STEAL This Trading Champion’s Exact Strategy - Math Based Models for Prop Firms

A model that I actually built for prop firm.

This is how to master prop firm.

This is how to bring prop trading to a next level using math. This is so controversial because this method responds to put advantage on losing traders and cut legs on profitable traders. So what we can use, we will use them.

This guy is a futures trading champion at just 20 years old with over 200% in total performance in just 9 months.

Introducing Gianluca I was a discretionary trader. I lost money. I was a mechanical trader that made me profitable. When I turned to statistical trading, that is what put me on a podium multiple times. I'm getting a 60% win rate with 1 to 4 risk reward. First is my favorite one, and it's the only metrics that can tell you if your strategy will make money or not. The model that I used in the championship that is called—

This is someone who traded ICT, traded discretionary setups, and instead developed a statistical trading He has created automation around his entire trading, and for the first time, he breaks down step by step the exact statistics and process to building your own automated trading model.

Statistical trading doesn't make a strategy profitable. You take a strategy that already works and put it on steroids. People will think that a strategy with a 75% loss rate is really bad. I think that this is a blessing.

Why? Because Not only that, he also shares specific models that he uses on prop firms, and in this episode, he shares them all with you.

Let's go to the most advanced risk model that I built on my strategy. This is the part that breaks people's brains. It goes deeper on why I do some things. So what you have to do—

Not only do we have a verified trader, but we have a world champion, someone who has ranked 3 times in the Robbins World Cup of Trading in the Forex division and actually won twice. And in a 9-month period has achieved over 200% return. And not only that, that's obviously an incredible accolade, but on top of which he's also done payouts on prop firms, multiple 6 figures at that, using this exact— and I say strategy, but that might not be the right word. I'll tell you why, because everything we're going to cover today is extremely unique. This is someone who has really dived deep into the automation and, more importantly, mechanical trading systems. He actually was an SMC trader, but most of all, he found a way to make it all objective and make it all statistical. And we're going to find out how he did that. This is a masterclass. Make sure you're ready. And I think you know who it is, and if you don't, you're about to know. It's the one and only Gianluca Bruni.

Thank you so much for the introduction. It's really an honor for me to be here. And today we will deep dive into the model that took me from a losing trader to a world champion trader at 22. So it will be a little bit different from everything that you bring here because my approach, I think that it's really unique, but I think that it will be very helpful for all the audience looking here. We will see basically 3 main things: how to take your prop trading level to a next level. So how do you gain payout even if your strategy isn't doing well? The second model is the most advanced risk model that I ever built on my trading. And the third one will be a surprise. But trust me, you will know— trust me, you'll want to stay for it because it's a really amazing method that I use. So let's start with the beginning. So I was— I started as a losing trader, as I told you, as I told to you. I started as a discretionary trader, and, you know, I started trading 6 years ago, so when I was 16 or 17 years old. And it was so difficult at that time to find information because, you know, now the educational world in trading is also thankful to you, is really a better place. But when I started trading, there were some guys telling me all different kinds of things, and it was so difficult to recognize what's true and what is not true, what's right and what is not right in trading. Because you don't have a traditional path. You don't have like— to be a doctor, you need a degree.

Yeah.

That's it. In trading, you can be profitable even with your feelings. Okay? Because you see the market, you feel, and I know some trader that does that, like Niku, I don't know if you know it, or Fabio is an incredible discretionary trader, or even Andrea. But I always thought that it was not a good approach for me. So what was the problem with discretionary trading? That every educator, every YouTube channel was saying something different.

Mm-hmm.

One swore that the truth of trading was by harmonic scanner. I don't know if you remember that, but it was really a—

I think big thing.

Yeah, big things. Another one said forget harmonic scanner, it's all about Fibonacci and Elliott. So you just have to wait for the 61.8 level. For sure you remember it. Yeah, I know it.

As do you.

Okay, okay. And another one say forget all of that. Just use ICT. So I trusted all of them, and I lost all the money that I have. And when I was a teenager, and it was so bad because I was doing something and everyone was watching me losing money, and they thought that I was like a gambler or something like that. And maybe they weren't wrong. So I understood one thing that Every educator out there were selling you concept.

Yeah.

But I understood with my pain that concepts are a scam. Why concepts are a scam? Because a concept tells you nothing about how to be profitable in trading. It tells you really nothing because remember that a concept, knowing a concept makes you feel smart. A strategy instead makes you money. And that's the really difference. What's the point of knowing what is the fair value gap? No point. Okay? You understand that? A concept can be useful once it's united to a strategy. So multiple concepts into a strategy that can be tested across thousands of trades, and this can be useful. Because, you know, a concept is something that we talk on a meeting, on a trading meeting. Yeah, I use this concept. Yeah, it's good. You should try. A strategy is something that we can help another person, we can teach another person, and he can actually make money with that. So that's the big difference.

Would you say the difference between concept and strategy is that concept is subjective? You know, there's an idea there, but, and, you know, it's open to interpretation.

Yeah, it's open to interpretation.

With strategy, As you say, you build it, it's based on statistics and data. Even if someone else has it, it should be the exact same, like for like.

Yeah, that's it. And I think also that a strategy tells you so many things, like when you have to enter on a trade, when you have to get out on a trade, when you have to wait. It's multiple concepts united. Okay? That's the big difference. So let's talk about... discretionary trading and mechanical trading, because we also— this debate between discretionary, mechanical, and one say that discretionary is the future of trading. You should spend hours and hours on the chart until you feel it. That's, for me, it's not working. But the other part is mechanical trader. But for me, there is a third part, statistical trader. Today, I would define myself as a statistical trader. I was a discretionary trader. I lost money. I was a mechanical trader, and that made me profitable. And when I turned to statistical trading, that is what put me on a podium multiple times. So after losing money for more than 2 years, I finally clicked. I started to understand what I need to make money actually in trading. Because, you know, I was always guessing on the chart. I was, okay, these setups look good. I saw so many times, maybe it's time to go long. What does it mean? Nothing. Okay? So I needed objectivity. So I had to turn all of my concepts into an objective way that I can trade in the same way every time.

Mm-hmm.

And I thought that would be easy. But it wasn't for sure, because take something that is discretionary and turn it like in a scheme of rules is not that easy. But I did it. And today I will show you how I did it in all the model that I actually use.

Yes.

Because once you objectify certain things, you can move on with the cooler things. And let's talk about the difference between mechanical and statistical, because mechanical is objectify all the concept, as I said you before. And the mechanical made me profitable, as I said you before. But I know, but I knew that when I was trading mechanical, I was just looking for the tip of the iceberg.

Okay.

What is the difference between statistical trading? Statistical trading doesn't make a strategy profitable. It takes a strategy that already works and puts it on steroids. That's the big difference. Because mechanical asks, responds to the question, you look at this and you follow always like this. Statistical responds to the question, oh, you can squeeze your strategy to the limit.

Okay.

That's the huge difference that is between mechanical and statistical trading. So why mechanical is also, I think, the best way to start trading or even to get profitable in trading, because it also removes interpretation and emotion. Because once you are allowed to interpret something, you can actually modify it and you can mask it as interpretation.

Mhm.

In mechanical trading, if a certain condition happens that we call A, I just do B. If I'm doing C, I'm getting it wrong. Okay? That's the big deal. That's the huge difference. So mechanical, it's also good for your emotion, but later on we will talk about trading psychology and why it's the most controversial, not thing, how can I say it? The most controversial topic in the trading industry.

Let's take a break for a minute there, guys, because a quick word from our official platform sponsor, NinjaTrader. If you've been following along on this channel, you know how much time we spend breaking down charts and market structure. And NinjaTrader is an award-winning futures trading platform built for exactly that. You get advanced charting, depth of market tools, and market replay so you can work through historical sessions and practice in a fully simulated environment, all without risking real money. Here's how to get started. If you're new to the platform, use the link in the description below to download NinjaTrader and explore it in the simulator. You get 2 full weeks for free to test everything out. And if you're already trading prop, choose NinjaTrader as your platform at checkout. It's the best platform for the job.

So go check it out and let's get back to the episode.

But let's talk about how you objectify a strategy because you have to start from the beginning. How do you read market structure? Because it all started like that. And we will see on the whiteboard. And we have to remember that market structure is an art. It's not like, I like this setup. It's not like, this, I think that is an IRI. We need a certain rule that objectifies it. And we will start with this. First one, I show you how I do it.

Mm-hmm.

I need to understand that I need to get a context of the bias. So I will read the market structure in 2 layers.

Okay.

First one, I can just do that. Okay.

And would you say that even for yourself, even as a statistical trader, market structure is still the foundation of what you're doing?

It's still the foundation of everything, because if you have market structure and if you know how to objectify it, you can move on. If you interpret market structure, everything else will be— I wouldn't say fucked, but—

I think so. I think so. Yeah.

Okay. You understand. So first one, we should know what is P and A.

Okay.

I think that maybe in the SMC concept is like the same because we look at the peripheral structure and actual structure. So we need to set up some rules. So when we see a market that is doing that, we need to follow the last true breakout. So we always follow, I will call it BO. So we don't see this, okay, because it's too far away from our dealing range.

Okay.

And we will see it Just this point and this point, the latest two breakout. This is our context.

Okay.

Okay. We will start to trade here, and we need to understand the latest breakout that is actual structure. In this case, it will be this, and the peripheral structure that it will be this. This sounds like easy, right? I think that most of traders that are watching this, they are doing. like the same, we need to set up condition. Okay? So when to follow peripheral structure and when to follow actual structure, because when they are the same, it will be easy. We just go along.

Okay.

In this case, we don't— we never go against that. But it gets complicated when we have an actual structure that is not the same of the peripheral structure. So I will try to make another level. So let's analyze, for example, this. And another important structure is the internal that I forgot to mention. So what we can see here, First, we have to get the context. So this, we don't look at it because we just look at the second, the last 2 breakouts. In this case, we have for the peripheral structure this and this. Yes. Okay, because they broke that level and they broke the level. But what we can see that this level has been actually broken. Yes. From here.

Yes.

Okay. So we know the concept, we know the context, and the context, it will be like this. It's important to not confuse timeframe because this in a timeframe can be this, in another timeframe can be another. Okay?

Yeah.

So what we can use, we will use a mono-time strategy. That's what I do. Okay? If you need to objectify something, you see the breakout at 15 minutes and you don't see it maybe in another timeframe. So which one do you follow? You need one. You need one. Okay? That's it.

Do you just stick to one?

I'm sticking to one totally. And this is the 15-minute timeframe.

15.

So what we actually do like this, we follow actual or we follow peripheral? It depends on one concept that is directional potential. Okay? What does it mean? I will write it. What does it mean? That we can follow the actual structure, but we have a limit of it. So the rule is we follow the actual structure, but we couldn't go against the peripheral structure. So in this case, what we can do? We have 2 main choices. First, use the discount and premium zone.

Yes.

That is really common with SMC or ICT spaces.

Yes.

In this case, it's the easiest one because we just track a Fibonacci here and here, and we check where is the 75 level.

Okay.

In this case, it can be this, for example. So we perfectly know that we can follow, in this case, this structure till this level.

Okay.

So we can take some kind of trades here.

Okay.

Okay? Once this level has been reached, what we can do, our directional potential finished.

Yes.

So we need in this case to follow the peripheral structure. And even if the actual structure is actually against us, we can just take a trade from here for searching for a long.

Okay.

That's the easiest one, and I would say that is not the most precise one. Okay. Because the other tools that you can use is volume profile. Okay. Okay.

So that's a prime example there where if the price is, even though you're using one timeframe, but shorter term is doing a pullback potentially.

Exactly.

Because the overall peripheral bigger picture structure is still bullish in that example.

Yeah.

you're able to create that A and B scenario just in that simple format.

Yeah.

So the most current, the actual structure has gone bearish, but until we reach said discount level, so 75.

Okay. Yeah.

So you associate that 75. So you have your A and B options there. So it hasn't reached 75, so we're able to short or potentially have a trade down. But once we hit that 75, now the rules dictate we can't short now, we have to then focus on the peripheral structure, the larger picture.

That's it.

And so therefore, potentially, the long position can then come in.

Because we can follow the actual structure, but we have to deal with this directional potential. So we need to know that once this is finished, we have to go with the peripheral structure. So the most accurate model is using a volume profile. I'll show you how it works. Let's see, for example, this.

So would you, with the volume profile, is this something in combination with what we just talked about?

