Transcript
In this free video course, you will learn everything you need to know about AI trading for stocks, crypto, forex, or any other financial market. By the end of this course, you will not only be able to analyze any chart using AI directly in Trading View, but you will also be able to build, back test, and automate your very own trading strategy step by step. All right. So, hello everyone and welcome to this full course on AI trading. And before we begin here, I want to say something. This right here is probably the most important course I have ever made on this channel. And the reason is simple.
AI or artificial intelligence is becoming a bigger and bigger part not only of trading but of basically our lives in general. And in the coming few years, the difference between a trader that really knows AI and a trader that is bad at using AI will be massive. But the good news is that right now as I'm recording this video, we are still so early. And my goal here is that by the end of this course, you will be on the right side. You will be one of these traders that are very good at using AI and as a result, you will make more money. But okay, so now let's begin here by taking a look at what you can expect to learn from this course.
And before we can start using AI, connecting our AI to Trading View and start developing trading strategies with it, we of course first need to know what is AI trading in the first place. You probably already by clicking on this video have a little bit of understanding what we mean by AI trading. But after chapter 1, you will probably have a very clear understanding. And it's so important to have that understanding before we move on to take a look at how we can actually connect Claude. Claude is an AI. It's a LLM uh used as Shatbt. Uh but in my opinion, right now at least, it's actually the best uh AI out there, especially its latest model called Fable 5 is, you know, mindblowing in my opinion.
And we will actually take a look at how we can connect claude directly to trading view. Trading View is the trading platform. If you have been following my channel, you know that that is the trading platform I use for all my analysis. And it is, I think, at least in the west the largest trading platform. So, we will be using Trading View throughout this whole course. And if you don't have Trading View yet, I do have a special link. Uh you will get, you know, a $15 bonus and you can try out the premium version for 30 days. So check that one out if you don't have trading view because I do think a good way to use this course is to have you know to basically follow along.
So to have your trading view up and you will also need that as I said to connect to cloud. But after we have connected our AI to trading view we're going to start taking a look at some practical AI workflows. And with workflows, I basically means mean you know specific step-by-step ways in which we can actually actually use our AI. So I will try to show examples of how we can use the AI to find market structure. So things like support and resistance. Uh but we will also use the AI to you know use Trading View indicators. We will even try to do some multi-time frame analysis. And after chapter three, we're actually going to start taking a look at some, you know, more complicated things.
We'll take a look at both building and back testing, you know, a rulesbased trading strategy with AI. And it's kind of magic because the AI can actually do all this on your own. Uh, and last, but definitely not least, we're going to take a look at AI risk management and how to set alerts. And we will also take a look at an example of a sort of daily routine uh that you can start uh that you can start using. Uh but here it's very important especially for this course that you watch the chapters in order. So in order to get you know the most value out of chapter 3, chapter four and chapter 5, you of course first need to install uh you know claude and trading view and you need to know how to connect them.
Uh, and before you do that, you know, it's super important to know about what AI trading is. So, I highly recommend to watch the chapters in order, but if you need to, you know, rewatch a part or jump around in the video a bit, don't worry because I will include YouTube timestamps uh so you can jump to a specific part of the video. But now guys, this is a very important video. So, make sure to really lock in, remove all distractions, you know, go grab a coffee if you need one. uh you know we do have if you didn't know we do have these coffee mugs with you know candlestick patterns and chart patterns that you can check out in the mymath money store.
Another thing that you know really helps me when I try to learn things is to actually use a physical uh pen and paper. This is the mymath money trading journal that you can check out if you want to. So yeah guys grab a pen and paper if you want to maybe grab a coffee and without further ado let's jump right into it. But right so now then what is AI trading? Well in very very simple terms AI trading just means that we're using artificial intelligence in your trading process and using AI in your trading in trading processes has actually been around for a very long time. In fact, in today's financial markets, most of the buying and selling usually happens by an AI.
And this was true even before the Shatbt era and the era of LLMs. But what is very very exciting now is that the power of AI is starting to become accessible to you know normal traders like for example me and you. Right now, as I'm recording this video, you and me have the same access. We have access to the same powerful models as you know the top traders uh on Wall Street. It has never been more equal. You know the access we have to models. Of course, the institutional traders and smart money traders have their own advantages. They have more capital. They usually have more advanced data and so on and so on.
But both you and me can actually access the most powerful large language models. And in this video uh later on in this course, we will take a look at how you can connect, you know, the best model directly to trading view. So I'm super excited about that. But here when we talk about AI trading, we're talking about everything. It's basically a range from AI assisted analysis to fully AIdriven systems. So on the one hand you can be a trader that do maybe you know 90% of the work on your own and you use maybe AI you know for the 10% that you need help with. So maybe you need help with a certain aspect of your trading maybe it's risk management maybe it's back testing testing your strategies and so on and so on.
So you can be almost uh you know fully human but using a little bit of AI but it can be all the way to you know 100% of your trading being fully done with AI. And in this course we're going to do a little bit of both. So so we are going to use AI a lot but at the end of the day in this course we're going to focus on that you in front of the screen will take the final decisions. So you will use AI to sort of make your analysis but it's you who decides you know whether you actually press the buy button or whether it whether you actually press the sell button. But here just to repeat institutions or in other words trader with with lots of money.
Think about things like hedge funds or big banks and so on and so on. They have have used AI for decades. This is nothing new. But now as I said that edge is now open to everyone. AI the same AI that is used in Wall Street can now uh we retail traders can now use. And once again I can re-emphasize that in this course we will mainly focus on using AI as your trading assistant where you take the final decisions. And if we quickly look at this uh image right here, you can see the spectrum here from AI assisted analysis all the way to 100% AIdriven systems. And this right here is so important that the edge uh from AI is now accessible to everyone.
