178: Nick Radge – The Blueprint: Create a Simple Trend Following System

Chat with Traders is brought to you by Trade the Pool. Did you know that every decade the market reinvents itself? Online brokers opened the doors, mobile apps made trading seamless, and commission free trading erased barriers. Now a new era has begun. Meet, trade the pool, limited risk trading. And now you also have unlimited time to reach the profit target. From now on, your trading risk is capped. and your trading opportunities are limitless.

Trade the pool funds home-based stock traders with up to$200,000 in buying power. That means you can trade larger positions and scale your strategies without risking your own savings. It's time to trade with more capital, making it truly worth your time and effort. Ready to trade the pool? Click the link in the description and join the stock trading revolution today.

Are you ready to get serious about trading? Then join Tasty Trade, Investopedia's best platform for options trading in twenty twenty six. Options, futures, and more. Tasty Trade has everything you trade all in one platform. Get low commissions, including zero commissions on stocks. So you can keep more of what you earn. Trade smarter with advanced charting tools, a pre-built strategy selector, risk analysis tools, and more features. Visit Tastytrade.com slash.

Chat for more information. Tasty Trade Inc. is a registered broker dealer and member of FINRA, NFA, and SIPC. This is the first one. Podcast.

Folks, welcome back. It is great to have you here. On this episode, for a third time now on Chat with Traders is a trader of 30 odd years, Nick Raj. Nick is a systematic trader of trend following, momentum, and mean reversion strategies, which he runs across Australian and US equity markets.

Now when I asked Nick if he'd be willing to do another podcast, I said to him, This time around though, I'd like to discuss something very specific with you, and that is how to create a simple trend following strategy. So that's exactly what we did, a complete breakdown of Nick's process. This includes details about strategy objectives and idea generation, portfolio testing and analyzing results.

position sizing and risk management, and execution in a live market. I'm sure this episode will be really helpful for anyone who's interested in writing those big market trends, regardless of asset class. And even if trend following doesn't overly appeal to you, then I think you'll still pick up a few useful things regarding system design.

Please remember this is not to be treated as financial advice, and the examples given are just that. They're examples. So do your own testing and do your own due diligence because you're entirely responsible for your own trading decisions. Now one last thing, and this is most important. I've got new edition Chatwith Traders merchandise available right now. So if you want to rep your favorite podcast on a t-shirt, go to chatwithraders.com slash shop.

Go there, get some, support the podcast. ChatwithTraders.com slash shop. And on that note, that concludes the intro. Here is returning guest Nick Raj.

The goal or the I guess the purpose of this episode would be how to create a simple trend following system. So ultimately if someone's listening to this and they go, you know, I'd like to create a simple trend following system or trend following strategy. They can go tomorrow and pretty much know exactly what needs to be done. So I guess first things first. Maybe something we should clarify is what's the objective or the characteristics of a trend following strategy? Because if someone's

not clear on that, then the rest of this might be a little bit confusing. So when you when we talk about a trend following system or a trend following strategy, um, what does that look like? Well trend following or a trend is simply price persistence in in one direction, whether that be up or down. Um it can happen, it's a fractal, so you'll see it from one minute charts to sixty minute charts to dailies to weeklies to monthly.

Um trends don't occur all of the time, but they do occur enough of the time to exploit and to profit from. Um And You know, you can see trends in every kind of instrument. We've got the uh the CTA kind of managed futures trend followers who I guess would be the classic trend followers. They trade long and short and they trade a very diversified type portfolio.

and their trading instruments right across the spectrum from foreign exchange to uh grains to interest rates to long bonds to indexes and more recently they've gone into individual stocks. Um, then you've got other more serial correlated trend followers like myself that trade long only equities looking for those kinds of trends.

So as a long only equity trend follower, my goals are quite simple. Um, for example, eighty percent of my assets are in trend following style systems across the world. And the goals are simple, um, to beat by and hold. Uh to protect the downside during sustained bear markets like we saw in the GFC. and to do so with minimal workload and using a very systematic approach. Um, you know, I'm now at the point where I simply plug my account balance into the computer. Push the button.

It generates the orders, it does the position sizing and it places the orders with the broker and that's basically it. So it's very automated. Um took twenty years to get to that, but you know, that's the way it does now. I don't look at any charts or anything like that. So when we talk trend following, all we're talking about is price persistence. And the important thing with trend following is the ability to easily

attain a mathematical positive expectancy, which is how we generate a profit. Okay? And that comes from the very simple um phenomena of writing winners and cutting losses. It's as simple as that. So we're not necessarily trading for accuracy. Indeed, most solid trend following strategies have a win rate of somewhere around, you know, the forty five to fifty percent. That would be reasonably good.

But where the money is made is where those winners really outperform the losers. Um there's no prediction involved in it and for a lot of people that's quite difficult to understand. But the analogy which I use and have been known to use for many, many years, is that of like a hitchhiker. So say for example, uh we want to travel um from Brisbane down to Sydney.

Um if we're going to hitchhike. We're going to stand in the southbound lanes. We're not going to stand in the northbound lanes because there's a no chance that a car heading north is going to Sydney. Um so we stand in the southbound lane. And what that means with with regard to trend following is we want to buy strength. We wanna we're not trying to buy weight.

a dip. We want to buy as the mo momentum of the stock is moving up, or if we're trading on the short side, we want only want to initiate a trade when the momentum is going down. The next thing with that analogy of a hitchhiker is, well, we don't know what car is going to stop and pick us up. If we stand there long enough, a car will come along and will stop and will pick us up.

And same when we're following a trend. We don't know which stock is gonna run next. We don't know which one it's gonna be, but if we catch enough of them, well, one will run a long way and that's what we're after. The next point with the hitchhiker analogy is we don't know how far that ride is gonna take us. Um we might get a ride

In Brisbane and it might take us just to the Gold Coast. It might take us to Byron Bay. It might take us to Coffsara. It might take us all the way to Sydney. We just don't know. What we do know is when the ride finishes, we just hop off. And same when we're trend following. We don't know how far it's gonna go. We don't predict how far it's gonna go.

And we just keep following it until it turns. So the key concept of trend following is to buy strength or sell weakness. We know that price tends to persist. There's plenty of academic evidence out there about that. We ride those winners for everything they're worth, and we cut those losses to create that positive expectancy, and we don't predict. There's absolutely no prediction about it.

What is it that you like about trend following inequities? Like you said earlier in your answer there. Um, there are some CTAs and some fund managers who run trend following strategies across all types of asset classes. Is there anything that you particularly like about trend following in equities? So I have a history in the both in both. I spent the first seventeen years of my life um trading Uh trend following in in commodities. That's what I did. I finished up

running a managed futures fund uh that used trend following. And dare I say this, but I moved to equities in two thousand and one because it was easier to raise assets with trading or investing in stocks than it was trying to sell a commodity fund. People hear commodities, they hear pork bellies, they hear these kind of

esoteric products and they run for the hills thinking, Oh, that's some kind of a scam or whatever else, too much leverage, you get blown up doing that kind of stuff. But with equities they kind of understand. So it was a bit of a catalyst for me to move across albeit a business catalyst. But personally, look I like the equities approach. Um, you know, you have these growing businesses around the world all the time.

Um, you know, you just gotta look at these tech stocks in the last ten years and you've got to think that these tech stocks were going s through some kind of new industrial revolution and it's going to last a lot longer than what we've seen. I am super bullish, um, the US, especially the tech side for for the next ten, twelve years.

Um and I I think that's a key a key thing. You've really got businesses that are growing and as they grow they're gonna be driving those trends for you and it's just a matter of capturing those growth parts of the market.

Okay. And one other thing we should probably just make sure everyone's clear on is that for you, your trend following systems are all um end of day strategies, aren't they? They are minimum end of day. In other words, I do not look at any intraday price action. I don't even look at charts. Um, I'd now run a couple of models that run on a month to month basis.