Yeah, you can either use the Fibonacci for see the discount and premium zone, or you can, if you want to be like more precise, you can use the volume profile. So let's track a volume profile. Maybe we'll see the value area here. Something like that. So in this case, we know that we can follow the actual structure because what we can see that the value area low in this case is not taken yet.

Yes.

You see that this as a risk. So in this case, we don't use Fibonacci, but we use the value area high and the value area low as a target for this. Directional potential.

Yes.

Okay. So in this case, for example, we can go long and this in most cases doesn't agree. If we track the 75, maybe it's this.

Yes.

Okay. So it's a little bit a different concept that you can use. And I tested all of it and the results are not so different, but this has got more results.

Okay.

So we can still search for a trade, okay, for example, short till here. But once we got here, we just follow the peripheral structure.

So is this a good example of the difference between mechanical, being a mechanical trader, and then the statistical bit, putting it on steroids, in a sense that mechanically you could say 75, but statistically the edge is improved, even if ever so slightly, from the value area low?

Statistical is more than that, really more than that, because all we build A statistical strategy we'll see later on in the video. Okay, this is just, I prefer this or I prefer this. The results are similar, but this should be better.

Got you. Okay, got you.

Futures traders, it's time to hear about Apex Trader Funding, the largest futures firm in the industry.

They have completely changed the game with their new evaluation. So far, they've already paid out over $600 million to traders around the world, just like me and you.

Now, not only that, Apex actually holds the record for the largest payout to traders, which is of course with JadeCap, no stranger to Chart Fanatics, with $2.5 million in a single payout.

And the person before that was Trader Kane with $2 million in payouts. Now, what sets Apex apart? Apex, you can get up to 20 accounts, which no other firm Offers. Now, not only that, they have extremely clear rules. With their revamp, there's no MAE rule, no 5:1 risk-to-reward rule, no hidden rules, no gray areas, clearly defined drawdown models. So you have 2 options now: end-of-day drawdown and intraday trailing drawdown, which means no payout denials, no payout reviews, no discretionary decisions. And no surprises.

What sets them apart as well?

Traders receive 100% of approved payout requests, up to 20 accounts as I mentioned, and 1 day to pass. Now the other thing that I love about Apex is not only can you get 20 accounts, but there's different account sizes. So they have $20K accounts all the way to $150K accounts, so you can get 20 of any of the account sizes that you choose.

Not only that, you have Various trading platforms from WealthCharts to Tradovate to Rithmic.

The choice is yours. So make sure you use the code CF to get up to 90% off on evaluations right now using the link in the description below. Trade at Apex today. Let's get back to the episode. Do you know, in terms of the volume profile, one question that comes to mind, and I'm sure will probably be easier to to show on the charts. But, you know, whether it being CFDs or forex, uh, in terms of volume, like, what are you using? Is there a particular chart you will decide to use? Because there's no centralized data, right?

Yeah, there's no centralized, but you can still get some good results with also volume profile of TradingView. Of course, if you are trading index, for example, the Dow Jones, you can use the centralized volume profile and it will be better. And for my strategy, it doesn't run only on one pair. It's an edge. It's not an edge on that pair. I my strategy works on 33 asset. Wow. So when I trade, I trade multiple asset and it works on forex. It works on index. I I didn't got any good result in metal. Okay. But beside that, it works on mostly of the forex pair.

Amazing.

So this is what we look, but we have to get some rule. Okay. Because we objectify peripheral. peripheral and actual structure. But there are some conditions, and when we follow like actual against peripheral or something like that, and we need to know what is the internal structure. So let's take another example. First one, in this case, it's easy to recognize peripheral and actual. As I told you before, this is peripheral, this is actual.

Yeah.

But remember that actual is the latest breakout. What does it mean that this has been broken by this?

Yeah.

If we have this, okay, the actual shift from here to here. So this will not be a point that we consider.

Okay.

Okay. And we have to know when the structures are aligned. Okay. So in this case, what we can do, let's put, for example, that the actual structure is this. Okay. And we have a breakout of this level.

Yep.

Okay. When I say breakout, of course, I mention with the body, not with the wick.

Okay.

That's really important.

So you need the body to close.

Yeah. And also a certain level. So we need to get a number here, but we will talk about numbers later. Okay. Because we need to make sure that we are trading with a breakout or without a breakout because we don't trade without a breakout. So in this case, let's put, for example, that the value area low is here.

Okay?

And the actual structure is here and the peripheral structure is here. Okay? So they are aligned, but we still have some kind of magnet here.

Mm-hmm.

Okay? But It doesn't interest us because we have the peripheral and actual structure aligned. So if we can take, like, we have like Asian session here, I use mostly Asian session for my trades. I codify different schematic in the post-Asian session, so London and New York, and I trade that. Okay? We will see later on the trigger.

Yeah.

And we can see that maybe we can take a trade here. Okay? We will do that. And the next question is, where do I put my target? And the response to this is statistical. We will see later on with a concept that I created. I don't know if I created it or someone already told you, but it's called the optimal target.

Okay.

So what we can do, we just follow the structure. So in this case, maybe we will have Another breakout. And remember the main rule that I told you before: latest 2 breakouts. So in this case, which is the latest 2 breakouts? This.

Yep.

And this.

Yep.

This is not the second breakout anymore, so we will stop considering this zone. Okay. And we will start to look at here. Yep. Okay. That's another example where peripheral and actual structure, they are always in one zone. Not true. So let's make, for example, another example of peripheral plus actual structure in only one zone. We can do like this. Okay, let's take this, for example. What we can see that the latest 2 breakouts were In this case, this and this. After what we had, this has been broken by this, and again, this has been broken by this, and this has been broken by this.

Yep.

So let's contextualize it. The peripheral structure was that.

Yeah.

Okay. We need to also know where is the value area low or the 75 level. In this case, for example, it was there.

Okay.

So they took it.

Yep.

That's the main difference. And what we saw, that the market came in this zone, in the periphery, in the peripheral structure, and changed structure.

Yes.

So in this case, we can say that this zone has made its work because when the price came here, we had a reaction. In this case, this level is more important than this. This is not important anymore because as already did its work. In this case, we will say that this zone is B plus A. Okay? So why this is important? Because once this level is broken, the market shifts. So before we were in a long setup, now we will be in a short setup. Also, if this has not been broken.

Mm-hmm.

Okay? That's the main difference because we have to follow the market always. And it's not like last 2 breakout. This is a general, okay, condition. In this case, we have to recognize the most important breakout. And this was the main important breakout. We just went from long to short, and we can continue to look at this.

So when the mindset of it being short, that comes in once we break below this level, because now at this point, based on our conditions, we shouldn't be breaking this level, right? This should sustain. And if it does break, now we've obviously got a bearish scenario.

Yeah, because we cannot count anymore on this because it has already been a result from this.

And do you find like, If we're talking, you know, SMC or just discretionary traders, they'll see this. They may have gotten in here or they may have missed it, right?

Yeah.

So let's say either one. When they get to here, people are either getting taken out breakeven or those who missed it are starting to try and get in.

Yeah.

It breaks, no problem if it breaks, 'cause, you know, generally it's an okay area to look for longs. Once it breaks, however, because let's say if that hasn't been taken, they'll start loading up more 'cause they're like, oh, it was a liquidity sweep and now...

We don't do it like that. If we have a breakout, we cannot guess if it's a breakout or if it's a liquidity sweep. But based on my test, if we look at this as a breakout, we will get best result than seeing this as a liquidity sweep and still go long.

Yes.

Okay? So every question that you can make has been verified across thousands and thousands of trades. That's the main difference. I don't guess. I execute my data. That's the difference. So let's analyze this and let's follow, for example, the market. Maybe the market is doing like that. Okay? In this case, for you, I would say the value area high in this case is here. Okay? So it has not been reached by the price. What we will have? We will have 2 zones or actually 1 zone? We will have 2 zones because here we have the last breakout and here we will have the peripheral breakout.

Yes.

So we can say that the peripheral didn't help this for getting a breakout because it will help only if it's reached that level.

Okay.

Okay? So we will still go with this. Let's make, for example, another one. In this case, we have another breakout. The value area high was here. And what we can do, we just shift that zone. This came here.

Okay.

Because this will be the 3rd breakout and we don't follow the 3rd breakout, we follow the last 2. So in this case, we will have actual and peripheral structure.

So it shifts down.

Yeah, that's correct. And if we have this, so the price, the price, Andrea, ritraccia? Retrace, retrace. Okay, the price retraces and it gets our premium or discount or our value area high that was here. In this case, this zone helped this breakout.

Yes.

We can say that.

Okay.

And in this case, this will be plus B plus A.

Now it's aligned.

Yeah, it's aligned. Okay. Now we have to look for another kind of structure that is internal. Okay. And it's a little bit more complicated, but we will try to make it simple because our work is to take difficult concept and make it understandable for everything.

Of course.

Okay, let's be honest, most traders don't fail because of bad setups. They fail because they have no one in their corner telling them the truth. TradeZella gives you just that, and it never stops working. This isn't just a trading journal, it's the all-in-one platform built to make you profitable. Powered by an AI copilot that is literally by your side every single session, watching every trade, learning every pattern and guiding you toward consistency in real time. The second you close a trade, your AI has already captured it fully. Fully automated, synced directly from your broker, nothing missed and nothing manual. It lets you replay any trade tick by tick, pip by pip, so you can see exactly where your mind went wrong or right. And it runs your strategies through years, of real market data so you know your edge before you risk a single dollar. And through all of it, your AI Copilot is connecting the dots, flagging what's costing you, protecting your account, and doing the analytical work so you can focus on trading better. And that is what it feels like to have an AI working for you. Now, if you want that AI Copilot and see your trading excel, the link for TradeZella is in the description below. Use the code CF20 for 20% off your yearly subscription, or CF10 for 10% off your monthly subscription. If you're a trader, you have to know about Chart Academy. Chart Academy is the first all-in-one free trading education platform. That's right, absolutely free. No catch, no upsells, no links to sell you something. All free. Proven world-class traders providing an A to Z masterclass. Masterclass on their strategies, their concepts, and their processes that make them the best. From futures, options, crypto, stocks, and forex, everything is covered on chartacademy.com. Whether you want to fix your trading psychology or learn from a market wizard directly, you can do it all on Chart Academy. Not only that, by just spending time on Chart Academy and bettering yourself as a trader, you can get points that unlock exclusive discounts Free trials and free accounts with your favorite prop firms and trading softwares. Don't take my word for it. Chart Academy is live. The link is in the description below.

Now let's get back to the episode. Do you want to maybe keep this and we draw it?

Yeah, of course.

And then we can like maybe translate it into internal.

What we can see here.

Okay.

This is a week. Okay. Because The seventy-five, the value area can be taken also by a wick. It's not important that it's a body. Yes. Okay.

So for a breakout, you want a body.

For a breakout, I want a body with a certain level.

Okay.

We will later talk about that.

Level will get to you.

In this case, what we have? We have the in this case A P P because this is was the breakout. Maybe when it was like this. Okay. So it touched the 75 or the value area low in this case. So this is A plus B.

Okay.

Okay. But if we have an A plus B and we don't have any breakout here, this is not actual structure. Look at the difference. This here, we have another breakout and the retracement. Yeah. Here, We don't have any breakout. We have just a retracement, and it's a different concept. This is called internal structure. What we do with that? Easy. Let's take it simple. We just follow with the same rule. So we look at here, our value area low in this case, and we see if price reached it.

Okay.

In this case, for example, the price doesn't reach it and make a change of structure. Okay. Okay.

And do you just for the audience, is it still one time frame?

So even when you're still one time frame, we don't shift time frame because if we shift time frame, this will not be a breakout. It will be something different, etc., etc. So you have to marry one time frame.

And do you your for you it's fifteen minutes, right?

For me it's fifteen minutes, but you can use also five minutes. 30 minutes. Um, I tested all on 15 minutes. If I want to shift on 30 minutes, I need to get my test again. It will be really long.

Okay, you see, you stick to the 15.

Really long. So I stick to this 15. So let's analyze this. What we have, we have a breakout in the directional of the A plus B.

Yes.

Okay, because we have a breakout of internal structure. So this became the actual structure.

Okay.

Because it's the latest breakout.

Yep.

So even if we don't have a breakout on that level, we will have on the internal structure. So this became the P and this became the A.

Okay.

Okay. What we have right now. If we have like another breakout, okay, that follow the directional of long in this case that are aligned, we just follow the market. So we forgot about this zone, okay, because it will be the third breakout and this will be A and this will be the B. Okay. Yeah, in TradingView it's easier because you can just delete it.

I know.

