Transcription
In this video, I will take you behind the scenes of my risk management protocols, from how I manage a prop firm account, to how I manage a competition account, to how I manage my personal account.
The prop firm risk geometry. It's different from what they sell on social media. Okay? It's not about sniper entry. It's not about high risk-to-reward, low win rate. This is the blueprint for failure in a prop environment. But this, and all the way up to hedge fund risk management protocols, because while retail only focuses on technical analysis and strategies, professionals' only holy grail is risk management. So I'm sure this video will help you to set up your risk management and money management better in your trading. Let's get started.
Let's go over the risk profile blueprints. Let's start with the first one. And the first one that we will cover today is the hedge fund approach. What is the goal of the hedge fund? The goal of the hedge fund is to maximize sustainable risk-adjusted compounding over time, at the same time achieving the highest possible long-term return for a given level of acceptable risk and survivability. The rule should be that you need to do this following all the legal guidelines of the regulators. But as you know, there are thousands of cases where the hedge fund maybe takes a fine of 200, 300 million, but they make a profit for market manipulation of one, two, three billion. Okay? So this one should be like this. Okay. But the top five, the top 10 hedge funds in the world have multiple cases related to this to market manipulation. Okay.
And let's go over the hedge fund structure and risk framework. They are diversified. Okay? Okay. They are risk-aware and they point at asymmetric return because they want the minimum amount of risk and the potential upside, okay, related to this. So they start from the risk protocol. They don't start from the raw return protocol, for example, like a competition environment. Okay.
And let's go over how hedge funds diversify. They diversify by strategy. You can have global macro, you can have driven, you can have quantitative, you can have trend following. You can have a blend of these. You can have a multi-strategy. You can have arbitrage, you can have HFT. There are thousands of strategies that you can implement in a hedge fund. And they point at having multiple of them to have the best possible diversification. Recently, we saw how much this one is important when we saw the conflict in the Middle East, no? Because if you were completely exposed, let's say, for example, you are a hedge fund that is exposed in the financial markets of the UAE, you would have taken a huge hit recently. But they have a diversified global exposure. So they point at minimizing this.
Also, the legal structure. We will not go over the legal structure, but they are divided in the portfolio management side. They have research and analytics. They have operations in the back office. The trading, the execution side is divided. They have a department that is for risk profiling. So it's all separated. Okay? They have auditors. They have their own infrastructure to execute. They have the legal side. So it's a whole corporate entity, okay? It's not a strategy that you put there and you open the hedge fund. It's a different entity. And this will be my dream because this excites me so much. Right now, I only managed to be in the research and analytics side for the hedge fund, but I still didn't manage to open mine. And I hope in 2027, 2027, 2028, I will manage to do it.
Now, going on, we can see, for example, the 2025 performance of multiple hedge funds using aggregated strategy. Okay? They can use quant macro, they can use CTA, they can use credit, they can use event-driven, they can use convertible, volatility, discretionary macro, all of them. And you can see the index performance. And here comes the usual retail trader, okay? Okay. And the usual retail trader sees this. They only watch the performance, the raw return, okay? Like they do in the competition. And they see, "Oh, you did 20% with MSCI World, you did 17.5% in the S&P 500." So it's useless. All the infrastructure for this year, 2025, of the hedge fund is not like this. And I will explain you why. Because the goal of the hedge fund is not to optimize for maximum return. The goal of the hedge fund is to optimize for maximum risk. Okay? So they want to take the minimum possible amount of risk and they put a maximum on it. Okay? And they see how much they can do with this maximum amount of risk. Maybe to make this 20% in 2025, they had a maximum drawdown during Trump tariffs, for example, of 30%, 20%, 18%. And maybe to do this 14%, they had the maximum drawdown of 5% on the account. So you can understand that when you introduce the concept of risk-adjusted return, everything changes. And if you optimize for maximum performance, raw return, you are focusing yourself on only one metric of the trading world.
