Live Trading? Build the Confidence to Take Your Trading Strategy Live!

Thank you.

Thanks, Dennis. So how many of you, just so I can understand the room a little bit, how many of you are actively trading right now? Just by show of hands? Okay, great, thank you. And how many people are yet to start trading? Okay, great, thank you. And how many of you trade systematically? So you have defined rules that are objective that you follow each day? Show up hand. Okay, great. And how many of you are discretionary technical? So you look at charts, make judgment calls. Okay, great. Fantastic. Thank you. And we have some options traders and presumably some long term investors in the room. Is that fair? Okay, great, thanks. So, as Dennis said, my name's Adrian Reed and today I'm going to be talking about how to build go live confidence in your trading system. Now, I use the word trading system because I'm a systematic trader. You may use these concepts or some of them at least for your discretionary technical trading. But in any case, take what I've got as ideas to build confidence so you can trade consistently and make money. Is that okay? Sound good? Okay, great. So before I get started though, I just wanted to take a moment and just thank each and every one of you because I know it's a Saturday and I know you've taken time out to be here to learn and grow and develop your trading skills. So I want to let you know that I don't take that lightly. It's a big commitment and I want to do my very best to make this next couple hours absolutely worth your while. I want this to be quite an interactive session, so if you have questions, please feel free to raise your hands. I may take it on the spot or I may go a little further and come back to your question, if that's okay. Just depending on where I'm up to.

Okay. So since we're going to be spending a little bit of time together, would it be okay if I just spend a couple of minutes and just tell you a little bit about myself? Would that be okay? Yeah. Okay, thanks. So my name is Adrian Reid and I have been trading now for.

About 15 years.

I've been first. I was first interested in the markets when I was about 8 or 9 years old when my parents brought home a board game called the stock market game, called the Share Market Game. And I don't think you can get a game anymore, but you've got the game. And look, all I can, I can remember a lot. I actually still got the game. My mother found it in the attic under a pile of dust. And I have really great memories of this game because I just remember becoming a billionaire really, really quickly. And it was amazing because I bought stocks at the bottom of the cycle, I sold them at the top of the cycle and between times I got 2 for 1 splits, 3 for 1 splits, 4 for 1 splits and lots of dividends. And it was pretty cool, right? And so When I was 8 or 9, I remember thinking that, wow, the stock market is. That's where it's at, that's where you make the money. And so that was my first exposure to the stock market. And I hope your experience with the game was similar. I haven't actually used.

So look, if you have an opportunity to play it, do it. It's fun. It won't teach you many great lessons, but it'll at least get some interest. And as a kid, that's what first got me interested in the stock market. Fast forward quite a few years now. So I'm just over 40 years old, I live in Sydney, I trade full time and I've been trading full time since 2012. Full time is a gross exaggeration. I don't actually spend very much time on my trading because it's purely systematic. And what that means is I have rules that I follow every single day and they're codified. So as Dennis said, I use amibroker, I press the button, it tells me the trades to take today and it tells me the position size. I take the trades and then I close my computer and that's it. So from a systematic point of view, that's how my trading runs. And I spend about 30 minutes a day doing that. Yes. Are they long term trades? Yeah, good question. Thank you. I trade very quite long term. So long term trend following is where most of my money is made. I have some medium long term, sort of more swing trade type systems and I trade the Australian market and the Hong Kong stock market. Ordinary shares. I've also traded in the past, US Stocks, Shanghai, several other markets, currencies and so on. But most of my success and what I like the most is, is ordinary shares. I trade Australia and Hong Kong because the lifestyle and time zone is very convenient from where we are and so it works really well. I did trade US stocks for a while, but I really hate staying up late at night to place trades, so I don't do that anymore. Yes. Is the Aussie market not sufficient that you need to have the Hong Kong market as well? Really great question.

I'm just thinking whether I'm going to get into it. Look, the Aussie market is sufficient. I had one system on just the Australian market for six or seven years and made very good returns. I took my account from $7,500 where I started to about $350,000 $400,000 over six or seven years with just one system on just the Australian market. So yes, it's sufficient if that's what you want to do. But I found to take it to the next level and to diversify. As my position sizes grew, I wanted more, more markets and more diversity. One of the things I found as I got into system development more deeply is that the diversification in multiple systems, multiple markets really has a dramatic impact on performance. So much smoother, better risk adjusted performance. Does that answer your question? Thank you. So full time trading since 2012. I was working for Suncorp in a strategy and finance job in 2012. I was working about 14, 15 hours a day on average, stressed out of my mind, traveling, writing tenders. And it was just the worst corporate experience. I mean, I had fired a good team and it was, you know, a good job, but from a lifestyle perspective, it was disastrous. From a relationship with my kids perspective, it was disastrous. And in 2012, my training systems made more money than my job after bonuses and all of that. And clearly that was a sign for me. I was like, okay, it's enough. And so we had the, you know, the.

Conversation at home about, okay, we're going to change something. And so I left the corporate world and just focused on my trading, developed some more systems. And then in 2013, after having so many people ask me how do I trade, what do you do? How did you get there? I decided to start teaching. Yes. Did you find any hiccups going from just trading whilst you were working to being full time trading? Like, did you find you overanalyzed or did having more time create any problems for your trading? That's all. Thank you. Really, really great question. I'll cover it briefly. It's not what I'm going to get into the presentation, but we can absolutely talk about it and I might even do a blog post on that and share it with you because I think it's really valuable. When I was working full time, I was following my system and I designed it so I could follow my system while I was working full time. So it took very little time. It was once a day. I could do my analysis anytime between when the market closed and the daily data became available in the next morning. And I would just jump out of a meeting at 10 o', clock, place my trade, go back into the meeting and keep Working. So that was my routine and that worked fine. I didn't make too many mistakes, a few, but it was enough to profit. And when I left my job and was at home, I had.

A lot of soul searching to think about how I would trade from then on. And I realized I tried day trading quite a few years earlier and I tried it when I had a slow period at work and gentleman in the front row shaking his head and I know why I share your experience.

I was taking about 10 trades a day for a period of three months and I had nothing to do at work because I was off project. I was a strategy consultant at the time, between jobs. So I was taking these trades following this day trading system. And.

What I realized was I was so busy and so stressed following this system that didn't suit me because I don't like making quick decisions, I don't like having constant pressure. And.

If you've ever looked at a five minute chart, you know what I mean? It's just harrowing on the brain for me. And so after three months I realized that I hadn't stopped and taken stock of where I was at and how I was feeling about it and how it was working for me. So I stopped. And at this time I was trading my long term stuff as well. So I had the day trading thing because I thought everyone, you know, everyone talks about day trading, I should try it because I'll make a ton of money. And I was doing my long term trading and after three months I stopped. This is ridiculous. Let me see what's going on. And so I analyzed my performance and this is my experience. Your experience may be different, but I realized that in three months I'd lost $30,000 day trading. I was still up overall because I'd made more than that in my long term system. But I had added so much stress to my life and I added so much uncertainty and anxiety and I'd lost a ton of money. And so I realized day trading wasn't for me. And look, if you take nothing else away from this, I'm going to talk about some deeper content as we move into it. But the most important lesson I found on my journey, going from hobby to profession and full time income, is that you have to have a method that suits you. And so back to the original question. When I left work and I started trading full time, how did I change my approach? I didn't, not at all, because it suited me.

So how many of you feel that you've really got a method that suits every aspect of your personality right now? How many of you are there? Okay, good. A few. And how many think you might have a little way to go to get that? Okay, great. And so the lesson there for me anyway was like, you absolutely have to, have to, have to. Because I found when I had went anything that didn't suit my personality or my objectives or my lifestyle, guess what happens?

Make mistakes, lose money, have stress. And I traded. You know, I was trading for freedom. I don't know about you guys, but I started trading because I wanted to be free. I didn't want to have to work. And so when I left my job, I thought, why do I want to have another job? I don't want a trading job. I just wanted to trade to keep my freedom up. So I kept following my same system. Does that help? I think following your rules, your approach that suit your personality is the most important aspect of all of this, because that's how every one of us becomes successful. But every one of us will have a slightly different approach. Make sense?

Okay, so today we're going to talk about.

How to build go live confidence. Now, some of you are already trading, some of you are learning, some of you have traded before. And maybe on a pause right now, that's okay. And whether you're systematic or discretionary, when you start, you need to have this level of confidence so that you can follow your rules. Because if you've got fear, if you've got anxiety, if you're not sure if you're uncertain or if you're stressed about it, what's gonna happen as soon as something goes wrong?

Thank you. So as soon as. And the market is there, let's face it, the market is there to cause stress, to take your money, to give it to someone else, Unless you're the someone else that is getting the money. I mean, that's my view of the world. It's, you know, the people with the edge, the people with the approach and the discipline are the ones that get the returns. And the people that are not sure, that are stressed, that are driven by emotion, that don't have this.

Unfortunately, are the ones that will tend to lose money.

So who wants to be in the confident go live group? Okay, great. Anyone not? Okay, good. Okay.

So what drives it all?

Try that again.

What's the word?

So at the core of it.

When we start trading, whether it's following some technical rules or following a system or investing for the long term to get dividends.

Any false start, any mistakes, any.

Uncertainty about actually launching into it and committing our Capital is usually driven by some sort of fear. Fear of something. So help me out a little bit. What are the sorts of things that we fear when we're trading? Loss. Loss. Good.

What else? Being wrong. Being wrong.

Sorry.

The negatives.

Missing out. Awesome.

What else?

Time. Wasted time. Great.

How many of you ever sat on the sidelines and watched a big move go up, feeling like you should have gone in and you wasted the opportunity? Yeah. Thank you. Anything else? The unknown. Fantastic. Really good.

So all of these things prevent us from taking the bold action, taking a step, launching into it, that will allow us to ultimately profit consistently.

So what do you think we have to do to get past all of this and.

Eliminate the fear?

System. Good. Confidence.

What else?

Have a system, gain confidence. Anything else?

Discipline. I heard knowledge as well. Thank you. That's great.

So for me, one of the big things, because I trade systematically, I have a set of rules. Now, that set of rules might have come from a book. Well, probably not this book, that's a psychology book. But.

It could have come from a website, it could have come from a teacher, could have come from me. Or you could have made it up. But the key to making getting past the fear.

Initially at least, is to gain confidence in the system or the rules you're following. Again, whether it's a systematic approach or whether it's indicators on a chart that you're using a discretionary manner, it doesn't matter. You with me? If we've got that confidence, then we can take action, we can launch and we can get into our new strategy.

So what do you need in order to get that confidence? If you're going to have a set of trading rules that you're going to follow day after day, week after week, month after month, draw down after drawdown.

Loss after loss.

What do you need to have? What do you need to be sure of?

Belief in the system. And what brings belief.

Back? Testing. Outstanding. What else?

Deep pockets could be. It depends on how big the drawdown of the system is. Absolutely. How many people have had a drawdown?