Here we have to Cancel everything. So this is B. Yeah. Okay. Let's assume that we have a breakout of this level. We do the same thing. So the actual structure will be short. So it will change. The A will be here because the latest breakout will be here. The value area low in this case maybe can be here.

Okay.

Okay. So in this case, for you, what we follow if the price is here, okay? It didn't break out the peripheral structure and the actual structure is short. What we can do for you, for maybe your trading, you go long or short in this case?

I think I know the answer for yours, but for mine, um, I'd, you know, probably be looking at long.

Okay, same, because my potential direction finished once I reached the value area low. In this case, I won't draw anytime the volume profile because it will take so much time. But let's assume that the value area was here.

Yeah.

Okay. So in this case, we just go and search for a long trigger.

Got you.

Okay.

But I can imagine though, a lot of people see this breakdown here, they see equals, they say this has been rejected, the sellers are in now. They're probably still especially if this high hasn't been taken, probably go, okay, we'll buy down here after this low's taken.

Yeah. But we have to follow the price. If the price got 2 breakouts without touching this level, it means that it doesn't need to touch that level. It has a strong for gain, 2 more breakouts, and we follow the strong of the market always. So this looks easy because we are drawing lines on a whiteboard. You perfectly know that in trading is not like that. We have to deal with candles, we have to deal with the wick, and it's not that easy. It will be good if it was that easy.

I wish, yeah, I'm sure.

It will be good. And so we need to go one layer deeper, okay? Because once you have the market concept, the market context that is always like that, we don't guess, we don't do anything like that. We just see a market and we, Draw this because our goal is to mark all the condition. So we cannot see a chart and say, what I should do? I don't know. It's choppy. Is not trading. Is not trending. And even with a choppy market, you can recognize the P and A. Yep. Always. Okay. When I went to TradingView, yeah, and it was so difficult for me to recognize demand and supply. Okay, and I show you why. I was looking for a setup. I was looking in a chart, and I saw something like this. Okay, that's strange because I was like, okay, this is the breakout, but this doesn't look like a breakout because it's small. This maybe can be the breakout. You see all this kind of zone, and you doubt why because you say, I'll take this. That is the beginning. I'll take this. this, this, this, this, what I should do. Okay. And if you want to become a mechanical or statistical trader, you can't have doubt.

Yeah.

That's really important. So you need to understand how to objectify every concept. And for do it, I needed to create my own concept. So I created this cool concept that is called Triple print. This is a way to objectify at the 100% the way you take a demand of supply. Okay. Okay. So how we do this? It will be easier on the trading view because I'm not very good in drawing, but we will try to do also here. First, we have to understand one. context, a valid candle and a not valid candle.

Okay.

What does it mean? What I mean with that, we will say that a candle is valid if the length of its body, let's assume that this is 10. Okay. Is higher than the longer wick.

Okay.

Okay. So if let's do this because they look the same. Maybe the longer wick between this and this, it's this, and this is nine. Okay. So in this case, the body has been major than the wick.

Yep.

So this will be a valid candle.

Okay.

But what that what I What does it mean with the mean 10 and 9? We need to understand how you can measure the market because you can easily do with 2 tools, magnet and ruler. Okay? So you will have like, how do you measure pips? Okay, that's it. It's the same. So you measure with a ruler this.

Yep.

This zone and this zone. Okay? That's really important to me because I need numbers. And I need this for classify a valid candle. Instead, if the candle looks like this, so we don't see this because this is are longer. This is longer than that. And then maybe the body is four. Okay. And this is seven. Yep. This would not be considered a valid candle. Okay. Okay. So. The idea between these, it's that I need a good pulse for takeout zone. That's it. So let's analyze, for example, let's try to go deeper on the demand. Okay, let's zoom on this. Okay. This can be true demand zone, okay? But we need to see which one we have to take, okay? Because this cannot be valid, this cannot be valid, and maybe the only valid is here and the breakout starts from here. So let's zoom this and we see that when we are looking for an uptrend, we need to see the sell candle, okay? So the sell candle, we will do Oh, shit. Red. We see that in this pause, so this retracement, we have something like that. One candle that is like this. We don't need to measure it. This is not valid. This is bigger than that.

Yep.

So maybe we have another candle here, and this time, It's valid because this is bigger than that.

Yep.

And another candle that is like this. And also this is valid. Later we have that part. Okay? So the impulse. Okay. So this is the zoom in of this. Of course.

Yeah.

It's important that this is bigger than this. Okay.

Okay. Than the whole—

Than the whole retracement, of course.

Okay. In one, it has to be one candle?

No, no, no. Doesn't mean, it doesn't need to. Okay. It can be multiple candles. But the rule is called triple print. So what does it mean? That we need 3 valid candles to classify a zone. In this case, this is not valid. This is valid. This is valid. So two out of three. So this will not be our demand zone. Okay. This will be nothing. Okay. This will be just a little retracement. And when we take a demand zone, we need a bigger retracement.

Okay.

Okay. So we zoom in in another part. Okay. This. Let's zoom in and we see that maybe. We found this, we found this, and we found this. We will maybe have lots of more candles.

Yeah.

Okay. But we can just measure 3. These 3 are all valid. Okay. For the same concept. So this will be our zone.

Okay. But let's say if that 3 was in the middle of this.

Middle? Yeah, it's okay. It can be from here to here.

Anywhere within it.

Anywhere. We just need 3 candles. If there are more, we don't need to count them.

Okay.

Okay. It's really important to understand that.

And is the zone the candles itself or just that whole pullback?

I'll show you. I'll show you.

I'm rushing.

I'm trying. No worries. No worries.

Getting excited.

That's what it is. Yeah, I know it's a little bit different.

No, I like it, honestly.

So it's important.

You're seeing the difference, as you say, is like getting specific.

Yeah.

You know, and actual reasoning, the numbers behind it. And we're even, I'm sure, very early into this still.

Yeah.

So yeah, let's continue.

So how we have to take that zone is the next question. Okay? We have to take the latest 3 candles closer to the impulse.

Okay.

So in this case, that zone will be from here to here.

Okay.

Okay? This will be that zone. So in reality, we have to check from all the candles and we will start from here, here, and we go upwards. Okay? And we will take maybe here's the first, here's the second, here's the third. We will take just this zone.

Okay.

So we won't take this. Okay? And the interesting point is this, the level that once it has been broken, the market shifts. This is not really important where it starts.

Okay.

Okay.

So in case, just to go back to what I was saying in terms of 3 candles in here.

Yeah.

That does the triple print.

Yeah.

But the zone will be the last 3 candles before the impulse.

That's right.

So it doesn't have to be the 3, the triple print candles, but it has to be the 3 candles before the impulse.

Yeah. The latest 3 candles before the impulse.

And those candles don't necessarily have to triple print.

What do you mean?

As in, as—

The 3 valid candles. It's important to understand this. So if we have a candle here that is not valid, we won't take it. We will take the 3 candles. Even if in the middle we have a not valid candle, it's still okay because we are looking for that retracement.

Okay.

There are 3 candles and that's the important thing. And it's tricky because if you want to test it, you have— if you want to test as a mechanical or statistical trader, You cannot have doubt, as I said before. And if in this all impulse, and I mean, this will be always on the 15-minute timeframe, this is really important, we couldn't find any triple print, we just go down.

The next one.

Yeah. We saw if we have maybe here.

Yes.

Okay. In this case, we don't, we go down again, but it's really rare.

So in this case, we take this example, right? We're up here, we're trying to work out where's our area for us to get our pullback and where we can look for a valid setup. If we've seen there's no triple print here, that's already—

That's not a breakout.

Yes.

Okay? Because it's important because maybe you see this and they say, oh, actual structure short. No, that's not correct. This is the actual structure and we have to see if here we have a triple print for gain a peripheral structure.

Yeah. So then if we do that, then we then have our P&A back.

Yeah, that's cool. So once we objectify it, we can test it. Um, I don't know if you have some other question, but the rule is always the same. It doesn't change. Okay, let's analyze one more concept. So the next question maybe can be, how do you create a concept? How do you objectify the demand and the supply? Because this I did it by myself.

Yes.

All that I did, all that I will explain to you, mostly of that I did all by myself. Okay. I also, in the later on, we will see an amazing risk model. It's not mine. Okay. Because I didn't invent any of it. And, but I adapted to my own strategy.

Nothing.

So, oh, we create concept first. We start with observation. So we look at the chart and see, hmm, this could be interesting.

Mm-hmm.

Okay. This maybe can work. Okay. And that's what I did with the triple print, with the triple print, because I needed a way to objectify supply and demand. Second, we have the recognition. Okay, so let's move on and not just observe it, but recognize the pattern that you want to validate.

Okay.

Okay. Third one. The scheme. We have to make some rule for this pattern that we actually recognized like this. Okay.

Okay.

For test. So we just don't take a concept and we test the concept. We need the context. So we implement— we try to implement that context in our strategy. Once we do that, we can test it. Okay. 5, validation. We have to test it long enough that variance can't lie to us.

Okay.

Because you know that if you take a sample and you validate for like 50 trades, 100 trades, variance can lie to you. Okay? That's an important thing. So you need 500,000 samples for validation, for validated.

What's your minimum? Would you say 500?

I would say 1,000.

1,000.

1,000 is the minimum.

Okay.

Because below, variance can still lie to you. That can still be a coincidence. It's rare. Okay? If you have 900 trades and It's rare that it can be a coincidence, but it can be. 6, celebration.

The moment everyone skips to.

Yeah. Yeah. If you found something that actually compounds your strategy and gets your strategy better, I will celebrate and say, okay, let's add this concept to my strategy. This is all the things that I do when I create a concept.

In terms of testing, what does that look like?

Boring. I would say totally boring because our work, maybe people think of me as a trader, of a guy that stares in front of a screen for hours. And the reality is I did it. I stayed for like a month locked in, in my apartment building my strategy. Now my work is really different. I just read data and I just adapt my strategy when the market shifts. That's what I do. Okay.

Do you have to manually test this and collect the data or you have programs that you use?

For, yeah, for go to mechanical trader, to go in a mechanical, how can I say that? For be a mechanical trader or a statistical trader, you either have to be good in math or you either Going to be good on Excel. Okay, I'm good in Excel. Okay, perfect. So I don't run the calculation. I know all of the formula that I use, and we will later see home. But I don't run it for. I don't run it manually. Okay, but my backtest, yeah, I input the data manually.

Okay, okay.

That's important because even if I am a mechanical trader, a statistical trader, there's still a little layer of human interpretation on top of the. on top of the data. It's really important to understand this because otherwise we can all— we can always go statistic, we can always go algorithm with algorithms.

Okay.

Okay. And behind my strategy, there's a brain that adapted, that take decision like that can— there's a brain that changes something meanwhile. Okay. And the One algorithm just fires the same setup. It doesn't look at anything. Me, I have lots of experience.

Yes.

So not—I'm not a discretionary trader, but I looked at the chart like every day till six years. So I think that I have that level of experience that can actually gave me—I built a certain subjective edge on my statistical trading. But This is not the main part of the strategy. It's a little layer. Okay.

So would you say that there's, to always be optimized and to try and always perform at your best, you will still, even as a statistical trader, look to leverage that experience? If you notice, I say patterns, but more so maybe market cycles or volatility or slight changes within the market itself. Yeah. Where you know, okay, if this is starting to take place, You still, the actual execution and all your decisions are still going to be statistical-based, but you know that there's a slight shift.

Yeah, that's correct. I know when I have to shift from a concept to another. I know when I have to move from a risk model to another. I know all of that.

Okay.

Okay.

And those models are all based still on statistics?

All on statistics. And trust me, it will be one of the coolest models that you ever bring here in Chart Fanatics. Let's talk about one more doubt that I had once I started trading my strategy. So that was the triple print. That was the way to objectify supply and demand. Let's start with another one. This model, this concept is not a model that I created, is called wick gap.

Wick gap.

So the problem was that while everyone else debate on what's liquidity on the charts, because maybe someone see a double top and say, ah, it should sweep. Or maybe someone looks at this and say, this is a sweep. I need to objectify it.

Okay.

Okay, so let's assume that I have all my context aligned, okay, and I want to take a trade. This is long candle. This is longer candle, and this is the short candle. Okay? Let's try to understand that this can be maybe my setup. So where do I enter? Also, where I enter is statistical, but let's say the inefficiency here. Okay? So I would like to put a short here. So a short position. We would drag the target here, and the stop loss here. Okay? Of course, all of my conditions were aligned on that trade. What I saw left me speechless because it's something really interesting. I saw that when the closer wick to my stop loss were close. So in this case, this is the higher wick, okay?