Now, going on, how, what will you think about this? Okay. In 1988 and 1989, the Medallion Fund, someone should know it, did 15%, okay, and 1%. So if Jim Simons was there online, okay, talking everywhere about his model or about his infrastructure and about his enthusiasm of starting this quantitative journey, he would have been attacked. You'd say, "Look, you cannot even beat the S&P 500 return. Okay, you did 1% here and 15% here." And this is the reason why professionals often laugh at the comments of retail traders. No, because they don't understand that maybe he had this with 10 times less risk. Okay? And in trading, you can have a negative week, you can have a negative month, and in the worst case, you can also have a negative full year. Okay? Specifically if you have something that is called the black swan, something unpredictable. Let's take, for example, what happened in the UAE now, okay, on the real estate side. There was a black swan. Okay? You had a flash crash of the prices on the value of real estate. Already recovered, but you had something like this. And if you were exposed in geopolitical conditions on this field, you would have experienced something that is not measurable in the in-sample. So you can test everything, you can make all the valuations, but you could have not predicted something like this. Okay? So the concept of return, they correlate from the market and the risk-adjusted return.
The outline of Medallion. Now let's see how they behave. As you can see, if you take only this portion, okay, you will think, "No, it doesn't make sense for me to invest in the Medallion Fund." But then you go in the next year and you see the S&P 500 closed negative, okay, and they closed 57%, okay, when the S&P 500 closed -26.6, they closed 43%. Okay? And then the momentum and the compounding also start to kick in. What you want to do with the hedge fund is minimize this. Okay? For an investor to have three negative years, okay, on the S&P 500 return will be a disaster. Okay? In 2008, when the market did minus 46%, they did a 92.5%. Okay? And this is important. You don't need to take only the isolated return. You need to take the risk-adjusted return related to it. Okay?
For example, this is the project that is only B2B. Hedge Forge, uh, in mainland Abu Dhabi, that I'm using to provide research to the hedge fund. Okay? So I'm in this side here. I don't talk about this on social media, but I'm also active, okay, in the hedge fund perspective, not operationally to build a structure like this. I think it will take me still one or two years, and I will need to do an audit out-of-sample of each model I trade independently if I want to implement here. And also the liquidity capabilities of the model. This is something that no retail will think about, but every model has a maximum liquidity capability. Okay? You cannot put billions in a scalping model. Okay? You will take too much slippage, and the model will not be profitable. So you need also to do a lot of measurements that either in the retail world are completely ignored.
Now, let's go on the prop risk model breakdown. The goal here is high pass rate and high probability of payout.
Now, the prop firm risk geometry survival blueprint. It's different from what they sell on social media. Okay? It's not about sniper entry. It's not about high risk-to-reward, low win rate. This is the blueprint for failure in a prop environment, but it's understanding the risk geometry based on the rules of the prop firm. Okay?
And here I did some infographics to explain why high win rate risk geometry is the best for prop firms because the goal here is to optimize for the variance of the results. Okay? So you don't want to have a huge amount of profit, but variable. Maybe the strategy in the month after does like this. So does a crazy amount of profit, but you have a higher probability of getting consecutive losses. And over a large number of samples, this one goes way higher because the black swan is always there. And maybe the strategy is profitable, but the variation is too high of burning the account. What you have here, the high win rate geometry is going on models that have win rates above 60%, 70%, and are capable of getting you a higher probability of pass rate, a lower risk of ruin, and a lower max drawdown. And you will think, "Oh, but it takes me more time to pass the prop firm because I cannot take the sniper one-to-10 that happens one time every week." Yes, I agree with you. But maybe in the attempt to take that sniper, you go in the low variance and you lose. Okay? You lose your account. Remember that is not a personal account. You cannot manage it as a personal account.
And here I show some tests to pass two probabilities. And you can see that both strategies are profitable. So you can have a strategy that goes around this line, okay, or around this line, or around this line, but you have a higher variance. Why? Because you could incur a large amount of drawdown. This is a simulation on only three months, okay? Not on the long term for scalping. So these are both valuable strategies, but this one has a lower cone of probabilities, okay? So the amount that you will incur a streak of stop-losses that will make you waste the money for the prop firm is lower. And that's the reason I strongly analyze that from a probability perspective, the best approach that you can have is the low variance risk geometry. So having a high win rate, okay, moderate risk per trade, and going for moderate risk-reward, one-to-one, even one-to-0.75, maximum one-to-1.5. This is the best for prop firm survival, okay, specifically in the pass rate.
