YT YouTube2Text

YouTube2Text → Transcripts → MASTER AI Trading Bots in 61 Minutes (Claude + TradingView)

MASTER AI Trading Bots in 61 Minutes (Claude + TradingView) — Transcript

by Mind Math Money · 10,848 words · 1,582 segments · language en · Watch on YouTube

Open in transcriber Summarize with AI Download SRT/VTT

Full transcript

0:00In this free video course, you will
0:02learn everything you need to know about
0:04building your own AI trading bot for
0:07stocks, crypto, forex, or any other
0:11financial market. By the end of this
0:13course, you will not only have built
0:15your very own trading bot with Claude
0:18and TradingView, but you will also know
0:20how to backtest it, check that it
0:22follows your rules, and of course, to
0:25switch it on.
0:28All right. So, hello guys and welcome to
0:31this full course on how to build,
0:34control, and use your own AI trading
0:39bot. And here of, you know, all trading
0:43courses I have made on this channel,
0:45this course is, you know, high up there
0:48on, you know, the most important courses
0:50I have ever made. And the reason for
0:52this is that, you know, AI, or in other
0:55words, artificial intelligence is
0:57becoming more and more important, not
1:00only in trading, but as you as you guys
1:02know, in society in general. So, if you
1:06want to,
1:07you know, be able to stay ahead of the
1:09curve, stay ahead of other traders, it's
1:12very important to be able to start using
1:14AI to your advantage in your trading.
1:18And yeah, this is what this course is
1:20going to be all about. I will in a very
1:22sort of step-by-step manner go over
1:25everything you need to build your own AI
1:29trading bot. So, at the end of this
1:30course, if you watch this course from
1:32start to finish, you will be able to,
1:34you know, build your own AI trading bot,
1:36and you will also be able to use, you
1:38know, Claude, which at this, in my
1:40opinion, is the best AI right now. You
1:43will be able to use Claude together with
1:44TradingView, and TradingView is the
1:46platform I use for all my trading and
1:49also investing actually. So, you can
1:51connect these to and you can also create
1:53your own AI trading bot.
1:56But yeah, guys, and this is very
1:57important, to get the most out of
1:59course. I want you all to really really,
2:02you know, lock in and focus. So please
2:05guys, you know, if you have your phone,
2:06you know, put it in another room. Make
2:09sure to, you know, put this uh YouTube
2:11video to full screen and remove all
2:14distractions. You know, if you have
2:15anything that gives you cheap dopamine,
2:17you know, fast food, uh you know, games,
2:20everything like that, try to remove it
2:22to really focus on this course. But
2:24yeah, the first thing I want to do right
2:26here is to quickly go over what you can
2:29expect to learn in this course.
2:33And the very first thing we of course
2:35need to learn before we can start, you
2:37know, using and building our AI trading
2:39bot is to first define, you know, what
2:42is an AI trading bot in the first place.
2:46And very importantly, we will take a
2:47look at what an AI bot can and can't do.
2:51You know, many people when they hear
2:53about trading bots, they think they are,
2:55you know, sort of magic and that they
2:57will make you rich, you know, overnight.
3:00And while in some sense I actually think
3:03that AI is a magical thing on its on its
3:06own, I actually think it's kind of like
3:08magic, you know.
3:10The fact that you can just take up your
3:11phone, talk to it and basically get
3:13answered anything, in some sense is that
3:16that is magic.
3:17But
3:18in this course I want to try to make it
3:20as realistic as possible. It will be
3:22grounded in reality.
3:25So we will learn what an AI trading bot
3:26is. And after that we will actually
3:30learn about something you may or may not
3:32have heard about. We will talk about
3:35you know, a group of traders known as
3:37the turtle traders and their breakout
3:40rules. And the reason
3:42we will do this is that we would use
3:44these turtle traders and their rules to,
3:47you know, later on in the course when we
3:49will actually build our AI trading bot.
3:52So that chapter is very important. Don't
3:54skip skip it. And to be honest guys, I
3:56have built this course, you know, in a
3:57step-by-step manner, so I highly
3:59recommend to watch the chapters in
4:00order. But here after the chapter about
4:02the Turtle Traders, we are finally going
4:04to take a look at how Claude, which is
4:07the AI we will use in today's video,
4:11basically can build an AI trading bot in
4:14TradingView. And very importantly, you
4:16need absolutely no coding. And this is
4:19also something that is becoming, you
4:21know, more and more true in general. You
4:24know, nowadays even, you know,
4:27developers start to code less and less
4:30and less because the AI the AI's are
4:34I would say probably better than 99.99%
4:37of coders right now.
4:39And it's just a matter of time before
4:41they completely, you know, outsmart
4:44humans on the coding arena.
4:46So we'll take a look at how we can build
4:48a trading bot in TradingView without no
4:51code. The next chapter, super super
4:53important, we will take a look at how we
4:55can check and backtest your bot. So
4:58before we start using a bot, we want to
5:01backtest, you know, how would this bot
5:03have performed in the past. And then we
5:06can use this past data to try to figure
5:09out how the bot will perform in the
5:11future.
5:12The next step here is to
5:14to actually switch the bot on. So we'll
5:16take a look at exactly how we can do it
5:18so that you will get, you know, a signal
5:21from TradingView every time your bot
5:24finds a buying or selling opportunity.
5:27And also last but definitely not least,
5:29this is very important, we will take a
5:30look at how you can, you know, review
5:33and basically stay in control of
5:35everything your bot does. And I can't
5:38stress enough how important this is.
5:40Yes, it's very cool and it's very, you
5:43know, amazing that we have the
5:45technology we have today so that
5:47everyone can build their own trading
5:49bot. But at the same time, it's so
5:51important to realize that the only one
5:54responsible for what your bot does is
5:57you. So you need to take full
5:59responsibility for everything that your
6:01bot does. We can't blame it on Claude,
6:03we can't blame it on anyone. So, before
6:06you deploy a bot, you need to, you know,
6:08review and make sure
6:10that everything is, you know, working as
6:13intended. But yeah, guys, now we are
6:15ready to jump into the course here, so
6:17make sure to really lock in. As I said,
6:19remove all distractions. Maybe you want
6:21to grab a coffee. And without further
6:23ado, let's jump right into it.
6:27But all right, so the very first thing
6:29we of course need to learn about is what
6:32is an AI trading bot?
6:35And here, in very simple terms, a
6:38trading bot is simply a program that
6:42follows exact trading rules on its own.
6:47So, it's any type of program that
6:49follows trading rules on its own.
6:52And what does a bot do? Well, it watches
6:55the market. It binds signals, both, you
6:58know, buy and sell signals, and it can
7:00also place orders automatically. So, it
7:03can automatically buy and sell. And one
7:06of the biggest benefits of trading bots
7:09is that the same signal always get the
7:13same action. This is one of the main
7:15benefits of a trading bot and one of the
7:18main sort of disadvantages of human
7:21traders. I would I would say that, you
7:23know, discretionary traders or traders
7:24that trade based on, you know, their
7:27own, uh one of the hardest things about
7:30that is your trading psychology. So,
7:32think about emotions like fear, greed,
7:35you know, FOMO and so on and so on. Uh
7:38you know, that is that is one of the
7:39hardest part of trading. And when you're
7:41using a bot, the good thing about that
7:43is that these emotions will be put away.
