In this episode of 2Traders Podcast, Darren and Walter zoom in on trading system diversification. Which offers you better diversification, multiple trading systems or multiple markets? In other words, will you get better results (and smoother equity curves) when you diversify by trading many (uncorrelated) markets or by trading (uncorrelated) trading systems?
They also touch on automated trading and what are the things to consider when making the leap into “trading with robots.””
Walter gives some useful insight on modifying system according to what you want to achieve. Darren also suggests mixing up your cocktails and looking at different menus.
Download (Duration: 32:20 / 37.0 MB)
In this episode:
00:33 – a lengthy break
02:43 – rolling out
05:07 – flips the switch
07:14 – subjectivity
09:03 – gray areas
11:57 – a constant argument
14:15 – the automated trader
17:05 – path to take
20:12 – breaking off
23:30 – correlation
25:25 – jumping out
27:37 – core system
31:03 – the friendly ones
Go with the one that you think is going to be the loser. [Click To Tweet].
Trade different markets with different systems. [Click To Tweet].
Add more strategies rather than adding more markets. [Click To Tweet].
Darren: Trading is really difficult and even when you think you’ve got a crack, it is always going to come around and shake your confidence a little bit…
Announcer: Two Traders, Darren and Walter, pull back the curtain on profitable trading systems, consistent money management, and profitable psychological triggers. Welcome to the Two Traders Podcast.
Walter: Hello, Darren. Welcome to the Podcast. How are things in your world?
Darren: Hi, Walter. Just getting back to trading after quite a lengthy break. Actually, I’ve been on holiday and although I normally trade on holiday, had a bit of a break this time and really switched off from trading completely.
It was not really a conscious decision. It was just I felt I needed to and I’ve really wanted to stop reading about trading or talking about trading and thinking about it completely. It’s been quite productive for me. I mean, I feel refreshed.
I really feel good for having a break but also, I’ve been reading so much trying to improve my trading and it did give me a bit of a moment to look at things slightly differently.
One little thing that has come out from that is, recently, I’ve been trying to find a way of adding some diversification to my trading because my strategy either works really well or, in flat periods it works really badly.
I was forcing myself to try and find a way of changing the strategy to remove those bad effects. This little break made me clear my mind so I could think about it from a different point of view.
So, I am looking now at having separate strategies. My strategy that I trade the most works really well when price is moving, when the market is moving, and it does not perform so well when price is consolidating.
I started looking at those consolidation period and is there a way I can use my ideas but with a slightly different approach that will work well in those areas. I’ve noticed quite a few traders will have different strategies for different market conditions.
That is where my focus is right now and I just wondered what your take was on that. Should we a have strategy that we could roll out over the long term or should we have strategies that balance out the bad effects of each other?
Walter: Yeah, it is a good question. I guess most traders fall on the camp of either I am going to trade both of these strategies. One might be a reversion-to-the-mean style of trading that should do well during a consolidation when the market is not really moving, just like a flat line and consolidating.
Then, you might have one like yours, I guess. I would consider yours almost like a swing trading — or it is not really break out. It is more like a swing trading or you are trying to catch these big swings or maybe like a trend. You are trying to take bites out of the trend or something so they compliment each other.
The question always becomes are you going to try and decide whether or not you get to flip the switch and say “Okay, we are consolidating now so I am going to trade my mean reversion system”.
Or, “We are trending now, I am going to trade my trend system” or whatever. Or, you are simply going to trade them both all the time knowing that one of them, the performances is going to suffer when the market is running off. In trending, your mean reversion system will probably not going to do so hot.
It is tricky. What I’ve basically come to the realization is that I am the guy that flips the switch. What I do is I go and I’ve tried to be the guy that trades them all the time, that trades the system all the time.
Instead, what I do is I take a step back and I go “Okay, does it look like to me…” — and usually this is on the daily and weekly chart so — “Does it look like to me that this market is trending? Is it going to a clear direction?”
If the answer is yes, then I’ll go ahead and slap a filter on that system and see the market through the eyes of a system and I’ll try and trade like that. Otherwise, if it is directionless, I’ll either go down to lower timeframe or just kind of I do not really want to trade it.
I guess consolidating on the higher timeframe would be the way that I would say it. Even if a market is consolidating on a higher timeframe, that usually means you could drop down to the four-hour charts and find some trades where the market is bouncing off these different levels.
I tend to do that but I typically — when the market gets really quiet unless I think it is consolidating in the midst of a really strong trend — I typically will stay away from that market.