Um so uh let me clarify those two kinds of models if you like. There's two types of trend following, and I guess if we're going to get pedantic Trend following in its classic sense is a diversified portfolio of commodities and uh traded on an absolute basis. So there's two styles of trend following in my view. One is called absolute. Absolute trend following is when we look at a single instrument or a single symbol.

And we trade that symbol for what it's offering. So if we're looking at Apple and it's trending up, we want to be buying Apple. if we're looking at uh NVIDIA and it's trending down, or we want to be selling NVIDIA. We're looking at them on their own uh on their own merits if you like, individual stocks, whether it be cotton uh or commodities, whatever it may be, on their individual that's called absolute trend following. That's the classic one.

The other one we would call relative momentum. It's still the same philosophy as trend following in that it buys strength, it sells weakness, it rides winners, it cuts losses, okay? But relative momentum is different in that it's looking at the strength of one instrument measured directly against the strength of a basket or a universe of similar interests. So rather than looking at Apple on its own, we would look at Apple compared to all the other stocks in the Nasdaq one hundred as an example.

So we would measure its trend somehow. Usually we would measure it using the rate of change over a period of time. And then we would rank that rate of change or that. power of momentum of Apple compared to all the other stocks in the Nasdaq one hundred. And then we would buy the top five or the top ten or whatever it may be. So that's called relative momentum.

Um I put it in the same basket as trend following because it has the same return distribution as trend following. That is, you know, you've got the outliers there which is making all the money. Um and you can do that on stocks. Uh you can do either or on stocks.

Okay, cool. Well, just for simplicity's sake for this particular episode, uh let's focus on the first example, so absolute. Okay, an absolute. Yep. Yeah. Sure. Before we get into it, I will ask you, do you feel as though uh trend following strategies uh best suited for certain types of traders or investors.

Because they're I guess they're quite um slow moving, uh quite systematic, uh, like removes a lot of the decision making once you've got an actual strategy uh built. Um, do you find that There are certain people who tend to gravitate towards these strategies or who are better suited for.

Yeah, well first of all, trend following is no different from any other strategy. Okay. It has it has its pros, it has its cons. Uh you can pick holes in it, but then again you can pick holes in any kind of strategy. Um, in terms of the kinds of people that are um uh uh uh attracted to trend following I would say those that have more of a uh mathematical head for numbers, uh, engineers, um, those kinds of people that want certainty. They want to be able to back test.

Um, they you know, they generally will have a professional life, you know, a professional career that keeps them busy and they will tend to have family uh which keeps them busy. So they want some certainty. They want something that doesn't take a great deal of time. And trend following does kick uh tick those boxes.

Um, you know, as I said, I run a monthly uh two monthly models and it literally takes me ten minutes a month. And um, you know, it's it's nothing better for a busy person and you the the benefit with that is that you don't have to worry too much about the day to day ups and downs. And I think all trend following models, because you need to give them room to move,

Um the ups and downs and the noise of the market, you just you can start to ignore that. And I'm not saying it's easy to ignore that, but it's very beneficial that you can't ignore Uh, knowing so for example, if I wake up in the morning and the Dow's down two percent, well it's not really gonna impact my portfolios a great deal. Okay, they're gonna be down on the night, but so is a buy and hold portfolio. But

It's not like, oh my gosh, the Dow's down two percent. I've got to take defensive action, I've got to get out of everything and I've got to go short. You don't it's not that quick, you know? So it certainly appeals to a certain kind of a person. Uh I I guess that kind of a person would be someone who's busy, someone who's more mathematical minded, wants a bit more certainty. Um, yeah, that's as a rough guy.

I was waiting for you to say something about uh it gives you leaves plenty of time for fishing. Yeah, well, you know, uh that too. But you look, you're quite right. I mean I spent I spent five years on the trading floor of the Sydney Futures Exchange and it was the best place on the planet to to work when it was busy. But for the other ninety eight percent of the time, you just sat there

drumming your fingers, waiting for something to happen. So Um smaller markets like Australia, that tends to happen a little bit more than something big and broad in America where there's always kind of something going on. But, you know, as far as my life is concerned, I've got better things to do and watch numbers roll over a screen. But that's that's the way the markets go, you know, everyone's different. Um it it wouldn't be it wouldn't work if everyone was the same.

Absolutely. Okay, well let's let's start getting into some details around this. So I guess the first step would be Coming up with an idea. So how do you come up with an idea? How do you come up with your strategy rules for how you're going to catch a trend? Well, how long's a piece of string? At the end of the day The whole concept and that goes for any kind of strategy, not just trend following, is, you know, your beliefs about the market.

Um, so for example, if we if we think about the beliefs someone like Warren Buffett value investors, you know, they're looking for great companies at at a bargain price. That's their belief, that's where they think they make money and that's fine. But with trend following our beliefs are that every trend starts from a breakout. So

Um if prices are going to trend high, well, they have to keep they have to be making highs, right? So the first thing to look at if you're building a trend style strategy. is some kind of a breakout, which when we think about it, that's exactly all the turtles were doing, you know, thirty, forty years ago. Um, they were just trading a a breakout because their philosophy was very simple. Every trend must start With a breakout.

So that would be the way to start, looking for some kind of a breakout. Um So my recommendation as a starter would be looking for if you're using daily charts, looking for a longer term breakout. A lot of people say we'll just divert back here a second. A lot of people tend to say that, you know, the turtle method doesn't work these days. Um

I'll I'll prove that wrong by the end of this podcast. But a lot of people say that because they're still looking at the old turtle rules from thirty, forty years ago. But markets change over time. And if you listen to some of these guys, and some have been on your podcasts in the past, if you listen between the lines, you can hear them actually telling you that the markets have changed.

They're more or less doing the same thing as what they were doing 30, 40 years ago, but just a slight variant of that. In other words, they tend to be trading longer-term breakouts rather than the shorter term breakouts, which they were trading back then. So let's use an example. We could use uh say a 100 day breakout rather than a twenty day or

one of, you know, a well known system that I created uh about seven or eight years ago, which is very, very popular around the world and traded on a lot of markets, is the weekend trend trader. So That's a weekly breakout. It uses a twenty week breakout um as a trigger point. So that would be your first port of point of call or point of reference, a breakout.

So a twenty week breakout, what does that mean? You've just got to close above the highest high of the past twenty weeks? That's right. So we would look back For the last twenty weeks that we would only do this at the end of each week. So this would be deemed w a weekly system. So the close of business on a Friday as an example. we would say, right, what stocks have broken out of their twenty week high?

Now you probably wouldn't want to keep that um as basic as that. Okay. So again, let me sort of digress a little bit here because this came up the other day. There was some uh discussion and uh an article written on Twitter which I don't particularly agree with and it revolved around the fact that some people claim trend following is actually not a very good strategy. Um But my argument was they were using a very, very simplistic bedroom kind of mechanism.

Um and look, simple things really aren't going to work that well. They'll still work. But they're not going to work that well. I think you've got to be a little more sophisticated than say buy when prices cross a 200-day moving average, right? That's really not going to hack it. So I would say the same for something that breaks out of a twenty week high. We that would be very basic.

Um and you'd want to improve on that. The easiest way to improve on something like that would be to add a confirmation filter if you like. All right. So Step number one, we'd want to see that twenty week breakout. Step number two, we want to see some kind of confirmation. And a good way to see uh have confirmation is the rate of change over those twenty weeks uh is above a certain level. So we're not just looking for a gradual

increase um over the 20 weeks, we want a reasonable rate of change for that period of time. So it's some acceleration. So what we could do as an example is we could use the rate of change over those 20 weeks and make sure that when we got a 20 week breakout, that rate of change was above a certain hurdle. So let's say the rate of change was above 30. Which is what we use in the weekend trend trader strategy.