7:47So, you will have no emotions and no
7:50guessing.
7:52And here, an AI trading bot is a bot
7:55built and tested with AI. So, first we
7:58talked about trading bot in general, and
8:00AI trading bot is built and tested with
8:02AI. And in this course, as I said, we
8:04will use Claude. And as I said, Claude
8:07is right now, at least in my opinion,
8:10the best AI out there. And the very good
8:13thing is that you can try Claude
8:15completely for [snorts] free.
8:17So here, as you can see on the image
8:20right here, it watches the market, it
8:22follows exact rules, and then it uh you
8:25know, produces buy and sell signals.
8:29But right, so now then, before we start
8:32building our AI trading bot, it's very
8:34important to be grounded in reality and
8:37to understand what a bot can do and what
8:40a bot can't do. So let's take a little
8:43bit
8:44uh look here what it can and can't. So
8:46what an AI bot definitely can do is that
8:48it can watch, you know, many markets
8:5124/7
8:52and it can follows follow its own rules
8:55without emotion. So this is two big
8:58advantages of over humans. You know, we
9:01humans, we need to sleep, we need to
9:03eat, we can't watch the market 24/7, and
9:06AI trading bot can do that. The other
9:08benefit is that the bot, at least as far
9:11as I know, don't have any emotions. So
9:13this is very, you know, good things
9:15about bots.
9:17What it also can do is that it can run
9:20fully automated,
9:22[clears throat] which means that it can
9:23send orders straight to a broker or it
9:26can run semi-automated, which means that
9:30it doesn't buy or sell, you know,
9:32automatically. It sends you a signal
9:34first. And this is what we will focus on
9:37on this course. And you know, it's very
9:39important, I think, at least when you
9:41are new to building trading bots, that
9:43you always have this sort of safety
9:45that, you know, the bot gives you a
9:47signal and then you approve the signal
9:49or not. I think that is very very
9:51important at least for beginners. And I
9:53have built this course, you know, from
9:54beginners to advanced traders. So
9:57we will mostly focus here on
9:59semi-automated.
10:01What is also very important to know is
10:03what uh an AI bot can't do.
10:07And what an AI bot can't do and what
10:08many, you know, people think and you
10:11know, when you search on it on YouTube,
10:12you can get this sense that, you know,
10:14some people say that they build bots
10:16that predict the exact future. And this
10:18is something it can't do. Nothing can
10:20predict the future, at least right now,
10:23nothing can predict the future with
10:24certainty. And it can also not turn a
10:26bad strategy
10:29into a good one. You know, an AI trading
10:31uh an AI bot can help you, you know, uh
10:35you know, produce buy and sell signals,
10:37but it can't predict the exact future.
10:40And here, and this is so so important,
10:42and you need to remember this throughout
10:44the whole course, you need to remember
10:47that you are responsible for every trade
10:49your bot takes.
10:52Okay, but now, and this is very very
10:54important, now we're going to take a
10:55look at exactly what you need to set
10:59this bot up. And it's pretty simple, but
11:02you need to pay attention, you need all
11:03of these things. The first thing you
11:06need for this workflow is that you need
11:08the TradingView desktop app. This is
11:11very important. Right now, this workflow
11:13does not work on browser, and it does
11:17not work on mobile. So make sure to get
11:19the desktop app.
11:22And you also need a plan that includes
11:25technical alerts. So you need at least
11:28the essential plan for TradingView. And
11:31here, don't worry because I do have a
11:34special link for you guys. I do have a
11:35special TradingView link, and you can
11:37use that link to try out actually uh I
11:40recommend to try out TradingView Premium
11:43for free for 30 days. And then if you
11:45later on want to continue, for example,
11:47if your, you know, AI trading bot is
11:49performing very well and you want to
11:50continue, you will also get a $15 bonus.
11:53So make sure to use that link before
11:55this step.
11:57But yeah, after you have clicked on the
11:58link, you want to then make sure that
12:00you are on this page,
12:01tradingview.com/desktop.
12:04And then simply you go down here and
12:06download it for Windows if you are on
12:08Windows, or if you are on Mac, you click
12:10here. If you are on Linux, you click
12:12right there.
12:14But right, so now you hopefully have
12:16downloaded the TradingView desktop app.
12:19Now, the next step is that you need the
12:21Claude desktop app. And here I recommend
12:25that you have a paid plan for Claude,
12:28but the good thing about this workflow
12:30is that Claude, the free version of
12:31Claude will allow you to do certain
12:34prompts. So you will be you will be able
12:36to test this both both out completely
12:38for free. But if you want to use this a
12:40lot, you will need to eventually, you
12:42know, get a paid Claude plan.
12:45But here to download Claude, you simply
12:46go to claud.com/download
12:49and if you want to download on Windows,
12:51you click right here.
12:53And if you have any other type of
12:55system, you find all of the systems down
12:58here.
12:59All right, so now the easy part is done.
13:01You have you hopefully have at this
13:03point in the course, you have the
13:04TradingView app, you have the Claude
13:05app.
13:06The last thing that you need is what is
13:08known as the TradingView MCP.
13:12And this right here in simple terms is
13:14the bridge. It's the free bridge that
13:17lets Claude see and control the charts.
13:21So if you look at this image right here,
13:24you can see that you currently have the
13:25TradingView desktop app, you have the
13:26Claude app, but now we need to make this
13:30connection. This is the most important
13:32step. We need to be able to connect
13:34Claude uh to TradingView. So now
13:38make sure to really pay attention. I
13:40will show you guys my free guide on how
13:42to do this. All right, so what you want
13:44to do now is that you want to go to my
13:45website. You go to mymathmoney.com
13:48and then you want to click here on free
13:51educational guides. I actually also have
13:53some premium guides here, one that is
13:55associated with this course. Uh it is
13:57available for channel members gold and
13:59higher, but
14:01this free guide here you can find right
14:03here, free educational guides. Then you
14:06go down here to how to connect Claude to
14:08TradingView.
14:09And what I do recommend here, I do
14:12recommend that you read this article if
14:14you have the time to understand, you
14:16know, what you are downloading. I also
14:19recommend that you actually go to the
14:20official page.
14:22So, as you can see right here, this is
14:23the official GitHub page uh for the for
14:27the MCP. So, you know, this is a very
14:30popular one. As you can see, it has like
14:327K stars. I have been using it. Many
14:35other traders have been using it. But I
14:37do want to mention right away that this
14:38is a common theme throughout this course
14:40is that only you know download things
14:43that you know uh you fully trust and
14:46understand. So, don't trust, you know,
14:48uh a random GitHub page without doing
14:50any uh research. Don't trust, you know,
14:53my video without doing any research.
14:55Make sure to do your research before you
14:57download anything. So, I want to
14:59recommend that you read this whole page
15:02before you do anything.
15:04But now with all of this all of this
15:06said, you fully you fully know uh you
15:09know what it is.
15:10Then you can actually scroll down here
15:12and I have created a prompt.