That is the way I view it. I am the guy that flips the switch and says the JPY is trending right now, I am going to go ahead and look for buy entries because it is going up. That is how I view it.
I know I would suspect that you would probably have a different idea or you would be trading both systems. I suppose, at the same time, not really trying to corner the market and say well, it is trending now, it is not trending, or whatever.
Darren: I was looking more at say, when price is ranging which is not good for my primary system when it is in a tight range. I was looking for a system that would generate entries when that price action is going on but then, when price does move in a trending fashion, then the system does not actually generate any entries.
I’ve been looking for the way you setup the entry. You don’t really have to make a judgement of whether the market is trending or ranging because the entry system does not generate those entries when the condition has changed.
Walter: Yeah, that makes sense. I guess you would basically see that the market is trending and then say “Okay, we are going to use this entry now”.
Darren: The secondary system just wouldn’t generate any entries when the market is trending because it would have to stop ranging to actually create a condition that would trigger an entry, if you like.
It’s not easy to do, obviously. I am slightly sitting on a fence which is the best way to go because with the traders that I’ve been trading with this year, we’ve tried a lot of things. Rely on judgement, or some way of measuring the market and using that approach. That’s been fairly successful but I just feel it is not quite elegant enough.
Whenever you rely on your judgement to measure the market condition, there’s always going to be that element of that subjectivity however you shake it, really. That is always going to be there and my drive quite often has been to try and reduce those judgement calls as much as possible. Whether that is a good idea or not, I couldn’t say. It is just my gut feeling of how I should approach it.
Walter: Exactly. For example, when you get it on automated trading, the big question always becomes how do we define this and what sort of filter do we put on this? You’ve got to decide, basically, what is this type of market.
How do you define that and that sort of things and, I find that talking to people who are into automation, the good ones are always leery of putting too many conditions on the system. That is the tricky part, isn’t it?
For example, for you, if you said, “Look, when the market is really moving, I am going to get these entries and when the market is not really moving at all, then I am going to get these entries.” But, I guess the question is how do you know?
Darren: That is why you need to draw up some simple rules to stipulate when you place your entries and this is always where you get the problem. Let’s say for instance, my rule was price needs to make a new swing high before you set an entry. Then, you’ll get into an area where, “Okay, what defines a swing high and how many bars are you going to consider that it is a proper retrace before the new swing?”
You always end up with those annoying gray areas where, at the end of the day like you say, you do have to make some judgment calls. There’s a lot of areas in trading that have that and some people do go down the route of automation and removing those gray areas completely.
I’ve been driving towards that this year and, has it been successful? I am still bit on a fence, really. I think this year has left me a lot of unanswered questions and I think I’ve made some progress as well.
One thing remains certain, trading is really difficult and even when you think you’ve got a crack, it is always going to come around and shake your confidence a little bit. It is a continual battle and I really need to plan well and I am going to go next year.
Walter: Yeah, I think you are right about the gray areas. I’ve been thinking a lot about automation too in the last four, five months. Getting into it, I find that what happens is sometimes, you’ll hear people say things like, “I see something on the chart and then I go and code it up — or have it coded — and then, it is not nearly as good as I thought it would be”.
I’ve been talking to a trader about this recently and we’ve been going back and forth and it’s like, “Well, is this because you missed something and it’s not in the codes? There’s something that you are processing when you see these trades on the charts and it is not in the codes.”
The codes does not actually reflect how you make these trading decisions or is it because we have these inherent — and we do have these inherent — biases where when we can scan through the chart, we pay attention to the ones that are good.
We scale over or miss the ones that aren’t so good. It could be part of that too. That is really what is going on because we are not really paying attention to the losers.
Darren: Yeah. then you start to realize how much feel and gut goes on in the background with trading. Let’s say for instance, right now we could argue that after all the uncertainty, that the interest rates in the States are going to rise and the USD is going to be in some sort of a long trend.
That is, to anyone with any trading experience, that is a bit obvious at the moment but could you define that in some rules to put in an EA? There’s a lot of gut there and just feeling right when the time is right now for the USD to go on a bit of a trend.
There’s a constant argument of whether using your gut and your feel is the right way to go or whether it’s like a completely defined rule base or rule set.
Walter: I believe that in the beginning our gut is awful. I think you can cultivate it as a trader but I believe, rather strongly, that with very, very few exceptions if you are new to trading — and when I say new, I mean first 3 years at least — and you look at the chart and that chart says “Yeah, I’ve got a buy”, I would seriously consider selling because you are just not setup.