All right, so you basically got two entry criteria. You want a weekly breakout and you want the rate of change for the last twenty weeks to be above thirty. That's a bit of a confirmation. Because that way it gives us a strong breakout. Okay, we don't want weak breakouts, we want a strong breakout. We want an explosive move. How are you measuring the rate of change? And you mentioned thirty there. Is that what is that, thirty percent or thirty

What's the unit of measurement? Um so rate of change is simply the percentage change price on price for the last twenty weeks. And yeah, thirty would be thirty percent. Okay. Okay. Gotcha. So all basic all basic technical analyst um platforms will have a rate of change indicator in there. And it's simply the percentage move of the price over the last 20 weeks. So we just want to set it up to a hurdle. So we'll set that hurdle to thirty as an example.

So twenty week breakout, if the rate of change is above thirty, then bang, we've got a buy signal right there. Mm hmm. Now, as you've been doing this for many years. And I'm sure in that time you've tested hundreds, maybe even thousands of different ideas. Got a big graveyard back there. Is there anything that you've learned or kind of stands out to you about

Entry rules. Because obviously When you when a when a new trader is trying to think about how to create a trend fine strategy, they're probably trying to think about how to get the the magic combination or like You know, how to define the best breakout. Having tested so many different entries, is there anything you've learned from h running so many experiments? Well stop looking for the best breakout. It doesn't exist. All you want is a breakout. The money's not made by having the best breakout.

The money is made from riding the winners for the long haul and cutting those losses very quickly. That's the key right there. We could use any kind of a breakout essentially. I'm not gonna go down the path of saying a coin flip, but At the end of the day, every trend, as I said, and as I I'm only repeating what the turtles said forty years ago, every trend

starts with a breakout of some type, i.e. a price movement above a recent high. Okay? Whether that be a fifty two week high, a one hundred day high, a twenty week high. Um, I'm doing some um uh work at the moment for one of my talks um at the upcoming conference.

Uh, because one of the things that's been thrown at me lately, I've seen it twice, I think, in the last three days on Twitter, is when a stock hits an RSI of ninety, well, it's got to be a sell. So let's put that to the test, which I went and did. And for the first three days, you're quite right, the stock tends to pull back when the RSI. But if we go and test that and hold that out to

one hundred days, the other ninety-seven are all profitable when you buy an RSI when it goes when it hits ninety. So again, an RSI hitting ninety is a form of positive momentum. Um, my main retirement account that I uh have been trading since the late nineties, long only equities in the Australian market.

It uses a Bollinger band breakout. Nothing special there, you know, but it's still a breakout using some volatility filters. So The key is I don't think there is any special type of breakout that is better than any other, ultimately. What makes the money is not the breakout. What makes the money is holding on to those winners and riding them and then cutting the losses quickly. Um that that would be the core ingredient. I I think too many people um

use too they think they think it through too hard. You know, one of my mentor students, for example, just this week. She sent me her first attempt at a momentum style model. And there was about four or five parameters in there, things that seemed logical to her, and the performance was quite terrible.

And I said, Let's strip all this garbage out. Let's just go back to simple stuff and and start with simple breakouts and see how we go from there and she's now building a very good model on that. So I I think the key is not to keep trying to find the holy grail. It doesn't exist. I think the key to successful trading is apply a very simple method for the long term. Most people can't do that.

Are you ready to get serious about trading? Then join Tasty Trade, Investopedia's best platform for options trading in 2026. Stocks, options, futures, and more. Tasty Trade has everything you trade all in one platform. Get low commissions, including zero commission on stocks. so you can keep more of what you earn. Tasty Trade is packed with advanced charting tools, backtesting, a pre-built strategy selector, risk analysis tools, and more features to help you trade smarter.

See equities and derivatives with high trading volumes, dividends, upcoming earnings reports, and more with their pre built watches. Or create a custom watch list to keep an eye on the companies and sectors that matter to you. Manage your positions with speed and precision using Active Trader Mode one click trading and smart order tracking. Plus Tastytrad's Stellar Trade Desk team offers live support during trading hours if you need it. Visit Tasty Trade dot com slash chat for more info.

KC Trade Inc. is a registered broker dealer and member of Finra, NFA and SIPC. Have you ever watched a stock explode and thought, if only I had the capital? Or sat on the sidelines because your account balance felt too small to matter. Good news. With Trade the Pool's limited risk platform, you don't need millions or even thousands to start trading the U.S. stock market. Bypass the PDT and tap into over 12,000 U.S. listed equities.

From penny stocks to big caps, ETFs, even the newest IPOs, and short anything you like, with zero locate or hard to borrow fees. Start your evaluation, get funded with up to$200,000 in buying power so you can go big without risking your own savings. And now you can also have unlimited time to reach the profit target. It's a game changer. Not ready to trade yet? Trade the pool offers a free demo and educational resources.

Practice on live data, master the platform, and build confidence risk-free before you even pay a cent. Click the link in the show notes to start trading with Trade the Pools Capital.

Okay, so perhaps I should flip that question a little bit. Ask you about trailing stop losses because ultimately that's how a trend following strategy operates. Buys the breakout and then there's a trailing stop loss in place um and you just keep I guess moving that up until it ultimately gets hit at some point in the future. So just before we go there, can we just step back one second?'Cause I think we've just missed an important step there.

We've got the twenty week breakout, we've got a confirmation filter which is say a rate of change over the last twenty weeks has to be above our hurdle, which we've set at thirty. The another key ingredient here is when we're trading stocks on the long side, we need to ensure that the broader market is also trending up. Okay, that would be called a regime filter. and all my research, without a shadow of a doubt,

suggests that you should only buy stocks when the broader market is trending up. So if we're talking, if we're if we're trading, say the Russell one thousand. We want to ensure that the S P five hundred or the Russell 1000 itself, the index itself, is actually trending up. When the index is trending up, it puts the odds in our favor. It's like

you know, a rising tide will lift all boats. A falling tide will tend to drop all boats. You know, look at the GFC, yeah, there might have been a couple of stocks that went up in that time, but not many. So adding a regime filter is a key ingredient to putting the odds in your favor and greatly enhances the risk-adjusted return of all strategies. So a regime filter again would be something simple.

Um, it would be we're only gonna t take buy signals if the S P five hundred is above its two hundred day move moving average, as an example. And that will work wonders. It will make a big difference to the bottom line.

Okay. So um you the part about uh the trading stop loss. Yeah. So the trailing stop and this is a this is a tricky one because there's a lot of information out there that You should have your stops uh set at eight percent or five percent. Um but you wanna follow a big trend. You wanna catch a big trend. You're never gonna make money.

By taking small trends. You know, you want to make the big money and follow the big trends. That means you have to give yourself or the positions some room to move. All right. The markets just don't go straight up. They zig, they zag, they congest sideways for a while, and then they go again and then they come back a little and then they go again. If your stop, if your trailing stop is too tight,

you're gonna get knocked out of those trends just by purely from the noise of the market. No, I would have to say that the markets are getting noisier and noisier and noisier. I know for a fact that my trailing stops over the years have got wider and wider and wider. And that's just to keep you out of the noise. Again, I've done research on this. I did a whole webinar on this. If any of your listeners are interested in it.

But we recognised that a trailing stop somewhere in the vicinity, the I think the optimal number is actually eighteen percent, but a trailing stop that's runs about twenty percent behind the position or the recent closing price will allow you to follow long term trends. Just on the surface, that sounds like quite a wide stop. Twenty percent. Sure. So let's put it into the context of a portfolio though. So let's say we're going to have

Uh let's keep it pretty simple. Let's say we're going to have twenty positions and we're going to allocate five percent of our capital to each one of those positions. If we run a twenty percent trailing stop and that gets hit Then

your loss on your total portfolio is only gonna be one percent. All right. So to give you an idea, my main trend following strategy that I use for my retirement account, let me just pull up the numbers here. I should have had them uh Let me just pull them up here just to give you an idea, because that's a very common question.

And some people say, oh my God, you know, you take you lost twenty percent. Well it's not twenty percent of the whole portfolio, it's twenty percent on that position. And that's only if that stop actually gets hit, right? Uh we are buying strength. Strength does tend to persist. Um but sometimes, yeah, we can buy the absolute top. I've certainly done that. No no dramas there, but you know, you do this long enough, it's gonna happen.