15:15This prompt right here is made to make
15:17this process as easy as possible. So,
15:20what you simply do is that you want to
15:21copy this prompt and then you actually
15:24want to open up uh
15:26chat here in Claude. And to make this
15:28process as, you know, uh you know
15:30similar as possible to you guys, I'm
15:32going to first write uh or I'm going to
15:35first speak to Claude that, you know, uh
15:37hello, can you please act as if you
15:40don't have any MCP connections and that
15:43this is a completely fresh Claude
15:45account. Act as if, you know, you have
15:47no history, no memory, or anything like
15:50that.
15:52Then I can press enter right here. The
15:54reason I said this is that my I want my
15:56Claude to look at as your Claude. So,
16:00sure from here on I have a fresh start,
16:02blah blah blah blah blah. If you want to
16:03use say so. Uh
16:05but yeah now Claude right here should be
16:07fresh. What I now do, remember I copied
16:09the previous prompt. So, now I insert
16:11the prompt right here and then I simply
16:13press enter. What should happen now and
16:16if Claude, you know, I actually already
16:18have this connection, but what should
16:19happen now is that Claude will guide
16:22guide you through the whole process. As
16:24you can see, it says happy to walk you
16:26through this. First, here's what we're
16:28building. So it explains everything
16:29you're building. It explains all the
16:32steps you need to answer if you're on
16:33Windows or Mac, blah blah blah. And here
16:37you will basically be guided step by
16:39step how to do anything everything to
16:42get this connection going. And if you
16:44anywhere in this process,
16:47you know, stumble upon, you know,
16:48something doesn't work or anything like
16:50that,
16:51what I actually recommend is instead of
16:54Of course, you can feel free to drop
16:55comments on YouTube and ask me. I'm very
16:58happy to help. But many times it's
17:01actually, you know,
17:03it's actually easier to just ask Claude,
17:05you know. Maybe you get an error code
17:07and then you want to give the error code
17:08to Claude and
17:10and work through the problems with
17:11Claude. Another tip here is that I
17:14actually recommend to use as powerful
17:16model as possible. So here you can see
17:18we have different models.
17:20You know, what if you have, you know,
17:23maybe the max plan, you can try using
17:25Fable 5.1, but Opus here, Opus 5.5 is
17:29almost as good. So what I recommend is
17:32actually for most people to use Opus
17:345.5. But if you have, you know, a plan
17:37with lots of usage, you can actually
17:39choose extra. Extra means that it will
17:41think a little bit more
17:43before,
17:45you know, giving answers. And I think
17:47that, you know, when it comes to trading
17:48and everything like that, the more
17:50intelligence we have, the better.
17:52But yeah, guys, make sure to do this
17:53whole process and it should work, you
17:56know, very well.
17:59All right. So at this point in the
18:01course, you should have Claude app, you
18:03should have TradingView, and you should
18:04also now have the bridge between Claude
18:07and TradingView. So now we're actually
18:09ready to see if this works. So as you
18:11can see, right now we are in the
18:13TradingView app and
18:15as you can see up here, we're currently
18:17looking at Bitcoin. We are on the daily
18:19time frame right here, which means that
18:21every candlestick on this chart
18:22represents one day. And as I already
18:24mentioned earlier, if you don't have
18:26TradingView yet,
18:28you know, I do recommend at least the
18:30essential plan for this workflow or if
18:33you, you know, don't, if you don't have
18:34TradingView yet, I highly recommend you
18:36do the TradingView premium trial. But as
18:39I said, I do have a special link for
18:41that. I will make sure to link that link
18:43both in the description and the pinned
18:45comment and you can use that link to get
18:47both the 30-day free trial as well as
18:50TradingView premium.
18:52But now we want to check that this
18:54connection works. I'm going to open up
18:55Claude right here and then I'm going to
18:57simply ask,
18:58you know, hello Claude, can you tell me
19:00what chart are am I looking at right now
19:05in TradingView? Tell me the chart and
19:07you know what the price is right now.
19:11And here we can see it was thinking for
19:13a little while, but now it says I'm
19:15looking at Bitcoin, which is correct.
19:17It's the exact correct market. It says
19:19that the price is right now at 78,000
19:22and as you can see, the price is at
19:2378,000. It's up 72% for the day. As you
19:27can see here, it's indeed up 72% for the
19:30day. So everything looks correct right
19:33here. So yeah, our Claude is connected
19:35to TradingView and we are now finally
19:37ready to start building our trading bot.
19:41All right, so now before we can start
19:43building our AI trading bot, we first of
19:47all need to know about the five rules
19:51that every trading bot needs. And here
19:55the very first rule we need to decide
19:58is, you know, what market and what time
20:02frame we are going to trade.
20:04This right here is very important
20:06because different trading strategies can
20:08perform different, you know, depending
20:09on what market you are trading and what
20:12time frame. So for example, a trading
20:14strategy that works very very well on
20:17the daily time frame might not work at
20:20all if you try to, you know, scalp on
20:23say the one minute time frame or even
20:25the one second time frame. So,
20:28market and time frame is
20:30very important to decide. And in this
20:33course, we're going to use Bitcoin on
20:35the daily time frame. This is one of the
20:36most popular and one of the most
20:39universal markets. And I think the
20:41technical analysis works very well on
20:42Bitcoin. But I do want to be clear here
20:44that you can trade basically anything.
20:46So you can trade, you know, stocks,
20:47forex, crypto, or as I said in the
20:50intro, any other financial market.
20:52The next rule here
20:55that we of course need to decide before
20:57we build our bot is our entry and our
21:00exit. And here we need, you know, exact
21:04conditions or in other words, we need
21:06exact rules to know, you know, when to
21:09buy and when to sell. And this can be
21:12anything from, you know,
21:14normal risk to reward ratios. And it can
21:17also be things like, you know, stop
21:20losses and target levels based on price
21:22action. And in this course, we're going
21:25to actually use the strategy that the
21:27turtle traders used. I really like that
21:30name.
21:30And here we actually have, you know, a
21:32mix between fixed and trailing stop
21:35losses and targets. And you will Don't
21:37worry if you don't know what that is
21:39right now. You will learn all about it.
21:41Um
21:42and here of course super super important
21:45is the stop loss. You know, us traders,
21:48one of the most important things we can
21:51learn about is, you know, risk
21:53management. That is how do we basically
21:56how do we protect the downside? Because
21:58in order to, you know, make money on a
22:01on a consistent basis, one of the most
22:03important thing things is to protect the
22:05downside. Because for example, this
22:08principle is so important to know. But
22:11for example,
22:13if the market drops
22:1550% right,
22:17that actually means that how much do we
22:19need to recover? Well, in order to
22:20recover a drop of 50% we actually need
22:23100%
22:25to recover. And this one of the most
22:27important asymmetries in trading. So
22:30that is, you know, why risk management
22:32is so so important. And in this video,
22:34we don't really have time to dive super
22:36super deep into risk management, but I
22:38do want to mention already that I do
22:40have a full risk management course right
22:43here on YouTube. I'll make sure to link
22:44that course up in the video card as well
22:46as the description. I can't stress
22:49enough how much I recommend that course.
22:51And in that course, we talk about things
22:53like stop losses, but we also
22:56talk about things like position size.