Yep.

These 2 that are closer to my stop loss, when the distance between them were close, okay?

Okay, yeah.

My setups tends to stop loss.

Okay.

Okay? So I will do it a little bit better because it's so small. So let's assume I'm Taking this. When these 2 wicks were close enough, even if my setups were valid, so I have like my position here.

Yeah. And it's stop loss above there.

Yeah. When the distance between A and B were close, so in this case, maybe they can be close, My setups tend to stop loss, tend to do some play.

Do you have a specific amount that would be classed as close?

Of course, because you know what does close and far mean? Nothing. Okay, you, you cannot say this is close, this is far, because it's based on intuition. Yeah, perception. As a mechanical trader, don't like intuition. So what I did instead is measuring it. Okay, so I would like to measure the distance between here and here. We will do like with the ruler and the magnet, same tools, and we see different levels. For example, we see that the difference is 0.2 pips.

Okay.

Okay. What I saw, that I tested all my strategy and I took all the setup that were 0.2 pips or above.

Yep.

So when they were closer So like 0.2 or less, I wouldn't take the setup.

Okay.

Because it tends to stop loss. When they are higher, so minimum 0.3, I take the setup.

Okay.

That's really important.

That's how close you're talking?

Yeah. That's how I objectify close and far by numbers. What I saw that the result of my strategy, we're using this kind of range, Were not profitable.

Okay.

Okay. So what I did instead, I started to higher this range.

And would you class the wick gap as more starting to identify your entry based on what we've covered so far? Yeah. Okay.

That's what I'm doing.

Would the entry go along with everything, the triple print and the structure that we've covered so far?

Yeah, of course we have to every single line.

The next step.

The next step. That's like the checker. We are not looking at the trigger itself, but we are looking at, oh, I classify a valid versus a not valid trigger. So it's really bad, this work, because you have to test it from one range. You see that is not profitable. You feel bad, you move on. Okay, so let's assume that we do now 0.3. So what does it mean? That if the difference between them is 0.3 or less, We don't take the trade. If they're higher than 0.3, we take the trade.

When you say 0.3, do you mean 0.3 pips or you mean between 0 and 3?

I mean between 0 and 3. No, no, 0.3 pips.

0.3.

Because 0.3, we have 0.2, 0.1, 0.4. Okay, that's pips. That's what I mean. So if the distance between these is 0.1, I don't take the trade. Or if it's 0.3, I don't take the trade. And the results were, I would say, breakeven. Okay? So we still have to go. We do higher. We don't do 0.3, we did 0.4.

Oh, still at 0.3. Okay. When you say 0.3, it's 0.3 pips.

That's right.

Right?

Yeah.

So like a decimal.

Yeah. We will see later on the chart because It's a little bit difficult to understand this, but it's pips.

Yeah.

We will see that same result, same test. When there is 0.5 or less, we don't take the trade. Above, we take the trade. And the result for the first time went profitable. Okay.

At 0 point— above 0.4.

Yeah.

Okay.

This is a B. Okay. But I need to know how much I can squeeze it. Because maybe with 0.5, I have better result. Okay? So I won't stop here and say, okay, I'm profitable, let's do it. I want to know which is my threshold of profit. Okay? My maximum profit that I can reach. Because once we go up, we will see worse result than maybe this. And that's where our limit is.

Okay. So we are— okay. So you then, Keep going until where it starts to fail again.

That's right.

Then you find your threshold.

That's right. So we do 0.5 and we see better result than this. Okay. Okay. So we can still move on because we have better. So we, we move on. We just say 0.6. And with 0.6, the results were worse than this.

Okay.

So I know where I reach my limit. So when they— of course, this works for long and short. I'm actually drawing a long setup, but it works the same as in the opposite. So I need to understand that if 2 wicks, and can be also this, okay? It can be also this wick, but the closer wick. So if the distance between this and this are higher than this and this, We'll just consider this.

Okay.

Okay? So we need to get the 2 closest wicks to my stop loss.

Okay.

Okay? It doesn't need to be one buy candle and another buy candle. It can be any candle. Okay? But it has to be minimum 2. So what I did, I understood my limit of the strategy in liquidity. We can call it liquidity. I don't care how do you call a concept. You can call it liquidity, you can call it in any way. That's not the purpose of this.

Yeah.

Okay? We are not— we don't name things. We just execute things. That's how I objectified my own liquidity. So now what we have, we know the context, we know how to read peripheral, actual, internal structure. We know how to create a concept because it all started with observation. I saw that When the wicks were closer, I take stop loss. My setups tend to stop loss. So I recognize it and I recognize that it's only 2 wicks, the closest 2 wicks to my stop loss. So I build a scheme that 0.1, 0.2, 0.3, et cetera, et cetera. I test it for every number.

Yep.

And it's really a long test because imagine taking 1,000 trades for 0.2. The result sucks. And you take 0.3 and you move on.

When you do the test, you're obviously testing the criteria we just talked about, but is that on top of the other criteria we've set out so far? Does that make sense? So like you've already gotten the data for the structural side.

Of course.

And then the data for the Triple Print.

Yeah.

So you're adding it all and then this is now added data.

Added data. Why I say triple print? I tested, I tested two double print. It wasn't working. Okay. I tested triple print. We got good result. I test four print.

Yes. It wasn't the same. Same thing.

Same pieces. Yeah. But the result were not like from unprofitable to profitable. Okay. For the triple print, it was from good result to better result. That's what I mean. For here, when the distance were so close, like 0.2, we have not profitable result. That's the, the main difference.

Yeah.

And that's how we classify a concept. We test it, we validate it, so we know where we are profitable, we know where is our limit, and then we celebrate because we now found a concept, a mechanical concept, and we can add it to our strategy. Okay, let's talk about now what backtest is, what backtest really is, and how to make your backtest correct. Okay, because a lot of people dismiss backtest. Okay, and if you want to be a mechanical trader or a statistical trader, you have to get data. It's the only thing that tells you if something works or if not works.

Yeah.

That's what you need to understand. I can bet that you saw some guys that shows you his backtest and their equity line looks like this. Okay? And you see, and maybe when I did education with Andrea and Fabio, I had lots of students that shows me this.

Of course.

I said, bro, I'm getting a 60% win rate with 1:4 risk-reward, and say yes, and I'm the tooth fairy. Okay? Because what is the problem? They do the backtest, they get amazing results, but once they go live, they burn their account.

Yes.

And they say, why? Why me? What is the problem? And they quit trading and they say trading is a scam. Okay? That's one of the main patterns of a losing trader. What is the problem? When you backtest and you validate an edge, you have a hidden enemy that is called overfitting. What is overfitting? It's when you— it's one of the trickiest concept in backtesting. Because you see, you optimize your strategy so good for the past that you don't let your strategy breathe. Okay? So you build a perfect model based on the past, but when you go live, all changes.

Yes.

Because you don't know how to backtest properly.

Yep.

So we will today look at some models that avoid this. How to avoid overfitting and how do you know if you are overfitting a strategy with backtest?

What is, if you were to define backtest for you, so I'm sure we'll look at, you know, on the chart and actually on the screen, but for a lot of traders, backtest to them is just bar replay, right? They just go back and they press play and then they try and, you know, see their strategy and then they collect the data as best they can from that. But from your standpoint, what is backtest?

Backtest is everything. Is the validation of your edge. Is the response of what you were doing makes money or not. Okay? So you have to be really serious about that because backtest is like your trust. You can trust your backtest. You cannot trust a feeling. Okay? So if you do backtest for get experience in the market, that's not what I do. Okay? I just Search for the condition. As I told you before, we see for market structure, you objectify in order that in every backtest it's the same.

So would you say your backtest is more automated in the sense that you set the criteria and then let it present the results?

Yeah, it's not automated because as I told you, you still have to manually input and you have also to judge it. Okay. But mostly of the strategy is the same.

But you're not having to go bar replay and then try and look for those conditions.

Yeah, I do buy replay.

Oh, you do it?

Yeah, you do buy replay because of course if you see the other part, you can have a look-ahead bias. So you can get conditioned because maybe you are trying to backtest the NASDAQ, but you are also trading it. So you know that for the following 3 months it will go long. So you are unconsciousness conditioned to go long. That's what I mean. So you have to be fair when you do backtest.

Okay.

But it's the main validation of the strategy. Okay. How we recognize that our edge is an actual edge and not a coincidence. So edge, actual edge, and not coincidence. How you can say if you got a profitable strategy or you got a good fantasy for you? I don't know how do you do it, but for me it's a really strict process.

Mm-hmm.

Okay? And I called chunk optimization. I don't know if this is a concept that I created or not because I don't really See so many trading content. Okay, but it's it look like this. So chunk. So let's take our trades. Let's take the basic result. So we rate PF that is profit factor and avg.

error. Mm-hmm.

Maybe we have 1,000 trades on it, okay? And we see that we have a certain win rate, a certain profit factor, and average risk-reward. Later on, we will also do a little bit of basic statistical analysis on what profit factor is, on what you should consider, because if you look just like this on your backtest, You are not doing the things well. Okay, because there's a lot of there's a lot more because with only this you can say that if you're profitable or not. But let's take for example just this to be to simplify, and we know that we have one thousand trades. Okay, in one year. Let's take that the condi let's assume. that the win rate, let's take that the win rate is 60%.

Okay.

The profit factor is 1.5, and the average risk reward is 1 to 1. This on paper looks profitable because 1 on 1, I win 60% of the time. If we do this for 1,000 trades, we should get a good result. Okay? We don't calculate yet. Profit factor, I will explain you In a few words, over 1, you're profitable. If it's less than 1, you're losing, you are donating money.

Yeah.

You don't beat the benchmark, you should invest it. Invest, not trade. We just do that our 1,000 trades, we split into 10 chunks. So we will have 10 different results, okay? So for the first chunk, we will do, of course, 100 trades per chunk. Okay. This is trades. So what we need to see if our edge is actually an edge and not a coincidence, Okay. that in each chunk, in each chunk, 1, 2, 3, 4, 5, 6, 7, 8, 9, The result doesn't really move a lot. Okay. Okay. If in one chunk we are not profitable, not profitable, because you see, if the win rate is below 50%, of course we are not profitable. And then not profitable, not profitable, not profitable, maybe profitable. Yeah. Profitable, profitable, because of course this is profitable in a bigger way because it compounds. where we are not profitable. Okay? So we see that here, maybe we have 20% 1:1, 30% 1:1, 40% 1:1, 80% 1:1, 20% 1:1, 90% 1:1. That's a beautiful coincidence.

Okay.

Okay? That's not an edge because in certain market conditions, our setup works, But mostly in all conditions, this does not work.

Yes.

So how do you know if you overfitted that strategy? Using this. Okay? If your result stays stable across the chunk, it doesn't mean that you need to have 60% win rate 1-on-1 every single time. Okay? Or the profit factor that in this case we are just imagining. This is not the right profit factor probably. Okay? We are just putting some value. Okay? If you see that maybe you go 50% win rate 1-on-1, 60, 70, 50, 40, et cetera, et cetera. So the results shift, but not so much. Then you have an actual edge.

Okay. Okay. How do you find that balance of like the overlap? You know what I mean? In terms of where it becomes coincidence and too far away. Yeah. So let's say if like that same example, it was 50 or 70, it was, oh, sorry, 50, 45, 50 again, 70. And that, but then there's one month that is 20%.

Yeah. You know how I do this? Of course, I'm a methodical person. So I look at the number. If the win rate shifts more than 20% in 50, in the half of this.

First half, okay, yeah.

First of all, if it shifts more, we have a coincidence.

Okay.

If it stays In this range. So 40 till 80.

Yep.

It's okay.

Okay.

Okay. But it has to be on the majority.

Yes.

If you have only on 2, it's another thing.

Yeah.

Okay. That's the first thing to avoid overfitting. So if you that you're watching this right now have an edge that looks like this, but when you go live, this actually falls apart. What you should do, you should do the chunk optimization and see if you add an edge or if you have add a beautiful confidence on a period. That's what I mean with chunk optimization. Okay? Okay. Second thing, forget your result more likely to the live one.

Yeah.

Okay? First, you are trading in CFD, in futures, It's not this the problem. The problem is that when you trade, you have cost.

Mm-hmm.

This cost can be spread, commission, slippage, and swap. Okay, these are the 4 common ones. I know that you that you're watching this right now, you don't put this on your backtest. I perfectly know because you think that you are trading with the exact price without any cost, and you think that commissions are really low and they don't get your results so different.

Mm-hmm.