I guess my line of thinking around this was you have your entry and then you've got your stop about twenty percent below. That's kind of your risk on the trade, right? Yep. With such a wide stop And you have to spread your risk along between that. it means that your position size is obviously not as big as if you could get a tighter stop on it. So, um, you know, you need I guess you need to make sure that when you do catch those winning trends

um that you can catch a large enough move to justify having um quite a wide stop. Would that be correct? You're probably overthinking it a little bit. I understand where you're coming from and it may well be that the argument for a tighter stop could be made if you had a much more accurate entry mechanism. If you had a much more accurate entry mechanism if you like, then you could certainly may have a tighter stop.

Um, but invariably what that's gonna do is um, especially when the market's g noisy, is chances are you're gonna get knocked out more often than not, you know. So I'm just gonna read some real statistics. These are my actual statistics. throughout the last four or five years, the average trading loss, right, on position, single position loss. for me is eleven point nine seven percent, which roughly is what, point six percent of point six percent of total portfolio.

Okay. And that's nothing. You know, you we hear the old two percent rule. Well, I'm risking point six of one percent. So, you know, those those twenty percents, look, they do come along, but not as often as you would think. Um, and to give you an idea, my um average win is 31%. So two and a half, two point six times that amount. And there's that edge I was talking about, right? My win rate is only forty four percent.

But my win loss ratio is two point six. And that's exactly how we get a positive expectancy. I'm outperforming the market over the last four or five years by, you know, a considerable margin. Um so it's about giving it enough room to move so you can ride those big winners when they come along. Um so it's hard to it's a hard to conceptualize.

Um, but as I said, my goal here is to outperform buy and hold. And w where's your stop loss with buy and hold? Well, you don't have one, right? You're buying and hoping. So, you know, we've got to put it into the context of what I'm trying to do.

But this is proven, you know, it's it's you can mathematically prove this. Um these numbers we're talking about or the system we're trying to build here, I'll actually run that through. I've got it coded up. We'll run it through and we'll look at these numbers at the other end.

Um but here's a little twist to that. The trailing stop is or could be dynamic. So let me explain. Assume... We use a uh a generic twenty percent trailing stop loss, right? One would assume When the market does turn and head lower, that all of those stop losses are gonna be triggered and each therefore each position is gonna give back twenty percent, correct? Mm-hmm. So a more dynamic approach to that.

Is we spoke about that regime filter before. When that regime filter goes from uptrend to downtrend. Rather than using a twenty percent stop loss on our current positions, we ratchet that up to ten percent. So what's going on there is we're leaving the door open in case we're just seeing a period of consolidation or a small pullback rather than a full on bear market, we're leaving that door open.

For those trends to continue or consolidate. But rather than giving back 20%, we're now only giving back 10%. Does that make sense? It does make sense, yeah. So we've got this dynamic stop loss procedure going in place that Is allowing us, and here's one thing I've learned over the years, okay? When you're following strong stocks and the market tends to have a pause or a pullback. those strong stocks will tend to just consolidate. They won't pull back very far.

And they are the ones that are going to take off again when the market starts to head higher. So we've tested things like a time stop where we'd say, right, oh, if this position hasn't moved X percent in N days, then just cut it and go on to something else. What we found that was actually detrimental to performance because we would be in the leaders, and the leaders wouldn't pull back, they would just consolidate.

It might be for two weeks or three weeks, but they wouldn't pull back too far. And the broader market would pull back a little bit, but the leaders would consolidate and then they would pop again. So the whole idea is you don't want to get out of those leaders. You want to stay in there. So by moving that stop loss up to 10% from the recent high. What we're doing there is leaving the door open.

And sometimes we see a lot of this. Strong stocks keep going up, even if the broader market is moving sideways, and that's what we want. So We're taking some defensive action by halving our trailing stop loss and locking some of that in, but we're also leaving the door open in case we're just getting a brief pullback or a brief consolidation period.

Okay. And just going back a couple of steps, um well not a couple steps, a couple of moments, um, you gave some stats. Were those stats what account is that? Is that your sort of longer term um Uh uh retirement account. Yeah, this is my growth portfolio. Uh this is a portfolio I actually started trading futures with back in the late nineties, not as my retirement account, but the same strategy.

Um and it hasn't changed much since then. A and which strategy is that one? Is that the Bollinger Band breaker? This is yep. This was d outlined in my book Unholy Grails, which was written in two thousand and twelve. And this is the Bollinger Band breakout strategy within that. There's a couple of little special Raj tweaks in my own personal one, but it's more or less the same kind of strategy. Okay. So if someone wants that strategy, they gotta get the book.

Look, y you can get the book on on Amazon, but really if you just Google it, um, Google Unholy Grails plus um Bollinger Band strategy, you'll you you'll see a ton of people are out there are using it and putting it to the test. So you don't you don't out there I say you don't have to buy the book. There's eight different strategies in the book if you want to look for some different ideas but

That's the main one that I trade and have traded for twenty twenty plus years. Okay. Well I actually do have a copy of the book and I would say it's um it's worthwhile. It's definitely um it's good to get some sort of ideas for So some strategy ideas that you can actually test'cause there's hard and fast rules in there that can be tested. That's right. And look, for a certain type of person coming up with ideas

um, you know, i is not easy. So it certainly hasn't been easy for me and that's why I continue to read and and look through, you know, certain journals and that kind of stuff because there's a lot of smart people out there doing very different things to what I do. And you can always learn something new. One of the little tweaks in this portfolio, which we added in, I think in about 20 twenty twelve, um made a big difference. To the bottom line.

And that was something I learned from somebody else. So you can never stop learning, you know, and I've been doing this for almost thirty five years and still, you know, things pop up and I can test them out and say, Oh yeah, well that's that's okay, but it's not really making a big difference or that makes no difference. Whatever it may be.

Um so yeah, that's right. Yeah. Okay. Well let's get into that. So I guess this is the next step. We've spoken about how to sort of come up with ideas and and different things to consider. How do you actually go about testing your idea? Like where do you begin? What's the first step there? Well ideally you want some software to do that.

Um, personally I use Amy Broker. It's off the shelf, it's cheap, it does very good portfolio testing and that's important, okay. We want to test on a portfolio level. So uh Amy Broker is does the job for me. There's plenty of other kind of um software packages out there. Um you've got Trade Station, you've got multi charts, um, you know, a lot of people use Excel.

Um can I just interrupt you, Nick? Sorry. Um when you say it's important to test on a portfolio level, can you just explain what that means? So A portfolio level means we want to look at a universe of symbols. Okay. So for example, my retirement account it trades the All Ordinaries Index, which is the same in Australia as the S P five hundred. So there's five hundred stocks. We do not want to test individually on each stock, okay, and pick out the best ones.

um because personalities change over time. So what may have been a very nice trending stock um for the last ten years may not necessarily be the same trending stock in the next ten years and a lot of people fall into that trap. So what I wanna do is I wanna take the whole universe um of S P five hundred stocks or all ordinary all ordinary stocks. And I want to test them all together.

In other words, I want to say to the computer, I want uh twenty positions at any one time and I wanna allocate five percent of equity and I wanna run this test from nineteen ninety nine through till today for the last twenty years. Um and it will do that for you.

There's a couple of interesting things you just said. Um one of those is testing across twenty years. Is that something you would actually do or would you how would you segment your data to have in sample, out of sample and you know, going back twenty years, is that like a um is that something that you would actually do? Yeah, absolutely. Um so we're starting to get into some deeper parts of backtesting. Um In terms of in sample, out of sample, that kind of stuff, I'm gonna go out and be um

I'm gonna go out on a limb here and some people aren't gonna agree with this and that's fine, I don't really care if you agree or not, I'm just telling you the way I do it. But

I've been doing it too long, it just doesn't bother me. It's not gonna change my opinion. Um it's not gonna change the way I do things, but it's worked okay for me for a long time, so I don't think there's too much wrong with it. But Generally speaking, generally speaking, when people go into the depths of in-sample, out sample, they tend to be more orientated to optimization. Okay. In terms of optimization, we do it to see the robustness, but

At the end of the day, all my systems use pretty linear kind of input parameters and not many of them. For example, let me give you an idea of a linear input parameter, 200 days. 100 days, 50 days. Okay? So for example, we talked about a regime filter before. We'll use two hundred days for that. Is it the optimal one? Probably not. Um If we have a look at a 20-week breakout, well, that's effectively um 100 days. So it's half the index filter length, right? So we're keeping the numbers linear.