22:58This is another thing.
23:00This is another rule that your bot
23:02needs. So in other words, how much of
23:06the account should you risk on each
23:08trade? You know, it's often very smart
23:11or all pretty much always very smart
23:13that you or it's kind of, you know, says
23:16itself that you should not risk
23:17everything on one trade. You can have a
23:20trading strategy that in theory works,
23:22but if you lose everything on one trade,
23:25then you are completely out of the game.
23:27And that is
23:28probably the most important thing we
23:30want to avoid as traders.
23:34All right. So now, guys, the time has
23:35finally come to take a look at the
23:38turtle traders. And what on earth or who
23:42are the turtle traders? Well,
23:46let's take a quick look at this because
23:48this will be the basis of the trading
23:51strategy and trading bot we will build
23:53later on in this course.
23:55But first of all,
23:56in 1983,
23:59Richard Dennis, which you can see on the
24:01image right here,
24:04actually set out to settle a debate with
24:08another trader called William Eckhardt.
24:11So here is an image of William Eckhardt.
24:15And they basically debated if, you know,
24:18can trading actually be taught? Is this
24:20something you can teach? And of course,
24:22by having this channel, I think that
24:25trading definitely is something you can
24:27learn. I also think that trading is
24:29basically a a mix of things you can
24:31learn, but you also need experience. You
24:33also need, you know, to put in the reps,
24:35of course.
24:37But that that was basically the debate
24:39between Richard and William.
24:41And
24:42what Dennis then did to try to
24:45demonstrate this what that was that he
24:47recruited completely beginners, so noob
24:49traders that he called the Turtles, and
24:52he trained them for only two weeks.
24:55And here the Turtles traded a trend
24:59following system. This simply means that
25:02you trade in the same direction as the
25:04trend. So if you, for example, have an
25:06uptrend, the goal here
25:08is to, you know, trade and make money as
25:11the price goes up. A very common
25:13strategy is to buy pullbacks here, but
25:15you can also buy breakouts. But yeah, so
25:19he did a trend following system
25:21and in this case he actually it was all
25:24about buying breakouts and buying
25:27specific breakouts over specific highs
25:29and then sell the price when we fell
25:32below
25:33you know recent lows. And you know as I
25:36said with when you need when you build
25:37[clears throat] a trading bot you need
25:38specific rules and I will talk all about
25:40that very soon.
25:43And here in this course we are going to
25:44use a simplified version of their rules
25:48as an example bot because my goal with
25:51this course is that you know to show you
25:53guys how you can build basically build
25:55your own trading bot from scratch and
25:58you can also you know
26:00AI is so powerful so you can actually
26:02build your own trading strategies and
26:04you can back test them and you can build
26:06bots around them. but I do want to
26:08emphasize that this example I will show
26:10in this video is a
26:12it's first of all a simplified version
26:14of the strategy and it's also a you know
26:16just an example.
26:18But all right guys so now it's finally
26:20finally time to build our trading bot
26:24and as I said the trading bot we will
26:26build is the I call it the Turtle
26:30breakout 5520
26:33and you will soon understand
26:35the name here and you already know where
26:37turtle comes from. It comes from the
26:38turtle traders. But now let's begin here
26:41by
26:41you know stating the rules for this
26:44trading bot.
26:46So the rules here rule for this bot is
26:48as follows.
26:50The entry we enter or in other words we
26:52buy when the daily close is above the
26:56highest high of the previous 55 days. So
27:01with a daily close I mean that if we
27:03have a candlestick right here the close
27:06of the candle is where the candle stops
27:09on the daily time frame. For example
27:11this right here is the close. This right
27:13here is the open. This right here is the
27:14high. This right here is the low.
27:16In this trading strategy if we get a
27:18close that is higher than as I said the
27:21highest high of the previous 55 days. So
27:24let's say that this right here back then
27:26was the highest high and we close above
27:28this is our signal to buy. This is our
27:31signal to enter the trade.
27:34But here and this is you know at least
27:37as important as the entry. This is our
27:38stop loss.
27:40And here what we are going to do is that
27:42we're going to place a stop loss two
27:45times the ATR which basically stands for
27:48average true range. This is a very
27:50common indicator used for stop losses
27:53but we're going to use the average true
27:54range over 20 days and ATR basically
27:57measures the size of a normal days move
27:59or that is a simply way
28:01to think of think about it but we're
28:03going to do two times the ATR below the
28:06signal days
28:08close. So we're going to do two times
28:10the ATR here below the close. So two
28:13times
28:14the ATR 2x right there. And
28:19also guys don't worry if you don't know
28:21about the ATR I highly recommend
28:22actually to understand the strategy
28:25you're using. It's very important to
28:26understand the pieces and I do have a
28:28video on the ATR indicator. I will make
28:30sure to link it up in both the video
28:33card as well as the description below.
28:36Uh, now we have the entry, we have the
28:38stop loss, but what about the exit?
28:40Well, here we're actually going to sell
28:42when the daily close is below the lowest
28:47low of the previous 20 days. So in this
28:52case the exit of the turtle breakout
28:55strategy is actually like a trailing
28:57exit. It will not be determined on day
28:59one. It will actually follow the price.
29:02Uh, so that is very important.
29:05Last but not least here for position
29:07size, uh, this can be, you know,
29:09different depending on your your sort of
29:12risk tolerance and your overall goal.
29:14But to keep it simple, a very simple
29:16rule is to risk 1% of your account. And
29:20once again, I'm going to do a note here
29:22that this is a simplified version of the
29:25turtles rules. Uh, you know, for
29:27example, in this case we only go long to
29:30keep it more beginner friendly, meaning
29:31that we only bet that the price will go
29:34up.
29:34We use closing prices, we do no
29:37additional trades and so on and so on.
29:39And of course, as I mentioned in the
29:41disclaimer, this is of course, you know,
29:43only an example and not financial
29:46advice.
29:47Okay, so now the time has come to build
29:50an AI trading bot together with Claude.
29:55And we're going to use Claude and
29:57TradingView.
29:58And the process here is pretty
30:00straightforward. What we're going to do
30:02is that we're going to give Claude the
30:05three, uh, I mean the five trading rules
30:08from our trading strategy in very clear
30:11terms. You know, nowadays I would say
30:13that Claude is very good at
30:14understanding you even if you are not
30:16exact. But when you build a trading
30:18strategy, it's of course better to be as
30:21exact as possible when you formulate the
30:23trading strategy.
30:25And then when you give the prompt,
30:26Claude will write the bot as a Pine
30:30Script strategy. If you didn't know,
30:33Pine Script is simply, you know,
30:34TradingView's coding language. So it is
30:37the coding language used to, yeah, build
30:39scripts and indicators and so on and so
30:42on. And it also adds it to your charts.
30:46And Claude is so impressive. It can
30:49actually fix its own errors until the
30:52strategy works. And here a pro tip to
30:55sort of double check it yourself.
30:57As I have been talking about throughout
30:58this course, I always recommend to you
31:00to double check everything. Yes, AI
31:03nowadays is so so powerful, but we are
31:05still in a phase where AI can make
31:08mistakes. It can hallucinate and so on
31:10and so on. So I highly recommend to
31:11actually double check the work.