And here I gave you a summary of the optimal low variance, okay, and the high variance path. Okay? They can make way more profit than this, okay? But they can be inconsistent. So maybe you have a period like this, and then you have three amazing months. But in the period like this, you burn the prop firm. Okay? So you need to understand that this needs to be treated following the game theory. Okay? You need to understand the goal. You need to understand the rules. I strongly think that prop firm trading is way different from real trading. Okay? Because you are not trading the real market. You are trading the rules that another entity put on you. So you are trading a synthetic market.
Here is the prop firm risk geometry across phases. Okay? You can see that on the evaluation phase, the best one is the low variance zone. Okay? And on the funded phase, you need to optimize for high outcome because you want to get a payout. Okay? The goal here is to get a payout. And there is a thing that no one considers here. Okay? And I will share one experience about my life. Some years ago, there was one prop firm in the CFD space, I will not make the name, that was based in Canada. Okay? And I requested from them more than $50,000 in one payout. Okay? When I received in the bank account that payout, this payout got flagged for review. Why? Because the prop was having problems with the regulator. So you also need to input the probability that the prop firm has way more risk than a brokerage. Okay? So compounding on the prop and keeping all the money on the prop is not the best option. Okay? The best option is to withdraw as much as you can and put it in your personal account.
Here I did some projection of one approach that I've made. This one, I think was The Trading Pit. I traded live the prop firm some years ago with the community, using the order flow approach, and I tried to go on the side of the low variance spot. This one, I use a risk-reward that it's even too high, but I had a really high win rate on this one. It was a phase where the regime of the market was completely directional. And I did some stress tests to pass the prop firm. And after a lot of tests and also a lot of review on the community, I noticed that the high win rate model with low variance and with low risk-to-reward is the best one for probability of passing it. Okay? Then you can go on this side on aggressive optimization. And yes, if you have a high probability model, you could try to optimize for the maximum amount of withdrawal. But then you need to consider also the rules about the prop. Okay? Because some props have consistency rules.
Now, let's go in the CME Group Equity Cap breakdown. The goal here is to build equity fast with minimum downside possible, extremely limited time, and symmetric incentives. Okay? The rules: single account, so you cannot test on sequential real accounts, no test. Okay? And here you have an extremely limited time. So in the Equity Cup, the CME Group, for those who don't know, the CME Group is the biggest exchange in the world, and they organize a cup one time per year. Okay? One time I decided to participate because their platform is really amazing. Okay? And what I did here is I had only one approach. So I didn't have time to test regimes, and I didn't even have the possibility to test the live market environment. So what I did, I used the best model that I have. And the best model that I have is this one: the order flow, high win rate, scalping, low variance, the one that you saw me using in all the podcasts in the world where I traded live or in the live trading sessions. Okay? And the goal here is to unlock margin to go heavier. So, as you can see here, the goal is to keep the drawdown as low as possible because the time is limited. So if you go in drawdown, you have even less margin. So it's almost guaranteed that you will not be in the top percentile of the competition. Okay? And I got a positive drift in the beginning. Okay? I was positioned in the top 0.7, 0.6%, I don't remember, 0.6%. Okay? And the risk-reward payoff is amazing. So everyone that will see this, okay, will say, "Okay, you did 50, 60 executions, you had a net profit of 54.6%, a maximum drawdown of 0.42, 42%." You are better than funds. No, no, no, no, no. First of all, the risk protocol and approach that got used here is aggressive. This protocol here, everyone will be crazy to apply this protocol in a hedge fund. Okay? Second of all, okay, you are approaching here in a positive drift. So in this case here, probably I took this. Okay? I took this positive. If you continue with this risk when you take the negative, okay, maybe you will approach a drawdown that is 10 times. This is still acceptable, still acceptable. But maybe for a certain prop firm that allocates a maximum risk per model, not prop firm, sorry, hedge fund, a maximum risk per model, is not acceptable.
Here are all the executions that I took in the cup. And as you can see, okay, you have the execution compounding on the abet. When I release margin, okay, the more I was releasing margin, the more I was getting aggressive. Okay? And on this one also, what you can see is that the incentive, okay, is to optimize not only for performance but to optimize also for maximum drawdown because the leaderboard and the competition is composed that everyone, okay, can see the maximum profit and the max drawdown of the user. Okay? So it's a completely different competition environment here. And also you cannot test, you don't have time to test the regime and to test sequentially. You need to go there with the model that you are the most confident, and you have one opportunity. Okay?