A couple of people haven't really.

I've had my share as well. I can give it some back. Testing, deep pockets. Got to weather the drawdown so you can come out the other side. But in order to do that again, you need the confidence in the system. And it comes back to this. For me it's, do I have an edge?

Who can tell me what an edge is? What does that mean? Anyone?

Sorry, I can't hear it back. An advantage. Okay, good. I heard something else. Here. A system. Yeah. So if you, I mean, if you, if you read Van's books, you know he'll talk about expectancy, right? Positive expectancy. So if you risk a dollar many times following your system or your rules, you expect on average to make more than a dollar back. So you expect to have profitability. So I want to ask you this.

How do you know if you've got a real edge when you're backtested or whether it's just some data mining fluke and it looks profitable in the past, but it's not going to work in the future? So that's a key question because when I started trading with the system, I wasn't sure.

Consistency of the.

Yeah. Over time. Right. Okay, great.

So the edge has to be sustainable, has to last. So what I want to do first is I want to find out all of the bad ways to come up with a system so that we know what not to do.

Because we all know what to do wrong. Because let's face it, I mean, if you're anything like me, we've done it right. I've come up with lots of systems that don't work that break as soon as I started trading with real cash. Anyone else done that? Yeah. Okay, cool. So what are.

Some poor methods coming up with trading rules, a trading system.

What would you not do? What shouldn't you do?

Use a dartboard? Well, it depends on who you believe. I agree. I think. You know.

Alex, I mean, this is like picking stocks randomly, right? Tongue in cheek. Now you can actually make money picking stocks randomly and have good exits, but.

It'S not really the best way to go. You can have a good entry and a good exit and make a ton more money, but. Good. Thank you. What else? What other ways should you not develop trading rules? The system developed on a single stock.

Okay, good. So.

You're challenging me. Yeah, but it's a common one. Yeah.

I'm going to look up, I'm going to put this in a broader bucket, if that's okay. Because what we often do is we see a pattern.

We go, hey, look, it was profitable. See, if I bought when this thing happened and sold when that thing happened, I would have made X thousand dollars. Yeah. So whenever that happens, I'm going to trade that again and again and again. Right. It's a small sample size applied to a broad universe.

Understand what I mean? So extrapolating a single observation or an optimization around a single stop too far is a sure way to fail because you don't have the data to Prove that it works sustainably over the long term. Make sense. Okay. What else? Any other ideas? I'll seat you with one. A good one. I was just gonna say not so much system stop me. Especially if you get too greedy, jump in too early, even though you won't have a system. Okay. But you initiate it, you think it's about to happen. Anticipation. Yeah. Okay. Anticipation of some rules.

No, no, of course no. And I've never done it either. Who else has never done it? Great.

Okay. Let me give you another couple of examples since you're a bit more system focused. Oh yes. If you don't have proper reflecting, proper market conditions over all types of cycles, if you're going to end up hurrying forward results. Right. So cherry picking what's going on and applying that. Yep. Is that effect? Yeah. Or if you do your testing over a bull market, for example, it's not reflective of good. Yes.

I'm also going to. I'll add. So as cherry picking. Also add like a well chosen example.

Now let me just give you another couple. From my perspective as a system trader or as a technical trader, you might read about an indicator. Okay. Let's just take a crazy example, a stochastic oscillator. And when you look at it on a chart and you optimize it to find an entry and an exit signal and you find that when it crosses above 23.25 on XYZ stock, you buy and when you sell it, when it crosses below 71.83 over the last six months, it didn't have a losing trade. Good way to trade.

Over the last six months. It would have been over the next six months. How well is that going to go for you?

Probably pretty rubbish to be honest, because it's. And the point is in the. That the green of optimization, you know, it's so specific and so focused on a particular period of time. And if you take that off XYZ stock and to Mark's point, put it on ABC stock, it's never going to work. You're just going to lose money constantly because it's just a well chosen example or a data mined solution. You have just basically optimized to find something that worked in the past and there's no sound basis for any future profitability.

Does that make sense? Okay, good. Okay, here's another one. You go to my website or someone else's website and there's a blog post and it talks about this system. And so you take the rules off that system, you type them into amibroker and you see that there's a bunch of parameters in there and you set them all up as optimization parameters and you hit optimize. Amibroker comes up with a warning that says warning. There is 736,000 combinations that you're about to test. It will take a long time. Do you want to proceed? You say hell yes, because I want to optimize this. And the computer runs all night and then you go back to your computer the next day and you find the perfect parameter set. Has anyone done this? Yeah, I've done that too. Why is that not a great idea?

Because things change.

Because what you're doing is you're finding what the best parameter set for when the past we're not finding the parameter set which is most likely to work in the future. Good. So these are some poor examples of how to develop systems now. Data mining versus real edge. If you've approached it like this, you know you don't very unlikely to have a real edge. It's probably not going to be sustainable. Make sense? Okay, so out with the bag.

What are some.

Solid approaches for creating trading rules that will actually work and will actually make your money with real money in the future, not in the past. Any ideas?

Good.

Thank you. What else? Being on the right side of the moving average. Being on the right side of the. Can I say train. Is that okay? That's the moving average. That's good. Cuz I use moving averages as well.

Oh, that's a very good question. We'll get to that.

The one that works the best did I hear?

So being on the right side of the moving average of the trend. What else? Stop loss. Stop loss. Good.

So what I'm hearing stop.

That says stop loss. What I'm hearing are components of a good system. If I just take a big step back and think about how do I find a good system that will actually have a real edge rather than just be data mining Mark.

Good.

Data, lots of it.

And when you see large data set, can you elaborate, can you elaborate on what you mean by large data set? So you've got a lot of samples to take your right use from. So if you're testing on a really small range of information.

Okay, great. So let's say many trades.

Long time.

Many stocks.

I clearly wasn't a spirit teacher in a past life. I'm sorry for my writing something else. I was just going to say accurate data. Accurate data. Great.

Okay, thank you. So when I'm teaching a new trader how to build a system and I go through this in my program, one of the first steps, there's a bit of prep work, but one of the early steps when it comes to looking at the market and charts is to find the sort of move you actually want to catch.

So when someone says to me I want to trade.

Mean reversion using the stochastics, I sort of think, okay, that's fine, maybe that makes sense.

But can you show me on the chart where this move is? Can you demonstrate that it's actually there and it actually happens?

So what is the point is you must have a real market behavior that you're trying to catch. Now what's an example of real market behavior? What can you see when you look at a 2 or a 5 year chart? Any ideas?

Trends? Do stocks trend?

Yes or no? Yes. Can you make money following the trend? If you had a set of rules that caught 6 month, 12 month, 18 month trends, could you make money? Hell yes. Absolutely. What else, what else do you see on a chart? Do they go in a straight line in their trend.

Pullbacks? Okay, right. So is there a regularity with which the market corrects within the primary trend? Yeah. Could you make money trading that observation? Yeah. So my point is.

To ensure you've got a real edge rather than data mining. I always like to start with what you can actually see.

Not some mathematical calculation of an indicator.

Because yeah, the crossover of this, that it might make you money provided the strategy and the concept that you're trying to construct to trade and follow is sound. You with me?

So when we go and we go to a website or a book and we get a trading system, the first step is not to optimize that trading system for our market. What's the first step? Take big steps, step back.

And see if it's in the market. See if it's in the market. Yeah. What are these rules trying to do? Is that real and is it sustainable? And can I see it over many stocks and many markets? And does my market exhibit this behavior? Because how many of you realize that different markets exhibit different behavior? Anyone seeing that? Okay, great. Do U.S. stocks trade the same as Australian stocks? No, not really. Do stocks in Hong Kong trade the same as Australia? I don't know if any of you trade Hong Kong, but I do and they don't.

They trend, but the characteristics are different. So if you're looking for the Australian pattern blindly and applying it to Hong Kong stocks or US stocks, maybe that won't work.

So the first step for go live confidence is to ensure that you are absolutely certain that you have a real edge. Yes.

The market.

Know your market. Okay. Can you tell me more? What would you want to know? Well, what's it doing next?

Understand the animal you're working with. Good. Great. Yeah, absolutely. I like that.

Okay. Now I don't. When I think about my market and yours could be different, and yours could be different, and yours could be different. I think about ordinary shares on the Australian Stock Exchange or on the Hong Kong Stock Exchange. Right. For me, I don't think about ABC stock or XYZ stock as my market because I trade a very diverse portfolio. But your approach may be different if you are very focused on one instrument or another. That could be your market. Right. And you can trade a totally different concept that is a real edge for one particular stock, if it exists.

Or commodity or currency or whatever it is. But as long as you can see it and it's actually real, before you go and develop the rules and the math around, hey, how you're going to trade. Make sense? Okay. Is everyone good with that? So what's the first step? Ensure you've got a real edge. Thank you. Okay, cool.

Okay, we covered porcelain. Sorry, I missed a couple of slides. That's fine. These are just placeholders to stimulate your mind.

So let's say.

We have a observable market behavior. Let's call it Trends in Small Caps in Australia, just for an example. Because that happens, right? And we're going to develop a system around that. We've got some rules and indicators and we read some from a book and we put them into our software. Metastock, amibroker, whatever it is. And what do we then do? We're going to try and.

Optimize the system to make it better. So why do we do that? What is the purpose of optimization?

Pardon?

Give you confidence.

Oh, good. Make more money.

Come on. Let's be honest. That's why we are here, right? It's also fun and interesting and it's a good mental challenge, but it's fundamentally about that. So to make more money, to improve the edge. Did I hear or get an edge? Confidence. Confidence.

Improve our edge. What else? Looking for something.

Cut down the risk. Okay, good.

What else?

Make sure it suits you. Suits you good.

I'm just going to call that fit. Is that okay.

What else?

Oh, no, no. That is cheating. Is it to reduce the size of.

Because you're optimizing, you gain and reduce the size of what you're actually looking at the whole data set concentrated.

So that doesn't finish up.

So.

Is it the. To focus on the best moves or the best. Yeah. Okay, good. So let's say. Let's. Let's call it to focus.

Cool. Tell me more about that.

Good. Okay, I like that. So we don't want to get 50 signals.

If five of them are the good ones. So we want to save that time. We also don't want to have 50 trades and hope that one of them is a good one if we can focus and narrow it down to save time in the market on bad trades. Is that correct? Save your dogs.

Okay.

Okay. Huh. To ensure our ideas work in the future.

Which to the be honest is I think, and I still have to get it out of my system optimization initially, gut feel is about maximizing the dollar return.

And I think the distinction between the two is subtle, actually.

Yeah, absolutely. Because when we're optimizing. Let's face. Thank you, Mike. That's fantastic. When we're optimizing, which data are you looking at? Present, past, future?

Well, the data today, maybe. So. So, yeah, but predominantly the past.

Very good. Yes. You have something. Well, is it fair to say we have no idea.

Okay, good.

So if you have a set of rules that you've tested in the past.