31:14All right. So now we are back here in
31:16TradingView and I'm going to open up
31:19Claude here. Remember, you need to use
31:20the Claude app.
31:22And once again, just to confirm that
31:24Claude is connected to TradingView, we
31:26can start by just asking,
31:29uh, hello Claude, can you tell me what
31:31chart are we looking at right now?
31:35Uh, and press enter right here. I also
31:37recommend to either use Fable 5.1 or to
31:41use Opus 5.5. And if you use Opus, I
31:44prefer the extra option right here. The
31:46max option is a little bit overkill. As
31:48you can see, it takes
31:50four times or more usage. So it uses
31:53lots and lots of of tokens.
31:56And, you know, I have the max plan, so I
31:58have the the the the sort of
32:00plan with the most tokens. But if you
32:02are not on the max plan, it can quickly
32:04get eaten up.
32:05So I recommend, you know, extra or if
32:07you are tight on tokens, you can even
32:09try high or medium. But if you have
32:10tokens, go for the extra right here.
32:13And as you can see,
32:14it says you're looking at Bitcoin. We
32:16are on the daily time frame. It says,
32:18you know, the last daily candle,
32:20November so far. You know that the price
32:23was last at 78,400. Everything looks
32:26very correct right here.
32:28Uh, but what I'm now going to do is that
32:29I have actually prepared the prompt here
32:32to build the trading strategy.
32:35So as you can see, it's pretty long
32:37right here. Build this trading as a
32:39TradingView Pine Pine Script blah blah
32:41blah blah blah. Of course, for this
32:42step, you can build whatever trading
32:44strategy you want. This is just an
32:46example for this course.
32:48And what I can also mention if is that
32:51if you want this exact prompt and
32:53basically everything you need for this
32:55whole course, I have a like a study
32:57guide that you can download. It's a PDF
32:59guide. It's currently available for all,
33:02you know, our channel members gold and
33:04higher. So if you didn't know, I
33:06actually we actually have a community
33:08right here on YouTube where members, you
33:11know, get access to things like, you
33:13know, this study guide. You get access
33:15to
33:16all the guides for my other courses as
33:18well. You get access to my own
33:19TradingView indicators and so on and so
33:21on while at the same time you support my
33:23work. So if you want to become a channel
33:25member, just click the join button next
33:28to the subscribe button and a short
33:29video will pop up where I explain the
33:31different memberships. But
33:34you know, you don't need to become a
33:35member. All you need to do here is to
33:37follow along. So now I'm going to click
33:40enter right here
33:41and now
33:43as you can see Claude will work here to
33:46build this indicator.
33:49I mean this trading strategy. This can
33:52take a little while, but
33:54as I said, I much rather it take, you
33:57know, a minute or two and that Claude
33:59actually works to build the strategy
34:00correctly rather than, you know, rushing
34:02it and getting it wrong. So now I will
34:05just let Claude cook for a while and I
34:08will be back when the strategy is
34:10complete.
34:12Okay, so now after thinking for like two
34:14three minutes, we got this back and here
34:17what they actually wrote in the prompt
34:19is that
34:20before he will actually build the
34:23indicator,
34:24I wrote that he should explain
34:27everything in, you know, plain English.
34:29So it says, you know, what chart Bitcoin
34:31direction long only, you know, exactly
34:33when to buy, exactly stop loss, what
34:35when to sell, the size. So everything is
34:38listed here. So I actually recommend to
34:40read through your whole strategy. You
34:41see that everything is correct. You see
34:44the assumptions. You see, you know, some
34:46notes about the stop loss, position
34:48size, and so on and so on. And, you
34:50know, it says reply go or tell me what
34:53to change and I will build it. So
34:56I didn't read through all of this, but
34:58this is just a demonstration. So I'm
34:59going to say, you know,
35:01uh uh let's
35:02let's build the bot go.
35:06And now
35:07Claude will start actually building the
35:10bot using the Pine Script coding
35:13language.
35:15I can also just mention here that it's
35:17kind of cool to let, you know, or to
35:19look at, you know, Claude working here
35:22in
35:23you know, live. So we can see here that
35:25it has already started to build the
35:27script right here. And you can see that,
35:30you know, Claude is working right now.
35:31You can see everything it has done here,
35:34how it like failed a task and then found
35:37a solution. Right now it's working.
35:39Yeah, you can see it's working really
35:41hard with different things.
35:43You know, I'm not going to say that I'm
35:46understanding anything of this, but
35:48yeah.
35:49Like sometimes, especially if you're
35:51using here incognito mode, you will need
35:54to click allow.
35:56There are some like checks here that you
35:58need to click allow or there is usually,
36:01you know, if you don't use incognito,
36:04you can choose here to, you know, always
36:06allow
36:07Claude, you know, everything. But
36:10you might need to be able to be at the
36:12computer to allow certain tasks.
36:15But as with everything, you know, before
36:17allowing things, you need to, you know,
36:19understand what you are allowing. And
36:21now if we take a look here, we can
36:23actually start to see the trading
36:24strategy appearing here on the chart.
36:26Really cool here. We see buy signals, we
36:28see sell signals, buy, sell, buy, sell,
36:31and so on and so on.
36:33Okay, guys. So now the the bot has been
36:37built right here.
36:39So let's actually close down Claude and
36:41we can see just like that the whole
36:44trading strategy is built uh right here
36:47on TradingView. And the script we can
36:50see down here where it says uh
36:52you can't really see because my camera
36:54is a little bit on the way, but you know
36:56Claude built all of this code right here
36:59beautifully beautifully.
37:01And what we can also see is that down
37:05here
37:06we actually have a backtest of the
37:08trading strategy. And we will talk more
37:11about this backtest later, but you can
37:13see here that this trading strategy made
37:15around 700% in profit. You can see here
37:18the max drawdown. And one interesting
37:21things you can also also already notice
37:23here is that the amount of profitable
37:25trades was actually only 45% of the
37:28time.
37:29But what this means is that you know yes
37:32we lost more than we won, but when we
37:35actually won the trades
37:37we won a lot more. So for example this
37:40trade right here you can see that we won
37:41massively.
37:43Uh
37:43so yeah, but as I said we will talk more
37:45about the backtest later on because now
37:48I want to switch the attention and take
37:49a look at what do we need to check
37:52before we start trusting this trading
37:55strategy.
37:56But so now the time has come for step
37:59number two right here and that is to
38:01actually check the bot by hand. And the
38:04reason we want to check you know at
38:06least one trade but preferably more uh
38:09is to just you know confirm that the AI
38:11trading bot actually did what we asked
38:13it to do. Uh as I mentioned already yes
38:18today's AI are super super powerful and
38:20I would say it's better than 99.9%
38:23of coders and it's better than most
38:25people at most things, but it can still
38:28make mistakes. So it's very important to
38:31still double check your work especially
38:33when it comes to things like you know
38:34trading and so on and so on.
38:36So
38:38here what I recommend is to check three
38:40trades on the chart. So double check you
38:43know the entry is the entry correct
38:45according to the rule we gave it.
38:48Also, make sure to check the stop loss.