And that is why this course is so so important. But right. So now then now I want to give everyone a little bit of a reality shake. We need to understand what an AI can do and what an AI can't do. Because if you have made it to this part in the course, you are probably already locked in. And as you can see on me, I'm super excited. But I want to give you guys realistic expectations. My channel here on YouTube is not the one to watch if you want to get, you know, become a millionaire overnight. There are already so many channels claiming that out there on YouTube. I really want to set realistic expectations before we continue here.
So, let's actually start here by taking a look at what it can do. What are the upsides of AI and and it already can do a lot. So, it can do things like analyzing live charts. Yes, AI can now read, you know, every single candle on your chart. It can also apply uh technical analysis concepts like uh you know support and resistance. It can use indicators and so on and so on. It can do any math you need to do on the chart and it can help you stay objective. But what I want to say right away here and this is very important. It can't predict the future or guarantee profits. And when I mean that it can't predict the future, I mean that it can't predict the future with 100% accuracy.
In fact, no AI system has ever been able to do that. But what an AI can do is to help us with our analysis and make the odds in our favor. So it can improve the odds of us taking successful trades and as a result make more money. What is also very important to notice here is that AI is improving every single you know week. If you are following the AI space, you might be confused about all the new models. There are basically coming uh coming new frontier models every single month now and they are improving and improving. And this is why it's so important to learn AI as soon as possible because if you don't learn it uh right now, there is a big risk that you will get left behind.
But what is very important about this course is that what you will learn here will be useful in the future as well. So when new models inevitably come, you can still use this workflow and all of the things you learn here can make you stay ahead of the curve while AI is improving. So my goal here is that this course will actually be be useful years and years from now. But right so now then now now the time has come to take a look at how we can connect claude which is our AI to trading view which is our charting program. And this part is very very important. So make sure to follow along step by step.
And first of all to make this connection we are going to use the trading view MCP. And what this basically is is that it's a free open-source bridge or connection between Claude and Trading View. And this allows Claude that is our AI to directly see everything that happens on the screen. So it can re read the real time live uh live price data and it can also control the chart automatically. So what Claude actually can do is that it can for example draw support and resistance lines without you even touching the key the the keyboard. It's actually amazing. And here the connection consists of four pieces.
This is not really something you need to memorize or something you need to learn but it consists of basically a node.js the MCP which is the uh connector and we're actually going to use a debug mode. But as I said uh don't really worry about that because I have created a a guide and a prompt that will guide you through how to connect trading view to cloud step by step. But what is very important that is that for this to work you need the trading view desktop app. So you should not use trading view in browser. You need to download the trading view desktop app app. And I actually think you need a play paid plan for this.
And here, if you don't have a paid plan, don't worry because I do have a special link to Trading View. You can use that link to get Trading View Premium for free for 30 days. And if you later on decide that you want to upgrade, you can also get a $15 bonus. So, you need Trading View and you also need to get the Claude desktop app. But now, I'm going to guide you through how to connect Claude to Trading View step by step. And to do this, I have actually written a guide here on mindmathmoney.com. I will make sure to link this guide in the description of the video or you can just go to mindmathmoney.com.
You go to free educational guides right here and you scroll down and press right here. How to connect Claude to Trading View. And here, if you're interested more deeply on how this works, I actually recommend to read through this guide. But if you scroll down a bit here, I have actually created a prompt that you can copy. So what you want to do is that you want to copy this prompt right here. And then you actually want to place the prompt within Claude because here is the magical thing about AI. The easiest way to install this connection is to actually use Claude itself. So by using this prompt, Claude will guide you guide you through the installation process step by step.
This was exactly how I did to connect a trading view to Claude. And I have made this prompt to be beginner friendly and step by step. But here real quick before you make this connection, I also recommend that you actually check out the official uh GitHub page for this Trading View MCP or in other words for this connection between Trading View and Claude. And I also want to uh really emphasize right now that this uh this uh connection between Trading View and Claude is not an official tool made by Trading View. This is not made by Trading View. As you can see here on the GitHub page, this tool is not affiliated with or endorsed by Trading View.
And this GitHub page uh explains everything. It explains, you know, how it works and why it's safe to run and so on and so on. So before doing anything, I actually really recommended that you read through uh you know, the disclaimers and the important parts of this page so you really know what to do before you do it. But as you can also see here, this page has, you know, 4.3,000 stars. So this is a connection that is already used by, you know, thousands and thousands of traders. But yeah, a good habit to develop, you know, as soon as possible in the AI age is to always, you know, read through everything and really understand, you know, what you are installing before you do it.
Okay. Okay, so what you now need to do is that you need to download the Claude app and you do that on claude.com/d download and you also need to download the tradingview desktop app and you do that on tradingview.com/desktop. Then you want to open up a new chat inside claude and then you want to paste the prompt that I showed you guys earlier that is in the article. This is the prompt that will guide you through the installation process step by step. So you paste this prompt right here and here depending on your claude plan. As you can see right here, I'm currently using the best model which is Fable 5.
But depending on when you watch this video, Fable 5 might not be available. Uh but what I always recommend doing when you use AI uh is if possible you want to always use the best model because the difference between you know a model that is at the very top and a model that is just a little bit below there can actually be a very big difference in efficiency. So if you can um uh I recommend to actually use the highest model uh especially when you do things like installing things and you know uh using it for trading I highly recommend the the highest model. So I'm going to use Fable 5 here. And as you can see uh in this installation process I will not show the whole process but it should be able to guide you through everything you need.