Now, let's go on the World Trading Cup competition. Okay? The goal: terminal rank position, convex upside, and symmetric incentive, limited time. Why terminal rank position? Because in this cup here, and you can see from the composition of the leaderboard, the final goal is the percentage. Okay? So the approach is totally different from a personal account. Okay? Also from a prop firm account and from the hedge fund approach. Okay? And I explain this because I request, I see, I receive a lot of requests related to, "But why, if you can do 500%, okay, you don't do this in a hedge fund that you don't manage capital?" Guys, you don't understand the goal of the competition. The goal of the competition is not to optimize for risk of ruin. It's not to optimize for maximum drawdown. It's optimized, okay, for terminal rank positioning. And in game theory, we will go over this. Okay? You need to understand first of all, the goal of what you are doing is different from personal, is different from hedge fund, is different from prop firm, is different from active management certificate, is different on investment. It's completely different on everything. And you need to understand this. So the rules: you don't need to use simultaneous accounts to avoid hedging, for example, one account long or one account short that can influence the performance or illegal trading tactics, for example, I don't know, arbitrage or something like this. And sequential accounts are allowed. So you are allowed to test multiple models and strategies. Of course, in a short-term competition, you will be stupid, okay, to test too many strategies because you don't have the time. So you don't have the time to appear in the leaderboard. Then, from a game theory perspective, this completely changed the optimal equilibrium strategy. So you need to take controlled enough, controlled variance to maximize profitability of top percentile outcome, at the same time minimizing the probability of total elimination. Nash equilibrium implies that the optimal strategy depends on how other competitors are expected to behave under the same incentives. Okay? Most participants deploy aggression too early. If you open the leaderboard in the beginning, it's 200%, 100%, and then they disappear. Okay? Because they optimize for immediate ranking instead of long-term positioning. Okay? Early uncontrolled aggression increases the variance before you have enough information, first of all, about the regime of the market. So if the market is compressing like it was last week, okay, or if the market is expanding under the peace treaty of Trump and Iran, you need to use completely different models. Okay? And the leaderboard structure, okay, here is known. Okay? So at the second, third month, you know what's your goal. And if your goal is to beat, for example, hi to my friend Nazrian, that did that 220%, you cannot use a conservative approach. You will not even appear in the leaderboard. So you need to input the maximum amount of aggression in the late stage. But before, you need to know who is your competitor. Accounting for World Trading Cup commission and real execution cost prevents false positive expectancy models. For example, I deployed models, I tried models in the World Trading Cup that comes from the quant, for example, the automated. And you can see that with the markup on execution, okay, and this slippage, they are making profit, but never enough to win this kind of competition. Okay? So the first one framework: optimize information acquisition, validation, and then asymmetric payoff deployment last. This is the three-month competition structure. This is the blueprint. Okay? For this kind of competition, any competition or environment can have it different.
Phase one is regime detection and testing out-of-sample. So you have strategies that are profitable on the long term. But remember that here you have only three months. So you need to test in a live market environment. So in the cup, okay, three strategies. In my case, I used one account, the trend following strategy. I tested one account that has the mean reverting, and one account I tested the volatility breakout. Okay? And I have a matrix of the one that is sustainable to scale risk. If I see that the regime is compressing, the trend following strategy, even if it's profitable on the long term, in this phase here, will not give me a leaderboard positioning. So this one is discarded if the market is rewarding the compression. Okay? And I see in the first month, okay, here I use this one. Okay?
The second month is validation and moderate risk. So what you need to do on this first phase test. Okay? The mean reverting strategy. Perfect. So you open the account. Okay? And you deploy risk on the mean reverting strategy. And you say, "Oh, but you lost one month for testing." No, you didn't lose one month because first of all, now you have information about who is your enemy, the leaderboard. And second of all, you have the perfect model for the regime. Okay? Now you can test also this one. Okay? In a simulation environment, but in a simulation environment, you will not have the exact commission, and you will not have the exact live slippage. And as you know, for models that have like the one that I use, 500, 600 executions per quarter, the slippage and commission are a huge part of it. Okay? So in the second part, you use the model, you validate it. Okay? And you start to scale. You start to scale using this risk protocol.