38:49Is the stop loss correct according to to
38:51the rules? And also check the exit. Is
38:54the exit correct according to the rules?
38:56And in our example,
38:59what we need to look for here is first
39:01of all that the breakout compares the
39:04close with the previous 55 candles. So
39:08remember, our entry signal is when the
39:10price breaks above the highest point of
39:13the last 55 candles.
39:16The second thing we need to check here
39:18is that the stop loss is is set once
39:21only at entry. So we don't want to see
39:24our stop loss move. In our case, in our
39:27bot, the stop loss would be an orange
39:29line like this. And if the orange line
39:31moves, we know that something is wrong.
39:33The orange line should be fixed at a
39:35certain level.
39:36The next thing you can check right here
39:38is that the position size is indeed 1%
39:41of the account
39:44divided by the distance to the stop
39:48loss.
39:49Okay, so now we're back here in
39:51TradingView and I can once again mention
39:53if you don't have TradingView yet,
39:55hopefully at this point you have
39:57TradingView Premium so you can follow
39:58along with everything. But if you don't
40:00have TradingView Premium yet, I do have
40:02a special link. It will be both in the
40:04pinned comment and description. Make
40:05sure to use that link if you haven't
40:07taken advantage of it yet.
40:10But as I said on the slide, I recommend
40:12that you check three trades. But for
40:13this course, let's just double check one
40:15trade here. The first thing we want to
40:17check is the buy signal. So as you can
40:21see
40:22right here, we have a buy signal. And
40:25remember, the buy signal happens when we
40:26close above, you know, the highest point
40:29of the last 55 days. And this is
40:31represented by the I hope you can see
40:35the green line right here.
40:38So if we extend this green line, we
40:40should hopefully see, you know, the
40:42point we're talking about. And as you
40:44hopefully can see right here, this was
40:46the highest point of all of these
40:48candles.
40:49Uh, so that is why we we get the entry
40:53signal when the candle closes. And
40:55remember, it closes up here. So when it
40:57closes above, that is when we get the
40:59entry signal. And remember, the second
41:01thing we want to check here, which is
41:03very important, is the stop loss level,
41:05which is set two times the ATR below the
41:10candle close. So so this distance right
41:13here is two times the ATR. And what is
41:15very important to check about this stop
41:16loss is that it doesn't move. You can
41:18see the line here is straight. Compare
41:20that to the red line.
41:23You can see the red line is constantly
41:24moving. The green line is also
41:26constantly moving, but the uh the orange
41:30line should be straight. So this is also
41:32correct. So now the entry is correct,
41:34the stop loss is correct.
41:36The last thing we need to check here is
41:37the exit.
41:39And you can see here that the sell
41:40signal came all the way up here.
41:44[snorts]
41:45And the reason we sold was that the
41:46price dropped below this low right here.
41:49You can see this red line right here. If
41:50we extend it,
41:52we can see I think the low is actually
41:54from this low right here. So this was
41:55the lowest point of the last 20 candles,
41:58and that is why we automatically sell
42:00when the candle closes below right here.
42:03So just by looking at this trade, the
42:05bot looks completely correct. But as I
42:08said, what I recommend is to check
42:10maybe, you know, at least two or three.
42:12Uh, but that would be pretty boring in
42:14this course, but you just do the the
42:15same method. Another reason why I think
42:18it's, you know, so important to check
42:19your bot is because that is like a test
42:22if you actually understand the strategy.
42:25So the rules you have given it, it's
42:26important that you understand the rules
42:28because if you don't understand the
42:29rules, you will not really know if the
42:31bot did anything uh you know, anything
42:34wrong, if that makes sense.
42:36All right, so now the time has come to a
42:38very important step when it comes to
42:41using and trading with AI bots. And that
42:44is of course to back test the bot. And
42:47luckily, we have actually already done a
42:50back test, but it's very important to be
42:51able to understand the backtest. So
42:53let's take a look at what a backtest is
42:55and why it's so important and what you
42:56need to basically have in mind. So in
42:59very simple terms, a backtest runs the
43:02bot on past data. So it looks at, you
43:05know, the chart and the past candles and
43:08see how the rules or how the training
43:11strategy would have performed in the
43:13past. So it says, you know, how would
43:15this strategy have performed, you know,
43:17in on the last in this case of Bitcoin,
43:20maybe the last 10 years.
43:23And
43:24to make it realistic, what we want to do
43:26is that we want to add commissions and
43:28slippage. So, you know, when you
43:31backtest something, it can often look a
43:33little bit better than what it costs in
43:36real term because every time you buy or
43:38sell something, you need to pay a little
43:41bit to the bank. You need to pay a
43:42little bit to your broker in
43:43commissions. And another reality of
43:46market is something known as slippage.
43:49That is basically, you know, when you
43:50place a buy order, especially if the
43:52market is not very liquid, if you try to
43:54buy something, you might not get the
43:56exact price you were looking for. Uh you
43:59might have to pay a little bit more and
44:00that is also a reality of trading and
44:02something we, you know, have to account
44:04for.
44:06Another thing that is good to do with
44:08backtest is to actually compare your
44:11result with a buy and hold strategy. And
44:15this can actually be a very sort of nice
44:18aha moment, you know, especially for
44:20Bitcoin. Bitcoin has performed very very
44:23well by just holding Bitcoin. To be
44:25honest, it has probably outperformed
44:27pretty much every trading strategy. You
44:29know, if you held Bitcoin from, you
44:31know, 2013 or uh you know, from even
44:35more recent, you would have made lots
44:38and lots of money. So it's important to,
44:41you know, compare, you know, how would I
44:43have performed by just holding the
44:45assets versus using the AI trading bot.
44:49Another thing you can do is that you can
44:50test the bot on other time frames. So
44:53for example, look at, you know, how
44:54would the
44:55trading strategy perform on the five
44:57minute chart instead of the daily or on
44:59the one hour chart instead of the weekly
45:03and so on and so on.
45:05Another thing, a quick note here is to
45:07just of course understand that all
45:08trades are simulated. So TradingView you
45:12know creates sort of simulated orders.
45:15This is not you know, but the backtest
45:16is not something that has actually
45:18happened. And in real markets, if you
45:22know,
45:23it doesn't really matter if you're a
45:24small trader, but when you start to have
45:26lots and lots of money, you can actually
45:28affect the market itself. And that is
45:30something to have a little bit in mind,
45:32but if you trade something like Bitcoin
45:34and don't have a lot of money, you don't
45:36affect the market much, but you know,
45:38it's still something that I think is
45:40worth mentioning.
45:42All right. So now then, now let's take a
45:44look at how to read the results of a
45:47backtest and there are a few important
45:50things we need to learn about. The first
45:52thing very simple is win rate. This is
45:54the percentage of trades that made
45:58money. And
46:00as I have mentioned earlier, you don't
46:03have to win more than 50% of trades to
46:05make money. You can actually win let's
46:07say 25% of the trades
46:10to to and still make money because it
46:13all depends on your average, how much
46:17you win versus your average, how much
46:19you lose. So how big are the winners
46:22compared to the losers? If your winners
46:24are much bigger than your losers, then
46:26you can afford to win less and still
46:29make lots of money.