So for example now it asks ask you you know are you on a Windows or Mac? So I'm on Windows so I press Windows. Uh it asks you you know do you have the trading view desktop app installed installed? Yes it's installed. Yes, cloud is installed and so on and so on. And it will start to work through work you through this whole process. And as you can see here, step one, you want to check if node.js is installed. It shows show you everything, you know, everything really step by step. But all right, so I really hope the installation process went well. If you have any questions regarding the installation process, I actually highly recommend to, you know, ask Claude about it.
it's, you know, much more intelligent than me in many cases. Uh, but of course, if you, you know, stumble upon any problems also, you can leave a comment down below and I will try my my really best to help. But now we have open up Trading View. Here we are, as you can see, we are on Bitcoin on a daily time frame and Claude is connected to Trading View. And here, as you can see, I usually like to have uh this on my second screen. So, I usually have Claude on one screen and Trading View on another. But so you guys can see everything at once, I will keep them both at the same screen. Uh so here I'm just doing a simple test.
I'm asking is Claude uh connected to Trading View. Uh sometimes you need to allow for things to happen. And as you can see right here, yes, Trading View is connected and healthy. I can see your open chart, Bitstamp, Bitcoin USD. So it actually sees it. Uh it sees that we are on the daily time frame. Uh it says I can control your charts, your indicators, p uh pine script alerts, everything. So it already starts to feel a little bit like magic. But we are just getting started. But right so now then now when you have training view connected to Claude, let's start using them together. And let's begin here by taking a look at how to prompt when Claude sees your chart.
So these are basically some tips on what to think about, you know, when you prompt Claude. It can be very overwhelming in the beginning because you can pretty much ask it to do to do everything. But it's important before you prompt to understand what Claude can uh what Claude sees and knows and what it doesn't. So here what Claude already sees is basically the assets. So for example, are we looking at Bitcoin or are we looking at Tesla or Forex or whatever? It can also see, you know, the time frame. It can read every candle and so on and so on. So the way to prompt it here is to go straight to the question.
And what I like to do here is that I like to ask, you know, one specific request at a time. So ask for one task at a time. And what you can also do is that you can share your view and let Claude challenge it. So you can for example, you know, say that if you think the chart is in an uptrend, you you can say, you know, you can ask, for example, Claude, are we in an uptrend right now? And basically see if it agrees with your analysis. And what I also like to do is to name the concept clearly. So for example, when we prompt Claude, are we talking about market structure? Are we talking about supply and demand?
Are we talking about liquidity? Be, you know, concise, but be be very precise. So, these are basically my beginner tips in how to prompt Claude. So, let's try a super simple prompt just to get started right here. Let's just write, you know, what is Bitcoin trading at right now. And once again, while Claude loads a little bit, I do once again want to emphasize that I personally highly recommend to use the highest model when it comes to using Trading View because I I have been trying to use a little bit of lower models. I tried Opus 4.8 and set 5. And even though this is just one step down in the quality of the model, I can actually already notice a pretty big difference in uh you know the accuracy of of the outputs.
But here you can see that uh it has loaded uh completely and it says Bitcoin is trading at 63,187 right now on the bit Bitstamp spot price and as you can see that is actually exactly the price Bitcoin is trading at right now. So it's super super cool. It really does have the uh the live data. But right. So now then now let's take a look at the first workflow or in other words the first way in which we can use claude inside trading view. And my first recommended way to use it is to use it for market structure analysis. And if you didn't know, market structure is basically when we look at the chart as a whole.
So this is when we are answering questions like, you know, are we in an uptrend, are we in a downtrend or maybe in a sideways market. We look at things like trends like break of trends. You might have heard of, you know, break of structure or for short BOS and change of character or for short CH, OC. So these are all things we can you know uh ask Claude to analyze. But here, if you are completely unfamiliar with market structure, in order to get the most out of your AI, I really think it's important to have a good understanding of market structure yourself. And luckily, I do have a full course on market structure.
So, I will make sure to link that course both up in the video card as well as the description below. Uh, so I highly recommend to check that course out if you don't know market structure. If you have a good understanding of market structure, the AI will be much more useful. So, we are still in an era where the best traders are basically traders that have good knowledge on their own and they use the AI. If you don't have knowledge on your own and use the AI, you can still make devastating mistakes. But here, just as a few examples, you can, for example, verify, you know, swing points. Uh, but your structure knowledge is the quality check.
So your own knowledge is still you still need your own knowledge as I said and if you need more market structure knowledge I highly recommend as I said to check out my market structure course but all right so now let's try a market structure prompt right here so I'm going to write find the most important swing points on this chart use the visible candle candles and By visible candles, I say that Claude should look at the the chart that I can see right here and analyze just this chart. Uh mark the major swing highs and swing lows and tell if we are in an uptrend or downtrend and then press enter.
And now we can just sit back and relax and let Claude do its thing. uh currently sometimes you will need to you know uh click allow uh for some actions for cloud to do some act some actions and because we are using the best model here it will take you know a lot of time to think but I really really think it's actually worth it you know it's worth to wait you know extra time to get more intelligence and better analysis and now you can actually see that without me doing anything uh the chart is uh claude is actually starting to mark up you know major swing highs and even the levels and major swing lows and even the levels.
So you can see here it has marked you know multiple highs and lows here. So here you can see all in the chat you can see all the major swing points and you can see here that the verdict verdict is that we are in a downtrend right here and here to be honest for a first try I actually think that this is very good. I think it found this major swing high super clearly. You know the swing high right here super clearly. This is how I would have drawn it as well. It found the lower high right here super clearly and then yet another lower high. So very clean uh on the highs here on the lows. I wonder a little bit about this one.
I'm not sure why it drawn uh draw this higher low right here. But it clearly found the swing low right here. It found a lower low and then a lower low. But here, remember that my prompt was pretty vague. I just said, you know, find the most important swing swing points. And that that can be hard for the model to understand. So if you want more specific rules, you can be more specific with your questions. So for example, what do you actually mean by a swing low or swing high? And you can also be more specific about the range. So maybe start finding swing swing highs and swing lows from all the way back in, you know, October.