Okay, good. So there's some techniques you could use to give you some confidence that might work in the future, like maybe a walkboard test. But what if you do a test in the past? What sort of behavior would you like to see to give you some confidence? Not absolute certainty, because we don't have that, but some confidence that it might work in the future? Yes.

Yes.

Present, today. And so it gives you greater confidence that you'll be doing it tomorrow. Nice.

If The Edge worked 15 years ago, and it worked 10 years ago, and it worked five years ago, and it worked two years ago, and it's still working today, and it's based on a real visible, observable concept that you can see on the charts.

Is it likely to give you some degree of confidence that it might work in the future? Not absolute confidence, because I don't have that. You don't have that. But is that going to help? Sure. Yeah. I mean, it's not going to give you, like every day, I wonder, is this going to keep making money? Is the system going to keep working? But I follow the rules anyway because they're the best rules I've got. And they don't make money every day, but every. Almost every year they make money. And over 10 years, they make a ton of money. So designing the system so that it works in the future, because it Has a stable edge is really the concept that I'm. I want to. I want to get across to you. When you, when you're. Because you're using an angle breaker, the first thing you look at is the equity curve and the drawdown. Right. Looking at that trade off between the two, optimizing.

The equity up. Yes. And control it down in lower. Good. As best as possible. Okay, so that's very good. Very good. So I've got an exercise coming out where we can talk about that. Okay. But before we do that.

I want to talk a little more about.

To reduce my font size.

About what? Stability. So for me, in my trading and when I'm kind of teaching and helping people develop systems, stability is one of the most important, important concepts. So what do we want to be stable? If we think about our trading rules and apply them to the market, what has to be stable about that? Any ideas? Yes. Position size. Okay. Position size has to be stable. Good. What else?

Say again? Sorry. Your mental concept of how you're trading. Your mental concept of how you're trading. Can you elaborate a little more? Well, if you don't. If you're not able to actually focus and stick to your rules. Yep, good. Okay. So your application of those. Yeah, good. So your mental state and how you actually apply yourself and follow those. Those rules consistently. Great. What else? Pardon? Expectancy. Okay, good. You want expectancy to be stable over a long period of time.

All of those other things are correct as well, because we need to have constant position size or at least a stable position sizing method that will keep us in the market. We need to have the stable psychology and application and discipline to apply it. But when we're thinking about just our rules, the system or the technical indicators that we're measuring.

The edge or the expectancy needs to be stable over time.

Great. Consistent. Moving up. Now, I would love it if my equity curve looked like you just drew.

That'd be great. But sadly doesn't.

But not a lot of volatility. That's exactly right. And so when you zoom out, at least when I zoom out and I look at my equity curve over three to five years, that's what I want to see. Okay. Yeah, there's some bumps, but it's got to be stable and it's got to be making money consistently over time. And the edge has to keep working over time because if your edge comes and goes.

Now, I'm not talking about little market cycles, but like if it worked in the 1980s and the 1990s and then in the 2000s it kind of was sort of break even and then from 2010 to now it was losing money. It's an unstable edge. That's what I'm getting at. Okay. A bull market to a bear market. Okay. Long, long side trend following may not work that well in the bear market. That's not what I'm talking about. I'm talking about a more macro sense because I test over 20 or 30 years of history.

So I want to see that over the long term. It works. It's robust, it's stable. Good.

Okay, great. What else has something else really important has to be stable. When we're thinking about our buy and sell rules and where we place our stop loss and all of that, there's something else that needs to be stable. What do you think it is?

Ourselves. Ourselves. Good. That comes back to the psychology. Yep, absolutely. That's in the application of the rules. Something about the rules? No, the system metrics. System metrics tell me more.

Good.

So yeah, I completely agree. And a lot of that, not all of it, but a lot of that is about the edge working over time. Right. And how do we know it keeps working? Because our system metrics are stable at least somewhat. Right now this year you might have 30% which. And next year you might have 45% winners and the year after that you might have 42% winners. That's okay.

Because it depends on how in sync exactly with the market your rules are. But if it's going from 50 to 40 to 30 to 20 to 10, year by year by year, you know you have an unstable set of rules. Make sense.

There's one more thing.

When you think about your rules.

We'Re buying when this moving average crosses over that moving average.

What two terms in that sentence determine when you're going to place the trade. This moving average and that moving average.

The parameter values.

If you're saying a 10 day moving average and a 200 day moving average.

That'S what makes money. If you're trading a crossover, say just to take a simple example.

We want our rules to be stable over time.

But you could develop and back test to see system over time. That's still data mined.

And curve fit to the past. And it could be stable over time because you've done a really good job curve fitting it, right? Yes.

Different cycles. Yeah, absolutely.

Yeah. So if I told, let me take a little simple example. If I told you I was going to trade it dual moving average, crossover system, it's not a great system, don't trade it by the way. But just as an example, if I was going to buy when the 10 day moving average crossed above the 200 day moving average and I was going to sell when the 10 day moving average cross back below the 200 day moving average.

If I optimize that, or if you optimize that and it made Money with a 10 day and a 200 day, what do you do? You celebrate. Woohoo. Right. System works, get to make a ton of money. But if I change it to a nine day moving average and 198 day moving average and it loses money, what does it tell you about the parameters?

Yeah, great. They're not. What?

Yeah, they're not resilient, they're not stable.

Right. If I'm going to trade a system, I want to know that not only does it make money over time, but I want to know that the parameters are really, really stable.

Because if they're not stable, if it's like the 10 day works, but the 9 day doesn't, and the 11 day sort of works, but the 13 day loses money, you got a real problem because what it's telling you is you've curve fit to the past. You with me? Do you understand what I mean by curve fit? Yeah. Does anyone need me to explain that or is it okay? No. Okay, good. So we want stability over time and we want parameter stability so that.

We'Re sure that it's likely, more likely to keep working. Now I've designed heaps of amazing systems and then I came to do the stability test, which I'll talk to you about. It fell apart.

And all it shows is that I, I'm really, really good at curve fitting because I like maths and I like the optimize button. And when you just use the optimize button in the wrong way, you end up curve fitting. You get unstable parameters. And when you trade with unstable parameters in real life, what happens to your money.

Goes down. Because the past is not. Sorry, the future is not exactly like the past. Good with me.

So.

Let's talk about stability over time first. Is that okay.

I'm going to draw a couple of equity curves. So this is all this is conceptual, right? I mean we can do the charts and optimization, coding, but kind of not really in a format like this. So I want you to take some concepts home so that when you go do your testing, all of your charts, you've got something. If you have any questions about how to apply it, when you get home, I'll give you some information about how you can Contact me later. And I can kind of help you with some of this. Is that okay? So, equity curve.

If I designed a system and my equity curve looked like that.

What would it tell you?

It's like a lottery ticket. Good. Tell me more. Well, you can come up with a system which generates several losses before you go. Yes. Gain all sensing losses. Yes. It's unstable to sensor. It can be long run or that loss, you don't ride it correctly.

Great. Very good. Okay, good. Thank you. What else? What else? Seasonal. Seasonal. Good. So it works sometimes, but not others.

So it could be, you know, maybe every up is the Santa rally. Anyone ever heard of the Santa rally? That's not terrible.

True, it's not terrible, but it's not great either. Good.

Is this the next best use of your money? Because it's all about opportunity and opportunity cost. If you've got a better system, use a better system. If this is the best you got, use that. Yes.

Missing out on a lot of drawdowns. You're missing out on a lot of drawdowns. Tell me more about that. I agree. Because you're avoiding the falls while taking advantage of the rise. Good. So maybe you're neatly sidestepping the bear markets. That's right. Could be. Absolutely not going short. Good. This is another question. Yep. I'm assuming. Let's assume low. Manly. For simplicity. Yeah. So this is maybe, maybe you've got a really great way of detecting that a bear market started and you just get out, go to cash, wait, and then step back in and the market keeps going. Yeah. It probably won't look this beautiful in real life, because they don't, let's face it. But it's absolutely valid. What else.

Might turn that into a 50% return over five years?

Absolutely. In fact.

My equity hasn't gone down, so, you know, initial response.

Yeah. Now look, I mean, we've just got one picture on the chart, so we don't know if it's good or bad yet, but it has potential. I guess.

Management of risk is excellent. Yeah, it could be. Yeah. Now, it depends on what we've done with position sizing as well. I mean, if we've, if we've, if these are lots of little losses in here, like if that was starting to be a little bit of a drawdown and then we had a big win, lots of little losses and a big win. If you dug into this and you found that that's one trade. That's one trade. That's one trade. But the whole system takes 500 trade trades.

Over that time period, how good is it? Fair, fair, fair, fair at best. Because what if you miss one of those trades?

Shit.

Right. Because you have a very high dependency on a very small number of trades. Now can you make money like that? Yeah, absolutely. But you want to have the discipline to follow the rules. Rules until that trade comes along. Right. How do you trade?

Still learning, still learning. Okay, good. Well, you've got the discipline thing down pat. If you have that in the front of your mind. That's the key. A rule with a system with an edge or an approach with an edge and the discipline, consistency to follow it. Because if you had this, and I had a system like this once and I did manage to make a fair bit of money with it, what this usually looks like is a very, very heavily filtered trend following system with tight stops. Very, very tight stops. So I had a system that had about 80% losses, 20% gains.

The losing trade was on average about two days.

But when I happened to get the right move.

I let it run.

Run and run. And so it made so much more money than all of the losses. The downside is the equity curve is a little ugly because we'd like what, A bit more of this.

Make sense.

Now this could be. Okay, the point is we're looking for stability over time. So yeah, the pattern keeps repeating. The equity curve still going up, it has some degree of stability, but it's not there all the time. It's somewhat seasonal or dependent on very specific market conditions. Good.

Okay, let me draw another equity curve.

Now, I trade pretty long term. I hope that's okay. Let's assume that this horizontal axis is, let's assume this horizontal axis is 20 years. Okay, just, I just want to get you to step way back is that right now my trades, my best winning trades might be around 12 months long.

I like this is not tax advice and I'm not a financial advisor, but I like most of my profits be more than 12 months long for reasons you can discuss with your accountant. Okay, so.

I hold my winners for a long time and I cut my losses pretty quick so I can try and get the money back into a new winner. So let's just assume this is like a 20 year backtest.

Now if I had an equity curve that looked like this.

Okay, what does this one tell you?

Say again? You've got a problem. I've got a problem. Good. What's the problem? The problem is you're in a loss, right? Losing train. Yeah. Yeah. Well, 10 years, could be five years, two years, whatever it Is, but there's.

The system works right? Now you see this all the time with car curve fit systems. Right? Chances are this is what happened.

Around here. A book by a popular author was published showing the backtested results of a system.

And.

