary sources of alpha are going to disappear. You’re going to need some really weird crossmarket sources of alpha. The simple stuff that people are doing even with AI is not going to work. Let’s get into this. This is a black box of a topic. AI and the future of trading. So uh I just came to know that just a few days ago Nvidia has launched a supercomputer that you can have at home and locally host a an LLM. So now everyone there’s just a democ democracy in access to supercomputes which was usually potentially an advantage for institutions that is now access to anyone at an affordable rate. How does that influence where the markets is headed? So well there’s several things that I I’m pretty sure going to happen. So the first thing that I’m fairly sure is going to happen um by the way this is all in the back of the Kindle edition of my book the science of free will I actually have a presentation of the future of AI in trading in it um the the first thing that I think is going to happen is that you will get both volatility compression and expansion simultaneously. So that is to say that on average or most of the time you should expect volatility to be compressed. you should expect not much volatility, not much to be happening. But then when it does explode, you should expect it to explode even higher than it used to. The reason is that again, everybody’s going to be doing the same thing at the same time in the same way. And when you do that, you actually are going to get exactly that pattern. Stuff is going to get arbitrageed away arbitrage arbitrageed away until something really bad happens and then bang here comes the risk. So that’s the first thing that is very likely to happen with AI. What kind of a cyber horizon would you expect a loadable change for this? It’s already happening. You can already see it. I mean um um even just the advent of computers produced things like that. Remember the flash crash? Just ordinary computers produce that. So you can just imagine with AI running more and more how much that’s going to happen. So that’s the first thing. The second thing is your ordinary sources of alpha are going to disappear. You can’t run around anymore saying I’m a value guy or I’m a growth guy. I’m a growth at a reasonable price guy. or whatever it is that that’s just all going to get arbited away very quickly and you can already see it. What is it 90% of mutual funds that are underperforming the S&P? I mean it’s some absurd number. Um right so already that that’s a problem. Um so that’s the second thing that’s going to happen for sure. The third thing that’s going to happen for sure is that and it’s already happening again is that there will be more and more meme stocks like AMC. Oh, right. because at this point retail traders are going to start to actually gain some power relative to the institutional traders and AI will help level that playing field particularly if the retail traders start to use AI. Um it’s already true that they’ll coordinate on Reddit and stuff like that and that’s their power over hedge fund managers so much that uh there was a guy that blew up famous guy because he shorted something and kept shorting it. I can’t remember the last four or five years. Some famous guy. Anyway, he blew up. Um uh because he had 50% of his portfolio in some short, which is madness. I don’t know why you would do that, but anyway, never mind. Um so that’s the second thing. And then the and then the other thing is those obvious sources of alpha because they’re going to disappear mean that you’re either going to have to have some knowledge, which is what I said earlier. You know, I’m a doctor. I know something. That’s one thing. Or you’re going to have to find weird sources of alpha. So that is to say sources of alpha that are coming because uh the stock market is is is being disrupted by something over there or by there or here whatever not something internal to the stock market itself and you need to figure out how to take advantage of that. Um and in all of those things actually I think retail investors have the upper hand. Um again because of this massive siloing that’s happening in institutional investment retail investors if they really want to put in the time are definitely in the driver’s seat. Would you say in this era of AI empowered alpha decay and the markets navigating towards a very predictable way as you were describing almost like elastic band being pulled back and then when you let go it flies with increasing volatility that momentum would be the purest form of edge. Yes. Um, so I recommend to everybody if they’re beginning traders that they start with momentum. The reason is that it puts you on the right side of every trend. You you if you’re actually trading momentum, you’re never going to have large losses. You’re always going to have small losses, which is a good thing unless you lever yourself through the roof, which you shouldn’t be doing. Um, and it puts the base rate in your favor. So that’s where you should always start. And then you can decide what you want to do, but you should always start with momentum trading of some kind. It’s easier to test. It’s easier to trade. And if you’re an individual invest investor, you don’t have execution problems because you don’t have that much size to push through. That’s where you should start. And then after that, add to it to your heart’s content, but start with something that’s going to work and not blow your head off. Curious to know in general your thoughts