And just to show you guys a little tip if you are lazy is that you don't need to, let's say you want to remove the drawings. You don't need to do this on your own. You can just say remove all the drawings you just did. And then just press enter and Claude will automatically remove everything. And now, as you can see, all the nine drawings, the seven lab labels and the dashed lines are removed and the chart is clean. Thank you for that. All right, so now let's take a look at our second workflow, our second way to use AI inside Trading View, and that is to draw support and resistance, or in other words, key levels.
And what I like to do here is that I like to use claude to find strong levels or the most significant levels. These are levels that, you know, many traders are paying attention to. Um, and the reason I like these levels is that these are the moments where we can really start to see some predictable behaviors. So, I really like to draw out these strong levels. So, you can use it to, for example, find, you know, tested swing points. You can fi find areas with maybe high volume. You can find for example round numbers or psychological levels and in simple terms claude will find them. They can explain why they are important or you know why they are not important and of course it will draw them automatically on your chart.
But here it's always a good exercise to, you know, not trust the levels blindly. After Claude has drawn the levels, you want to pause and check the levels yourself if they sort of make sense. Okay, so now let's jump back in Trading View. Let's actually draw out the chart a little bit more so we can get the more significant levels. Let's open up Claude right here. And one thing I really think is useful in Claude is that you actually don't have to write your prompts. If you want to, you can just speak instead. So now I'm going to use the recording. So uh wait, I'm going to uh remove it. So I'm just going to uh get it clean.
Okay. So now I'm going to give my prompt. Okay, Claude. So I want you to find the most significant levels on this chart. I want you to start from the beginning of 2024. And what I want you to do, I want you to draw out the levels that are really important. Think about the levels that many traders are paying attention to and levels that are likely to get, you know, uh, very strong reactions in the future. So, think about things like all-time high levels or levels that have been clearly tested multiple times. uh also think about uh you can also consider volume. So levels where we haven't seen lots of volume are more significant you know take your time and make sure that each level make sure that you have good reasons for drawing each level and then press done.
So as you can see now when you speak you can get much more sort of precise with your prompts. So let's hope that Claude does this well. Let's press send message right here. And don't worry if it works for a while. You know, when you give this kind of prompts, I actually like to see it working for a while because that means that it's using, you know, a lot of its intelligence, uses a lot of its test time compute, which is a fancy way for basically saying that the AI is thinking. All right. So after a lot of thinking uh the model is now done and we have multiple multiple levels here on the chart.
We have a total of 12 drawings and here uh uh Claude draw the resistance lines with red and the support lines with green. So the levels to the upside with red and the downside with green here. And what is very very nice about this is that you will get the explanation for each level. So for example, the first resistance is pretty clear. It's the level at around, you know, 126,000. Uh this is the all-time high level. And as you can see, it even says right here it says here you have the all-time high. It's triple tested here in between 24,000 to 126,000. And as you can see, it was tested one time, two time, three time.
Maybe if I did this myself, I would preferably have drawn it so that it touches, you know, maybe draw it out as a zone like this would make so make sense in my opinion. But very nice that Claude found it. Another very important area is the one here in between around 112,000 to 108,000. It's these two red lines right here. And once again, I would have probably drawn this out as a area right here. But I think that overall, this is actually super super impressive because as you can see, this was indeed a super important area. You can see we had one touch, two touch, pretty much uh three touch, four touch.
We broke this one and then we sort of flipped this previous resistance right to become support for a while. It also found this resistance right here, this clean touch right here that previously acted as support. Uh, and here it actually drew out a zone. Here you can see it actually drew out drew out a major support zone where we have multiple touches here recently. So, to be honest, guys, for being, you know, a one-time prompt. This right here is, you know, in my opinion, it's so good. You know, it's not perfect. It's not, you know, exactly how I would have drawn the levels, but it has found the most significant levels.
Maybe it overdraw, maybe it drew a little bit of too much lines. Uh, in my opinion, but it found the most important levels. Um, and here you can always improve. So, for example, next time I prompt it, I might instead of ask asking it uh for drawing lines, I might ask ask it to draw zones. So, I can write something like, you know, focus on support zones like you drew the one at, you know, uh, 57,000 to 60K. And this is one of the magical things about AI is that you the more you prompt the AI, the more you will get used to it. And uh the better the AI will become and it can make you better. So it can basically be a positive cycle where you make the AI better and the AI make you better and so on and so on.
But overall, I think that Claude gets a big thumbs up for the support and resistance workflow. And now, as always, if you're too lazy to delete all the drawings yourself, you can just write remove the drawings now, please. And as you can see, this was super quickly. Just after like 5 to 10 seconds, all the drawings looks removed. It says done. The chart is now clean and the drawings are removed. Okay, so now it's time to take a look at our third workflow. And this right here is using Trading View indicators and indicator analysis. So here it starts to get really really interesting because you can use Claude to for example automatically add indicators but you can also automatically generate for example buy and sell signals.
You can automatically change the settings as well. Claude has the power to pretty much do anything a human can do. So, as I said here, Claude can, for example, add indicators by just typing or even using voice like I really do. You don't need to worry about any menus and stuff like that. And what Claude also can do is that it can read the live values and also compare them to price action. So, it can not only read the indicators, but it can also tell you what the indicators actually tell you about the price. Yet another thing it can do is that it can check confluence. And in simple terms, confluence in trading is basically when you have multiple signals pointing in the same direction.