That system may or may not have worked. But the point was it was published and it was back tested up until that point. And then five years later I've gone and read the book, hey, cool system. I'm going to try that one. And I back test it and what do I see? Well, it works during this period. I go back and look at that year, it's like, hey, funny, that's the year the book was published because it was designed up until that point, totally curve fit make sense. So you'd be really careful of this. This is why stability is so important because you get a set of rules or an indicator from the website or a book or something like that. You want to know that it's stable and it's not just put there because it's. Is a really great example.

Yeah. Because the system is broken, the performance, the edge isn't stable over the long term. With me. Yeah. Okay, great. Has anyone seen one of them? By the way, we got a strategy now for options, 10 years. Yes.

Okay, good.

Okay, good.

Fantastic. Good. So what, so if we can characterize that the. What was the, the profit driver of the system.

The actual stock used to move before earnings. Moved before earnings? Why? Why did it move before earnings? The assumption is the CEO was leaking. Okay, Right. So if that's true, I don't know. But if it was true and some information was coming out into the market a little before earnings. So the market was anticipating the result.

Right, good. So after the earnings announcement. Great announcement. And the stock does nothing. Why not? It's because the market already knew. Right. So what, you have a direction neutral strategy on earnings for that stock. But like what was the stock? Okay. Pre earnings you go, okay, good. Okay, great.

Okay, great. So the profit driver was presumably the leaked information or the hinted ad or whatever. I'm sure it was all leaked. Yeah, it was just. I didn't know what was happening.

Okay, good, good. So can a proper driver go away?

Absolutely. Can a low interest rate environment go away?

Doesn't feel like it's been here for a long time.

Right.

So if the profit driver changes.

The system could break.

So this is why I'm really big on what we talked about earlier, which is have a concept which is you can observe. Right. It's fundamentally there. I like trends because it doesn't matter what market you open up. You can see trends.

Make sense. Now, a strategy like that, where there's a different profit driver, that's okay because you could have made, probably did make a ton of money on that. Right. But you've got to know when that edge goes away. So the first, the real lesson, which I haven't written down was in my talk, but that's awesome. Please write this down. Know the profit driver of your approach.

If it's just as simple as stocks tend to trend and a trend tends to continue, that's great. If it's the behavior of XYZ stock, two weeks pre earnings.

Is indicative of something that's going to happen over the next two weeks or there's going to be a move pre earnings because some information is likely to leak out. So if I position myself anticipating a move, then I'll make money. That's a good edge too. Fantastic. As long as you can prove that it's there and that it's stable. And you know when it goes away. Right. Because what happened co changed, went away. Yeah. Well.

Two in a row, which became the.

Back testing. Right.

Third cat.

Great. So edges will come and go depending on what type of edge you trade. Now, I've been trading trends for 15 years and stocks still trend. So I really like that edge. It's really stable. But you know, stocks also swing around the primary trend. That's true too. And there's plenty of other sustainable edges. But the point is, if you know what your edge is, what the driver of that edge is, you can also know if it happens to go away.

So you know, to do what? What are you going to do if, you know, the CEO that's leaking information changes and that was the edge that your system was relying on. When the CEO changes, what's the most sensible thing to do?

Pause the system. Wait and see if the next guy is as dodgy as the first one.

Does that make sense? Yeah. Or if your system relies on interest rates falling. When interest rates start to rise, what's the most sensible thing to do? Pause the system.

Or if your edge is, you know, I observe market trends and trends have been here for the last hundred years. And if I look back on all of those charts, they always trend and there's periods of lull in between. But if I keep buying breakouts, eventually I'll catch a trend and then I'll make money and over time I'll be profitable, you know, then, you know, that's a pretty sustainable edge and you just need to Watch the macro market conditions to see if any structural changes that makes that go away.

You with me? Is that okay? Good. Okay, great. Is this helpful? Interesting. Okay, good. Alright, great. Next type of equity curve.

You run your back test and you see this. Is it still making money? Yeah, but what does it tell you?

Okay, good. Achieved your asymptote of growth. So can you explain a little more what that means in the system context? It means that either the market or your system don't fit quite the same way. Ah, good. As it did. Maybe here, very good down there. But what has changed? So how stable is the edge? Not so good at the top. No, the edge is decaying over time. Now, have you ever seen, have you ever seen an equity curve where it was amazing in the 90s.

Or the 70s, like whatever, and then over time it just like this is a bit boring now, it's still profitable, but there's better ways to make a living. I might go get a job.

So this is a decaying edge.

Now this is a trap, right? This is. You've got to look carefully to make sure your edge is not decaying over time.

Because.

Let me see if I can explain this.

Let's say your equity curve is compounding, right? You're risking. You're putting 5% of your account into each trade. So as your account grows, the trades get begun and you're reinvesting all the profits. And so your account compounds. Right. So we're getting constant percent growth, not linear growth. Yeah. So over time our equity should accelerate up. Making sense. But here the position size is much bigger than it was here and here. So if you back test.

Only with a position sizing algorithm that compounds your profits, if your system is any good at all, you'll see a compounding equity curve and it will accelerate up and at the end there'll be billions of dollars.

Right?

And we know it's not totally realistic and we're not going to get to the billions of dollars that are at the end of the equity curve. But the problem is the compounding equity curve, the position size is getting bigger and bigger over time, can hide ugly stuff under the surface.

Because if. Let's talk about expectancy. Everyone know what understand expectancy. If the expectancy, the expectancy of this, I'm going to draw it on a different scale. Okay? It's not the same scale, but let's say the expectancy of this system is gradually falling.

You can still get a compounding equity curve that looks like that because it's still profitable. But under the Surface, something ugly is happening, your edge is decaying. Does that make sense?

So it's really tempting to want to do this back test or where you go, yeah, risk 1% of my account on each trade optimized backtest, and we get, you know, again, the billions of dollars at the end, which is, you know, gets everyone excited, including me, I can't lie.

But if you don't look under the surface, if you don't look under the hood of what's going on for stability.

You can miss this because the compounding equity can might look wonderful, but your.

Expectancy is getting worse and worse and worse and worse. And the system is decaying, it's dying. What that's showing you is you're in this bit.

And in the future that's probably going to happen. Does it make sense?

So when we're doing system testing, it's not all obvious. We say, okay, great, we're getting great compound annual return. And year by year, they're looking okay. The numbers are all right, but they're very. It's always erratic. Right. You know, plus 20%, minus 5%, plus 15%, plus 35%, you know, and it's really hard to see the trend, particularly if you're compounding everything. So what I like to do, as well as testing my doing my backtest with a compounding equity curve with like 1% risk, perfect trade, or half a percent risk per trade. I also like to run it over the last 20 or 30 years, risking just $1,000 on every trade, every single time for the entire backtest. Because what that will do is what show you if this is happening.

Or if this is happening or.

That's happening.

If you know what.

Oh, no, true. But if you, if you're, if the expectancy of the system was increasing over time, it could look like this. If the system is getting more in sync with the market, if the system is better now than it was 10 years ago, like if there's a. If the profit driver is sort of enhanced or magnified or something like that, but really the blue line is the ideal, right? You want a stable, somewhat stable expectancy so that you can compound it consistently without fear it's going to decay or die.

Does that make sense? Okay, cool.

Whack that system.

It was compounding to the stage where you buying more shares.

Oh, yeah, of course. Absolutely.

Yeah. Now look, I mean, that's a. That back test.

True. Yeah. I mean, look, that's a really long discussion. How do you back test and how do you. At what point. Point do you start to take that into consideration versus just looking at the price moves and all of that? So I probably won't. I probably won't do that justice today.

Oh, yeah, absolutely. Yeah. It's really important.

Do you not compound to get rid of that problem? Because it is a problem. You start off at 10,000.

Absolutely.

Okay, good. Let me talk about it. I'm just hoping I can change the money comes in that I can get. Right. Okay, cool. So I think about it in a couple of ways. The first thing I want to know is, do my rules work? Is the edge stable over time?

I don't want to know. If I've got $10,000 to invest, I don't care. If the System can trade 5 million bucks today, I don't care.

In 10 years I might, but right now doesn't matter. So when I'm designing a system, I want to make sure that it's going to work with the amount of cash I'm going to have for the foreseeable future.

But I don't really care at the tips of the equity curve right now, as long as the equity curve shows me this sort of stability.

In the expectancy, right. If the expectancy is stable.

Over the entire period, if the signal work.

Consistently over time, then I'm pretty happy because I know that I'll be able to compound that into something meaningful in my account.

And when my account has 50 million bucks in it, which it doesn't yet, but one day I'll worry about that problem. How do we worry about that problem? Well, there's lots of ways, but we could just develop a second system and then split our capital and then the third and fourth and fifth, and we can go from Australia to Hong Kong and then to Shanghai and then to the US and then we can go to futures and then we can go to currencies. Like, there's lots of ways to deal with the issue of this system. Can't support my account.

But before we get to that problem, we just have to grow our account and make sure the system is stable and works. At least that's the way I think about it. So I want to see that the edge is stable over time. The expectancy doesn't decay, it hasn't died.

It doesn't have too much jaggedness in it. Because I want to make money regularly, smoothly.

And I also don't like to see the expectancy doing that.

Because what happens after that?

Right? Because I want to know that it's going to continue working and it's not going to blow up my account. Because fundamentally, you have to be in the game to win this game. If you blow up or lose too much money, you can't win. The way I believe you win is you have an edge that works, and you apply it consistently with low risk and good risk management diligently, day after day or week after week. So I first test for the stability, and then I worry about the liquidity and how much money it can support. And I'll play with the liquidity to make sure it can support my funds now and for the next couple of years.

And then as my account grows, typically, I'll diversify and split my money, spread it between different systems and markets so that I can continue to grow and don't hit that problem.

Good. Does it help?

There's a lot in there, so I just want to pause. Any other questions on that.

It's very heavy for a Saturday. I'm sorry. We'll have lunch soon.

Does that help? Does it make sense? Yeah. Yeah. Okay, cool. One last equity curve. This is a good one.

Because remember, we're looking here for stability, which is my big thing. Okay. If your equity curve.

Okay, so if you do your back test and your equity curve looks like that.

What'S it telling you? Huge dream. Huge drawdowns. Good. And what's happening to the size of the drawdowns over time? They're increasing. And what does that signal?

Failure, Imminent death. Stress. Stress. Good. Certainly lack of stability.

Now I see. I don't know about you guys. I see this a lot.

An edge. There is an edge here, right? Because the portfolio is going up, so the trend is in the right direction. It's making money, but it's getting less and less stable over time.

Now, that could be the market structure changing. It could be the market participants changing. It could be lots of different things. But the point is, the edge is unstable, and this is very dangerous.

So you want to look at your compounding equity curve with like 1% risk or half percent risk per trade, and check that this is not happening. But I also like to look at the constant risk per trade, like a thousand dollars risk on every trade over 20 years and see if this is happening. Because I want to make sure that there's as much stability as.

And what's ideal? Well, that's pretty obvious.

Ideally, our compounding equity curve should look like what.

The compounding equity curve.

Exponential curve.