46:32Another very important thing you need to
46:34learn about is what is known as max
46:36drawdown. And this is simply the biggest
46:39drop in the account from the peak. So
46:43how much did your account totally drop
46:46and check that it counts the open trades
46:49too.
46:50Once again, I do want to emphasize here
46:53and I want you guys to remember that a
46:54backtest shows the past and it's not a
46:57promise of the future. So even if you
46:59have a backtest that has performed, you
47:01know, very well in the past, it doesn'
47:03mean that it will perform very well in
47:05the future. Yes, it makes it more likely
47:08that this strategy will perform well,
47:10but it's no guarantee. You know,
47:12anything can happen. Let's say you built
47:13a trading strategy that performed very
47:16well on the stock market before, you
47:18know, the COVID crash, and then the
47:20COVID crash happens, and that is a
47:21completely new market environment where
47:24your your strategy might get completely
47:27obliterated. Black swans happen. We need
47:29to understand that, you know, uh yes,
47:32backtesting is very good and important,
47:34but it's no promise. It's definitely no
47:37promise.
47:38All right, so now let's jump back here
47:40into TradingView, and you should be able
47:42to see your backtest should open up down
47:45here. So you just press down here, and
47:47then you drag it out. And here are some
47:51key sort of stats. First of all, you can
47:53see that our trading strategy made uh 70
47:57uh
47:58729% in profit, or in other words, uh
48:01$72,000.
48:04You can see here that our max drawdown
48:05was 3.29%.
48:08And this is not the max from the highest
48:10to lowest point. This is only of closed
48:12trades, so it's important to have that
48:14in mind.
48:15The other thing that is very important
48:16is that as you can see, our win rate was
48:19only 45%, but despite having 45% win
48:22rate, we still made lots of money. And
48:25that's because we had a very good profit
48:27factor. We had a profit factor of 7.6.
48:32And a profit factor of 7.6 basically
48:35means that for every dollar the bot
48:37lost, we actually made $7.6. So so we
48:40had a very good profit factor. And of
48:43course, right here we can see, you know,
48:45the graph. You can see how the bot has
48:48performed. It's important to notice
48:49that, you know, most of the gains here
48:52or lots of the game gains come from a
48:55few very like like uh
48:58very good trades. So we have a few
49:00trades that really, really makes up most
49:03most of the profits. And then we have,
49:04you know, a few like slow and steady
49:07trades right here.
49:09Um, what can we see more? You can see
49:12here the gross profit, the gross loss,
49:15commission load. You can go down here
49:17and take a look at the distribution. So
49:20you can see, you know, losers, winners,
49:22and how much you made on these.
49:25What is also very important, you can, if
49:27you click right here, you can see how
49:29would your trading strategy have
49:31compared with a buy and hold strategy.
49:34And if we click show right here, you can
49:35see that we get absolutely obliterated
49:39by Bitcoin. So what this means is that
49:41just holding Bitcoin would have beaten
49:44this trading strategy by a lot. So I
49:47have to be honest with that. You know,
49:49even though this is a decent return,
49:51Bitcoin completely beats that. And
49:54what I also have to say is that, you
49:56know, this is probably true for pretty
49:59much all trading strategies. It's very
50:01hard to beat the performance of Bitcoin
50:04from back in, you know, 20 2013.
50:09That is an insane gain from Bitcoin.
50:12All right. So now the time has finally
50:14finally come to, you know, turning our
50:17AI trading bot on. And what we want to
50:21do here is that we want the bot to, you
50:23know, notify us and give us a signal
50:26every time it finds a trade. And to add
50:31a layer of security to this, this is,
50:34you know, especially important if you
50:35are new to building AI trading bots. And
50:38if you're watching this video, I assume
50:40that you are, you know, you know, pretty
50:42new to this game.
50:44The important part here is that the bot
50:46will give you a signal, but it's you who
50:49actually click buy or sell. Of course,
50:52you can automate it this completely so
50:54the bot automatically buy and sell. But
50:57I think that especially in the
50:58beginning, it's super important that
51:00it's actually you that, you know double
51:02check and click buy or sell.
51:05So, what we're going to do here is that
51:07we are going to have it like this. Um, a
51:10strategy alert will fire every time the
51:14bot enters or exits a trade.
51:18So,
51:19we are going to have first of all, we
51:21are going to make you know we're going
51:23to have Claude create the alert and the
51:26whole alert system for you through the
51:29MCP. The MCP is simply uh the connection
51:32here between um TradingView and Claude
51:35that we have been using throughout this
51:37whole course.
51:38And
51:39here the alerts uh run on TradingView's
51:43servers and uh will can reach your
51:46phone, your email. It will uh
51:49you know reach your TradingView app and
51:51so on and so on. And it's very important
51:53that you don't need to have TradingView
51:55open. You will get these alerts from the
51:57bot uh you know if you're out and about
52:00you know you just have your phone with
52:02you and you will always
52:04get these alerts.
52:06And
52:07it's very important here that if you
52:08make any changes to the strategy you
52:11need to
52:12you might need to recreate the alerts.
52:15So, here on this image you can see a
52:17simplified sort of explanation of what
52:19we do. First we will have Claude create
52:21the alert system here and then when we
52:24have the alerts you can have them sent
52:26to your phone, to your email. The alerts
52:29will run as I said when TradingView is
52:31closed and so on and so on. But now
52:34let's jump back into TradingView here
52:36and take a look at how we can actually
52:37create this alert system and turn the
52:40bot on.
52:42Okay, so now we're back here in
52:43TradingView and I have the you know
52:45Claude window open right here. So, now
52:48let's create the alert system and one of
52:49my favorite way to actually prompt
52:52Claude is to use you know voice. I think
52:54that is much it's much faster and more
52:57efficient than writing. Um,
52:59so let's try to you know just talk to
53:02Claude here and create the alert system.
53:05So, all right. So,
53:08hello Claude. What I want you to do now
53:10is that I want you to create an alert
53:13system for the current trading strategy,
53:16the turtle breakout 5520
53:20and I want you to
53:22you know, fire an alert every time the
53:25bot buys or sells. I want you to use the
53:28bot's own alert messages and I want you
53:32to send the alerts to both email and the
53:36app notifications. Make the alerts last,
53:39you know, as long as possible and don't
53:42touch my other alerts. And when you are
53:45done, I want you to tell me the settings
53:49you're use you used. So, yeah, that's
53:52it.
53:54So now, as you can see, all of the text
53:57appears right here. I will simply press
53:58enter and let Claude do its thing.
54:02All right. So now, Claude is done
54:05working and as you can see, just as we
54:07asked, it shows all settings right here
54:10in the chat.
54:12If we take this chat down, you can also
54:14see that the alert is created right
54:17here. It should already be saved. So you
54:21don't have you should not have to click
54:23save again.
54:25But now, to find the alerts, you want to
54:26go up here to the upper right corner
54:28where it says alerts right here.
54:30And as you [clears throat] can see, your
54:31alert should be
54:33right here. You can double click on it
54:36and you can see here that the alert
54:39should be active. You can also see here
54:41on notifications
54:43that we asked it to send notifications
54:47to both app and email.