So for example, if you have one, let's say you have a bullish price action signal like a candlestick pattern, but at the same time you might have let's say you have a bullish signal on the MACD as well and a bullish signal on the Fibonacci. Now we have three bullish signals happening at the same time and that is what is known as confluence. The more signals happening at the same time the more confluence we have. So you know one type of confluence is if the indicator for example agree with the overall trend and so on and so on. And here the claude tool works with you know pretty much any indicator.
So you can for example use you know the RSI, the MACD, moving averages and so on and so on. Uh but when it comes to using just as with many other workflows when it comes to using indicators I highly recommend to learn about the indicators before you, you know, just blindly use AI. But luckily I actually have full trading courses on all of these indicators on the channel. Uh so I will make sure to pop up a playlist right here. In this playlist, you can find all my most important uh trading courses right here on on YouTube. And all of these trading courses are completely for free. Uh so I highly recommend to, you know, no matter what indicator you use, you know, dive in to the course so you really fully understand the indicator.
Okay. So now we're back here in Trading View. We are on Bitcoin and we have our Claude assistant ready to work here. And let's just start with something simple. Let's add an indicator to our chart. And to make it really simple, let's just add, you know, let's add a simple moving average. And as you guys might know by now, I prefer to actually prompt using my voice because you can get uh more uh details and it's also faster. So now, let me prompt our assistant here real quick. So, we press record. You can also click Ctrl D. Uh sorry about that start. Ignore what I said. Ignore Ctrl D. Uh what I now want you to do is that I want you to open up a simple moving average indicator.
And I want you to do this using the indicators tab. So you open up the real uh simple moving average, not using the API. And what I also want you to do at the same time is that I want you to change the color of the simple moving average to be purple. Um and I want it to be the thickest line possible. And remember it should be based on 50 days and make sure that it works. Uh and yeah that is your task. And then we just send this message right here. And hopefully cloud will be able to to understand this prompt. All right. All right. All right. It says done. The SMA is now live on your Bitcoin USD chart.
Uh so it looks pretty good here. It looks like it added the simple moving average. Uh to double check, we can go up here to the settings tab just to check that it is indeed, as you can see, it's 50 days. Uh we are on the daily time frame, so that is correct. And if you look here at the style tab, you know, fable, I asked for the thickest line and I asked for purple. Uh so it basically solved the whole prompt in one shot. That was very uh very very impressive. Uh but now okay let's sort of try to test this AI assistant to the limit. So what I now want to try to do is that I want to try to make it add a MACD indicator and without me telling it uh you know the settings I want it to research and find the right settings and change the settings for me.
So let's try this prompt right here. Hello Claude. very good job with the uh adding the SMA to the chart. That was really, really good. Now, I have a harder task, but I think you can do it. What I want you to do is that I want you to add a MACD indicator using the same method. Use the indicators tab and add a normal MACD, but then I want you to change the settings to Linda Rashki's settings. Uh, and to find out the settings, you know, you you will probably have to look it up. I'm not going to tell tell you the settings. Uh you need to find them out yourself, but I'm sure you can do it. Uh so add the MD, change the settings to Linda Rashki settings.
And yeah, that is your task. And then we hit enter and let Claude do its thing. Uh its thing. This is a pretty complex complex task, so I assume it will have to think for quite a while. Uh but hopefully it can uh it can do that. Uh so it's already looking here. It added a few tasks. You can see it's searching the web. Uh it's looking for the right settings here. So I think it it will be able to do it. Uh but I will be back when Claude has done its job. All right guys, so Claude is already done here with the task. This was actually pretty fast. Uh, I wonder if you can see the thinking time. Uh, it was pretty damn fast to be honest.
Done. The MACD is on your chart with Lind Dashki's settings. So, here it looks like it actually researched that the correct settings is the famous 310 oscillator. That is correct. It has a fast length of three, a slow length of 10, and a single signal length of 16. So, let's double check that the work here is correct. We go to the settings tab. We go to to inputs and just like that beautifully we have the 31016 MACD which is uh the Linda Rashki uh MACD settings. So Fable was able to do both of these things. And guys, this is just scratching the surface. I've showed you how to add indicators and how to automatically sh check settings, but you can of course also ask uh Claude any question about the indicators.
Uh so here you have I I highly recommend to you know experiment as much as possible. The the opportunities are basically endless. But right so now the next workflow I want to take a look at is actually multi-timeframe analysis. So we can actually use an AI to analyze a chart not only on one time frame but we can use you know two time frames, three time frames or even more. This can uh if we if done correctly this can actually improve our odds of success and as a result we can make more money. Um but here just some basics here with time frames. We usually use a higher time frame for bias and a lower time frame for entries.
And what I mean by this is that we can for example let's say we identify a you know uptrend on the daily time frame then we can actually zoom in to maybe you know the 4hour time frame or maybe the 1 hour time frame to get precise entries to get more specific entries. Of course, it's very possible to enter at the same time frame where you find the trend, but in many case cases, you can actually find better opportunities if you combine multiple time frames at the same time. And here the magical thing with claude is that Claude can switch time frames and uh compare the structures automatically. So for example, you can let uh you know Claude analyze maybe uh four chart windows at the same time and it can sort of uh put all of the different time frames together and give you the result of the analysis.
Okay, so now we're back here once again on Bitcoin. I will try to just keep on this chart the whole course to keep it consistent. We are still on Bitcoin. We are still here on the daily time frame. And now let's actually try something. Now I want to give it some kind of prompt that it can use to analyze Bitcoin on multiple time frames. So let's say something like this. Uh great job with the MACD. I have now deleted it to make the chart a bit more clear. But what I now want to do is that I actually want you to analyze the chart you're looking at right now on three different time frames. I want you to analyze it on the daily time frame.