Like constant percentage growth year on year. So the money just rockets up. And our constant risk equity curve. So when you're not growing your position, size, over time, you're just taking the same size trades every time. You want that to kind of look.

Like a straight line in an ideal world.

Okay, is the distinction between this clear or do I need to go back through that?

Pardon?

Yeah, this is.

Let'S say again, let's go back to the moving average crossover system. 10 day moving average crosses above the 200 day moving average, I buy, and when the 10 day crosses back below it, I sell. Okay, let's assume that's a good system, which it's not, so don't trade it. But If I risked 1% of my account on every single trade.

When my account is $10,000, how much am I risking on that trade?

100 bucks. When my account's a million dollars, how much am I risking on each trade?

Too much. Well, that's interesting psychology. What percentage?

One percent. Right. For how many dollars? 10,000. Okay, so it's a certain percent of your account. The position size is linked to the size of the account. So as your account grows, your position sizes grow and the compounding continues to go. So you get that exponential growth. This other one is on every single trade.

If I enter at $10, my stop loss is at $9, $1 risk per trade. I'm going to risk a thousand dollars on every single trade in my back just now over the entire 20 years of the backtest. Okay, so it's just a constant level of risk.

If you do the backtest with that type of position size and you don't get exponential growth because the profits that come in don't affect the size of your trades, which therefore don't affect the size of your profits. So therefore you don't get this exponential thing. Does that make sense? So that's the difference. And what, what I'm suggesting to you is to understand stability of your rules. I want you to test it both ways.

So that there's nothing hiding underneath this compound.

And if you need help with how to set up the different position sizing rules and stuff in Amy Rec or whatever, I got a program for that. We talk about that later. There's stuff on my website. There's also plenty of stuff out there on the Internet. But conceptually I think this is, is pretty important if you want to be sure that it's stable. Good. Okay, great.

Okay. It's almost five to one, is that correct? Yet another five. Five minutes. Okay, good. I do have a tendency to go into too much debt, so I'm going to apologize. I might skip a couple of things later. This next one is really, really important because I want to nail stability. So we talked about stability over time and we also talked about what type of stability. What was the second type of stability? Time and parameters. Good. So.

Just help me out. Who has done an optimization in a backtest like in Amy Broker or Metastock or something like that? Show of hands. Anyone? Not. Okay, good. So what. Let me explain it real simple.

Let's say this is the length of your moving average and you want to vary it from 10 to 100 in steps of 10. So you want the back test to say, tell me how profitable 10 is now, how profitable is 20 now? Her profitable is 30, 40, 50, 60, 70, 80, 90, 100. And basically you just get your software to increment each of the parameter values and report back how good was the system. Does that make sense? Okay, so when you do that, it reports back a whole bunch of what data?

Compound annual growth rate and expectancy maybe and drawdown, blah, blah, blah. A whole big long list of stuff. Let's just think about expectancy.

So.

When I vary the moving average 10 to 100, let's just call this expectancy.

If the data comes back and it looks kind of like this.

What does it tell you about that parameter value?

Irrelevant. Good. Makes no difference whether it's 10 or 100 or 50. Is that a good thing? No, no. It's kind of weird. Like why do I have this in the system if it doesn't make any difference at all? Right.

So it's stable. Yeah. But it's totally irrelevant. It's great. Now I'll draw another couple of pictures.

If I step my parameter value over a wide range and.

The result.

Is like this again. This is the value and this is the expectancy. What does that tell you? There's an optimum. Good. What else does it. Thank you. What else does it tell you.

Is what. Sorry, there's lots of. Lots of. Non. Very good. Okay, I like that. So there's optima, clearly. And there's lots that are much worse than the very sensitive. Very good. It's very sensitive. Is it stable?

Not at all. If you put this value into your trading system today and traded with that for the next five years, how well do you think that would go for you? Not well at all. Why?

It's true. Absolutely. We don't.

It could be the optimum for the next five years. It could. So could this, this, this, this, this, this, this, this, this, this, this, or this. If the market shifts now, the market may shift or it may not shift. I don't know, and the point is we don't. Okay, let me draw another chart to show the comparison. Is that okay if instead of this profile you did your optimization.

Is this helpful, by the way? We liking this? Yeah. Okay. If it looked like this.

And okay, is there an optima? Yep, there. Is it stable? Yes. Because here it makes money. Here it makes money. Is the parameter relevant? Yeah. Does it like, would you bother to include it? Yeah, I mean, this one you might bother to include as well. Right, but I would. My hypothesis is that this works well in a back test. If I trade it forward, I'm not likely to get exactly this level of performance. I'm probably likely to get something, you know, a little either side of it, which might be down here. Only because the past is not exactly the same as the future. Yes. When you get a result like that, that's indicative of one or two trades dominating the results. Absolutely. It could be. Yeah, yeah. It's heavily over optimized to one particular instance. And if that doesn't repeat, the future is not going to be the same, same as the past. You're not going to make that level of money. How about this one? Yes, perfect. Perfect. Right. Because we've optimized that. We see, hey, this one matters. If it's too small, it's not that great. If it's too high, it's not that great. If it's kind of in the middle, it's profitable. But it doesn't really matter if it's 55.234 or if it's 30 or 40 or 60. Does that make sense? So when you vary your parameter value one at a time and you see this, you go. When you see this, you go, hmm.

Caution. That's all I'm saying. But remember, this is trading, it's the markets and it changes over time. There are no absolute certainties. So you can get this and the system could fall apart. But all you can do is tilt the odds in your favor. Right. And try and test with as much data as possible. And try and follow good practice and keep your risk low so that if the edge is broken, it doesn't destroy you or your account.

Is it good? Okay, now another couple of profiles.

You know this volume that this often happens too. If you vary the value, you might find an optima, which is very sensitive. But if you push it a little further, you might find one of these stable areas. So which value do you select to trade with in your time? This one. Because it's the stable one, it's profitable and stable.

Good. So I had a couple of conversations over lunch and what it really highlighted and reminded me was that everyone trades in. You all trade in your own way. And that's totally fine and actually is absolutely necessary. So I have my system and my rules and that works for me with my personality. And anytime I'm teaching someone new, I talk about personality, lifestyle and objectives. And if your method fits your personality, lifestyle and objectives and you have an edge and you manage risk, then you've got a good chance of making money.

I guess my point is some of the stuff that I'm talking about may not be directly relevant to you. Like if you're a value investor or if you follow a newsletter or any of that, that's fine. But if there's some understanding or some lessons that you can take and apply, even better, or at least it'll store some ideas in the back of your mind for the future. And just because I trade long term trends, if you trade on five minute bars, that's okay too because I know quite a few profitable traders who trade very short term. It just doesn't work for me.

But these concepts will be when we're talking about system performance, stability over time and parameter stability, all that sort of stuff, that absolutely applies regardless of time frame or instrument. Is that good? So.

Just remember, we've all got our path, we've all got our own personality and different lives and objectives. So I think the key to success, or one of the big keys to success is to find your way. And if I can help a little bit, some of these ideas. Good. If you need some help or pointers afterwards, then reach out and let me know. We'll have a chat about it and I'll give you some pointers. And if I can help, great. But if I'm the wrong guy to help, then someone else will know about the path you're on and it's just a matter of finding the right instructor. So is that okay? So hopefully you get some stuff out of this. Take what you can, ask a lot of questions. If it's not making sense, put your hand up. But just know that I'm a systematic trader and if you're a fundamental or a value investor or a price action trader, you know, some of this is going to be interesting, but maybe not directly applicable. If you're a discretionary trader, some of this is going to be like, whoa, I haven't really thought about doing that sort of stuff before, but it might actually help you along the way to get sustainable and consistent. Okay.

So look, one Of I had an interesting snippet of conversation.

About yes, at the back. Significance of a rule at lunch. I forgot your name. Sorry. Howard. Sorry.

Howard. Howard. Okay, so I had a good conversation, or started a conversation with Howard about whether a rule, you know, like, if the parameter value doesn't matter.

Do I just choose like an average value or what? And it comes down to this concept of significance testing. Because, you know, Albert Einstein said, I think it was him, said that everything should be made as simple as possible, but no simpler. Right. So what does that mean in trading?

That means if the rule adds value.

Use it. But if it's kind of just there and doesn't really do anything, then you might want to think about taking it out because it's adding complexity for no additional value.

So when I've, you know, we talked on the previous chart about parameter stability.

And there's a few different profiles. So when you, if you do back testing and you vary the parameter value, like the length of your moving average and you find one that's stable and it makes a difference, then you probably want to have that in your system. But if you find one like this, where varying the parameter actually makes no difference at all, the first question I want to ask is, well, should I even have that rule or should I take it out to get a simpler set of rules with fewer, less complexity so that my system can be more robust?

Because if you can make money with five rules, that's better than 15 rules, in my view. Because if you've got 15 rules, there's so much potential for, you know, curve fitting, cherry picking, precise results. Right, that makes sense. But if all 15 rules are significant, then fine, you know, but just be careful. You know, simpler is generally better, I guess, in my view. So let's talk about significance.

I have a real font size problem today, in case you haven't noticed. Sorry about that.

If you think about your trading rules, whether you're discretionary or systematic, doesn't matter. What determines if a rule is significant and actually matters? How do you know a rule is significant? Changes, Changes outcomes. The.

Good. Thank you. What else?

Personal comfort. Personal comfort. Tell me about that. Don't like I just stretching, but like some people, it might make no real difference to the outcome. But I feel better having that. Okay. All right. Has anyone ever used or thought about a rule that might make you comfortable but doesn't necessarily seem to affect the outcome? Anyone done that? I have. By the way, why would you do that? Yeah. Stop loss. Okay, good. A stop loss. So why do you use A stop loss.

Makes you feel comfortable. So protection right now, it could be like disaster stop. Like it's really wide, it'll get me out if something really bad happens, but it's almost never going to get hit. That could be a rule for comfort and it may not show up in your back test, but you could imagine a scenario where that rule might become important. So I'll put comfort, I'll give it to you. I think that's sound. But the warning is just be careful.

We just want to be careful because if we're just making ourselves comfortable without a real profit driver kind of rationale, just this feels better. You know, I like this. You know, that's not going to help. But if it's actually sound and there's some underlying principle that you're trying to adhere to, great. Okay, what else? What else makes a rule significant? Impacts on the number of trades. Impacts on the number of trades. Great. Can you elaborate at all on that one?

Great. Okay, how about this scenario? You're exactly right, by the way. So let's say I have a backtest, I've tested, my system looks great and it has a thousand trades in over the last 10 years.

And then I think of this clever rule that might improve my performance.

And I add this rule in and the number of trades goes from 1,000 to 998.

And the average profit per trade goes up.

Is that a significant rule?

Okay, tell me more. Why is it significant? What makes it significant?

Pardon me? Accuracy. Okay, the result. Yeah, the result is the profit went up. The average profit per trade went up. Okay, but how many trades did impact? I went from a thousand trades in my backtest, I added this new rule and it dropped to 9982 trades.