But you're probably wrong because if you compare all this and you united all of this, you will see that your profit will be cut. And even if you are on futures, you still have commission, slippage, and you have like the order book. So you will fill maybe in another level.

Yeah.

Okay. So you don't have swap, but you have all of this. If you trade CFD and you take your position overnight, you have also swap that can be a cost. or a profit. It depends.

Can be a loss.

Can be also a profit, but can be really, can bleed your account slowly. It can, yeah. Yeah. So what you have to do with that? First, you are trading a prop firm or you are trading a broker? That's the question. Because if we are trading this or this, the condition changes. Let's assume that I'm trading a broker. You perfectly know that a broker, you have 2 types of account, Raw and Standard.

Mm-hmm.

Okay. What is the difference? In Raw, you have less spread and higher commission. In Standard, you have higher spread and no commission, usually. It's not always like that. It depends on broker on broker. And there's one thing that we cannot calculate even in the platform, that is slippage. Yeah, we can't really know. We can't because we cannot predict the futures. Maybe if our trade hits stop loss during a news, the slippage can be higher. Of course. Okay, the swap is a value we can calculate it. And you have to understand that some people maybe talk and say, oh, this time I got, uh, $10 of swap, and one day I got triple. And he probably don't know that on—

There is triple swap.

Uh, mercredi, the Wednesday, that on Wednesday there is the triple swap.

Yeah.

So you have to calculate if your trades, if you take trades for multiple days, that you can have also triple swap share.

Do you hold multiple days or?

Yeah.

Yeah?

I do. Uh, my trades runs from few hours to maybe 1 to 2 days. So I'm not a swing trader. On Friday, I close all, but it can be old. Yeah.

And do you ever have to make a decision based on, okay, this trade might be a 2-day hold?

No.

It's going to be over Wednesday.

Yeah.

Over the triple swaps, do you then have to just take into account?

Yeah, yeah. I know that I can handle triple swap.

Okay.

Okay, I know it. But I don't know if the trades will close in 1 hour or 2 days. I don't know that.

Okay.

So what you have to do in your backtest, first, calculate your average spread. Let's assume you're trading a broker or a prop firm, you still have spread on CFD1. So how I calculate my average spread, it's so easy, really. You can take a clock, you can take 1 minute in normal condition, and you can see that maybe in that minute, The spread went from 0.6 spread to 1. Okay? This is the range. This is the minimum, this is the maximum. So you just do this plus this and you divide. Okay? So you know that your average spread will be 0.8. So every time you take a trade on that asset, you should know that you have to move your entry or your stop by this.

Yeah.

Yes. Okay? That's really important. Commission, we have a fixed value. So we perfectly know that per lot, we have maybe $6. Okay? That's the— fuck.

That's okay.

Okay. You have maybe $6. It depends if you are trading on a standard or on a raw account. Because the standard, you usually don't have commission. But you should understand that if you open 3, 4, 5 lots, you will have commission if it goes to stop loss or to target.

Yeah.

It doesn't care. Slippage is the only, um, thing that we cannot measure.

Yeah.

And we will measure in a percentage. So how we can objectify The— how the slippage will bleed our account. I did this calculation based on experience. So usually if my data are 60% win rate, 1:1, I just nerf it by 5%.

Okay.

All.

In terms of slippage?

In terms of slippage. So I know that I can lose 5% of my gain if I gain 100 R in 1 year, I just know that I gain 95 in reality.

Okay.

Okay? That's how I classify slippage. The swap is a cost and you have to understand it if you trade multiple nights. So also the spread, you have to know. But the only problem is that you don't know if the trade will take or gain spread or don't take spread at all. Maybe it will be an intraday trade. Okay? So if your set, if your equity line look like this and you didn't mention the average spread, the commission, the slippage, and the swap, this is not a strategy. This is a fantasy that forgot to pay its bills. Okay? So we have to understand this. And the latest, not the latest, but you have to know the sample size. Okay? Yeah. Because as I told you, variance is a really bad enemy for us. And we have to—

Even outside of statistical data, you know, would you say 1,000 trades is the sample size you want to go for?

Higher is always better. You have to understand this, that sample size And variance, they are opposite correlated. So more sample I will have, the less variance I will have.

Yes.

Okay? So if you take a trade, if you backtest a strategy, You show to your mentor the equity line and you have won on the trade, you don't have a strategy. You have— because the variance is so high and you won't put any money on something that can be variance.

Yeah.

Okay? That's one of the most important things that you should understand. I know that it's boring. I know that it's not very good doing backtests because Being a discretionary trader is funnier. Okay? When I trade, it's really boring. It's the same thing day after day. So you should understand that the difference between trader and gambler is that when I click buy, I feel bored. When maybe you click buy, you get excited. And it's really important to understand. You should know all your trades can be the sequence of the trades. Okay. Because—

Yeah. Oh, 'cause they're like, you know, you might get one sequence with a certain win rate or certain, you know, winning trades like this.

Let's assume I have a 1 to 1, 60% win rate, and we have 1,000 trades and we have 1,000 lines. Okay. What we will have, we will have 100 different conditions of our trade shuffled together.

When you say a thousand, you say a thousand lines?

Uh, no, thousand trades, one hundred lines. What do you mean by lines? Lines is the expectation. Like every different lines is a scenario. Okay. Okay. So we will get like one thousand scenario of our trade shuffled. Okay.

Oh, so the hundred is the shuffled. Yeah. Okay.

Okay. Yeah.

So what we will have? What we'll have? No, sorry. The one hundred is the Not the shuffle, the combination. So the 100 is each one is a combination of result. So it's a shuffle, we can say that. But we can press multiple times the Monte Carlo for get infinite shuffle, maybe. Not infinite, but the more we have, because our worst case scenario, that is the most important thing that you should get by Monte Carlo, changes every shuffle.

Okay.

But because they change the order of the trades. So you know that you will get lots of lines and you will have the average line that will be the most common result.

Yeah.

Okay. And usually it will be the one in the center. Okay. And after we will have the best line, the best line.

Yep.

With the best result and the worst case scenario.

And when it's doing those shuffles and getting those trades in order, is that all taken from the actual trade data?

Yeah, the actual trade data. So this is the final part. Okay. Why I don't like it. Okay. Let's assume this, because if your strategy has a 60% win rate and the 1:1 risk reward, this works only on the trades that you put.

Okay.

Yeah. But if you go, you can see periods when your strategy can be worse or better than that. Okay. So you cannot define a worse result because it's just based on the data. If you trade, you can use, for example, FVT. That is another method that calculates your worst-case scenario based on not your trades, but on your possibility. is another maybe most advanced one, but the Monte Carlo usually works for the majority of traders. And it's really important that you know this, the worst-case scenario. If your strategy can survive this, you can have a profitable strategy. Of course, this will change based on risk that we put. Okay? If you put 1%, these numbers will be—

Of course, yeah.

Will be some kind of numbers. If we put less, will be another, another Case, et cetera, et cetera. So we can say that the Monte Carlo, we can use it for choosing our risk.

So like if you're, so let's say on 1%, your worst case scenario is survivable and acceptable, that's fine. But if you put 2, and then that worst case scenario has dropped significantly to a point where it's no longer acceptable, you know, to adjust the risk accordingly.

You can know also the longest streak, the longest winning streak, and the longest losing streak. And that's so interesting because my strategy can lose up to 13 losing trades in a row in the worst case. Okay. And I will use Monte Carlo just for that, for the average losing streak, not for choosing the risk. For choosing the risk, we have better model, better and more advanced model that we will later on see. we started to understand how we can get a good backtest.

Yeah.

And not an overfitted backtest, we should understand how you can read your metrics. Okay? So we will study a little bit of the basics of statistical analysis, because as I said to you before, if you look at your strategy and you check win rate, PF, and risk reward. That's the really surface level. You can't tell if you're profitable by only that. So we need to add more condition, more metrics.

Yeah.

Okay. Not too complicated. The complicated metrics we will later on see. First is my favorite one, and it's the only metrics that can tell you if your strategy will make money or not. It's called expectancy. And this, what does it look like? This represents the return of money per dollar risked.

Yes.

So for example, if we have an expectancy of 0.10 cents. Okay, it's cents, not dollar. Okay, $0.10. What does it mean? That per every dollar that we risk, we will get 1 plus 10. Okay, this will be our return. Okay.

Yeah.

And this is so bad because usually The cost that I told you before is this. Yes. Cost. Eat it. So you won't be an unprofitable trader, but you will be a breakeven trader if your expectancy looks like this. Of course, we can move on. Like the LT range is 0.10 between 0.30. This is an LT range. In reality, it's not. 10 is 20, because you can be profitable with that. That means that per dollar—

Would you say that's healthy? Would you say that's on the lower end?

It's, you can be not too much profitable. You can make, you don't lose money. Yeah. Okay. And that's one of a good thing. Okay. You can later move on with another range that is 0.40, I think. 40. 0.50, and this is a very strong expectancy. This is what you should aim for it.

Okay.

Okay. And if you have an expectancy that is above 0.50, it's amazing. Okay. So you have to understand that win rate is just vanity. It looks cool when a guy comes to you and says, I have a 70% win rate. That means nothing.

It's good for my titles and trailers. Say it again, Dan, and I'm done.

Yeah, because, for example, my strategy runs at 36% win rate.

Okay.

Okay. And I don't have a higher win rate. And maybe the people judge the trader by win rate.

They do, yeah.

They do, because that's the easiest thing to do. They should judge it about its expectancy, because win rate is vanity, expectancy is the truth.

Yes.

Of trading. Once we understood this, we can move on with the next metrics, that is profit factor.

Yep.

And I'll explain you why I don't really like it. Profit factor is so simple. It's basically your total profit divided by total loss. Okay. It can be in a range between 1 to 3, et cetera, et cetera. I'm not sure, but if— okay, yeah. If your profit factor is below 1, you're a losing trader.

Yeah.

Okay? If it's above 1, you make money. If it's above 2, you're an excellent trader, et cetera, et cetera. You can move on. So let's assume that in 1 year of trades, because you calculate it on dollar, In one year of trading, you made $10,000, okay, profit, and your total loss were $5,000, okay? You understand that?

Mm-hmm.

Okay, it's— you will have— you have the double of it.

Yeah.

So your profit factor will be 2.

Yeah.

And 2 looks amazing on paper.

Yes.

But which is the limit of profit factor that consider— that does not consider if in this $10,000 you had a lucky trade? For example, let's assume that in this $10,000 you made $17,000 by only one trade. One trade that didn't follow your strategy, that we can call it luck. Or you deviated from your edge.

Mm-hmm.

Okay? So if we try to remove that trade, we will have $3,000 divided by $5,000. That is— so if you remove your lucky trades, you have 0.6 of profit factor. And this means that you are not a profitable trader because your profit factor is less than 1.

Yes.

And with that lucky trade, you look amazing. Okay? So this is a very tricky, tricky metrics. And for be sure that this is true, you have to take out your best trade.

You have to?

You have to. Yeah. And if you survive, so I mean, okay, if you have a profit factor that is higher than 1 without your best losing trade, you have a solid edge. But yes, best winning trade, you have a solid edge.

And even let's say you reviewed that best winning trade.

Yeah.

And you saw that it's not lucky in the sense that it was with the plan. You still take it out anyway to make sure.

You still make it. Also, if this is stick to the plan, you still remove it.

Okay.

Because you need to understand, because maybe you had a trade that were—

Is it just one trade?

Yeah, just one trade.

Take one best trade, okay.

Because maybe you had the one trade that you take one time per year and you made a 1 to 20, but it's not common.

Of course, yes.

You need to understand if you're—

Like conditions, volatility, like something, everything would just work that day and it was a big trade, even if it was unplanned, but you remove it to try and see a more average look.

Yeah. So if you use profit factor, pay attention to this. Now let's see another condition. It doesn't judge you on the result. So maybe we can assume that we made 30% on a year. Okay? We have 2 traders that made the same result. We will call them Andrea and Fabio. Okay? Andrea made 30% in a year and Fabio made the same result. Okay?

Mm-hmm.

What does Sharpe ratio measure? It measures how clean the path was to get there. Because maybe Fabio was smooth, controlled. Okay, he made 30% in a very linear way. Yeah. Okay, you see this equity line looks good. Maybe Andrea got the same result, but meanwhile he had 3 heart attacks because his equity line looked very, uh, okay, volatile. The result is the same, but the Sharpe ratio is totally different. So we can say that the Sharpe ratio measures the consistency of your returns, because you can't judge a trader by result. You have to judge it by how clean the ride was to get there.