I want you to analyze it on the 1 hour time frame and the fiveinut time frame. And if you are unclear about you know what period uh I'm talking about or you know anything else, you can ask me questions before you start this task. So here I used a technique that can be very useful and that can make your AI assistant understand you better and that is to simply ask Claude to ask me questions so it's more clear about what I'm looking at uh or what I want because when I just say when when I simply say you know analyze the charts on multiple time frames it might be a bit unclear even even if you know I were to ask you to do that you would be like uh but yeah what what do you mean by analyze on different time frames?
Should I focus on, you know, uh, what I'm currently looking at on the screen or what is the time frame? Maybe you you would have some questions. So, let's put this to to to Claude and let's see what we work with here. Okay. So, here it's asking what kind of analysis do you want on each time frame? Uh, let's say, you know, full technical read, but keep it very short, simple, and concise. Uh, how should I deliver the analysis? Uh, let's use a chat summary. It can actually also have, you know, a big report file or chat plus screenshots, but I'm going to keep it simple. Uh, should I mark up the chart?
Uh, let's just keep the chart clean and get the full analysis in the chat. So, here it actually asked me three questions. I answer the questions. So, now we just can sit back and relax and let Claude do its job. All right. So Claude is done with the analysis. And while it was working, uh, yeah, it you could actually see it flipping to the different time frames and analyzing. Uh, but here we get a short summary. I'm not going to read the full one. Uh, but here you can see that on the daily time frame, it's bearish, but trying to find a bottom bottom. And you know, just by looking uh briefly, it looks like we might have some sort of bottom here.
Just from experience, I might uh it might be some divergence here. If we uh have a momentum indicator on the chart, it looks like we might have some divergence. Uh so that looks, you know, correct just by looking. And here on the 1 hour time frame, it says that we are neutral to a bullish range. And if you look here at the 1 hour time frame, you can indeed see that we are in actually this this was good. You know, we are in a pretty clear trading range. We can see that by marking our uh resistance and support and we can also see that of course by looking at the flat moving average. A flat moving average indicates that we are in a range.
Uh so very good job so far in my opinion. Uh last but not least it's uh on the five minute time frame it says that we are looking bullish uh intraday. So let's see how the five minute chart looking. And as we can see here, it's a pretty solid analysis because we have recently broken out of this sort of five minute range right here. Got a strong breakout. So we are indeed, you know, looking pretty bullish here on the 5 minute. So just from a quick look, I haven't looked at this uh in depth, but from a quick look, it looks like it has a pretty solid understanding even when looking at multiple time frames at the same time.
But right, so now it's finally time here for workflow number five with AI and that is how to build a rulesbased trading strategy. And of course when it comes to building a trading strategy, this is a long process and I do actually have a full course where I go uh you know step by step I give you guys a process you can use to build your own trading strategy. But even when it comes to building your own trading strategy and getting the rules, Claude can be super useful. Uh but first of all, let's just talk a little bit about the basics about the trading strategy. So first of all, a real trading strategy needs to have exact rules.
So we need to know when to enter, where to set our stop-loss and where to place uh our uh target. So we need you know some condition from for buying, selling and taking profit. And here this process we can actually start with just a simple sentence to Claude and then Claude can help us make the rules exact. And of course before we trade with real money and so on and so on we need to back test this the strategy and it's also a very good idea to first uh before you trade with real money you can trade with fake money that is called paper trading. And I do have a a video where you can learn how to get started with paper trading.
I will make sure to link that video up in the video card. and the description. Uh so here a key part of building your trading strategy is that you want to keep on improving it in until every rule is objective uh and testable at least preferably and especially if you're going to use AI to automate the the strategy. Uh you want to make sure that the rules are actually you know uh correct and that the strategy of course works. uh and here claude can help you make this uh process that's usually weak maybe done in an afternoon that is the power of AI yes in the beginning it will take longer but when you really master AI it can you know speed up so many workflows so many processes so don't underestimate this right so now let's actually take up claude right here you can use claude not only in trading view.
You can pop it up wherever you want on your computer. As I said, you can have it on your main screen. Uh I usually have it on my second screen, but to show you guys, I'm having it here on the main screen. Uh so now, let's actually just uh oh uh let's actually prompt Claude to try to build a very simple trading strategy uh using the simple moving average we already have. Uh so let's do it. Okay. Hello, Claude. Now, I want you to help me build a super simple trading strategy. And the trading strategy, let's make it as simple as possible. Let's use the simple moving average we have on the screen.
And let's just say that every time the price goes above the moving average, we want to buy and every time the price goes below the moving average, we want to sell. Uh, and I don't want you to do anything with Trading View right now. I just want you to set up in a very concise way uh my trading strategy in text uh so I can review it before we do anything else with it. And then I press done like that and enter. And now we just let Claude cook for a while and let's see uh the output. Okay. So after just a few seconds uh Claude is already done here with trading strategy. So the strategy SMA 50 cross uh so here it basically lists up everything.
So the chart is Bitcoin USD the indicator 50 period simple moving average of the close entry when the price crosses uh above the SMA open a long position and exit when the price crosses below the SMA close the position. Uh uh position handling uh always either long or flat. Okay so this was actually something I didn't specify. So maybe the model could have guessed that I meant uh go long every time uh it's above the moving average and go short every time it's below. But now it's that we buy every time it's above and we sell our position every time it's below. Uh so it's a little bit different.
And here signal confirmation. This is very important and this is something I didn't specify to the model but here it has chosen to uh uh base its calculations on the candle closes. So the candle need to close above the simple moving average to confirm the signal and this is very important. We need specific rules and it was good to see that fable added that specific rule. And here it's very important that I'm not you know here I'm not saying that this is a good trading strategy at all. In fact, I do think especially if you count with the commissions uh for trading, I do expect that this one will at least, you know, I think it will it might be a little bit profitable, but I think it will at least underperform, you know, the buy and hold strategy.