Okay, so is 2A very statistically significant proportion of a thousand.

True. Good. So the performance improved, but it only affected two trades. Ok, now I'm going to tell you something different. The rule was exit. If the date is October 1, 1987.

Okay, that's going to make you a lot more money if you had that rule on October 1, 1987. Right. But could you possibly have had that rule back then? No. So adding that rule, it might have affected a small number of trades. It might have improved your back test, but it's complete rubbish. It's not significant. You couldn't possibly have known or designed that at the time. It didn't affect anywhere near enough trades to have confidence that it was statistically significant.

So if I have a thousand trades in my backtest and I add A rule. I'm adding complexity. Is that a good thing or bad thing? Bad thing. To justify the complexity, it needs to be a really significant impact. So profit needs to go up. But also to Mark's point, the number of trades needs to drop significantly. I would put it to you that a drop of two out of a thousand is not significant. A drop of 200 might be.

So you want to know that that rule isn't just cherry pickings some bad trades and getting rid of them. It's got to be systematically getting rid of a big chunk of bad market behavior that you can't profit from. Good. So number of trades, very key. So if you add a rule, not only do you want the profit to go up or the profit per trade to go up, but you also want it to reduce your number of trades by quite a bit. Otherwise it's probably just Cherokee.

What else?

Timing. Tell me more.

Okay.

So like if the rule is a, about a really precise degree of timing, that makes it more or less significant. Well, the real.

You'Re not going to, for the most part, you're not going to get out through the day, you're going to exit the next morning the closing price.

Okay, cool. So if you're trading a long term system, maybe the intraday price action isn't that significant in the context of the move you're trying to catch. But if you're trading a short term system, where's my friend?

Oh, there you are. The back on three or five minute charts, then obviously that's going to be a really, you know, the intraday price action is really significant. So it's going to make sense. So if a rule is not significant, it's got no business in your trading plan or in your trading system.

Yeah. Like we were talking about Motley fool earlier. Yeah. And if I subscribe to Motley fool and I'm taking their trades, their suggestions, the system is Motley fool, their analysis, clear. How I feel about that stock is completely irrelevant. And I'm probably not going to add any value by adding a rule about, well, I got to know that company or I got to like that company or you know, that company has to have been in the news this week. They're not significant, they're irrelevant. So you only want the core relevant rules that make a significant difference. So how do you know.

How do you test if something is significant?

Say again? By removing it. By removing it. Good. And what are you looking for when you take it out? A big change. Big change in what? Result. Result and number of trades. Yeah. So what I do is I kind of dream, let's say I dream up my rules and if I've added a rule to improve performance and I've done the optimal everything towards the end, what I'm going to do is look at every single rule and say, huh, is it significant? And I'll say, let's say it's like an inequality. Like this has to be greater than that or this has to be less than that. So the price has to be greater than the 200 day moving average. To take a simple example, I'll test that rule.

So close greater than ma, for instance.

What if I do the opposite?

Close is less than the ma.

So the trades with the rule should be dramatically better than the trades without the rule or the trades with the opposite of the rule should be dramatically better. Different because otherwise the rule has no business in your system or trading plan. Not only that, I'll also test it without the rule to your point to make sure that not only is it dramatically better, but it also affects a lot of trades.

This making sense? This is. It does, it does. Because a lot of news comes out after close, right? And that can affect the.

Opening price.

And the trend there from, right? Sure. Now this could apply to any, I mean this could be a rule that says the open of the day has to be greater than the close of the day or vice versa. Like it doesn't matter what the rule is if you can apply this test to it, the rule has to be dramatically better than the open opposite of the rule. And also the rule has to be dramatically better than if you take the rule away.

Because what I notice when I'm teaching traders, particularly when they come to me and they've already been trading for a while, what I notice is they go, I do this and this indicator has to do that and this one has to be greater than that. And this price action has to be doing that and this has to be bigger than that. And, and, and, and, and, and okay, who knows what talking about, right? So what usually happens is there's a couple of really great ideas in there and a whole lot of what crap.

And so because there's a whole lot of crap in there, it doesn't work consistently over time. So the first thing you can do is go okay, well let's just significance test everything. Which of these rules actually matter, what's giving you the answer. Test with it, the opposite, without it, eliminate the rubbish and you simplify to a smaller number of rules that are actually making a difference.

Make sense.

Good. Okay, so that's significance testing. So whether you trade with discretion, as in looking at indicators on charge, or whether you trade systematically. I think this is really important because if you're looking at a whole bunch of indicators, let's say you've got 10 indicators that you'd like to tick off. And if the weight of the evidence of those 10 indicators says buy, you're gonna buy. Okay? You want to know that those 10 are actually significant and there's not just one random one thrown in there that makes no difference because that's gonna be skewing your decision. So find a way to test that. I do it with Amy Broker with backtesting, but you could also do it by looking at charts and paper trading or trading with your real account if you wanted to do that.

Good.

Okay.

Now.

We talked earlier about parameters, right? And we wanted the parameters to be, sorry, stability. We wanted the system to be stable over time. And what was the other thing? What was the second thing we wanted to be stable? Parameters. Good. So we've tested the parameters. Typically you test these things one at a time because I don't know about you, but if I'm. If I'm in a system with five rules in it, and I'm testing all of those five rules all at once, my head hurts because I can't think in five dimensions.

So I like to test one at a time to check that one at a time, they're all significant.

But if you've got five rules, are they going to affect each other? Are they going to interact? Right. So at the end of the process, once you've checked, you've got significant rules and they're all stable in isolation, you pull them together into a complete set of trading rules. I have this other step that I call robustness testing, which is I want to know that that combination of, of parameter values I've chosen.

Is robust. It's not going to break next week or next year.

Does that make sense? So I've already looked at the moving average and I've already looked at the atr and I've really looked at the width of my stop loss separately. When I pull them all together, I want to know that if a little variation in any one of those things, I want to know if it matters.

So.

Let'S say, you know, again, we'll just take a simple moving average system.

And.

We'Ve got a stop loss.

So let's say we have one parameter which is the width of the stop loss. Good. We've got another parameter.

Which is the Number of days in the short term moving average. And we've got another parameter.

Which we can set vary which is the number of days in the long term moving average. You with me? So let's say this was, I don't know, 210 and 5%. Just for argument's sake. We know that we've already tested for stability, so we know that if we vary the 200 on its own, we're going to get this nice beautiful shaped optimization chart that we talked about earlier. Right. Remember that we know the same for this and we know the same for this. But my last step, to check that the system is robust. Because I want to make sure that the system is not going to break. It's going to work work over a wide range of market conditions in the future. So what I'll do is I'll vary each of these values plus or minus some amount, let's say 20%.

So simultaneously, if we vary the long term moving average plus or minus 20%, the short term moving average plus or minus 20% and the width the of stop loss plus or minus 20%. So let's say we've got 1, 2, 3 values for that. Getting a bit heavy here. Sorry guys. Not too much. We've got three values for that. We've got three values for THAT. Is this making sense?

Three long term moving average values, three short term moving average values, and three stop loss widths. How many combinations?

27. Three to the power three.

Now, how many of those combinations do you think I want to be profitable in my backtest?

27. Why?

Right. So if I vary all of my parameters by a bit and the system still back tests well, it tells me what it's pretty robust.

But if I've done my optimization, my system design badly and I come up with my one set of parameters and that makes money in the past and I do this robustness test and 26 of them lose money. And the one that I chose makes money. What do you think I do now?

Throw it away, start again?

But the more of them that are profitable, the more comfort I get that it's likely to keep working.

It's actually doing a backtest on a back test. Yeah. Like how sensitive is my backtest?

Does that number really matter? Precisely. Or as long as it's in that general vicinity. In that general vicinity. In that general vicinity. Is it like do I expect it to make money? That's what I want. It's got to be stable and it's got to be robust. So the parameter values actually don't matter that much.

And obviously there's more than three rules in the system. Right. So I'll do this. And there could be a couple of thousand combinations. If it's a somewhat complicated system and if the vast majority of those are profitable, I'm good. Happy with that. Passes the test.

Yeah. Okay, cool. So we want a robust system. All right?

Actual trading.

Okay. So you've all got a new trading system.

I have your trading system for you here.

Okay.

Can you take them and pass it along?

So the purpose of this exercise is we're going to launch a new trading. Okay. Because part of the problem.

When are you most concerned about launching a new style of trading approach to trading, what are you most concerned about? First time, right at the beginning.

And once you're off and running and it's all natural and you're used to it, it's going to feel great. Right. But if you're feeling nervous and you don't know what to expect and you haven't got the right mindset about what this system will do or could do, then you're going to stuff it up because you're going to second guess. It's like, oh, I got two losses in a row. This system sucks.

Doesn't work. See? Losses. But the next one might have been the winner. So here's our system. 33% winners.

When you win, you win 500 bucks. When you lose, you lose 100 bucks. Will this system make money over time? Yeah, positive expectancy. Good. Here's what we're going to do.

Your dice is your trading system. If you roll a five or a six, you get a win. If you roll 1, 2, 3, 4, you get a loss. We're not going to worry about starting capital. You start with zero money. Every time you roll the dice, you'll either win 500 or you'll lose 100. Make sense? Now, you all have the same trading system, Correct.

Good. Everyone has the same trading system. Show me one dice. Good. So what I want you to do is we're just going to take a couple minutes. Now I want you to roll the dice and record the results and show your equity building over time. Don't worry about what anyone else is doing. You've all got the same trading system, so we're going to do it 30 times. If we.

And stop. Okay.

So.

What happened?

Lots of losing trades, lots of losing trades. Good. Why?

Because you weren't winning. Why not?

Why not? What's the. You got a.

2 chance to lose against every chance to win as a ratio, right? Yeah. The. The probability of win is pretty low. Okay, good. How did that make you feel? No good. When you got a string of runs losing. Okay, good. No, good. When you. Who else had a string of losses?

And a star. Good. Okay, let me ask you two questions in two ways. Who had a string of losses right at the outset? Good. How long, how many losses in a row did you get at the beginning? 4. 1. 4. Your dice must have really been broken. What else? 5. 5. 5. Good.

Actually, it was 6. 6.

Initial draw. Drawdown really sucks.

So what else happened?

Got a few winners. Good. So what happened?

Bank balance went up. Great. Whose bank balance went up? Okay, great. How did you feel rolling the dice, recording your profits and losses?

Good. That's one argument.

If it was real money, how would it have made you feel?

You would have retorted, tell me more. Well, particularly when, when I consider minus. Minus. Minus. Minus. Yeah. Okay, good.

We designed this system, we backed it, this system, extensively. We took, we did Monte Carlo simulations and walk forwards and optimization and we got a beautiful equity curve in our backtest. This system was fantastic.

First, 30% winners. Your wins are much bigger than your losses. Positive expectancy, stable over time.

What happened when you started trading it?

Lost confidence. Okay, good. What else?