If you come to me and say, Oh yeah, well I'm gonna use seventy six point seven days for my breakout and my e regime filter is gonna be um a hundred and twenty seven days. Well, to me that's optimization and it's not going to stand up in the future. What we do want to do when we optimize is to check the robustness of that parameter. And that that's, you know, we're going probably too far ahead at the moment to get into that. Um so You can Choose a small window of time.

Um, by all means. And you can choose a longer window of time and then optimize over a smaller window with the theory that the markets are or you're tuning the system to more recent data. So say for example, We run this test over the last 20 years and it stands up. then we're probably onto something reasonably robust. There's not too many parameters. We're not optimizing any of the parameters. We're using linear kind of numbers, 200s, 100s, that kind of stuff.

Uh and they're gonna stand up a lot better. We've got to start somewhere, right? We've got to start somewhere. Um so a two hundred day as a regime filter is a perfect place to start. Yeah. I mean that all that all sounds very reasonable.

Um, one of the other interesting things that you said which I wanted to pick up on was uh limiting to twenty positions. Uh is that something you also do? Yes. And we do that for a number of reasons. Um and we use for these portfolios um we use fixed percentage allocations. In other words, um five percent of your account.

um, you know, will buy five percent of this stock. Okay. So for we do that for a couple of reasons. It's simple, people understand it. Um And if people understand it, they're more likely to to do it. Now, I'm gonna get an a uh an opinion here, I have no doubt in the comment section, that someone says, Oh, you have to volatility weight your position.

Well, go and test it, because from my research, volatility weighting doesn't actually make a great deal of difference to the bottom line from doing something simple like this. It doesn't necessarily make the equity curve any smoother. It doesn't necessarily make the drawdown any less. It doesn't necessarily make the return any more.

So before you jump all over me and say you got a volatility match Well, go and test it for yourself because my research suggests it actually makes not much difference. And I think if you find equal weight indexes out there perform significantly better than cap weighted indexes, if we could kind of call it one and the same type of thing, which we probably can't, but

That's my point. So test it first. Don't think that that's the way it's gotta be. It sounds logical, but doesn't necessarily test out like that.

Okay. Now when you are limiting to twenty positions or a fixed number of positions How does that work? So let's say you've got uh let's say you've got eighteen open positions and then you have three signals um the next day. How does your your system determine which signal to take?

All right. So that would be called selection bias. And that's when you have more signals than in available cash about um cash available. Okay. So that doesn't happen. We talked about those two types of uh trend following strategies, absolute and relative. That doesn't happen with relative Okay, because they're all rebalanced on a regular basis, whether it be weekly, fortnightly or monthly in my case. So that doesn't happen, you don't have that selection bias problem. With an absolute

trend following strategy such as the one we're discussing here, yes, that would be a problem, but you would add a ranking mechanism in there. And again, the ranking would be something pretty simple like the rate of change. over the last period of time, say the last 20 weeks. So when you're presented with three new signals and you've only got two slots or you've only got one slot, then you would simply take the one with the highest ranking on that particular day.

That doesn't suggest in any way, shape or form that that one is going to perform any better than the other two. but it gives us a mechanical way um from which to make that selection and from which we can back test accurately. We always want to ensure when we run a back test. that we're always getting the same back test, we're always getting the same signals. So by using that rank of change, rate of change to rank those signals, we're always going to get the same back test results.

Okay. And I imagine if you're trading a universe of about five hundred stocks, uh, like you'd mentioned, uh

You're probably always gonna be in about twenty positions, right? Well, not necessarily. I mean, for example, it just depends on the market. For example, in the Australian market, I'm only fifty five percent invested at the moment. And there's a reason for that. If I was in a different part of the market, I only trade small cap industrials. So the small cap industrials are only industrial stocks that are outside of the ASX 100 and with inside the ASX 500.

Okay, that's that's all I trade, and I trade that for a reason, which is beside the point here. Um, but if we go and look at the top. weighted or the top cap stocks, the ASX one hundred, they're going gangbusters this year. They're having an absolute cracking year and significantly outperforming what I'm actually doing at the moment.

Um the reason for that, markets obviously somewhat cautious. Generally speaking, after a a significant correction or period of volatility, generally what happens, people move into the safer stocks, the big names, the names they can see out on the street. and they leave the smaller names, the ones I tend to trade, they leave them. So my guess, and it's purely a guess from my experience, is that after Fourth quarter twenty eighteen.

Um, most market participants um are steering clear of the higher volatility small cap stocks like I'm involved in. and they've been solely focused on those um larger name caps, which tends to mean we're still early cycle in this bullish run because usually The risk moves out the curve and that hasn't started to happen yet. So I'm only fifty five percent invested.

Um, if I was in the top one hundred, I'd be a hundred percent invested, no shadow of a doubt there. So when you're saying you're fifty-five percent invested, does that mean you've got eleven open positions? Correct.

Okay. Interesting. Now, when you're thinking about your universe that you're gonna trade across, Are there any other conditions or things you like to consider uh that kind of defines which stocks you're actually going to trade uh beyond just saying this stock is in the all ordinaries or the S P five hundred. Like do you have uh any sort of price filters or volume filters?

Yeah, absolutely. So for example, um The Australian market is quite small since the GFC liquidity has disappeared significantly in the Australian market and volume filters are certainly very, very important, especially if you have a decent sized account. Um I use a average volume filter. um over the last uh fifty days, I calculate the average. And I want to ensure that my order size

is, you know, not going to be any more than two, three percent of the whole daily average volume over that period of time. So that will leave me out of certain types of stocks. Um and that's a safety thing, you know. It would be irresponsible of me to throw a large water money into a completely illiquid stock. Not only'cause I'd move it myself.

But if I want to get out in a hurry, um, you know, there's gonna be a lot of slippage involved. So when uh I'm trading this particular universe, I do so because It has the highest volatility and the highest return. Um, this is all I do, so I'm happy to be aggressive. And there could be an argument that if you weren't as aggressive as me, you would trade perhaps fifty percent of your portfolio in these small caps and fifty percent in the large caps, as an example.

So you'd have to look at your own risk return. But again, the the technology that we have these days, you can test certain parts of the market. You know, you can test the Russell one thousand, you can test um The as I said, I can go and test all the resource stocks in the ASX on their own. I can test the large caps, whatever portion of the market I want to test. And let me also suggest

that in this day and age we can also test uh historical constituents as well. Okay, so survivorship bias is a big problem. And survivorship bias basically means, let's say, for example, we talk about the SP 500 today. The SP 500 today contains 500 stocks, okay, and they are the 500 that are currently trading today. But historically, there's been 1788 stocks in the SP 500 at one time or another in history.

So if we were to focus on a single universe, say the SP 500, and we were to go back 20 years and test that. Well, we would want to be testing the specific constituents of the S P five hundred twenty years ago. Not what they are today, because today is different what they were twenty years ago. And we can do that kind of stuff today. So when we test historical constituents, we're actually testing the universe as it was at a point of time in the past.

And it makes a big difference to the bottom line. Yeah. I remember you uh speaking about survivorship bias on the last podcast we did as well. Um so

I'd suggest anyone listening to this also goes back and and hears that episode. Uh episode sixty four. So it was a while ago. Yeah. Yeah, yeah, yeah. Yeah, okay. Okay. So So my original question on this uh on this area was how to test your idea. Short answer is get a good software package or some good skills in Excel. Yeah, look the problem with Excel is um if you wanna make a change it's i it can't be done very simply, you know. Um

Look, uh there's some great software packages out there that don't cost a lot of money and if you're not willing to pay for it, well, you gotta kinda ask the question, are really taking this seriously? You wanna make money or you wanna save money? Because It's like any business. You've got to invest some money into it. You know? Um you've got to invest money into decent software to get the right answers, because otherwise you're just going to be guessing.