So, I think that is usually a good benchmark to compare your trading strategy with either uh an index like the S&P 500 or with just buying and holding. But for now, this is all we need. All right. So now I want to switch the attention and take a look at how you can back test your trading strategy with AI. And if you don't know back testing is basically taking a look at how your trading strategy would have performed in the past. So how would your rules uh have performed in the past? And in this case we can use you know an instant back test. So we just you know look back uh basically using for example your AI assistant claude which we use throughout this whole video.
Uh but you can also use your own for example drawn trades. Maybe you have you know uh long and short position tools in trading view or you have your own data of past trades or anything like that. You can also use for example trading views uh uh bar replay. It now has I think it has actually a builtin uh back testing now. But here the magic is that claude can calculate the win rate, the riskto-reward and expectancy and you don't need to be able to code at all. So this is the magic of claw. But right so now let's jump back here into trading view because now I do actually want to use the trading strategy we built earlier in the course.
Um so the trading strategy it's a very simple one. It's simply that we buy when the price closes above the simple moving average and we sell when the price goes below the simple moving average. Now I want Claude to automatically back test how this strategy would have performed in the past. Uh so let me prompt it using my voice. All right Claude. So now I want you to do a full back test of this trading strategy. So, I want you to use, you know, Trading View's builtin uh strategy tester and I want you to use the exact rules we recently uh put out and you can actually do two uh sort of back tests or you can have two variables.
One can be used uh without any commissions. So, how would this one have performed without commissions? And the second test can be to to make it more realistic. So, have some commissions for every buy and sell. uh you can simulate it on let's say a $10,000 account and let's say that you based base it on a sample size of 100 so 100 trades uh so we have a good sample size and yeah that is your task make sure to do it properly and no shortcuts and uh yeah we'll see you soon but yeah now we need to let Claude the cook here for a while this is a pretty big task you know testing 100 trades uh but you can See, it's already starting to produce some code right here for the back test.
Uh, so super excited to see how this one turns out. And while you watch this, like you just mentioned that, think about all the sort of opportunities this opens up. You can just, you know, build your own trading strategies, back test them, and yeah, just having the power of Fable to, you know, normal retail traders like me and you. I think it's, you know, such a, you know, good thing. I'm super excited about the AI AI in general. And yeah, this is this is just the beginning. If you made it this far in the video, uh I always forget to say this, but you know, drop a like on the video. It really helps out more than you think.
Uh if you have, you know, found any help in this course so far, uh you know, hit that like button if you want to. And I will see you guys when Claude is done with this. All right. All right. All right. So, it looks like the analysis here is complete. And as you can see on the numbers, they look super super impressive. Uh let's see here how it went. As you can see, the back test is complete. Uh both of the back test. And if we take a look here at the net profit for this trading strategy, it was without commission. So without fees, it was a net profit of $6 million and uh without fees, a net profit of $4.8 million.
So, it's insanely insanely profitable. Uh, but what we can notice here is that this trading strategy seems to go from all the way back here in May 2015. And remember that Bitcoin has gone up a lot since then. So, I do want to ask a question uh later on. You can also see here for example max draw down here at around 50% uh average% per trade. And here it also explains you know the strategy. It says this is a classic trend following profile. It actually loses 70% of the time uh but makes it all back and more on the few very profitable trades. So you can see this in the graph here that it goes pretty much sideways, you know, losing a bit, losing a bit and then it spikes up, losing a bit, losing a bit, spikes up, losing a bit, spikes up.
So it's basically a few very very profitable trades that makes this trading strategy uh that make this trading strategy work. And here in trading view you can see that uh Claude Fable automatically built this whole uh back test. And here you can get some more information. You can see total profit, total loss, the commission here, uh the total P&L and you get all this information right here. But if you are unsure about uh anything you can always ask cla and one thing that is very important that I want to ask right now is this. Okay. So this looks very very impressive on paper you know uh it sounds amazing getting that uh type of uh annual return but please compare it with just a buy and hold trading strategy.
How much would I have profited if I simply just held Bitcoin during this whole time? Because this is super super relevant. So maybe if the strategy of just holding Bitcoin was better than this trading strategy, then you know it wouldn't actually make a lot of sense. If you can can just buy and hold the asset, then all of this sort of fancy back testing and buying and selling would maybe just have made it worse. So let's see here if Claude can compare uh this trading strategy to just a buy and hold trading strategy. Okay. So here straight from the strategy testing's own benchmark you can see that a buy and hold strategy would have also made you uh you know multiple millions.
You can see it would have made you uh 26,000%. But you can actually see that in this case uh the simple SMA 50 cross uh trading strategy even with fees uh highly outperformed the buy and hold and that was actually a little bit surprising for me. Uh so here simply buying for 42 bitcoin at $236 in May 2015 and never touching it would have made you a multi-millionaire too. The strategy roughly 1.8xed buy and hold. So the SMA strategy was actually almost twice as profitable. But here it also says the strategy's edge here isn't really buying. It's the not holding through bare markets. And this actually gives us a little bit of insights into the usefulness of indicators like the SMA.
So it says here that the reason the SMA strategy was more profitable was actually that it made us avoid the very very bad drawdowns. So for example the worst draw down during the buy and hold was a draw down of 83%. So as you know when it when it comes to risk management if you fall down you know 83% you almost need to 10x your account to come back and what the SMA50 strategy uh basically did was that it allowed us to avoid the worst draw downs. So here we're actually getting some deep insights into the usefulness of uh indicators like moving averages. And now I just wrote, you know, great, thank you.