on how the world will change with AI. not necessarily with the lens of finance but in general because there is one think tank that I read a couple months back which was basically forecasting how it look in 10 years and it was cool a lot of paradoxes that were presented that I don’t have an answer for number one is let’s say these companies that have huge valuations which are finding increased profitability because they lay off their workforce AI creates um lower cost increased productivity better profits better earnings and this becomes a positive flywheel but then eventually you’re eroding the white collar jobs the middle class that is the consumer of said products and therefore now you have a portion of the economy with less buying power what happens to these companies you extend it out even to the property markets where you have property prices are are held up by mortgages I think it’s $13 trillion in global mortgages for the property markets but that’s relying on future earnings of jobs that may no longer exist what happens then so basically long story short what happens when in Manhattan uh 10,000 jobs can be replaced by uh one computer in the future uh in in a farm somewhere. What does the world look like then when you have PhDs driving Uber? It’s a good question. It’s effectively a status competition. So, here’s the problem. um until the advent of AI, you could say, listen, if you follow the rules, you go to some um uh you know, fancy school, you get a fancy degree, then you can get um you know, high paying job as a lawyer or as a consultant or whatever it is, and you know, you push paper around and you get, you know, your million dollars a year, $2 million a year, whatever it is. um that sort of upper middle class cushy life is going to be disrupted massively by AI and these are the very people that you would expect are going to be screaming from the rooftops about AI but there is an massive offsetting advantage one is productivity yes but the second thing is that the people that are going to gain from this are the people that do things in the physical world and that’s all the people we don’t think particularly if you’re in the white collar world. So that we’re talking about mechanics, blue collar jobs, blue collar jobs, HVAC workers, um you know, techs and so on. So um a year or two ago, uh this guy came over to uh fix or or service my heating system and I started chatting with him. Um his income from servicing heating systems and he’s an employee is $350,000 a year. And so I started laughing and I said, “Dude, why are you working?” Because he also told me about his other businesses. He owns four gyms. Um he has three houses he rents and so on and so forth. And I said, “So what are you working for?” He says, “Oh, my dad always told me that you have to have a job because otherwise it’s not steady income and anyway, I want health insurance.” And I said, “So this money you make from this job, does it matter?” Oh, no, no, no. It doesn’t. It’s it does not even a drop in the bucket. Um, now he might be an exception, but it’s not as big as of an exception as you might think. Um so uh surgeons are going to definitely uh benefit no question because they have to work with their hands. Um nurses same thing. Um radiologists their death has been called millions of times. I don’t know. That’s that’s one of those that’s on the fence. Pathologist. My dad and my brother are radiologists and they’re going to be listening in. Yeah. I don’t know. It’s a good question. People keep saying that computers will take over. So far um work for radiology has increased. So I don’t know what’s going to happen there. But basically if you do anything with your hands in the physical world your income is going to go up. If you are basically in the pro in in the business of essentially pushing out pieces of paper and large quantities of words your income is going to go down. There’s just no question about it because we’ve just made it incredibly cheap to produce words. And by the [snorts] way they’re going to produce better words than you. I’ve spoken to a variety of guests on the show and a unanimous common denominator between all of them is the emphasis they put on data and actually knowing the inner workings and the insight of your edge and your performance. That’s why I’m proud to bring a partner of the show, Tradzella, the number one journaling, back testing, and all-in-one insight experience created by traders for traders. What Tradezella really gives you is deep insights about your trading that would ordinarily not be visible. Whether it’s through understanding your trade types and playbooks or even insights powered by artificial intelligence through Zela AI. Whether you trade forex, futures, cryptos, the stock market, it all seamlessly connects to Tradzilla. So there is no additional work. You’ve seen me reference it dozens of times and all of the benefits I’ve had in my trading from the insights I found from my Tradzilla. So join myself and thousands of other viewers of the show. You’ll get the best discount using the link in the description or code toot for titans of tomorrow. So I proposed I mean when I came across this idea um not to go on a huge tangent but basically I was trying to think how do I navigate myself financially and over the last year I’ve got a heavy into just investing in the stock market it’s doing great because we’re bullish investing into property the U in the UAE and then the war kicked off that wasn’t ideal but I