Oh, good. Okay.

Yes. Yeah. Give him a hand. That was. Yeah, good.

Okay, so the point here is this. When you're back testing and designing your rules.

What time frame are you looking at?

Yeah, I mean, it might be one year, it could be five years, could be 10 years, could be 20 years. 30 years if the market goes back far enough. Can you be pretty clinical and objective when you're looking at a 30 year equity curve? Like, yeah, it makes money over time. Time. Right. Who made money in this system today? Did anyone lose money? Net. No one lost money. Good system.

But who felt like crap at the beginning?

Truly, it does. It sucks. I mean, you can, you can take the knowledge that, yes, every trade has its own probability and you can have some losses and the probability is you're going to get a loss because it's 33% winners. I mean 67% of the time you roll the dice or you put your money on the line, it's going to be a loser.

67% of the time. We're going to be wrong.

How many of you liked getting a test back at school that said gray? 33%.

You want to enjoy that? No. That sucked.

Well, yeah, actually.

I did biochemistry. No, microbiology at uni. And my final exam for microbiology, I got 27%.

And I was like, ha. That's awesome. Because I looked at the exam, I thought I got almost nothing wrong, nothing right. And it turned out I did better than almost everyone else in the course. And so my mark got scaled up and I got a high distinction. It was awesome. So the point is the percentage that you're right means what?

Not that much, but it does affect how you feel.

So one of the big distinctions when you're going live with any set of trading rules is know how it's going to make you feel in advance.

So if you know that you're going to get 33% winners, 67% losses, and you know that the probability of your first trade probability is going to be a loss and probably the second one and the third one and the fourth one, right? You all rolled the dice and you probably all got almost decent different equities at the end of the 30 trades. But you're trading the same good system.

So here's the distinction I want to make. When you go from backtesting, which has a big wide macro view of your trading world, to launching.

When you launch a trading system, when you launch into the stock market.

What are you concerned about? Did I make money.

Today?

Right. Because we're what, we're human. We were trained that way by our teachers at school, were we not? Get good marks, get them right, get the right answer, don't get the wrong answer. Have to stay in a lunch and study if you get the wrong answer.

But when you're trading, you're going to get the wrong answer a lot. And it's okay as long as you're prepared.

So if you have a system that has 33% winners, 67% losers, you're going to have to be prepared for that drawdown, that strings of losses, those initial trades that go against you and know that it's coming.

So I ran this a couple hundred times and showed all the different equity curves you could get out of 30 trades.

Just based on the probabilities. Now you have to do it a few thousand times to really see the full range of what's possible. But you can see that for a system like this, the first 30 trades, you don't really know where you're going to land.

Because the result is, is very random.

So you've got to give it time, you've got to have patience, you've got to have faith that you've done the work right, the rules are stable, that over time it'll work. Because if you poor person who gets this equity curve.

Is probably going to lose faith in the system. If they don't, if they haven't done the work properly and they don't have their mindset right, the person who gets this top green one at the top, they're going to think they're the best trader in the world. They're probably already calling out their lawyer to start their hedge fund.

True.

So our psychology in this, in launching and starting is really, really important.

That first trade, that first string of trades, that initial drawdown is something you really got to understand.

So.

There'S probably a few other differences between live trading and backtested. Backtest trading. Sorry, between live trading and backtesting.

I don't have time to go into that because I'm running out of time today. But this is a big one. The psychology of wins and losses. One trade at a time is the killer if you're not prepared for it.

Because when you've designed a system and you've looked at 20 years of data and the equity curve looks great and it's got some drawdowns, but it's making 20, 30% per year over 20 years, you're going, I'm going to be a billionaire. It's going to be awesome. And then you start trading and go, holy shit, my first one's a loss. This system doesn't work.

Am I right? But it's not. It's just random. So purpose of this exercise is what, what does it show you?

Variables, variability of performance. You're going to let the probabilities play out. And if you bet too big, if you risk too much on any one trade, if you were that trader who got five, six, seven losing trades in a row at the beginning, and you've risked too much, you're out of the game of play. You even started.

You know, makes sense. All right. Useful. Okay.

All right. So read any books about gam?

Little bit.

Right?

Right. Great. That's exactly what we're about. Now, the only difference with us is we don't know the odds in advance because it's a little bit unpredictable, but we can know kind of where the odds sort of sit based on our system and our back testing. So as I say, you've got to manage your pot or your account and make sure that you don't risk or lose too much, and you've got to make sure a string of bad trades doesn't wipe you out, obviously. Okay, good. I want to give you one tool to understand this problem.

Now, I know a lot of you aren't necessarily system traders, but I just want you to think about this because I had this challenge with a bunch of my students early on when they started trading. It's like, oh, this doesn't feel right. It doesn't. Like I'm losing money and like when am I going to make money? Well, your System's only right 40% of the time. Like you're going to have some losses. Okay, so what I devised was a method for helping people understand.

What it could look like when you start so that you're mentally prepared that yes, I could have three weeks, six weeks, six months, whatever of drawdown because of just the way the probabilities might play out. Okay, so here. Would you like to hear what I do? Yeah. Okay, great. So in your back test, let's say you started testing in 1st of January, 1990. 1990, just say, and let's just say for argument's sake, you tested up to the 1st of March, 2018. So it's a big time span in there when you design and optimize your rules. I'm not going to worry about in sample and out of sample or any of that. If you don't understand what that means, don't worry. But let's say you just designed and optimized the system over this period of time. You've got a beautiful looking equity curve. You're now ready to start trading. Okay.

I want you to know how long it could be before you break even, how bad the initial drawdown could get, how many losses in a row you could get, what the probability is you'll be up in one month, three months, six months, nine months, 12 months. So that when it happens, you're prepared so that you don't throw it away because it's broken when it's not. It's just the probabilities, the way it's playing out, does that make sense? So this is what I do.

I will start the test, the back test, this is on the 1st of January, 1990. And I'll run, let's say a three month test, a six month test and a 12 month test. These seem to be the points in time people are worrying about.

If you have different time frames in your mind that you're concerned about, that's fine, use your numbers. But on the 1st of January, after three months, where am I up to six months, where am I up to 12 months, where am I? Good. One data point. Actually three data points, one for each time period, but one data point for each duration of trading experience. Make sense? So that tells you a little bit, what if you then moved forward in Time a little bit, let's say to the 1st of February 1990, and did the same, same thing.

And then 1st of March.

And then 1st of April.

And so on, all the way to the end of the backtest.

What this shows you is if you started at all different points in time.

How your system results could have played out. I see a few forward browse. Is this making sense? Do we need to step back?

Does that make sense?

Because if I'm going to trade a system, I want to know how bad could it get before I'm rich.

Before I've actually made a profit? How long before I have a 90% confidence that I'm going to be in the money?

Do I have to wait like three years before I'm pretty certain my account is going to be profitable? Or after one week, am I pretty sure I'm going to be profitable? Who would like to know the answer to those questions before you start trading a new set of rules? Julia. Right. This is the tool. So it's, it's start date stepping.

And so you can code this in amibroker or most other systems actually. Metas. Yeah, actually, even in metastock, I think you can do this. You step the start date and you step the duration of the test. And then when you get all of those data points, you can analyze it and it'll tell you.

Proportion of the time. After three months or six months or 12 months, your account will be at different levels and you can look at it to understand how will the first 3, 6, 12 months of trading this new set of rules yield? Just like your dice.

Because what we just did, you all traded the same system, but you all had a different. What? Different start date, effectively. So if we took all of your results and collated them and analyzed them, that would tell us about the distribution of what can happen at the beginning.

So that you're prepared for what's going to happen in the future when you start trading.

This is deep, I know, but on a Saturday afternoon. But this is a game changer.

Because when you've got a set of rules that you've designed and tested and you start trading it and you go into a dip at the very beginning.

You'Re either going to lose faith in it or you're going to stick with it. And if you stick with it because. Because you know it's within the bounds of normal, you can go on and you can make the money. But if you lose faith and you quit, you give it away and you try something different.

Something even worse happens.

Because if you Try a system and you get a few losses and you lose faith and you give it up. What do you do then? You try a different system. And what happens at the beginning of that system? Same thing. And you give that away. Get another system. What happens then? Same thing, Same thing, same thing. So it's the curse of someone who is not comfortable, not confident in their approach. So if you're not confident in your approach when you launch and you can't stick with it, you're doomed. You'll never make money, never. Because you're going to keep switching and switching and switching. And this initial drawdown that happens will kill you. And eventually, a couple of years down the track, trading's too hard. It doesn't work. I'll just give my money to a fund manager.

That might not work. It probably won't. Okay. Is that helpful? Pretty deep for a Saturday, but like, this really helps me.

Yeah. Cool. All right. So that leads me to something slightly different. Now, I got about 10 minutes, 15 minutes. I want to give you a choice.

So you read a book or you design your own system. You come up with a new set of rules. You can start trading it like the dice. Roll the dice, take the next trade that comes, roll the dice again, take the next trade that comes. Every new signal you can take from the beginning.

And when you do that, it feels. Feels like you just experienced you're at the whim of the probabilities of the trades. Yes. Is there another way to start up a new set of trading rules or a new system? Is there any other way you can do it?

Okay, cool. So assume that you did that and you've chosen the one you think is best, and you're going to start trading that one. You can start taking new signals that come from that system. Right. Like we just did with the dice. Is there a different way to start that new system which you think is the best one?

Okay, good. You can paper trade or do it live. Good. So you can phase in. What else, Mark?

Good.

Okay, good. So what I'm getting at, what we're getting to, is there's a few different ways. It's not we must just start taking new trades at full size. Okay. But there's another way which is an extreme I want you to think about. I'm not recommending it necessarily, but I do want you to think about it. Because when you take. When you've designed a system, you know that every new trade will have a certain set of probabilities and you're going to start trading it. So you'll take each signal. The other thing you could do is just buy into the equity curve.

What do I mean by that? Well, if you pretended that you had always been trading this system since the beginning of time.

What positions would you hold today?

And buy in the right, right proportions.

That make sense.

So rather than taking each new signal, take your back test and just.

Implement it from then on.

Now, the reason I'm just going to leave this with you is because most people don't think of that. Most people think, I've got a set of trading rules, I'll just take the signals as opposed to implement the portfolio.

Now, what might be the benefit of implementing the portfolio? Yes. The portfolio is going to show you ones that you would have previously been in. Chart of that.

Nice. So you've been some winning trades. You won't have made any money with them because you just bought in halfway through. But that trade has proven to be a winner. That signal proved to be good. It didn't get stopped out. Okay, good. What might be another benefit? Confidence. Confidence.

Yeah. Tell me more. What would. How I would. The fact that you're confident because.

The livelihood is.

Good. Yeah, right. Absolutely. Now, when we designed, when we back tested over 20 years and we step back, what are you looking at? The equity curve. What is it that we're trying to emulate with our account? The equity curve.