I do not know about you, Darren, but that is my experience and I keep harkening back to the Turtles in how they were taught this. They were taught that — their instructors told them– that “Look, if you’ve got 2 trades and they are just as good as each other, then you can only take one signal because you are fully loaded. You cannot take any more trades, you can take one more but you’ve got two. Go with the one that you think is going to be the loser. Go with that one”.
I was actually part of their rule and I find that fascinating because I think it is true. It makes sense. You’ve got all these people with the wild emotions and they are just going crazy. The market is losing money, hand over fist and there’s a few people that collects very slowly, vacuuming up the money because they are a little bit more deliberate.
They are a little bit more disciplined and they see the market through a filter of a system. I guess it makes sense that you would come to the party and just have no clue what is going on when you first start out in this stuff. What are your thoughts?
Darren: Definitely. We’ve come to trading with our natural biases and beliefs. Generally, they wear off. If anyone has been listening to the podcast last year then we’ve covered that ground a lot with naturally back traders.
A lot of traders is learning to recognize those elements that are bad traps and then build some system and methods, process, to avoid falling into those traps.
Walter: So, what do you do? I mean, let’s say that you are an automated trader. Let’s say that you are somebody that says, “Look, I know what my system is. I’ve got the rules here, it’s all been coded out. I backtested it, it looks great. It is going to be awesome. I am going to make bunch of money”.
Now, my question would be when are you going to turn this off? In other words, traders who use these automated system, they need to also have a set of rules about how do you implement the EA. When do you turn on the EA? When do you turn off the EA?
Is it always going to be on? Is it going to shut off after the Asian session? Is it going to shut off during the interest rates’ announcement? Is it going to close trades before the weekend?
How do you know what kind of drawdown before you say, “Okay, this is not working anymore”? All these things seems to be looked at beforehand. It is not just the discretionary trader that has to tackle these issues.
Just because you’ve coded up your system and it is a nice little EA box does not mean you are immune to all these thinking mistakes, these cognitive biases that we make because we get sucked up into the emotions even if the robots are doing our trading.
These are the things that traders who are thinking about automating need to really consider. How are you going to interact with the EA? What are you going to do with your robot? What sort of tipping point do you get to?
Do you go, “Okay, that’s it. Turn it off. It is not working anymore. You’ve got to calibrate” or whatever. These guy, sometimes every 6 months or whatever, they’re filtering with the inputs and stuff like that.
To me, that is funny because what they have is a system that is perfect for the last six months but now, trading it into the future, who knows how good it is? It’s really sort of tricky situation. What are your thoughts? I mean, how do you do this?
Darren: Firstly, the whole backtesting scenario, I think it is way more complex than we give it credit, for a start. It is so easy to find a system that backtest well over a period of time but the variants can happen and the results seems to always remain. And, you can backtest something for a year and then the next year, it can perform very differently.
You can backtest it for five years and then the next year, it can perform completely differently. I think we give too much weight to its future performance. We have too much confidence in its future performance.
Diversification is not something that I really looked into in the past but I wonder whether we focus too much on finding a single strategy that is always going to work. I am starting to question whether that is the correct path to take.
Do we need to be trading more systems off with smaller accounts rather than put all of the eggs in one basket, so to speak? I’ve got no answer for that but I find it hard to believe that say, hedge funds would have a single strategy. I can be wrong.
My gut feeling is probably “don’t”. They might have a preferred approach but are they using a lot of diversification in their strategies to deal with that variants? If one system that has been working really well drops off, there’s a greater chance that another one will pick up the slack. What are your thoughts on that?
Walter: Absolutely, yes. I think you are right. What I’ve been testing the last 3 weeks and what I found is interesting because these are right along the lines of your thought. You say “Alright, I’ve got the system. It is pretty good but I want to diversify. It does not work in all markets.”
By the way, the best description that I’ve heard of the different types of markets are essentially quiet consolidating so, it’s really like flat line and not really moving much at all. Just sitting there or sort of a volatile consolidating where it is not really moving anywhere but it is a little bit more swing up, swing down but still it is stuck. Or, you can have a quiet trend where it is very low volatility. It just gets keeps marching on or a volatile trend where there are some sort of deep pull backs. Those are the sorts of markets that you see.
You see basically directionless or markets with direction so you’ve got to decide what kind of system do you have. Is it going to work in two of those markets? Is it going to work in one of those markets?
When you test it, you need to test it in all of those different types of markets. Getting back to your question about what does the private funds do, how are they going to manage these risks?
Essentially, you can trade different markets. You might have one system that trades different markets or you could trade different systems. Those different systems could be on different markets as well or on the same market.