started to realize uh with the disruption of AI everything that I would consider my financial modes and passive investments will be affected by AI how it unfolds let’s see but then I kind concluded that the only timeless thing would be alpha or at least the the date by trading elements. Um would you say all forms of investing could be uncertain? Yes. But I would tell you one interesting thing. The chances are that AI is going to lead to more trading volume. We already see this with Robin Hood. That’s not even AI. It’s leading to more trading volume. AI is going to lead to more trading volume. So to the extent that there are people like you that are trying to scalp, the more trading volume there is, the better it is for you because there’ll be more icebergs, there’ll be more people driving things and giving directions or whatever it is. And if you’re trying to scalp that, you’re actually in good shape and you should do something with it. So that’s one very likely outcome of AI is that trading volumes are going to go up. So that’s one thing. Um the other thing is that suppliers to the AI companies, Nvidia is the obvious example but there are many others like Aino Moto in Japan um are going to benefit very substantially from the investment in uh in AI. Um it turns out Aenomoto which makes the monosodium glutamate also makes some polymer which they don’t want to talk about and that polymer is like the only source in the world for something that goes into AI chips. I don’t even know the exact details. So a genomoto is an AI investment. It’s not just monos sodium glutamate. So that’s the that’s the places where again you can find if you want to call it alpha that’s alpha um investments of that kind but otherwise just buy the market indexes. I mean the chances are AI is going to drive productivity growth and good productivity growth is generally good for corporate profits. the the productivity and growth I get uh but do you think like these AI companies the big ones will just slowly cannibalize the consulting firms it just cannibalize industries and then the money flows to a few fewer select group of hands yes but again with a caveat I’m sorry I’m full of caveats today um the caveat is that the sales functions within those consultancies are going to remain so I’ll give you an example um why does anyone to be brutal um hire McKenzie. They hire McKenzie 99 times out of 100 because the CEO has already decided what he wants to do. But he wants to have a piece of paper that says McKenzie and now he can do what he wanted to do in the first place. Right now maybe it’ll take two consultants instead of 60 to write that piece of paper. But that means that the guy that has the relationship with the CEO now became very important. So his income is actually probably going to go up. I.e. If you’re a partner at McKenzie, you’re probably laughing all the way to the bank because that’s what you do. You build client relationships. Um the same thing is probably going to be true in uh in other parts of investment banking and so on. The sales people, they’re going to make out like bandits. Yeah. So, I’ve undecided what my worldview is, but there is that dystopia angle. But then the other angle that that I’ve had this mental exercise with a few is number one right now the cost of computer is cheap by design to get mass adoption but eventually they’ll just jack up the cost and then it won’t be as competitive to replace with AI you might as well have a human if the cost is equivalent or or maybe even higher but then the other one was um I slipped my mind but yeah the cost of comput was was the main one where it maybe won’t be as detrimental. Oh yeah, the other one was um humans have always worked with tools and then now the tool is more advanced but it’s still the human working with the tool. So therefore it doesn’t eradicate all white color jobs. It just becomes less. Yes, that’s correct. So several things. The first is the cost of compute will also though be limited by the fact that these these companies are all competing with each other and there is some chance it’s becoming more and more possible that the compute will become the not only the compute but the models will become commoditized. So in other words, they won’t really be able to charge such a massive premium because if they do, somebody else will undercut them. That’s one possibility. Um the second is that Elon Musk succeeds in this idea of putting data centers in space and it’s actually cheap. I’m actually trying to write a Substack on that to figure out if he’s right or wrong from physics. Um that would also lower the cost of compute. And then the third thing is that no matter what AI does, like every other change in technology, every change in technology makes something cheap and makes something else scarce relative to what it just made cheap. So uh when the internet first came about um what happened is that it made moving data very very cheap. So if moving data becomes very very cheap then the scarcity is the stuff that’s on the edge i.e the the computers and that’s why you got Dell computer and Apple and so on and so forth that went up in value. It’s the same idea here. AI will make some stuff cheap and some stuff expensive. And the question is what? To bring it back to trading and maybe we can end off on this topic is I know last year when I spoke to you when AI was still in its primitive stages, a lot has changed but already you were running a very lean hedge fund because you were using