“I never really wanted to be a manager – the idea that you could compete against everybody else and make money was just brilliant.” – Barnaby Cardwell (Tweet)
Our next guest is a fund manager in the short-term space, and thus is distinct from many of the guests on this podcast. He discusses how he got into the financial markets from a young age, and why he decided to trade in the short term space in order to separate himself from the competition. Listen in for details into his program and how he started his hedge fund.
Thanks for listening and please welcome Barnaby Cardwell.
In This Episode, You’ll Learn:
- How Barnaby grew up and what got him interested in the financial markets.
“One of the first books I read was Market Wizards – at about age 20.” – Barnaby Cardwell (Tweet)
- Why he left being a broker and went in to a more systematic approach.
- How he met his future business partner.
- How they started his fund in late 2008.
- What it was about the Market Wizards books that captured his imagination.
- What he does to compete outside of trading.
- His first mentor and what he taught Barnaby.
- Why he decided to go short term.
“It’s very hard to distinguish ourselves in the long-term timeframe.” – Barnaby Cardwell (Tweet)
- How automation has changed his industry.
- How he gets through the hard times in the market.
“If you know you’re system is correct, you know when the market moves happen you’re going to make money.” – Barnaby Cardwell (Tweet)
- An overview of the program that his company runs.
- How he set up his company and attracted investors.
- The challenges that let to auto-execution of trades.
“I wouldn’t say our model has evolved too much.” – Barnaby Cardwell (Tweet)
- How he builds a strong culture in an organization.
- How he wants people to look at his track record.
- Why he uses time-based stops.
“We’re trying to deliver market movement on short-term timeframes.” – Barnaby Cardwell (Tweet)
- The details of the short term models he uses.
- How he creates a strategy.
Resources & Links Mentioned in this Episode:
This episode was sponsored by Swiss Financial Services:
Connect with Cardwell Investment Technologies:
Visit the Website: www.c-i-technologies.com
Learn more about Cardwell Investment Technologies on LinkedIn
Follow Barnaby Cardwell on Linkedin
“We want to trade across as many markets as possible – that’s how we know a system is robust.” – Barnaby Cardwell (Tweet)
Barnaby: We spent a lot of time building long term trend following models. They're easy to do. They're not that hard to do. The data is pretty clean to get. Daily data is pretty clean; it's easy. You can use it. I found that it's very hard to distinguish ourselves in that timeframe, and I just knew it would be very hard raising assets if you're very correlated to everybody else. It's very hard to distinguish yourself. There was a commercial decision in the beginning, "Hang on if we do it short term we can definitely get more uncorrelated." Then the second reason is because I've sat in front of one of those screens for so many, so many years. I know a lot of nuances and subtleties in creating the systems, these short term intraday systems.
Niels: Being different is not so easy in the crowded space of systematic trading. But never the less it can be important if you want people to pay attention to what you do. This was a clear motivating factor for today's guests when he was just starting out. Trading differently, using a less crowded time horizon and perhaps most importantly targeting a risk level in a very different and perhaps even unique way has certainly helped his firm to be noticed by institutional investors around the globe.
Introduction: Imagine spending an hour with the world's greatest traders. Imagine learning from their experiences their successes and their failures, imagine no more. Welcome to Top Traders Unplugged. The place where you can learn from the best hedge fund managers in the world, so you can take your manager due diligence, or investment career to the next level. Here's your host, veteran hedge fund manager, Niels Kaastrup-Larsen.
Niels: Welcome to Top Traders Unplugged, where my goal is to give you the clarity, confidence and courage you need to invest like or invest with some of the top traders in the world. It is the stories that you never get to hear set out in the most honest and transparent way that I can make of what goes on inside the minds of some of the best investors in the world. Today you're listening to episode 75. If this is your first episode you've heard, you might want to go back and listen to all the earlier conversations. If you want to start with something different, and even explosive, start with episode 73. Before we go any further, let's find out who's on today's show.
Barnaby: This is Barnaby Cardwell, CEO of Cardwell Investment Technologies, and you're listening to Top Traders Unplugged.
Niels: Thanks for doing that Barny. By the way, if you want to read the full transcript of today's episode and all the previous conversations I've had, just visit the TOPTRADERSUNPLUGGED.COM website and sign up by hitting the "I am In" button in the top right corner. It's that simple. Now let's get on with part one of my conversation. I hope you will enjoy it.
Barny, thank you so much for being with us today. I really appreciate your time.
Barnaby: Thank you, thanks for inviting me.