Uh and you're gonna invest some good money into some good quality data. And that's very, very important, especially with the likes. of historical constituents um and all the things that can happen with stocks. You need really good quality data to go into that and that way you can be assured your testing uh will be accurate because you know the old you know the old IT um

uh saying, you know, garbage in, garbage out. It's simple as that. If you if you code with post dictive errors, well, you're gonna get a wrong answer and you're gonna waste money you're gonna lose money on that. You've got to be careful. You've got to run it as a business and you've got to do it properly. Do you think it's also likely that kind of your average Joe may need some assistance from uh a programmer?

Well it depends on the software package, I guess. It's like anything, right? It's like taking up golf. The average Joe can go down the golf course and hack around golf balls and maybe in five years they'll get the good hang of it. Or they can go to the local Golf Pro

get a few lessons, get a grip, get a bit of a swing going, and within six months, you know, having a much more enjoyable time. It's like anything. It doesn't just mean trading. Tennis, coaching, Uh heck, I've paid for people to take me out fishing to to show me the tips and tricks on, you know, putting an anchor out and doing all of kinds of things because I can go out there for the next six months, twelve months and bang my head against the bow of the boat, but

you know, you get those shortcuts very, very quickly and you can't buy experience. I mean, you you just can't experience has to be learned over time. So if you go to someone that has a lot of experience in doing this kind of stuff, you're gonna get an answer very, very quickly and you're gonna save a lot of time and essentially a lot of money. Um, you know, I have a lot of people come to me and they've spent

um minimal amount of money trying to get things done. And some people are going around in the same circles for two years. Um it's a lot of time, especially if it's a strong bull market. You wanna be you wanna be invested in those periods of time. Um So there are services available. I mean, we do all that kind of stuff. We have a full on six month high end mentoring course, which we teach you how to program right from scratch all the way through.

to system design, build, testing, stress testing, and then implement implementation.

Um we do custom coding. So if you've got an idea that you want coded up, you don't can't be bothered to learn how to code, well we can do it for you. Um and there's other people out there that that do that kind of stuff as well. So Um, you know, uh in my view, having a good mentor that knows what they're talking about can save you a ton of money and save you more importantly, a ton of time.'Cause at the end of the day, everyone seems to be very busy these days without much time.

Um and sometimes a little investment goes a long way. Yeah. I think uh learning to program, as I've said multiple times on this podcast, is a very valuable skill. And uh I've done a couple episodes in the past actually which sort of uh give some insight for how to get started on learning to program as well. So I'll link to those in the show notes. Um so if you're interested in that sort of thing, might be worth worthwhile listening. Um because funny enough, I've actually uh

I actually coded a a custom back test of my own. Um, and I've actually tested some of the strategies in your book, Nick, uh which actually got me interested and and gave me the idea that we should um tee up and do another podcast together. So

Um yeah, it's it's been a it's been a fun experiment. Yeah. Now the next part is how do we actually analyse these test results? So I guess what are some what are some things which you look for? What are the checks and measures that you want to see in order for a strategy to pass? Sure. So obviously the key thing that most people look at um in my 34 years of doing this.

I've not met one person that doesn't like making a profit. I've met a lot of people that hate losing money, but not many people don't like losing um making a profit. So the key there is not to just focus on the return. the key is to focus on how you get to that return.

Um so for example, people tend to get dissuaded very, very easily. I've had people trade, you know, my growth portfolio, for example, come back after three months and say this is shit, it doesn't work. because it just doesn't suit their personality or they don't understand what's going on or they don't understand

how you're traveling the journey, you know. Um and again we can use an analogy about crossing a city. You've heard me use this one before. You know, s let's say we're traveling from Balgala and Sydney to Parramatta, you know, that's a fair hike across a big city. uh you're gonna run into problems. You're gonna run into red lights, you're gonna run into traffic jams, you're gonna run into road works, you might have to take a detour, but if you keep at it, you're gonna get to your destination.

If you get to the first red light and you've got the hump and you turn around and go home, well you're never gonna make it to your destination. And it's no the same for trading, okay? There's gonna be ups and downs and and whatnot. So the better you can understand the journey

the more likely you're going to participate for the long term. And that's what I said right at the very start. I think that a big difference between a professional trader and an amateur, is a professional trader is someone who's able to keep on keeping on when things get difficult. And this is one of the reasons why we started the mentor course. I can hand people a trading system. You can do that. There's plenty of them available.

The ability for most people to stick with it over the long term is quite difficult simply because it's the unknown. you know, they'll have five or six losses in a row, or they'll have three months where their account maybe goes backwards. and that pushes them over the edge. So if you can run an accurate back test going back twenty years, it can give you a very high level of confidence, especially if you've built the system from the ground up.

And a lot of our testing that we do is actually stress testing as well. We try and break this system. It's not just necessarily coming up with the best result. It's how can I stress test this and try and break it? If I can't break it, well maybe I'm actually onto something really, really good to give you an idea of a stress test, right?

So in our example that we've been using a 20-week breakout, let's say we randomly uh adjust the closing price on Friday at the end of the week there by a random amount between half a percent and three percent. And then we run the system again. If it's still profitable, we've got this variance of price going. Does that make sense?

Yeah. So we're kind of you know, the future is never gonna be exactly the same as the past. It'll be similar, but it's never gonna be exactly the same. And a lot of people say, Yeah, well a back test it's it's history, you fitted the data, yada yada yada. But it's better than having nothing. And if we can then adjust that history by randomising, for example, the closing price by half a percent to three percent on a random level.

and still run it and it's still profitable, well, we've introduced a variable to that backtest that now makes that historical data slightly different and we're still profitable. Um, so they're the kinds of things we can do to that, um, to get an idea of of robustness. You know, it the more that you can add to these uh tests like that and stress test and if you still come up trumps, well you're onto something. As you said, the past returns are not really gonna

Um you know, future returns are likely to vary from past returns. Correct. How much variance would you expect? Again, I think people slot system traders into a little hole and say, because you don't know, then we can't do it. Well, the same is true for buy and hold. You know, we just had a decade long period of time from two thousand to two thousand and ten or whenever it was, the lost decade in the US market. Now

Tell me anyone in 1999, buy and hold, that would have expected 10 years of sideways returns. It just doesn't happen, right? You're never going to know. Um, if you do your back testing correctly and you do it robustly and you stress test it and you go back in history a fair way, um then there's a higher probability that what you've done is going to continue to work to some degree in the future.

It's never going to be exactly the same. It could be a little bit better. Who knows what's going to happen in the next ten years? We could have a gangbuster. 10-year run, we could go sideways for another 10 years. I don't know. But so long as the strategy is robust, you're going to be much better off than having no strategy whatsoever.

We can't possibly know what's gonna happen. Okay. Um, now just before I asked you the question about um, you know, what checks and measures do you require for a strategy to to pass, um, you spoke about kind of the journey from

beginning to trade uh to the the finishing P and L figure, you know, can you stomach the drawdowns and can you actually ride the journey to get there? Is that realistic? Are there any other metrics which you like to look for? I know you said earlier on that uh your win rate is s somewhere between forty to fifty percent. That actually seems quite good for a a trend following strategy.

Um, you know, d is that one of your prerequisites? Is a win rate above forty percent or uh, you know, a profit factor ab above a certain level? You know, are there any like hard and fast metrics that you're doing? Yeah. Good questions. So Again we can talk theory and we can talk reality, okay? In theory, you and you'll see it banded around.

Um, you can have trend following strategies out there with win win rates of thirty percent that are still profitable. But I'll give you the tip from my experience. Once that win rate starts to go below forty percent, it becomes incredibly difficult for people to execute. So I would draw a line in the sand right away and say you want a trading system uh with a win rate in excess of forty percent. Yeah, you can be profitable at thirty percent, but mate, it's a tough nut to to trade.