Yeah.

Okay? So usually the Sharpe ratio will be higher with Fabio and will be, um, lower with Andrea. This doesn't mean that Andrea is a bad trader. This means that Andrea had a different path in in a type of period.

Yeah.

Okay. And usually we have a range for profit factor that is below 1 is shaggy. So it's not very smooth. Above 1—

For Sharpe ratio.

For Sharpe ratio. I forgot to say the formula of the Sharpe ratio, because even if you calculate with Excel, For me, you should know what you're using for. So how you measure sharp ratio is so simple. We can call it SP. It's just your return. Oh yes, how much does it win? So it's really important to understand if your profit factor is higher or lower than 1.

Yes.

Once we have this, we have the basic metrics of statistical analysis and we can move on with the cooler part. Because if you combine all of this, you can have a good result. Okay? You can have a good picture. If you have just this, you won't have it.

Yes. So then you add these layers on the Sharpe ratio, profit factor.

Yeah.

What was the first one we went over?

The expectancy, profit factor, and Sharpe ratio.

Yes.

Now, let's talk about target. Okay? How do you actually know how much you have to put as a target? Okay? Because as a statistical trader, there is a number for always. Okay? So let's assume that you're trading with a fixed risk, Reward. Okay? And I do it, I trade with a fixed, I don't vary it.

Okay.

And I called it optimal target. That is very different from optimal F. That is totally another thing. And how we calculate this? Let's assume that I'm taking 5 trades. Okay? Okay, we analyzed 5 trades. Of course, you have to analyze lots of trades, but we are here just for the example. And they are all long, and you get to the target in all the trades. This is the target, this is the stop loss. You feel happy because you won 5 trades in a row. Okay? And maybe you go and you celebrate for it. Okay? But you don't know how much money did you left on the table because you should know your RR max.

Yes.

What does it mean? That you need to understand how much the trades go upwards on your target. Maybe all of the 5 trades reach 1:2.

Mm-hmm.

Okay, so you gain 5R. Okay, but you close it too early because all the trades went to 1.2. So if your target was 1:2, you would have made 10R. Okay, and you will find a limit of it because maybe you see that 3 of these trades went to 3R. Of course, we are talking about 3. You have to talk about 100 and 100 trades.

Of course.

And if you do the calculation, this will— when we go to breakeven, because once we are in 1:1, we usually go to breakeven. Let's assume that they were on breakeven. So we made 3R, 3R, 3R, 0, 0, because they went up And they instead went to break even. Okay? Okay. Yeah. Yeah. So which is our best result? With 1, 2, 3, with 1, 2, 2, we made 10R. With 1, 2, 3, we made 9R. So, you know, as I told you before with the wig gap, that this is our limit. Very good. And the optimal target Is 1 to 2, not 1 to 1, not 1 to 3. Okay.

That example is so interesting.

Yeah.

Because you look at it, you would go, yeah, the 1 to 3 must be better.

Yeah.

Because of the, just the higher number, it sounds better.

It sounds better, but it's a combination of win rate also. So you maybe, if you have different results, you can see that maybe 1 to 3 can be better, not in this case. And you will go upwards and you see, okay, which is my limit? Let's do, Let's assume 1:4, 1:5, 1:6, and you find, you will find your optimal target. So in every asset that I trade, I calculated this. And in some asset I trade 1:1 because it's the best target for me. I don't really go 1:1, 1:2, 1:3. I go 1:1, 1:5, 2, 2:5, 3, et cetera. So 1:2, 1:2.5, 1:3. 1 to 3.5, okay? And in some assets, mostly assets I trade like this, okay? In different assets, I go 1 to 3, but never more than 1 to 3.

Was all data-backed?

All data-backed tested. Because I can go also 1 to 4, but the data will shift. So they are against me. So why do something that is not supported by data? Okay? That's how I Choose my optimal target.

Okay.

Okay?

Based on all the data you've collected, what are your thoughts on risk to reward? Have you seen, because a lot of people I've seen, especially within the prop firm industry, will trade a negative risk to reward.

Yeah, they can trade negative risk to reward. Of course, you will have a higher win rate. It's the same. Okay? You can do with this kind of method also with a negative risk to reward. Okay. For me, it's not the optimal target. I tested the negative risk reward and it's not the optimal target for me. I know that I can go 1 to 2 usually. Okay. But if your calculation works better on a negative risk reward, you can do 1 to 0.5, etc., and etc., and see how many R you gain on 100, on 1,000 trades.

Understood, yeah.

So this is what I do, but you can do also with a negative risk reward. Okay? So now let's go to the coolest part of the video, one of the coolest parts of the video, that is the most advanced risk model that I built on my strategy. But first, I have to convince you why it works.

Okay.

Because I traded without this concept for years. Later, I discovered that and I was It was my first AA moment. Okay, it was like, wow, let's put this on my strategy. For explaining you this concept, I need a coin. Okay, this is the head, this is the tail. Choose head or tail. Heads. Okay. I have to put here or? Yeah, yeah, yeah. Okay. It went head. Okay. Now that it went head, What are the odds of the next flip?

50/50.

Still 50/50, right? So it doesn't mean anything if you took head on the previous trade.

Yeah.

How do you use this concept on the market? For you, I'll ask you this for you. If I have a 1 to 1, 16% Win rate, and I took a losing trade, the next trade will have the same win rate or it will change? For you, how do you think about that?

For me, it's all variance, right? So it's all random.

Okay. So you think that it's like the same every trade? Even if you take a losing trade, You still have the same odds on the next trade, the same data.

Yeah, I would say so.

Okay. The difference is this. When we flip a coin, if we take a coin, as we can see, the table where we can flip the coin is flat. It doesn't dance through head or through tails. Okay. But when we take a trade, We don't bet on the trade itself. We bet on the market condition itself. And we perfectly know that market condition, they don't switch in every time. Yeah. I'll explain you better. If, for example, we know that market follow regimes. Yeah. Okay. This is, I think, a well-known concept. concept. And when we took a trade and when we took another trade, that regime stays. It doesn't shift. We perfectly know that calm follows calm usually. And I'm talking about also volatile. Volatile?

Volatility.

I'm talking about volatility. So what does it mean for our risk model? That the condition sticks. And if, for example, we trade a strategy that works really better in—

Trending.

In a trending market. Okay? We are in a trending market. So we know that the condition is not flat, it's trending, and we have a trending strategy. So if we take a trade here, And we took target, okay? And the next trade, we take a trade and we took target. It's not because— it's because the strategy likes the condition of the market. Okay. So a win in this case will make a win more likely. Yes. So it tells you something about the next trade. If, for example, we shift from a trending market, that is the best condition for our entry, for our strategy, and we switch into a choppy market, we will lose and we will tend to lose. So what does it mean is that when we flip a coin, we don't have regimes. Okay?

Yeah.

When we trade, we have regimes. So if we lose, And we lose again, we will be more likely to lose again because the strategy doesn't like debt regimes. Okay? That's what I mean. And when I win, when I win and the next trade is a win and it will be more likely that it will be a win, it has a name and it's called positive Autocorrelation. What does it mean? That our strategy, that our sequence coming cluster. So when this kind of regimes happens, we tends to win, win, win, win. When this kind of regimes happens, we tend to lose, lose, lose, lose. You see the sequence. This is called Positive autocorrelation, and in my strategy I have a positive autocorrelation. This is really an amazing concept that was discovered by Lo and Minkley in a paper in 1988. Okay, and the paper is called Stock Market Prices Do Not Follow a Random Walk, because we have this kind of debate that market is random, that there is a random walk.

And of course, for me, So would you say that's wrong? Would you say that the statement markets are random and each trade is a random outcome?

I don't think so. I totally don't think so because this would mean that you cannot be profitable for a long time. You can be, it can be luck. If I'm profitable, I'm luck. I'm lucky. And luck doesn't resist time. Okay. For me, it's not like this. And they demonstrate in a scientific way that the returns are autocorrelated.

Mm-hmm.

Okay? And that's really interesting because I know that when my strategy is in a certain regime, I will have a sequence. If I win, I win. If the strategy turns in another regime that my strategy doesn't like, I will lose, lose, lose. I will tend to lose. So imagine that If we throw a coin, this will be like this. Okay. And it tends to heads and this will tend to tails. And if I drop it there, it will be more higher chance to make head. And it's the same in the opposite. Okay.

In that scenario, so that, well, with this scenario, you know, say you have a system that is working in trending markets. So you have a more likely chance of the wins. Yeah. When, let's say, it comes to a condition which isn't suited for that, and would it be a case where you switch that off, where you obviously don't look to apply that, but would you then have a system that might be suited for?

We will see later on, okay? Because yeah, we will have a reply for this. Amazing. But we have to understand that the returns, our returns are autocorrelated. It can be positive, it can be negative, and it can be zero autocorrelation. Each one has a really important meaning.

Okay.

So if you that you're watching this right now, you don't know your autocorrelation and you are varying your risk size on every trade, you can be a gambler and didn't know that. Because autocorrelation responds to one thing: how much should I risk per trade? If I have to use— not how much, but if I have to use Fixed risk sizing or varying my risk sizing each trade.

So this is really what dictates it. So when people talk about A+ setup, should that really relate to positive autocorrelation?

Yeah.

Not towards like a setup, you know, like, I don't know if you hear people talk about, oh, I'll risk more on an A+ setup.

Yeah.

And normally when they say that, they say that's something that has a higher positive expectancy. But I've never really heard anyone Watching positive.

Because when you say this is AAA+ setup, you mean that you like the setup so much, that you saw that setup so much, and you risk more. But this can be also your— how we can say that?

Downfall?

Your beginning of the downfall. Because if you like the setup, what does I like the setup means? We are still talking about discretionary trading here. You can be experienced enough to find a subjective way out of advantage, as Fabio do. Okay? Fabio did this. But if you look at this setup and you say, this is AA+ setups, and you don't know your autocorrelation, and maybe you calculate it and you see that your autocorrelation is zero, you are gambling because fixed size is the only thing that you can do. The math says use fixed size, don't vary your risk size. You can be an amazing trader and you can still be a gambler on how much do you risk. And now we will do the scale. We will calculate autocorrelation because it's a little bit more complicated. You have to understand true concept. And we will see what autocorrelation looks like and what we can do about it. Okay? So once I convinced you, so I convinced you that Returns are autocorrelated.

Yes.

Okay? Because it's really important to understand this and why this concept works. We need to calculate this and we need only the win and loss. We don't need risk-reward, but of course with a less risk-reward, it will be better. Okay? And we should take a sequence of 10 trades, for example. Okay, this is my win and loss sequence on 10 trades. Of course, autocorrelation is a coincidence of 10 trades. You have to do with 100 or 1,000. But this, we will just explain this, how it works. First, you have to know 2 metrics. First, unconditional win rate. Okay? And conditional win rate. This is the easiest one. This is the normal win rate. Out of all your trades, how many were winners? We can just calculate it. This is 10 trades. 1, 2, 3, 4. So 4 out of 10. This means that my win rate is 40%. Yeah. Okay. This is the easiest one. The conditional win rate, I think that nobody uses this. Okay. And this is a little bit more complicated because you have to ask yourself one question. After every losing trader, after every losing trades, what is the outcome? It is a winner or it is a loser? So I will do the example for you. This is a losing trade. After this, the trade is a losing or a winner?

Mm-hmm.

It's a losing. So we will just say 1, okay, for calculate how many trades. This is a losing trade. After this, we have a losing or a winner? Losing. True. This is— this came after a losing trade. This is a winner or a loser? Winner. So it's been the third and it will be a winner. Okay. This doesn't come after a loser, so we don't calculate this. This doesn't come after a loser. This calculate this came after a winner, so we don't calculate this. Okay. This did. Did. Okay. So we calculate this. The result is a loser. We say four. This same came after a loser, but it's not a winner. So we will say this. And this came actually after a loser, and it's a winner. Okay. And we have 6 and we have 2. So the result is 2 winners out of 6 trades. Okay. And this should be 33%. Yeah. Yeah. Yeah. Okay. Okay. Let's see.

I'm just, I'm staying in there. I'm staying in there. Okay.

So 2 out of 6. Now we have to understand one rule, okay? We calculate our win rate and we calculate our— we calculate our unconditional and conditional win rate. So now what we have to do, we have to compare this matrix for see if we have a positive or negative autocorrelation.

Mhm.