What I found is interesting is that you are always going to be better off trading different systems. Why is this the case? Know that people will often say “Well, I am trading this system but I am trading on oil and on the GBP and I am all good. Everything is fine”.
What I found is that… Because, basically, it does not matter with the market. The market is either going to be trending in a quiet mode, not very volatile, or trending in a volatile mode, or it is going to be quiet and sort of consolidating in quiet mode or not very volatile or, it is going to be in consolidating and somewhat volatile mode.
Because all markets are going to be doing one of those things, your real advantage in terms of reducing your risk — which is basically what we are talking about here — comes with the approach of using more systems.
Now, I agree with you. I think you are better off breaking off your accounts into smaller accounts and putting them at different brokers because we’ve all seen anyone who’s been around retail forex trading for any length of time. I’ve had three of my brokers go bust.
Which means, basically, you lose all your money but, fortunately, for whatever reason in every instance, I didn’t have money with that broker when it actually happen. So, Refco went bust, PFGBest went bust and FXCM essentially went bust and then the loan sharks came in and saved them.
I had an account with all of those brokers but for whatever reason, I’ve pulled it before the ultimate house of cards fell down. I was lucky but I may not be so lucky in the future. I think it makes sense to break it up beyond that for broker diversification as well.
That is basically what I’ve found and I would recommend that if you are testing this idea of what to do to trade different markets with the same system or different systems, I found that different systems will usually provide you more advantage in terms of smoothing out your equity curve.
Of course, you need systems that are complimentary. It can have two trend following systems so I am talking about a breakout system and a trend following system or, a breakout and a reversion to the mean or, a swing system and a scalping or whatever you want to call it.
That is basically what I’ve found. I am not saying that this is a be-all-and-all answer but I think that if these are the lines of thinking that people are going through, I would encourage you to test it out for yourself.
So far, I haven’t been able to find anything better than essentially adding more strategies rather than more market.
Darren: I found out this year where crises being fairly flatten one market’s because of say, fundamental reason, it is kind of gone across all markets. And, even though there might have been some that was still moving, they are still a bit sluggish and not really making up for the really flat markets.
I’ve come to the same conclusion that, really, I need some more strategies. It is quite difficult because there’s this notion that if your strategy works, it should always work. If you start looking for another system that’s got some negative connotations and, really, it hasn’t.
You are just being smart and diversificate in different market conditions. That is encouraging that you agree with me on that. I have been listening to recordings of who’ve been around for a long time and trading long time and it seems nearly all of them have multiple systems as way of diversification.
I’ve very rarely heard them talk about changing markets unless so in completely different bulk part. Picking a different currency pair, I’ve never heard them mention that. So, yeah, that is quite encouraging.
Walter: Some of these guys, they just trade in DCs or whatever, or stocks. It really does not matter if you are trading the Footsie(FTSE) or the SMP500 or the Daks or whatever. A lot of these markets are pretty correlated.
If you look at the correlations, what you will find is more or less when the Daks goes down, the SMP goes down and that sort of thing and you are going to find areas of diversion which is not the case.
What these guys are just absolutely terrified of is they have so much money, they have to place it in all these different markets. So then, what happens is if there is some sort of a crash, they are screwed because all their stuffs is in there. They have to diversify by system, they have to have different systems.
Even the same thing like if you are trading the Index and you long the SMP500 and you’ve got all these stocks you are also long, even if they are an emerging company or whatever, they are also going to be correlated to the index too. Generally speaking, when the index is up, they are going to go up, most stocks will go up. This is the thing that you have to think about.
Like you say, most traders have found out that it really makes sense to figure out “Okay, this system that I really like and been trading for so long, it works with this market.” But, what would a system look like did not necessarily maybe work in the opposite type of market or different type of market that is going to be really challenging for my main system and so they had there. That makes a lot of sense.
If you do not have any ideas for system, my advice would be go into the forex tester, just watch the chart and see what you come up with and see what you notice. I was talking to a trader the other day. He said he basically realized he had this rules for the system that he was testing and then he just saw something else kept happening so often in forex tester. He finally said “screw it”.
This is like, “I’ve got to add this. This is a new system” and I think that is what you will find if you spend a lot of time getting your hands dirty with the data. You will eventually come up with some ideas to test because you will see something happening over and over again.
Now, for me, that is just price patterns or whatever but it might be something else for you. It might be the way that price interacts with moving averages or the MACD interacts with the moving average or whatever.