So the higher the win rate within reason, um, the better off you're going to be. So mine, for example, those numbers I've been t talking about on my own is currently running at forty four percent. Okay. Um, so you know, that's in the ballpark. My US momentum strategies, they run a little bit higher. They actually run in the mid fifties.

Um, so that'll be the first metric I would suggest. The second metric, obviously max drawdown, but look at all the drawdowns, not just the maximum drawdown, because maximum drawdown just could have been a one off type of thing. Um and you know, I'm just looking at a particular strategy here. Um it's got a max drawdown of 48%, which is a bit of extreme, don't get me wrong. But other than that, every other drawdown for the last 20 years has been sub-20%.

So we'd want to go back and have a look to see what happened um, you know, to to get that bigger drawdown. Other metrics that I look at, profit factor, as you mentioned. Um profit factor is uh I call it the comfort level. The higher it is, the better it is. So ideally you want a strategy that has a profit factor in excess of two. So profit factor is calculated by dividing gross loss into gross profit, that'll give you profit factor. Um

Payoff ratio, so that's your win-loss ratio. That's got to be as high as possible. Ideally, well over two for a trend following system. Mine's currently around at 2.56. Uh I'm behind the curve on that one. Longer term stats.

um my win-loss ratio is just a little over three. So, you know, that's kind of the swings and roundabouts that you go through. Um So profit factor over two higher is better, payoff or win loss ratio over two, again towards three would be good, and win loss ratio above forty percent, ideally closer to fifty percent if you can. They would be the basic ones I would look at.

Okay, great. Cool. Well that sort of gives uh anyone listening a a few things to sort of uh keep in mind. Now What happens if you go through all of this and your test results fail? They don't meet uh these checks and measures, and it's just a a strategy that you don't want to trade. Where do you go? Do you just go back to step one or do you have to uh approach things a little differently next time around? Well ideally you want to have a look to see what's going on and the best way to do that

is go and have a look at each individual parameter itself. Take take a parameter. So for example, we've used a couple of parameters here. We've used a twenty week breakout. We've used a regime filter. We used the rate of change over thirty. What you'd want to do is go back and look at each one of those parameters individually and see if they're breaking the system down or what's going on with them.

Um and the more you can understand what each different thing is doing, the better idea you've got of it uh of of what's going wrong. If you've got a big advanced, you know, lots of different parameters, lots of things going on, well when it goes

if it goes a little pear shaped, it's gonna be very difficult to ascertain what's actually gone wrong with it. And that's a good thing with trend following. You can you can ask the why very easily. You know, I know where my system's gonna make money, i. e. Upward trending market.

I know when my system is gonna be in cash and that would be a sustained bearish market. I revert one hundred percent to cash and I know when my system's gonna lose money and that's gonna be in a sideways market. Simple as that. I know what's going on. I can look at the market and see Why am I losing money? Oh yeah, we've just had that big correction and I was fully long into that. Bang. That's as simple as that.

So go back to those individual parameters, not uh have a look at them independently, seeing, pull one out, completely remove it, see what difference it's making to the system. I would be very hesitant to add too many more parameters in there. Less is more. Um and uh then from there you might want to have a look at uh the robustness. So for example, we've used a 20-week breakout. You might want to go and have a look. what happens if we test everywhere from ten week to fifty weeks. Okay.

And you would have a look and plot that and see what it looks like. And you might say, well, actually, thirty week breakout. Uh they're all profitable between fifteen and fifty, but thirty is actually a lot more uh profitable, a lot more optimal than twenty. Um you don't want to get into a position where you go and run that optimization.

And you see that uh fifteen is profitable, sixteen, seventeen is profitable, eighteen's a loser, nineteen's a loser, twenty's a winner, twenty-one's a winner, twenty-two's a winner, twenty-three's a loser, twenty-four's a loser, twenty-five's a winner. That is not robust. when I run my systems, for example my momentum systems that look at uh two hundred day rate of change, well, every

Every level between 50 and 300 is profitable. So if the market moves in the future, you're still going to be on the map to making money. Uh and that's the important thing. You don't want a a strategy where you've got a lot of lumps and bumps in the optimization, if you like. So Bottom line, pull it apart, take a look at the individual pieces, see what's going on, ask the question: what is this actually doing? How is it helping me?

Look at things like your regime filter to make sure that's working. To give you an idea, this little strategy with the regime filter on, the max drawdown is about 24% with the regime filter off. Max drawdown's about forty eight percent. So things like that can make a big difference as we discussed before. Which strategy is this? The twenty week breakout? Yeah, twenty week breakout. Okay. In that example, what sort of trading stop are you using? So that one um

That is a twenty percent initial uh with a twenty percent trailing stop and when the regime filter turns down, i.e. the market turns bearish, that ratchets up to ten percent. Okay. Cool. Gotcha.

All right, well let's say uh all our checks and measures pass and we decide this is a strategy that we'd like to uh deploy into a live market. How do you actually go about that? Ideally what you would do in this type of a strategy because we've used that ranking, what we would do is we can either initiate the current existing open positions. So we would select a back test period of time. So let's say we're gonna back test from um

first of january two thousand and five as an example. We would run the back test from then and that would become our our our um point of reference going forward. And we may decide, well, currently, in my case as an example, well I've got eleven open positions. So you could if you wanted to.

buy those eleven open positions right here, right now, and then you're on par with your back test. The alternate would be to wait for the next new positions to come in. So we've got uh nine more slots to fill. And therefore we would wait for the next nine buy signals to come through. But ideally you would buy those eleven existing positions, get yourself to that point, um, and then the new buy the new nine when they come on.

Okay. And I know you your strategies are they're entirely automated, right? There's no no input from you. Um, well I actually have to run them myself. So, you know, something like a weekly strategy like this one we're talking about, you don't only have to push the button after close of business on Friday, generates the orders for you.

And then you would just manually place them. I don't bother doing automated stuff on these trend following systems because, you know, I I might not have to do anything for three weeks. On my short term marine reversion strategies where you're placing forty orders every single night. Um, yeah, well that's a different kettle of fish. But, you know, if I like yesterday, no, today I had a sell order. So I placed that sell order last night before I went to bed.

It executed market on open this morning. That's it. You know, um it took 30 seconds, you know, it took longer to actually log into the account than it did to generate the order. So there's no point really automating these kinds of systems per se. Okay. And when you say you log in and you run it, w what are you actually doing there? You're logging into is it Amy Broker?

So yeah, the the actual process I use, all my orders are generated through Amy Broker or through the software. Okay. So it's actually programmed. I have an exploration. The exploration is generating the signals for me and doing the position sizing. Then once Amy Broker has my selected stocks that I need to buy today or sell today, I then log into my broker. and simply place the order. Amy Broker is not connected to the broker in this particular instance. Okay. And, you know

regardless of what software they use in it pretty much they just need to be able to scan the market for certain criteria, really. Okay. So there's Yes, to a point. So you can have uh a scan or an exploration if you like to find criteria. We've taken that one step further, right? We're actually tested that criteria to ensure that by following a set group of rules, that criteria is going to give us a positive expectancy. As opposed to

just scanning the market for a MACD crossover or an R. No, I meant like scan for the the parameters of your strategy. Yeah. Yeah, well that's all built into my software. That's all built into my code, if you like, you know.

The average system of mine's got three or four hundred lines of code and that includes exploration code, backtesting code, and all that kind of stuff. But I was just thinking like for someone who's a beginner, like if this is their first strategy that they're sort of trying to roll out here. How are they going to get the signals each day? Like, you know, if it's a very basic uh trend following strategy.

chances are they have a software package that can scan for certain parameters being the parameters of the strategy. Correct. Which is going to display all the stocks which meet that criteria, um which they can then go through and sort of confirm. and then place orders from there, right? That that would be a pretty sound way to approach it. Yes. So in our particular instance here, their first check would be

Uh is the market trending up or down? Yes or no? If yes, then continue. Um find any stock that's made a twenty-week breakout. Uh any stock that's made a twenty-week breakout where the rate of change is over thirty. Um

And what else what other criteria? That was it. That's the that's the criteria, right? So yes, you could scan for that straight away. And then you would rank them obviously. Okay. So you'd put the rank in there, check the rate of change for that period of time, the highest ones, and they would be the ones you would buy.