In the third phase is when you do convexity exploitation. Okay? So you want to be in the first lot, second slot, third slot. Usually, they are above 50 to 150%. Okay? So it's a huge performance for three months. So you need to deploy maximum aggression on this. And here is when the maximum amount of execution is concentrated in the second half of the middle month. And the third month is where I make three, 400 trades. I stay there six hours per day scalping with the model that I choose. And this is what gave me then the results. Okay?
If I go on the one-year competition, it's possible that I will do it. Okay? I will use the quarter for testing. Okay? So I will open an account with trend following, test mean reverting, volatility, and I will choose the winner. Okay? To then approach extreme risk and deploying capital in this. Okay? So I will probably make in nine months the total that I make in 12 months of competition. Okay? You will make probably more than 2,000 executions. And it's really important that you make this test properly because if you don't account for commission and you do this test in simulation, not real account, if you don't put the correct slippage, a profitable strategy can become an unprofitable strategy. Okay?
This is the validation. I already published this. This is the podium account reconstruction performance. The cumulative month is the 68.22% uh 22% maximum drawdown. We had 89.5%, 158% cumulative. Ah, this one went losing. As you can see, even if you make all the possible tests here, I lost two weeks. Okay? Not I lost, but I was in drawdown for the first week and I recovered for the second two. Okay? And then I deployed aggression in the late phase. It was not enough to win this performance. For a hedge fund, it's amazing. Okay? And this is why you cannot compare a hedge fund with a competition or a hedge fund with a personal account. Okay?
And for example, I give you the breakdown of the first one. The consideration that I make is, someone considers to join swing trading. Have is too slow. You have for the quarterly competition, you don't have enough sample. If you want to win, you should take a huge amount of risk in one or two positions. And from my point of view, winning a world cup with three, four, five positions, it's not a sample validation that will give you any kind of satisfaction. Okay? Because it can also be luck. And, so it's not enough sample. This one is discarded. And also the small account. So if you enter with $25,500, you pay more commission. Okay? Because you need to open micro, and your edge will pay a huge amount of commission. It will be deployed. Okay? It will, sorry, it will not survive the execution of 300, 400, 500 trades. Okay? This is all stuff that I tested. I understood. No, if you want to optimize for optimal rank positioning, you will never survive with this.
So the regime test, for example, I give you the example of the first one. The mean reverting, bad performance, the market was directional. Okay? The account two, trend following, amazing performance, directional market. This is the first choice. This is the second choice. This is the third choice. Okay? This one is discarded. This one is discarded. This is the model that is telling me, "You can deploy risk." Okay? Perfect. I open the account that will go to podium. Okay? Where I deploy risk. And on this one, I go aggressively trend following scalping. Okay? 500, 600 in one quarter, if I'm not wrong, 650 executions. Of course, these results are not typical. This doesn't mean that if you use this approach, you will have the same results. And all the disclaimer.
Now, let's go on the personal account. What's the goal on the personal account? Risk-adjusted return, psychological stability, and long-term compounding. Only moderate risk tolerance. You can compound profit. Okay? So if you are not super conservative, but you are moderate, like me, and you have a month where you make an exceptional performance, I don't know, six figures in profit, you could use part of this profit to input additional risk in the next sample, maybe the AAAA sample. Okay? This is, of course, a moderate approach. It's not a conservative approach. You also need to have multiple accounts and multiple brokerages on the personal account. And I explain you why. I had a trauma, I don't know how many of you remember, with FTX. Okay? I was ranking in the leaderboard of FTX. I was an extremely active crypto scalper. And in this case, FTX decided to shut down everything. So I tried to access to the portal. There is no access. Okay? This created for me a trauma on the single brokerage use for the field. So if I do crypto now, I don't use only one broker. Okay? If I do futures, I have four, five, six brokers. Okay? Maybe one is perfect for more scalping, one is more perfect for investment. Okay? But I use multiple of them, and they need to have insurance. You need to isolate the risk. Okay? Unpredictable situations like the failure of the brokerage, the default, this can happen, guys. Okay? So if you keep everything in one, maybe you have an extremely accurate risk management in your personal account, then something happens, and you lose everything. Okay? So the diversification is not with only models and accounts, but it's also with multiple brokerages and