Now remove the strategy because I'm lazy and I just I'm just going to let Claude H clean up the chart here. Right. So now let's take a look at one of the best use cases for AI in trading and that is to use it for risk management. When it comes to trading, one of the I would say the most important skill you can learn is risk management. Because even if you have a profitable trading strategy, you might have a trading strategy that actually would make you, you know, very rich over time. If your risk management is not in shake, you might never reach that point because you might, you know, uh, completely wreck your account.
Risk management is something you can't do without as a trader. So being able to enhance your your uh risk management with the AI is a blessing. But what can an AI do? Well, it can help you with things like for example position sizing math. So it can instantly calculate, you know, how much you should buy, uh you know, how much you should sell and so on and so on. Um it can also help you, for example, build a complete uh trading plan, you know, where to set your stop loss and everything like that. We talked a little bit about it earlier in the course. It can help you come come up with checklists.
This is actually a very powerful. It might sound a little bit old school but having checklist is an underrated thing not only in trading but you know just in general in my opinion. So it can help you create these uh pre-trading checklists as you can see right here. Things that has to be true before you enter the trade. And here, as I just mentioned, risk management keep you in the game long enough so your edge can actually compound and you can make money. All right, so now we are once again back in trading view. And to demonstrate what an AI can do when it comes to risk management, let's just uh for example open up an imaginary uh trade here.
So let's say that for some reason we identified a trade right here. Let's say we set our stop loss just below the low right here. And let's say we use a risk-to-reward ratio of three. So, I'm just uh putting up a random trade right here. And now I want to see if uh if uh Claude can actually help us with let's say position sizing. All right. So, I'm just about to enter the trade. You can see on the screen right here. Uh I hope you can see it. uh the entry level, the stop loss, and the target level. Uh I currently have an account of $10,000. Um and I'm uh I want to risk 2% per trade. Can you tell me exactly how much Bitcoin I should buy.
As I said, I want to risk 2% per trade. And as you can see on the chart, I have my target and stop loss right here. All right. So, I think that should be good enough. Let's see if Claude can tell us exactly how much to buy. Okay. Okay. So, we are done here. Uh it says I can see your long position tool. Uh let me pull the exact level. So, let's double check here. Entry 63,926. Uh and as you can see, it's correct. The target 71,752. That is correct. And the stop loss is correct. So here the position size for 2% risk on $10,000 is you know we want to risk 200 and risk per uh bitcoin is 2,69. And here it calculated the size to be 0.0767 0767 or in other words around 4,00 $4,900 worth of Bitcoin which is currently around 49% of our $10,000 account.
So we don't need to use any leverage. We can just use our own equity. And here it's actually smart enough to notice here that our position sizing tool is actually set to our 1,00 account. Uh so it actually asks the question you know do you want me to update update the tool settings automatically to change it to the to the correct size that is you know an amazing an amazing thing. Uh but for now let's not do it because let's continue here with the course but right so now let's switch attention to something very important and that is using AI safely. What are some best practices to to use AI in a safe manner?
And the first very important thing is that we are at least in my opinion some will disagree but in my opinion we are still in a stage where you need to verify. So you can let Claude for example you know find support and resistance levels but I think it's still always important to verify that these levels are actually you know good. So if you watch this video maybe you know five years from now or 10 years from now maybe by then you know AI is so good that humans don't even need to to verify but we are definitely still in a stage where you know uh an AI is used as an assistant. You might have heard the phrase that you you you your job will not be replaced, you know, by an AI, but it will be replaced by an by a human that uses AI better than you.
And trading is about to enter that stage as well. The best traders will be the traders that are not only good themsel, but they also have a good AI assistance. I think that is where trading is going. Um, but here Claude's live prices, but not really the news behind them. And I'm a bit hesitant if I should even write this because, you know, of course, Claude can nowadays also find the news. But if you only use the trading view uh and claude connection and don't tell Claude to you know uh, you know, go deeper and find out news and so on and so on, it will not do it automatically. So here once again this is kind of like a skill issue.
If you are good at using AI you know Claude can see you know both live prices and news and everything like that but you need to still be cautious and it's up to you to you know prompt claude in good ways and use cloud in good ways and so on and so on. Also here keeping on the same theme AI is very good right now at you know preparing back testing trading strategies uh even doing analysis as as you have seen in this course it can do you know basic support and resistance it can it can do market structure but it's still very important you are responsible for everything you are the final sort of boss that decides yeah okay here is claude's analysis and claude's back test but you are the one deciding if you should click buy or if you should click sell, you are fully fully responsible.
And by adopting these sort of safe practices, you can hopefully get the full upside of AI without, you know, these careless mistakes that can definitely happen if you are not careful. But all right guys, this video this course is getting pretty long. So I think it's time to actually wrap up. Uh first of all, I want to thank you all so much for watching. Uh I know this has been a little bit of a different course, but trading is moving more and more towards AI. So I thought it was definitely time to, you know, start incorporating AI more into the channel. And if you like this course and if you want to see more of it, please let me know that down in the comment section below.
Also, another thing, if you made it all the way, if you watched the full course all the way to the end, make sure to drop down below in the comments, write full course. Uh that has been a thing I have been doing for a while and it's super, you know, fun to see so many people confused about it. People are thinking, you know, why is everyone writing a full course? Uh and it's also a way for me to, you know, see that you watch the full course. So, it's always nice to see these comments. So, make sure to comment that down below. And also, if you have any questions, of course, let me know. Another thing, this one is so underrated, but if you did enjoy this course, dropping a like on it helps more than you think.
So, if you have a spare a spare like in your pocket, I think it's free. Uh, at least right now it's free. So, if you want to drop a like, do that. And I will see you in the next video. But, all right, guys. So being able to use AI is super important, but the other piece of the puzzle is of course the trader that is using the AI. So, it's very important to improve your core trading knowledge, and you can do that with this free playlist right