We talk about positive autocorrelation, that is the easiest one and the most common ones. What does negative autocorrelation mean? Negative means that my strategy alternates. So when I win, I will tend to have a losing trade after. After win, after losing, after win. So this is not coming sequence. Okay, this coming sequence and this is the positive one. Yeah. When I have a strong negative autocorrelation, my strategy will tend to reverse. And this is so common in the mid-reverse strategy. In the mid-reverse strategies, like if I trade here, not with a trending following strategy, it's more likely that I will have a negative autocorrelation instead of a positive one. So the rule is, If your UC wear rate is higher than a co-wear rate, you will have a positive autocorrelation. Okay. Okay. If this is higher than this, okay, and this is the most common. If the UC wear rate will be lower than your co-working wear rate, the conditional win rate, you will have negative. If these are basically the same, Yeah. it will have zero autocorrelation. And knowing that you have zero autocorrelation is really important also. So we compare this and we can see— or I can also cancel that— we can see that the UC win rate is higher. And the difference between them is 7. Okay, so we are in a positive autocorrelation zone and we have to scale it. Okay, so we have to know what we have to get with— what we have to do with the 7. 7 is a good or bad result? Let's see. The range is this: 0.2, no auto. I would say just no auto. So what this scale means, that if our result is between 0 to 2, we don't have an autocorrelation. If it's 3 to 4, it's weak. 5 to 6 is solid. 7 to 8 is stronger. And 9 to 10 is extreme. We have to pay attention if our autocorrelation lands on this.

Mm-hmm.

Because either you found a gold mine, Or either either you made an error with the data. Yeah. Okay. And if you have autocorrelation like this and you didn't commit any error, contact me because I want to give you money. Okay. Okay. It's really uncommon. I tried in so many strategy. My autocorrelation is always in this level. Okay. Yeah. Yeah. So what does it mean? The higher is the stronger signal. So for example, if I have an autocorrelation of 10, okay, and my strategy comes in streak, as I told you. So after I win, I have like not 100%, but 90% of possibility that the next trade will be also winner.

Because that's where you bring your variable risk in.

Yeah. And if I take a losing trade and I have an autocorrelation of 10 and it's correct, I shouldn't take the other trade because 90% will be a loser again.

So it's the same both sides?

Both sides. So I don't take the— if I have a loser, I don't take the next trade because it will be a loser, loser, loser after it shifts. When it will be a winner, I will higher my risk because I know that it will be more probable that it will be a winner again. Okay? This is for positive. And in this case, we are in the positive zone. Even with 7 is a really strong signal. So it will look like this, okay? And we know that we can shift our risk and we can higher or lower or don't take the trades if we have this, okay?

Yeah.

So this is how autocorrelation works. If we had, for example, a 7 but with a negative, what we have, we know that after a win, we shouldn't take the trade. Because in with negative, as I told you before, it alternates. Yeah. So we don't take the trade or we risk really less after a lose. We higher our risk because we expect a winner. Okay, etc., etc., etc. What is the problem with risk reward? This does not count the risk reward. You can have the autocorrelation with the even one to four risk reward average, but it will be less correlated because. If you have a higher risk-reward, the probability that this number will be higher, it will be really small. So if I am trading with 1:4 risk-reward, it will probably land here.

Okay.

Okay? It's not totally sure, totally always like this, but it's more probable. And that's what we trade. Okay? Now let's assume that my results were this. One. That does not mean that I'm not a profitable trader. I can still be a really amazing trader with 1 autocorrelation. 1 autocorrelation. What does it mean? That your result does not tell anything about the next one. Okay?

So you should be fixed in your risk.

Yeah. Because if you, now that you're watching this video, you calculate your own autocorrelation and you see that this number lands To one, the math is saying something clearly. Don't vary your risk size. That's it. Okay? That's what we need to understand about it, this. And if you are doing this, you are basically gambling, as I told you before. Yes. And that's how autocorrelation works. Okay? Do you have any question about that?

No, perfect sense. Yeah, it's great. And like I said, you know, a big talk over the last You know, year, 2 years is about A+ setups and learning how to size in, and that's how you're gonna make a lot more money.

Yeah.

But in reality, now that we've gone over this and really dived into the data side of things, and most trading decisions, if not, I guess for you, all trading decisions should be guided by data.

Yeah.

This is how you truly figure out if your edge is providing you a setup that you can then look to variable, use variable risk or increase risk. Versus something that you believe to have a higher win rate.

Yeah, because if you see some guy that is teaching you and he say, this is AA+ setups, there are either 2 possible ways. The first one, he built through experience a subjective advantage, like Fabio or Andrea. The second one, he does not really know what autocorrelation is and is gambling. Okay? So when a guy say you that, just 2 possibility. Okay? That's what I use and that's what I used in the championship. But now we have to respond to another question. I said that you can higher or you can lower your risk size by this number.

Yes.

Okay? If you have a higher number, you can risk more. And if you don't have a— like if you have a weak signal, you can risk less.

Mm-hmm.

But the question is, how much? Because we always need a number in every conversation. So what we do is we know you use another formula. We use another formula that does not tell you you should risk more or you should risk less.

Mm-hmm.

It tells you you should risk this or this.

Yeah.

Okay? And this formula is called Kelly. I'm sure that you heard about it.

I have, yeah. I had a semi-mechanical trader, so I got to hear about Monte Carlo and Kelly. Okay. They're the only two I heard of. Only two? The only two, yeah. You were never audible, really? I never, which is shocking really, right?

Okay, no worries. I should have by now.

Okay, no worries. Well, we're glad. That's why you're here.

You know, you're teaching so much more. It's a really different approach, but that's what I do. And that's how I became not profitable, but to a next level, I would say, because mechanical trading.

Have you seen people, I don't know if you teach or anything, but like, have people implemented this that you know that then seen that same change?

I don't teach anymore. So I did education with Andrea and Fabio like 2 years ago. Since 2 years, I stopped teaching because I like trading. I will probably start again because I now want, I now would like Would love to, but I just say this to 2 of my friends and they got amazing result. And the cool things is that you don't change anything in your strategy because people think that for getting a better strategy, better results with your strategy, you should— you just should change the— how do you enter in a trade, maybe the condition of your entry.

Mm-hmm.

The reality for me is that the risk model that you use is more important than the strategy itself. And this, for me, is the truth. And I'm totally sure that if you have an autocorrelation, for example, or a model that we will see later, and you have a losing strategy, like 1:1, 45% win rate, so by math, that lose money, with the right risk model, you can turn it into profitable. And that's so interesting because people are so focused on changing things when the only thing they should change is how much they risk per trade.

Yeah.

So for response to the question, we have a signal, but how much should I risk? We use the Kelly criterion. It's a formula that was initially invented by some guys for blackjack in the casino in Las Vegas. I think blackjack, I'm not sure about which game. And later it has been implemented in the trading space. And it tells you one thing, how much should I risk per trade before blowing up my account?

Yeah.

Okay. The formula is pretty simple. We will see now. And it's your win Percentage, so we rate minus loss rate percentage divided by average error. That's basically the formula of the Kelly, but there are lots of calculate of calculator online, so you can basically put this and you will find the Kelly. Which is the problem of Kelly? Kelly is really too aggressive, okay? Because it, it's right that it's survivable enough. It pretends that your strategy cannot shift, and we perfectly know that market shifts. Yeah. So we use a fraction of it. Which kind of fraction? It depends on the signal. Yeah. So I will tell you now some example, because how much should I risk is really It varies on if you're trading prop firm, if you're trading a championship account, if you're trading broker, if you have any limit per risk, for example, in a prop firm. And it depends. Okay, so what I use is this. With the— here you cannot use Kelly. We use fixed risk. When I have a weak signal, I will use Maybe 1.1 Kelly, here 1.4, 1.3, and here I don't calculate it because it's too rare. Okay? So what does it mean? That you divided your Kelly 5 times, 4 times, 3 times.

Okay.

That's how much you should risk per trade. But that's not a theory. That depends on your style. You can use different. That's what I use for trading the championship. Okay? And that's what I did. And that's response to the question, how much should I risk?

Yeah.

Okay? On a model that I actually built for prop firm, not for—

So this is how to master prop firm.

Yeah. This is how to bring your prop trading to a next level using math.

Mm-hmm.

And it is so controversial because this method that I'm about to teach you responds to put advantage on losing traders and cut legs on profitable traders. So the question that you can ask it now is that if you, that you are a profitable trader, that you live happy with trading, Why do you use this model? And the response, as I say, the response is easy. If you want to live up to— if you want to live up to swing trading, you would like to get a paycheck every month.

Mm-hmm.

But you perfectly know that swing trading is impossible unless you apply this. But you have to apply only if certain conditions are verified that we will see later. And the So I would prefer to cut a little bit of my profit, but get consistently payout even with my— even when my strategy isn't performing well. And why? Because it can be helpful if you that you are watching right now have maybe some kind of psychological problem.

Mm-hmm.

Like a mindset. Maybe you deviate from your strategy after a losing streak because you can't resist in a losing month.

Mm-hmm.

Okay? And for applying this edge, you have to actually study one thing, your equity line. So I looked at mine because it's the same. I created this model, so it's the same observation, etc., etc. So I looked out of my equity line in the past 3 years, and it looked like that. Okay? That's what a real equity line should look like.

Mm-hmm.

Okay? What I found left me speechless and gave me the idea of 2 things. First, I started analyzing All of this. Hold. Of course, it's not exactly like this. I'm just drawing this, but I started analyzing all of my drawdown over the last 3 years, almost 3 years of trading.

Yeah.

And what I found left me speechless because I found that my drawdown does not come in sequence. Okay? So for example, If I close a month in 4% drawdown, okay, what I found is really interesting because if I have to classify my drawdown in week, so for example, week 1, week 2, week 3, week 4, okay, it's not like this, minus 1, minus 1, minus 1, minus 1. My drawdown does not never look like this. I still have one or two profitable week in a losing month. So my drawdown looks like this: minus seven, plus three, minus four, and what's the four and minus four? It looks right. The total is minus four.

Mm-hmm.

No.

Plus 4.

Yeah, that's right.

Yeah.

Yeah. So the result is still the same. I closed the month at -4, but I had 2 winning weeks in the month.

Yeah.

Okay. What I can use, what I can do with that is so interesting because first I saw, I built this model that is called Drawdown Shield. So let's compare. Of course, I see this in all my drawdowns.

Yeah.

Not only one. So let's assume that we compare 2 traders. The first one is trading normally. So for example, with a $200K account. Okay? Me instead, I'm applying the drawdown shield. Okay? So instead of having one $200K account, I have 4 $50K. What I do, if for example here I risk 1%, okay? Mm-hmm. Forget the same result. I will still lose. I will still risk one percent. Yeah. Okay. In every account. But the difference is this: I will trade one account per week. Okay. Let's assume that I closed a month at minus four. Okay. Because my strategy went in drawdown that month. What had I had minus seven percent on the first week on that account, plus three percent on that account, minus four and plus four. Here instead, I just gain a minus four. So what my equity line should look like here? Seven percent is three thousand five hundred, right? Yeah. So it will be. $46.5 after 1 month. Here I gained a payout because I made profit. Yeah. So, uh, I got still $50K because— but I got $1,500 and 80% of that is, uh, $1.2K, $1.3K, $1.2K. So I gained $1.2K in a losing month. This went to 48k. This went to 2,000 profit. That 80%, usually because it's from 70 to 80, is 1.6. Yeah. So in a losing month, I made $2.8K, and I still have 2 accounts that are safe.

Mm-hmm.

Instead, here I, I will be at $192K. How much did I earn that month? Zero. How much I earned? $2.8K. Of course, this will reward me just if the strategy Lose. If the strategy wins here, I will get more money. Okay? But what is the interesting thing about this method? Let's assume that this trader is called B and this trader, that's me, is called A. What is the power of B of making money the next month?

It's lower.

It's lower because he needs to recover drawdown first. So his buying power, we can say that, Yeah. Is low. Mm-hmm. Okay. The trader A still have 2 account in break-even. So he has more chance, more chance to make money the other month. Even if we take another month of -5%, this will go almost, uh, into down. We'll almost lose the account because it will go -9%. So it will have like $182K left.

Mm-hmm.

If we split this down and for example, we have -4% here, let's assume, plus 4% here, minus 2.5% and minus 2.5%. Percent. What we will have? This will be lost. Yeah. Okay. This will gain two k, so another one point six k to add, and it will be a break even. Of course, it's not exactly this because we will lose this account and some of the drawdown will be transferred here. Because you need at least, I think, for losing this account, 1.5%. That's right. Because after you will lose, so we'll start trading the next account and you will get a little bit less than that. But it's the concept that you need to understand.