And then I guess the next part of that is obviously updating um the price of the trading stop loss order. Yes. So here's something else that we do. Um we don't use we don't put our trailing stop losses into the market. We use closing price only to determine the exits. So for example, we We said before that price had to close uh below that. Well, sorry, we didn't say that. I'm going to tell you now. That we we've been using a 20% trailing stop.

Uh that doesn't mean the price hits that intraday and bang, your order gets executed. It has to close below that intra uh at the end of the day, and then you would execute the exit on the open the next day. And again, we've found our research that that adds quite a bit of value using the um it does a lot of things actually. It adds a bit of value to the performance. Um, second of all It ensures that you don't have to place that stop loss into the market and be prone to slippage.

And thirdly, um, especially in a situation where, you know, you run a service like ours where you've got a lot of people um, you know, you're not going to have that slippage kind of problem. The opening options that we use are very liquid, the most liquid parts of the day, and as a result, slippage is greatly reduced by executing those orders

on the open in that opening auction. And the other thing I would also say is that some brokers, believe it or not, some brokers charge you to use a stop loss. Really? Yep. Mate, I'll tell ya. I won't say who, but there's a couple of brokers in Australia that actually charge you ten bucks to place a stop loss. But a stop loss is just an order. Well, mate. No, I'm telling you. Well, get a new broker. Exactly Absolutely right. Change brokers. Um

It wasn't that long ago where a lot of Australian brokers for that matter didn't actually use stop losses. That's still a relatively new thing, you know. A lot of people still don't even know what they are and how to, you know, how to execute them. So you're quite right. If you if your broker doesn't do it, change broker, simple as that. Absolutely. So yeah, that's one of the things we've learned over the years and you you tend to get a lot of price spikes down where they might be just

illiquid holds in the market and prices just run and then come straight back again. And we've found that a a solid close price below that trailing stop is the better exit criteria. Okay. So you're always uh sort of calculating things based on the closing price and then executing on the open the following day. Correct. Yep. All right. And that's both for entries and exits.

Well I think that's most of the the process. Um I don't know, have we missed any steps along the way or is there anything in there you'd like to add? I do have one other thing I would like to ask you about, but before I do, um, is there anything we've missed?

Look, there's a lot involved, you know, as I said. Um, you know, back testing, stress testing. coding correctly. There it is very there is, you know, there is a lot of in uh involved in that. A lot of people make very basic errors. You know, as I said, Um, that code from Unholy Grails, you know, uh most of those codes are two hundred and fifty, three hundred lines. People send me their versions and their six lines of code.

And it's like, okay, well, what are you kind of doing different to me if I've got two hundred and fifty or three hundred lines of code in here? Um so you know, garbage in, garbage out, you've got to make sure you're able to code it correctly. There's a lot of other nuances um with with data that you you really have to take into account. So for example,

We do not um we do not generate any signals with dividends included in the data. We remove the dividend adjusted data or we we generate signals on dividend adjusted data. Um we can do backtesting on uh data that has the dividends including, but you don't want to generate signals on that.

Um, things little things like that. And there's quite a few of those little kind of things out there. Yeah. But, you know, like all things, mate, it's like golf. You've got to learn the nuances of, you know, how to read the green and get out of the sand trap and

you know, play to the wind and so on and so forth. It's it's no different to that. Exactly, exactly. Yeah, it's a very fair point. And I think But I think nonetheless, like what we've discussed here is a very good starting point for anyone who's interested in developing a trend following strategy. Um I I think that's unrealistic to think that

you know, you're gonna be an expert on it from just having listened to an hour long podcast. But it's a good starting point, yeah? Absolutely. And there's a lot of resources out there. You know, you've had a lot of trend followers on on your podcast. There's a lot of books out there on the topic. Keep it simple, um, do the testing properly, take your time, um, you know, build a system to your own beliefs. Everyone is different. Build it to your own beliefs.

Um, and just be rigorous in your testing and then it's just a long term application, you know, it's as it is boring. It's boring, but boring's good. You know, I've been running the same system now for twenty two years or something like that. Um and not much has changed. Not much has changed in terms of the actual strategy structure.

Um you just keep rolling through, you know, just roll it and roll it and keep going and some years are good and some years are great and other years are mm not so good, but that's how it happens, you know? If it if it happens to Buffett, well I guess it can happen to me, right? The last thing I wanted to ask you and it's uh it's a little bit off topic, but Uh, you know, trend following strategies are strategies which are used by, you know, very large money managers, hedge funds, CTAs, et cetera.

What are some of the things that they do? You know, some of the more advanced things that they do, which we obviously haven't discussed here and, um, you know, might not be so commonly found in um you know, a lot of the literature and content uh on trend following that is out there. Like are there any kind of advanced tactics that they use?

Well the first thing that comes to mind would be something like swarm theory. Um What's a what's a way to uh to discuss that? So swarm theory, let's let's say that you've got um let's use our strategy. Let's call it a hundred day breakout strategy that we've been talking about, or twenty week, let's call it a hundred days. So markets change over time. Theory, common theory goes that Uh, every so often, once a year, you would optimize that hundred day breakout.

Make sure you've got a robust set of parameters and then pick a parameter that is kind of optimal but in a robust flat spot, right? So it might be this year we're using a hundred day. Next year we do that. And it moves out to one hundred and five days. Um, just like the turtles, you know, they've moved out from twenty days to whatever they're doing these days, a hundred days or whatever it may be. So over time, uh your your strategy will slowly change with the market.

So that's the way it's been done in the past. Um the more modern way of doing it is what's called swarm theory. And you're basically running the same model but using fifteen, twenty, thirty different parameters instead. So You would trade the same model with a 100-day breakout. You trade the same model with a 110-day, 120-day, 130-day, 140-day, 150-day. You trade them all as one. And the whole idea being is that is the market.

personality changes back and forth over time, well, you've got that best optimal base covered because your one of the systems that you're trading is using that breakout length. Does that make sense? Yeah. Yeah. That's a very basic way of explaining it.

Um that for a retail trader to execute would be pretty expensive, pretty complex. Um but that's I believe that's what some of the big guys are doing. Okay. Interesting. Well, something to look into, I guess. Mm. Mm. Cool. All right, Nick. Well let's leave it there. I think we've gone for a bit longer than um I'd anticipated and um probably you'd anticipated as well. That's all right. Um but

Man, awesome. Really, really good. Uh, thank you very much for doing this. Uh, I think uh, as I said just before, like. a really great starting point, uh, using the information that you've shared here for anyone who's interested in developing some, you know, very simple trend following system of their own.

Uh obviously uh also as you said earlier, it's not for everyone, but you know, there there are some advantages to trend fine strategies, as you know, all too well. So yeah, again, really appreciate you taking the time um to share this, Nick. Um it's been awesome.

No problems. And if anyone wants to contact me, uh best website is nickraj dot com or you can just drop me an email, nick at nickraj dot com and happy to answer any questions. So yeah, thanks Aaron. Great to chat again. It's been a long time. Um so um uh thanks for having me on board. No doubt. And do you just want to drop your Twitter handle? Yeah, the Twitter handle is at the chart. Okay, at the Chartist on Twitter and um some older episodes with Nick.

Episode sixty four and episode four, one of the the very original episodes. So my interviewing skills were We're pretty sketchy back then and um hopefully have got a little bit better since. But um yeah, don't expect uh too much from episode four. When was that? That was two thousand and fifteen. Wow. Oh yes, okay, well not too long ago. I can remember back then. Cool. Once again, Nick, really appreciate it, man. Um we'll talk soon. Thanks a lot. Okay. Appreciate it. Thanks, Aaron.

You've reached the end of this episode of Chat with Traders, but rest assured there are more episodes.