models. Okay? Discretionary, algo, short-term, long-term. For example, I have one portfolio that is not managed by me, but is managed by one person that is paid by me to do exactly this, and it's the options quant portfolio with algorithmic strategy on the option side. Okay? I have my discretionary models, and I have also an algo model. One model that Luca validated on the MultiCharts, and also we received an audit from a market maker that generated 30 million profit. Mate, he did an audit, no, of the model because I showed the model logic on YouTube for free, the IVB model, and he validated if the volume was actually contributing to improve the model. Okay? Did a video about it and also texted me to give me to say, "Look, it looks really amazing." Okay? And so if you need to have multiple models, okay, you need also to diversify with multiple platforms, with multiple brokerage accounts. Okay? And in this is the personal account strategy. I know I'm moderate, but if you want to be really conservative, it should be 1.25%, 0.5% per trade. Okay? And I know that now the retail trading world will say, "But you are crazy." But it's, I will never live with $1,000 if I put this risk. The problem is not the risk that I'm saying. The problem is the conception in the industry where you can live out of trading with $500, $1,000 because this is what your mentor told you. Okay? And the fact that he put a photo with a Lamborghini convinced you that you can live, okay, with $500. If this was possible, okay, if living with this small amount was possible, okay, I would be billionaire with the amount that I trade, okay, but it's not like this because the risk profiling is extremely important, specifically for survival rate. If you risk, for example, I saw some people risking 5%. Okay? If you incur in a streak of drawdown, you will destroy.
And as you see, look at how the pyramid is inverted. The pyramid of risk is the opposite of the competition. First point is stability. Okay? Second point is risk management, risk-adjusted return, and in the end, the return. Here, the goal: terminal rank positioning. So the return. That's the reason it's called a competition. Even, okay, on this, I think this one was the best experience that I've ever had of stress. Okay? Because you are competing against people that input a huge amount of risk, like big props to Nazrian. He's a footprint scalper. And you need to push yourself. Even if I will be happy already with the performance that I did the first two weeks, okay, but no, if you want to be in the leaderboard, you need to step feet on the gas. And you need, of course, to have a plan, okay, a tolerance of risk. You need to have an execution model that should be also automated in terms of blocking the maximum risk per day, the maximum risk per week. And you need to track. If there is no measurement, there is no improvement in trading. Okay?
And then there is all the adaptation and optimization part that quant traders are well aware of, but retail traders never do. So if, for example, I give you an example, no, on this strategy, the IVB, that is based on the fact that the nature of the US equities have predictable behavior during the opening, okay, and that the battle of the first 30 minutes is giving a lot of value in the direction of the day, can be improved, for example, can be improved with option flow. And this is the study that I'm doing. But a lot of people are not happy. Okay? They they don't want to do this. Okay? They are happy with what they have at the moment. So what they do is just, "Till it's profitable, I will not implement new concepts." And this is the reason then the strategy breaks on the investment side. Okay?
I have multiple models. One of the models is the earnings surprise, the drift portfolio. This is a well-documented situation that is post-earnings drift. Surprise, of course, when there is a mismatch between the expectation and the surprise. They use a model for this. I use blockchain intelligence. This is one model that I created for the long-term investment to create an accumulation plan on Bitcoin that is smart, that is based on this mathematical constraint and infrastructure. So it takes into account the miner also. I never talk about crypto in this channel. Please let me know if you would like me to make a masterclass about the blockchain intelligence model on the long term for free. Okay? And then I do a lot of data research. Okay? So I try to see who, which kind of holders, and how much it's heavy these holders are buying certain stocks. If it's in the congress members, so it's a politician. If it's, for example, a hedge fund that can deploy a huge amount of capital, like BlackRock, they have always the best analysis team. And these data are public, are completely legal. Okay? It's published from the government. So if you have enough time and you want to do research, this one is an amazing way to build a stock portfolio. Okay?
And this is the five-point of the risk protocol that I use in all environments of trading. I hope you enjoyed it. Let me know in the comments if you want more videos like this. And if you would like to give me a breakdown of the long-term model for free in the YouTube channel. Have a great day. If you enjoy this video, leave a like and a comment and let me know which risk protocol works best for you.