a lot of AI systems. Well, now so much more advancements. How do you use AI specifically in finance trading whether it’s through the alpha creation process or is it just efficiencies in the team and logistics of the fund. So the biggest thing that it does for you is it increases the rate at which you can try new ideas and test them and figure out if they’re if they’re nonsense or not. And to do that um what you need to do is create skills. So the future of human employment is not going to be just humans. It’s going to be humans plus their skills which they’ve created themselves which are text files in in this case that live within your coding agent right and one of the things that we as quants realized many years ago is that a discretionary trader keeps their information in their head a quant can write it down because the quant can write it down their employment contracts can be much more strict than those for discretionary traders this is why you get all these litigations you took my strategy whatever it is that problem is about to hit all employment where you can write down skills, right? But what you need to do then as a hedge fund manager is own your own IP and your IP is a combination of you plus the skills that you have taught your agent. So I have something like 70 skills that I’ve written myself now that teach my agents what I want them to do, how I want them to think, what mistakes they typically make, how to avoid those mistakes, what kind of back tests I like, etc. And so now because I’ve taught them this stuff, it’s like training a junior analyst, um they when they produce results to me, they’re generally correct and so I can I can most of the time trust them. So that’s what you need to do. You need to create skills and write them down and then accelerate your development process. I think everyone listening understands the power of AI and the possibilities with AI and trading. But even just this where you mentioned empowering your agents with certain skills and parameters which is your in your head you’ve which is in your [clears throat] head and you’ve put it down into the code sounds great. I can imagine the the power of the execution and the ability you unlock. But how do I start? Where does someone beginning doing this and and I guess you’ve been selftaught because it all is new to everyone. Yes. So the best way to do it is to just start. So pay 200 bucks a month to either uh OpenAI or Claude. Start with one of them. download their coding agent and then just start playing with it. Install it, put it in your command line because it’s done through the shell or you can use the app but it’s better done on the command line and just start asking it questions. Any questions you like as general as you want and it’s going to make mistakes. As it makes mistakes, write down or note the mistakes or make a mental note and then as you realize the kinds of mistakes it makes, you say, “Okay, listen. Don’t make these mistakes anymore. Here are the mistakes you made. Please make a note of this. Turn it into a skill.” uh and you just keep doing that and iterating and you re and if you do this relatively every day which I do in a month you’ll have taught it most of what you needed to know and the rest is just refinement but that’s what you got to do is one of the starting steps connecting it to a charting platform or connecting it to your historic trade data uh connect it to the historic trade data and you can do that basically I I like to have text files because they’re obviously most portable CSV is the best format comma separation separated values. Put all your data into commerce separated values. Point your um uh coding agent at it and tell it and ask it a question. Tell it to go into the data and find something. Just ask a specific question and then see what happens and just start playing and you’ll [clears throat] find it very quickly. If you play and you’re relaxed about it and no money depends on it because you’re just playing it, it comes very quickly. So it’s a it’s a reiterations of trial and error to find solidity with AI and trading. Completely very interesting. Okay. And to wrap up because it’s been a wicked episode. I want to give you an open mic. Any words, advice, wisdom for the audience listening in? Yes. Um I would like to expand on the last thing I said, which is um everybody needs to understand that it isn’t going to be just you anymore. It’s you and the swarm of agents that you’ve created. It’s time to get on the AI bandwagon now. Start creating agents for yourself. And by the way, on the flip side, if you are an employee at a firm and you start creating agents, those agents or or rather the teaching you give the agents is owned by your firm. So now you need to think about what that means when you’ve taught that agent the best practices. So let’s say you’re a litigator and you you’ve taught them the best way to look at discovery material. Now you’ve decided to leave the firm. Guess what? That’s the firm’s property, not yours. What are you going to do about that? So, you need to think about this. Um, quant traders have had this problem forever. Everybody else is about to face it. So, this is what I think is the biggest takeaway from AI. People need to think about the fact that it’s going to be them and their swarm of agents that they’ve trained. What an episode. Samir, thank you for joining us today. Thank you so much. I enjoyed that. That was great. Let’s go. Let’s go. Brilliant, man. Brilliant. Thank you.