54:49And if you're wondering, you know, what
54:50email does it go to, I'm pretty sure it
54:52automatically goes to the email you used
54:55when you created your TradingView
54:57account. So now,
54:59you basically have a bot that will
55:02automatically
55:03send you notifications every time you
55:05need to buy or sell. So, this right here
55:08is uh you know uh
55:09yeah, this right here is now a bot. Uh
55:13the only it's not fully automated
55:15because it's still you that are taking
55:18the final action of buying and selling,
55:20but as I mentioned earlier, I think this
55:21is a very good sort of uh safety step in
55:24this process, especially if you are new
55:27to uh trading bots.
55:29All right, so now it's time to take a
55:31look at you know some very important
55:33things related to risk and oversight for
55:37your AI trading bot. So, let's take a
55:39quick look at the oversight rules uh
55:43oversight rules that uh basically make
55:45sure that your bot behaves well as you
55:48can see right here. And step number one,
55:51as you can see on this image, is that
55:53you are responsible.
55:56No one else you know at the end of the
55:58day, no one else is responsible for your
56:01decisions, not the bot, not the
56:04president, not me, not anyone else. At
56:07the end of the day, you are the one that
56:09decides to buy or sell, and that is why
56:11I have you know built in into this
56:13course
56:14you know the final step that you know
56:17the bot never actually buys and sells
56:18for you. You are always the one that
56:20presses uh buy or sell. So, this is a
56:22very important like safety measure. You
56:25are responsible.
56:26The second thing that is very important
56:28is that you know you should you know
56:30Claude should never get access, direct
56:33access to your broker login, your API
56:37keys, or you know the order button order
56:40button.
56:42So, you know Claude already definitely
56:44has the you know power and capacity to
56:47be to fully automate your trading. It
56:49can place buy and sell orders while you
56:52are asleep.
56:54But as I said, if you're taking that
56:56step, you need to make it really sure
56:57that first of all that your strategy
56:59works, and secondly, as I said, you know
57:02Claude is very very
57:04is very very reliable nowadays, but it
57:07still makes mistake. So, if you are
57:09fully automating using Claude, you need
57:11to you know, have into your calculation
57:13that Claude can make, you know, a
57:15complete mistakes. There's no guarantee
57:17that Claude doesn't make mistakes. So, I
57:20highly highly recommend to not give
57:22Claude access direct access to
57:25you know, you know, anywhere where it
57:27can actually buy or sell.
57:29So, that is the important step number
57:33two right here.
57:35The next thing is that you know,
57:37definitely make sure to you know, pause
57:39the bot if its drawdown gets bigger than
57:43the
57:43backtest worst drawdown because that is
57:46a that is a you know, indication that
57:48you know, maybe we did something wrong
57:50with the backtest or maybe you know, the
57:54you know, this strategy performed well
57:55in the past but maybe something have
57:57changed about the market. You know,
57:59maybe the market is much volatile
58:02nowadays.
58:04And that makes the strategy, you know,
58:05not really work as intended. So, I would
58:08recommend that if it gets especially if
58:11the backtest if it gets, you know, much
58:13worse than the backtest worst drawdown,
58:15that is definitely a warning sign that
58:17we should, you know, pause the bot and
58:19reassess it.
58:20Um
58:22and another thing that you want to do is
58:24that you want to renew the alerts before
58:26they expire. So, on standard plans after
58:30around two month months
58:32the alerts expire. So, make sure to
58:35review them.
58:38And as I said, to make this workflow,
58:39you know, work as well as possible, I do
58:41recommend you know, trading the premium
58:44or even higher.
58:46Uh and as I said, I do have a special
58:48link for trading the premium. It will be
58:51in the description as well as the pinned
58:53comment. But all right guys. So, this
58:56course is actually, as you can see, it's
58:58getting pretty long but we have covered
59:01everything. If you made it all the way
59:04till the end here, I first of all want
59:06to you know, thank you all so much for
59:09watching.
59:10By now, you actually know, which is you
59:13know, pretty impressive. You know how to
59:15build a trading bot from start to
59:18finish. No coding. You know how to build
59:21a trading bot. I hope that is amazing.
59:24If you are you know happy with the
59:25course and feel like you learned
59:27anything new, the absolutely best way to
59:30you know support me, support the channel
59:32and show your appreciation is to drop a
59:35like.
59:37Another thing I want to mention is that
59:39you know
59:40as I think I mentioned earlier in the
59:42course,
59:44I have made a complete study guide with
59:47all the prompts from this course so you
59:49can follow along exactly and lots of
59:51bonus material and so on and so on. I
59:54have made that study guide. It's a PDF
59:56file. It's available for all our YouTube
59:59channel members, gold and higher. If you
1:00:02didn't know, we have an amazing
1:00:03community right here on YouTube where I
1:00:05share you know bonus material like that
1:00:08guide. I also have things like you know
1:00:10my own created TradingView indicators. I
1:00:13have exclusive videos and much much
1:00:16more. So if you did enjoy this video and
1:00:17want to get more and support the channel
1:00:20at the same time, please consider
1:00:22becoming a channel member by clicking
1:00:24join down below.
1:00:26After clicking join, I have a little
1:00:27video where I explain the membership. So
1:00:29if you want to do that, that would be
1:00:30awesome. And also guys, if you yeah if
1:00:33you made it all the way here, please
1:00:35comment down below and write full
1:00:38course. That is a good way for me to
1:00:40understand that okay this is someone
1:00:42that actually watched the full course.
1:00:43I'm very I'm much more likely to answer
1:00:46comments from you guys.
1:00:49And it's also a good way to kind of
1:00:51confuse the rest of the audience a
1:00:53little bit. They tend to you know be a
1:00:55little bit confused when there's so many
1:00:56full course comments. So that is
1:00:59awesome. And of course, if you have any
1:01:00questions or you know feedback, let me
1:01:02know down in the comments below.
1:01:06But yeah guys, I hope you enjoyed this
1:01:07course and for now take care.
1:01:10All right, so now when you know how to
1:01:12build an AI trading bot, you might be
1:01:14interested in how you can use this cloud
1:01:17and TradingView connection to do other
1:01:20kinds of technical analysis. To do that,
1:01:23I highly recommend to watch my full
1:01:25course on how to use TradingView and
1:01:27cloud together.

About this transcript

This page contains the full transcript of MASTER AI Trading Bots in 61 Minutes (Claude + TradingView) by Mind Math Money, generated from the public captions YouTube serves with the video. The transcript has 10,848 words across 1,582 segments, with the original timestamps preserved so you can click any line to jump to that moment in the embedded player.

What you can do with it

Use the transcript to take notes, quote the speaker, build a study guide, generate a summary with ChatGPT or Claude via the YouTube Summary tool, or export it as a timed subtitle file with YouTube to SRT. You can also re-open it in the transcriber to choose a caption track available on the video.

Free YouTube transcript tool

YouTube2Text is a free YouTube transcript generator — no signup, no daily limit. Paste any YouTube link and get the full transcript instantly, with timestamps, click-to-jump, caption language selection, AI prompts for ChatGPT, Claude, and Gemini, and exports to TXT, SRT, VTT, or Markdown.