So rather than just taking new signals and putting up with that initial drawdown, which sucks.

We could just buy into the equity curve. Now again, I'm not giving you a recommendation that you should do that. It's just something to consider.

You might want to paper trade for a while and then phase in. Or you might want to just take new signals at full size. Or you might want to establish a portfolio so that you don't have to take trade, trade, trade, trade. With the probability of so many of those first trades being a loss.

And just, you know, stretch your mind. I just want to kind of throw the concept out so you can stretch your mind about how to implement some of this stuff. Because every single trader that I was teaching was doing the same thing. Oh, good, I'm ready. I'll write my scan. Scan. It says buy XYZ stock, so I'll buy XYZ stock.

I thought this system was going to make me money and it lost.

But they weren't prepared for it.

Is it good?

So you have some options about how to start. The first thing you have to do is get a set of rules that work and then think about your own personality. Understand the Intricacies of how the system will play play out over time, like the dice game exercise. One of the great things about buying to the equity curve rather than taking initial signals, is that you don't have to place that many trades at the beginning. Yes, you buy the 10 trades or whatever that the system was holding, but then your account's full, right. So you don't have to take 10 trades to fill it up, most of which will be a loss. You establish the profile and then one of them naturally closes out. And then you take the next signal and you pop it into the portfolio. Yeah.

So what you're doing then is more emulating the equity curve rather than emulating the startup of the equity curve.

This is emulating the startup of the equity curve. Now you can do this, and if you're comfortable with it, great. It's easier and it's less psychological traumatic because you don't have to invest everything all at once. But you don't have to do it that way. If initial drawdown is going to be ugly, you can avoid it by just establishing a portfolio. Okay.

If you have questions on that later, you can send me an email or a Facebook message and I'll help you out. Okay, so this is. I'm a bit behind on my slides. This was all about understanding the system. Okay, so when you go from backtesting 20 of you to trading live today, did I make money or not today is what we ask ourselves. But is that what we should ask ourselves? There's a couple of questions we should ask ourselves when we're trading live today. What do you think they are?

Is did I make money today? One of the correct questions? No. Okay, great. So what do you think the correct questions might.

Did I follow my rules today? Now remember, I'm talking about if you're training systematically. Okay. If you have just. Actually, even if you're trading with discretion, you still have rules.

They're just not absolute. You know, you've got some flex in how you apply them. But you still need to ask yourself this question.

Absolutely. The wrong question to ask on day one is, did I make money trading today? Because you probably won't because the probabilities are you going to have some losses. But if you said, did I follow my rules? And the answer is yes, what do you do?

And if you lost money and you follow your rules, do it twice. Even better.

Okay, what are the other questions you should ask.

On day one.

Over time? Yeah, sure. At a point in time you want to check, do the rules still work? Okay, I Want to. Because what I'm doing is going from the 20 year backtest to today I'm starting trading. What do I have to ask myself, I'm not going to question on day one, is my rules wrong or are they broken? But in six months time, if they're not doing what they're supposed to do, would I ask that? Okay, good. What would make me ask that question? Because you're absolutely on the right track. But what would make me ask that question? Poor performance. The fact that you're not.

Ah, great. Not what I anticipated. So.

Is the performance within my expected bounds?

If you've got seven losing dice rolls in a row, is that within the realms of what could probably happen?

Absolutely. We've proved it. Darth's got it. But would you expect to have 58 losing dice rolls in a row? There's a possibility it could happen, but it's pretty remote. Right. So what could reasonably happen? What could I reasonably expect? What is okay behavior?

If you don't know that, you don't know when to question your rules. If you do know that, you don't have to stress about am I up or down today? Because it's within the expected bounce. Good.

So to understand this, you've got to analyze the equity curve and look at, ok, how many losing trades in rows? How many times is there a three, how often is there a three month period where I'm down? How long is it between equity highs? You got to look at it so you know what to expect and when you know what to expect. When it happens, you're okay with it. So you can keep following the rules. Okay, great. One last question. Did I follow my rules? Is the performance within the expected bounds? Did I make any.

No, we're not going to talk about money. Good try though. I like that. Money is the byproduct in trading. Actually, that's a really valuable lesson. Something that had affected me a lot early on is if you do the right things in the markets, what happens? We make money.

But are we making money in the the right markets? Are we making money in the markets or are we doing the right thing and money flows into our account?

We're doing the right things. There's lots of right things, lots of different ways to make money. But you don't fundamentally go out and say money, money, money. You know, you're not just grabbing money out of the market. We're following rules that put money into our account. So the first thing is follow my rules. Second thing is it's performance within expected bounds.

Third one is, did I make any mistakes? Thank you.

Okay, I'm ultra short on time. This is the only reason why in real time in your account, you will not achieve the backtest of performance.

So let's say you start trading a set of rules or a system today. And then in 12 months time, you run a back test to see what money should have made in the last 12 months. And then you look at your account, you go, huh, My backtest said I should have made $15 million this year, but I've got minus $100,000. What happened? What do you think happened?

Made mistakes because you didn't follow her.

Rules. Precisely.

You with me?

So every day when you start trading, whether you're trading already or whether you're starting a new system or whatever, these are the questions to ask. It's not did I make money? Because you do these things. Follow your rules. Make sure performance is within the expected bounds. If it's not, stop trading the system. Reevaluate if it is. You have to follow. You follow the rules and you make sure you don't make mistakes. If you made a mistake, what's your job?

Correct it. Analyze why you made it. Make sure it doesn't happen again.

Because one mistake made on every single trade is gonna kill you. If you look, if you place a market order instead of a slippage order. Sorry, place a market order instead of a limit order to get in to a small cap stock and you let more slippage into your account than you've modeled, it's a mistake.

And you're probably not going to make money even if you've got a really great positive expectancy system.

Make sense? So one of the big jobs, particularly early on until we stabilize, but even after that is analyzing your mistakes and eliminate them. I have this wonderful mistake that is still haunting me like eight years later.

Because the mistake was I didn't get out of a trade before a corporate action. There was a restructure that was going to give me equity in a stock in a different country.

So I just, I didn't read the announcements and I just. Whatever, I was busy. And so this restructure happened and I got a paper share certificate, if you can believe it, to a stock listed in Toronto.

And what do you think happened to the stock price? I couldn't sell it because it's a paper share certificate. And I'm not in Canada. I don't have a brokerage account. I can't prove to a broker over there who I am. They won't let me open an account can't sell the damn things.

And so as soon as that happened, the price of the stock started to go. Damn. It was an $11,000 position. And it just went down and down and down and down and down and down and down. I still got the damn things. I've only just managed to get an account in Australia in the name of the right entity so I can sell these damn shares that I got granted eight years ago because of a mistake that I make. So what do you think I do now when there's a corporate action? I check the details of exactly what's going to happen so that I know what action to take. Good.

So we're going to measure the right things, which is basically those in a very simplified fashion. Okay, it's three o'. Clock. I'm afraid I'm out of time.

But how many of you realize there's probably a little more to learn if you're going to progress on the system trading front?

How many of you realize there's probably a few more things I could teach you if you're interested. Yeah, so look, no pressure. There's different ways of trading. I got no hard sell. I'm not selling you anything today, but I'm going to offer you something. If you go to this link.

Bit.

You can do it. Now it's up. If you enter your name and email on that website.

Just type that into your browser or on your phone. If you enter your name and email there, I'm going to give you a couple of things. Would you like something free that will help? Yes, you can have the dice too. That's all right. But my kids want the dice back. So if you don't really want it, I'll give it to them. Okay, so I'm gonna give you a couple of things. I'm gonna give you a.

Trading assessment, series of questions you answer, depending on how you answer them, tells me a lot about how you're trading, how you can improve. So if you answer those questions, I can give you some feedback about some immediate things that you can do to make your results better. Would you like that? So usually that's about, I think $27 or something like that. I charge for that normally. I'll give it to you for free. You just enter your. Enter your name and email in that link, answer the questions. I'm going to send you an email with that link with the assessment. You answer the questions, I'll give you your feedback. Hopefully that will help you improve. Good. Second thing, I'm going to give you is I have a template for a complete trading plan. So some of you have been in the markets for a while. Do you realize that there's probably a whole lot of stuff that you've got to think about and plan for that maybe isn't front of mind?

What would you do if your Internet went down? What would you do if the US market was down 15% overnight and you still get a buy signal in a certain stock? What would you do if the power is out when you have to place your trades? What would you do if. Blah, blah, blah. All these things I've been through hundreds of books, lots of courses and all that, and I've collated the things that I think you need to think about to make your trading business sustainable. And I'll put them into a trading plan template workbook so you can go through it and improve your trading by planning in advance for different things that might be unexpected for you. Good. Okay, you have that for free, no problem. And then the last thing is because you've probably got questions and some of those questions you probably don't want to ask here. If you want to hang around afterwards, I can answer questions. But if you want to book a time with me, we can just have a chat. I'll tell you about, I'll answer your questions, I'll talk to you about you and your trading and how you can improve. And if you want some help, I can show you how I can help. I've got a bunch of programs and courses and stuff that can help you.

So full disclosure, I do have a trading education business. I do make money from that. I had a really great question earlier, like why do I teach if I make money trading, why would I give away my secrets? My trading makes a bucket load more money than my teaching does. But who realizes trading's a pretty lonely sport.

Right? So I looked around a few years ago and realized that I was trading profitably. I didn't know any profitable traders, none. And I wanted to hang out with some people. So I figured if I was going to find profitable traders, I better teach them.

So I started teaching. And so I developed an online program which teaches you how to build your own.

Mechanical technical training system. Do the back testing design, it fit your personality. Do the optimization, do the sensitivity, the stability testing, the robustness testing, all the stuff we talked about today. I developed that process to step you through the same way I did to become profitable and systematic.

And so that's what I spend a lot of my time on, helping people go through that journey. And Then for those people that actually go through the journey and really want to keep hanging out and take it to the next level, I've got a mastermind group where we meet every couple of weeks. We share trading systems. I challenge them, they challenge me, and I talk about different teaching topics at a deeper level. And Mark is part of that group. Group. And that's my favorite call of the week because they're my guys, you know, they're the girls. Actually, there's no girls in it now. There were, but she's not in the group anymore. I'm trying to get her back in. But that's my dream community because I created them, I built them, I taught them, and now they're trading and they're making money. So there's that as well. And I also have a collection of trading systems so that if. If you don't want to build it all yourself, you can get some ideas to start with and work with those. So a lot of stuff I can do to help, but I don't want to kind of burden you with any of that right now. I'm just going to give you some tools that will help immediately for free. And if you want to talk to me, I can help you with your situation and.

Tell you a bit more about some of those if they're of interest when we're on a phone call. Is that okay? So if you haven't written this down in any web browser bit ly.

Enter your name and email and I will send you some goodies and we'll go from there. Okay, thank you. I want to thank you very much for your time. Attention, participation. I had a blast. I hope you learned something and I'll see you in another meeting.