And so if you lose an account in this, using this method, you have to keep trading on the next one.

So in the reality, you will have You will lose this account as minus 1.5% and you will gain 4 minus the rest.

Yeah.

Okay. But let's calculate this and let's ignore this.

Yeah.

Here you will have 2.5% is—

$1,250.

What?

$1,250.

Yeah. So we will have like 46, something like that.

Yeah.

$46,750. And here, same, you will get in drawdown. But still, if you lost money, you got $1.6K. Okay, so in total, in 2 months of losing 9%, you made more than $3,000. Okay, and here you almost rolled the account and got zero. That's the drawdown shield. That's what I use for Trade platform. And that's for me what you, that you're looking at this right now, you should use. And also we have an interesting, really interesting part of this because you will keep your payout lower.

Yeah.

Because you will gain here $2K and you perfectly know that if you gain $2K in a $50K account or you gain 4% on a $200K account, the size of the payout. it's higher. And if you maybe broke a little rule here, they will probably still give you a payout. If you break a little rule on a $20K payout, $16K payout, they will probably be attached to that and refuse that payout.

Mm-hmm.

We need to understand that you are the revenue of the prop firm. I'm not here to say you that, you know that. And so keep your payout Smaller, it always are good things and you will keep your payout consistent.

Yeah. Plus on this side, your lot size will be smaller because of the account size. So less slippage, less, you know, all around it's the execution and chances of getting to that payout and having that payout and having smoother conditions are going to be higher than potentially over here.

Yeah. Of course, if we try to run it on a winning month, so if we, for example, change this, Here we will make less money than that. But if for which case, in which, in some case we made an error and we made a nice payout here and they refuse that, it's still better this. Okay. Because here, for example, if you close a month of 4%, you will gain $8K and the 80% is $6.4K. And here you will get Maybe a little bit more than half. Okay? So you have to consider which model you have to use. And if you have some psychological problem, you find difficult to stick to the plan when you're losing because you can't handle a losing month, maybe the drawdown shell is the model that I created that can work for you.

Yeah. Looks good. That would be interesting. It is very interesting.

Yes, it's very interesting. And that's the first thing that I noticed. The other thing is a little bit a layer, a deeper layer of autocorrelation.

Mm-hmm.

And I will show you now. When I saw that my streak of losing months were not like this, but more like this, I started to go deeper on the autocorrelation, and I built a model that I call Variance Accelerator Model. That is the model that I used in the championship. And it takes one concept that is a modern version of the theory of the autocorrelation. It was made by Hamilton, James Hamilton, and it was called— Markov Switching Autoaggression. This is the name, just for a little bit of paperworks. This assumes that if you analyze your drawdown and your data in different states, you know, you can get multiple autocorrelations in different— in the same strategy.

Mm-hmm.

What does it mean? That if I'm analyzing my whole equity line, I know that in all my data, I have the positive autocorrelation and that's it. If I analyze some kind of data, this trick, for example, I can get a totally different number of autocorrelation. And they saw that if I analyze this, my autocorrelation shifts from positive to negative. Mm-hmm. So, I started to test when the streaks will most likely come to an end by using the negative autocorrelation. For example, I analyzed that after a losing streak of 7 to 8 trades, okay, my autocorrelation became strongly— what does it mean? that the shift of the sequence will be more likely. So if the 7 and 8 are losses, yeah, my autocorrelation shifts, so I will probably get a winner. Okay, and this I used combined with Kelly for understand how much my strip, my streak go deeper in my data. So you can also have 2 kind of correlation, you should get enough sample for this. You cannot analyze on 100 trades, but you can also understand when your strategy reverses and when your streaks get longer.

So you can actually get a better indication of when they can take place, how long they'll last, and when you're getting towards the peak potentially.

Yeah. And that's what I noticed. And that's what I apply in this 2 models. the VAM, I would call it, because it's a combination of this concept and Kelly for understand how much should I risk. And for example, if my autocorrelation is really high, I can go from 0.5% risk that I'm using to 3%. Okay? Because the math says so. And that's what I use. Okay?

Mm-hmm.

And in the beginning of the video, I say to you that I had a surprise, a really interesting model that you can also use you that you're watching this right now. And let's see this final model. And after this, we're done. Okay. Because I understand that it's new things and it's really complicated, but I'm trying to explain as simple as possible. I don't know if I'm doing good.

No, you did a great job. Amazing job.

Yeah.

There's a lot of things that really open your eyes into how much you can dive into data and why data is so important. A lot of people talk about data, and usually that just refers to, you know, collecting the data on your trades and journaling, that this is a whole new world.

Yeah.

Which, to be fair, if we see the industry, and on one side you see obviously the sort of unprofessional side, but we don't really talk about that.

Yeah.

But on this side of things, we are seeing such a rise in, yes, more professional tools from like order flow or even options flow now. But even mechanical trading, automated trading, statistical trading. So this is needed, necessary. You know, it's the only way that we can start to see more of a shift is more awareness, more education like you're presenting right here. So, but it has been, you've done a great job of going through start to finish so far. And it's really eye-opening as well.

Thank you so much. Because people maybe ask me sometimes why you don't use order flow. And the reason is that I don't need to, okay? I use math. I don't use a footprint, for example, because I don't need to understand a context of an indicator. I don't basically use any indicator beside Asian session. That is not an indicator, it's just a range. And for me, the people who trust more indicator than the strategy or the Risk model itself is not so good. You understand that. So we will see this final model, and this is where— and I think that this is the part where, uh, that breaks people's brain really, because, uh, it goes deeper on why I do some things. So every trader is trying to predict price, if the price will go up or down. Okay, I don't need to predict price. I don't care about price. I care about the weather of the price. As I told it to you, market moves in regimes. Okay, and this is a verified concept, and most of hedge funds use this concept. And there is this tool called GARCH. What does GARCH tell you exactly? The concept behind is the volatility cluster. So, I know that if the one session were so violent, The next session can be the same. Okay? This is probability. And GARCH is the tool that tells me exactly what to expect about the weather of the next session. Not if the price goes up or down, but how violent or calm it will be. That's how you use GARCH. We won't see the formula of GARCH because it's so complicated. And compared to autocorrelation, for example, it's simple. Compared to GARCH. But you just need to know that GARCH is the tool that tells me what to expect for the next session. So what I did is basically simple because I codified 4 regimes. I did it, so you won't find this in any paper or something like that. And the We can set it in range. The first we will call calm. Okay, the second one is middle. This is I, and this is extreme. Why I bucket this? Because I need to know how my strategy performs in each regimes. And the concept between behind regimes is the concept that is between also autocorrelations because. We know that a strategy can— the sequence of our trades is different here and here. This is basically autocorrelation.

Yeah.

Yeah. But here, what really stunned me, because here is where 75% of my trade happens.

Okay.

Okay? And let's assume that we are talking with a strategy that is 1:1 risk-reward, because it's easier to understand for you that you're watching this right now. So we are using a 1:1 strategy risk reward, and we have a win rate of 65%. Okay? Here, I usually have a breakeven result. I'm not very profitable here, and I usually have a breakeven result.

Okay.

And when you have a breakeven result, you cannot do much about this. Okay? So we have Break even. What left me speechless is that when I follow my strategy in a calm regimes, my win rate shifts. I gain 75% loss rate. Wow! And it's so interesting because I lose so much money here. Okay, and I perfectly know with Gars. When this will happen. Okay? So people will think that a strategy with a 75% loss rate is really bad. I think that this is a blessing. Why? Because if you are trading with a 75% loss rate and your strategy tells you, okay, in this condition you need to go long, you can just flip it. So when the strategy said go long and you know that if you follow your strategy, you have a 75% loss rate, you will just go short. Mm-hmm. That's what I do. And people can ask me why this happened. And the reply is, I don't know. I don't care why my strategy is so dead here. My work is not to understand the market, is to be on the right side of it.

Okay.

So I'm— I don't care why price is doing this thing. I only care about building edges. And I don't care if the market structure is long and I'm doing the opposite. I'm doing what data tells me. If I have 75% loss rate with 1:1, I will have a 75% win rate going in the opposite side. And that's what I do. Mostly of my trades doesn't follow actual— not mostly, because here, like, they have like 50 or 20% of my trades up in here. But I later— before, I told you how to read market structure, how to use this concept, and you will use it. And when you find the signal of long here, you just go short. Okay? And that's, I think, the most controversial part, because what I do is not following the market, it's following the data. Okay? And asking why does this happen, I think that's not useful.

Mm-hmm.

Because once you know why this happened, how do you, how do you deal with it? It just happened. Okay? You cannot predict it. So don't ask you why something works. Let's see if it works and do what the data says.

Mm-hmm.

And that's, I think, the latest model that I will put here. I have to tell you that there is so much more. When you run a statistical strategy, you have another part that is called optimization. Because as you perfectly know, the market shifts. You have to monitor the data, you have to adapt it, you have to judge it and, uh, and say in a discretionary way how to stop it or how to change it. Because data is a bigger part, but remember that there's a brain behind it. And that's what I do basically. Okay. That's all of my— that's, that's not all of my arsenal, but this is, this is One, they are lots of, how I can say, this and the other model that I showed you are one of the best models that I ever found on my own career, I think. And yeah, that's it.

How much time would you say is spent on the data and optimizing and sort of refining and understanding versus execution?

So I would say a response By my own experience, I would say a few years. Okay? Because I took maybe 6 or 7 months to switch from discretionary to mechanical. So objectifying all the rules. When you have to go deeper and you want to be a statistical trader, you have to search first for the risk model that you want to use. That can be— it's not only GARCH. Okay? GARCH is a You see this, do this. There is model, can be autocorrelation, and you have to find for a good one because, um, I explain you this, okay? And this works, but, uh, I really like to read. So I wrote lots of books basically, uh, of statistical analysis. They tried so many of this and maybe 7 didn't work out well. And I tried it and I lost so many times doing it. So you need to understand what works and what doesn't work. And you have to gain data for it. So I would say that is a really long process. You don't actually need to do it because you— I became profitable just with a mechanical way, just with a mechanical strategy. And yes, that's it. I think that we live now in a trading content while everyone is searching for the better strategy, the coolest indicator.

Mm-hmm.

the coolest way to analyze a chart. And I instead think that the whole secret of trading is that admit you have— there is no secret. Okay? People always try to find the shortcut to profitability, but the only real shortcut is stop searching for a shortcut.

Literally, yeah.

Yeah. This is, trust me, this is really, True. If I have to give an advice for a trader that is not profitable yet, I would say find something that proves you that it works by data. Because most people trust other people's words and they put their money on trust. But is it really a good idea? Even if I'm saying you, if I say to you tomorrow, trade this because it works. You shouldn't trust me. You should test it. And if it works across large enough sample that variance can lie to you, you can put your money on it. So I think that's the best advice. Don't search for trust. Don't just listen to words and verify all that you find online in the trading space. Definitely.

Man, well, You know, as you said, I love the closing statement there in terms of it will take time. There is no shortcut. That is the reality. A lot of people don't like to hear it, but I feel like I always lean on, you can get into trading for the money. That's fine. That's natural. Yeah, we're human. But you have to stay in trading for the markets, for the intellectual side. The puzzle, if you will, needs to pat, you know, you need to have that passion for the puzzle pieces of the markets. And then Also, your goal might be money, but your goal is also long-term, right? And as you mentioned at the very beginning, is to live off trading.

Yeah.

And to do that, it takes immense work. It takes immense understanding. So whether you transition to be a statistical trader or mechanical trader, this is the foundation that you've laid out for everyone here. So I thank you for that. Well, there you have it, guys. Here we are in London. We finished the studio. You might have seen a segment in Dubai. You might have seen a segment in another studio. Um, but that is just the depth that we've gone into today. And I thank you, uh, Gianluca, for broke— breaking this down in such a step-by-step process with the statistics to go along with it. And, you know, I'm very interested for people to then follow up and see the feedback as well. So drop a comment with your biggest takeaway from this episode. We said it from the beginning, very, very different. But the levels that we're seeing in terms of the guests coming onto the podcast, we're talking world trading champions, verified traders. It's different. It's not the same as what you see on social media usually, and that's what we're trying to do here. And we're trying to bridge a gap, and it's thanks to our guests that allow that to happen. So I thank you again. The links for Gianluca will be in the description below, so make sure you check them out right now. Hit like, comment your biggest takeaway, let us know what you wanna hear. If there were any questions you had that I didn't ask, drop 'em in the comments below so we can look to maybe cover them in a future video as well. But other episodes are on screen right now. This has been Chart Fanatics. Take care.