It will be something that will just jump out at you and you’ll say this makes a lot of sense. To me, that makes much more sense to go down that path than to open up some book and go, “Oh, this guy says this works. I will just use that”.
No, no. It is better off if you come at it organically where you come up with the rules based on what you see because now, it is built in the foundation of what you see, about your beliefs. It is a little bit harder to shake that than it is to shake what someone gives you and says, “Oh, this works. Go ahead and use it”. That is what I found, anyway.
Darren: People who eventually become successful, a large percentage of them end up doing that. Eventually, bringing everything together and make their own unique board, if you like, to play on.
The other day you said most searches lie in the places where other traders are not looking and that is the way you need to go. You said something earlier, as well, about different timeframes.
I find that that is quite a good way to diversificate, as well. There’s a bit of an issue with managing trading onto different time frames at the same time but something like a daily time frame strategy with say, a 15-minute strategy that you use intraday, sort of peak market times.
Something like that can work really well. I’ve noticed that in my group where some traders are trading daily and others are trading the one-hour time frame. When the daily traders aren’t doing so well, the lower time frame guys seem to make it up. That can work quite well, mixing two different time frames.
Walter: Absolutely. It is a different system though in the end, isn’t it? Or, is it the same system that they are trading on different time frames?
Darren: I think it is the same core system but just modified to suit the time frame. The core idea is the same. Let’s say that they are both trend trading ideas but they have different methods so event turn outs in the market.
Walter: I believe that it is always sort of trading somewhere. I mean, if you look at one-hour chart, you’ll say this chart is so choppy but, if you broke it down and look at the five-minute chart, you might see some beautiful trends.
Darren: Yeah, exactly.
Walter: In other words, there is always a market. You can always find a market. The excuse should never be — and the people who over trade can tell you this — the excuse is never be like “Okay, there is nothing to trade”.
If you are a trend follower and you cannot find any trends, look at the different time frame but I hope you have your data to back up that works, that your system works, there because you might have to tweak it.
Darren: That is something, perhaps, we do not do enough all the time. Like you were saying, you can always find a market. We are drawn to decide where “Okay, this is the strategy that I am going to trade. I am going to trade it on the GBP/JPY and gold”.
You backtest it on the GBP/JPY and gold, you get locked into that being how are you going to trade. I think, perhaps, it is a good approach. Like you say, I’ve noticed stock traders do this a lot is to perhaps look at the markets right now and say, “This particular market is performing best in my strategy right now so I am going to trade that”.
I know you do this. You look at all markets, you look at all currency pairs and crosses and you can quite quickly decide, “This one looks like right, it is moving in the way I want to see.” And then, you go and look for possible entries there.
I think, too often, traders decide “Right, I am a EUR/USD trader, this is the strategy that I am going to trade” and they just do that. Sometimes, it is clear in the obvious that the EUR/USD is not moving in a way that is good for your system.
There’s a negative aspect to making changes and I think all good traders have constantly having into sort of making changes. Perhaps adjust the parameters, adjust their style of managing and, perhaps, trade in a different market for a while.
Walter: It makes a lot of sense. You can certainly lock into one. I know some successful traders who do that. They just go, “Look, this year I am just going to trade the GBP/JPY” and they have their reasons for that or whatever.
To me, it makes sense to open it up because, otherwise, you could be reaching like if you are just trading the GBP and your system does not work when it is consolidating and you know it is consolidating. Should you really be trading that?
Maybe you should just open it up and what is the NZD/CAD doing? What is the AUD/JPY doing? There is something else going on. For me, if I am looking for a trend. There is a very typical trend I am looking for which is a quiet trend that looks really strong, moves up and then, consolidates and then moves up and consolidates.
I don’t like the ones that do the deep, volatile pull backs. Those aren’t really the ones I want to play with. I want to play with the friendly ones that are really easy. For me, it is more of the look of the chart but, some traders, they like the idea of trading their home currency or one that they feel like has a better spread which I think is a horrible reason to choose a market.
I am going to trade the Euro because I get a .5 pips spread. That is like a terrible reason to trade but, whatever. Some people, that is the way they approach it. Alright, any closing thoughts, Darren? Are we all wrapped up here?
Darren: Nothing. We’ve covered some good stuff there and I am going to go through into the… Yeah, I am going to try and work more on diversification with my trading.
Walter: Cool! Sounds good. We will see you next time. Maybe next time we can talk about — I do not know if you want to do this but something I’ve been thinking about is flow. Trading with the flow. Maybe we could talk about that next time.
Darren: Yeah, that sounds cool. Good to be back, Walter.
Walter: Great! See you soon. Bye.