Niels: Yeah, no absolutely. You know I noticed a couple of things in my preparation that I found very interesting, and I believe it sets your firm apart from some of the other guests that I've had on the podcast. Firstly I noticed that some of the language you use in your information material. It really leads me to think about Toby Crabel and some of the things that he does when it comes to trading at certain times of the day. So I'll be very interested to learn more about this. Also, the way you define your risk and what risk targets you have - I thought that was quite different and interesting. Finally, you seem to have chosen only to offer managed account at this point rather than co-mingle funds. I'd love to learn more about sort of your motivations for doing this. Before we jump into all of those details, I have this sort of a simple question that I try to ask all my guests. It is something that I've also, myself found a little bit tricky at times to answer, and it goes something like this: When you meet people for the first time, it could be at a social event, for example, who don't know you, how do you explain what you do actually?
Barnaby: Okay, yeah people who don't know about futures markets, it gets a little bit tricky. I don't really want to go into the depth of how you define a futures market and how you can go short as well as long. If you see that the person knows that, I sort of tend to tell them that we use historical past data, we try and fit models on that data to come up with ways of profiting from the marketplace going forward. It's essentially how I explain sort of what we do.
Niels: That's fair enough. Let's stay with you for a little while longer, and I want to tell your story, but more importantly I want you to tell it really from the first time that you got involved in the business and to put a little bit of color on the story. Feel free to go back as far as you want, you know what were you like as a kid growing up or a young man growing up? Really try and take us back and tell us how this all evolved.
Barnaby: Yeah, okay so, well going right back as a kid I always liked competing. I used to play a lot of card games, domino games with my grandparents. I use to spend weekends with them playing these games. I used to love it. Not necessarily a gambling, but just games where you basically you know, you can win, and I loved doing that. Also loved numbers. I took quite a lot of mathematics at school. I went on to study math at a University. Sort of around that time at University, I was sort of interested in financial markets. I'd never traded, but I started to read a few books, and one of the books I read was Market Wizards, the first one.
Niels: Oh right! At what age did you read that? That's interesting.
Barnaby: Yeah, so I was probably about 20, something like that.
Barnaby: Loved it, great book, still read it to this day sometimes, and the others. So yeah, I went to the University, and I came out… while I was at the University, I did a few sort of internships with various industries, accountancies, actuarial practices. But I'm glad I did the internships, but I couldn't see myself taking a career in those areas, and it was really more and more about trading. And I felt that I'd eliminated the industries that I didn't want to go into, and I felt that I wanted to have a go at trading in some form.
So I graduated in 1999, and I tried to get a job in an investment bank, and it's pretty tough. I think that was when the .com bubble had just burst, in 1999, 2000, something like that. But I eventually managed to get a job, actually at a very small broking house in London who was backing traders to trade on Life in Eurex. They'd just gone electronic, they just basically… Life was on the futures pit exchange before, and obviously Eurex came about electronically and took some of the business away from Life.
So there was this whole new electronic trading era, arena sorry, opening up in London. And I managed to get this broking house to back me to trade on the exchanges. So I developed my own style back then. It was very short term. We were only trading a few products back then. You know, we traded things like EURIBOR, FTSE 100, futures, bond futures, banks futures. There wasn't too much to trade, to be honest. But I developed a short-term method. It was quite mechanical, but it wasn't completely systematic; it was still discretionary.
I started to do very well at that. I bought my contract out after a couple years and started trading my own capital. Then over the years more and more markets opened up electronically, so there're more and more things to trade, and I did very well at that.
Over the years I felt that I wasn't getting to the top of how I could be as a trader and I felt that taking a more logical, a more quantitative approach would get me to where I wanted to be. Especially I started following a few of these big funds out there like AHL at the time and Winton, and people like that, they were starting to make some amazing money and manage some huge AUM's. So that got me thinking. Also at the time I was backing a few traders as well, and this is where Tim Marchant comes in. He was one of the traders I initially backed. So we started working together trying to build up models, getting together loads of historical data and trying to come up with systems to trade in the markets, over lots of markets instead of the more concentrated markets, we were trading discretionary.
We embarked on this process and realized yeah, there's some money to be made here, but we need more help. Our program abilities are limited and our time is limited so we hired a few more people, and things just grew from there really and we started the fund in late 2008, November 2008… the program and we've just taken it from there.
Niels: Sure, sure. I want to go back in time from what you just said. I mean I'm very curious about how you got that first job. In a sense it sounds like you come out of University, you've never traded anything and then you get someone to back you to allow you to trade. How did that actually all sort of take place? Did you already at that time have anything to show them or describe to them what you would do with the money that they would invest with you?
Barnaby: Well, I had a lot of tips I obviously picked up from reading books like Market Wizards, and I basically sort of plagiarized the ideas in that book and told them that this was how I was going to do it, and they seemed quite impressed I guess.
Niels: Fake it until you make it I guess!
Barnaby: Yeah, maybe! I mean the rules made absolute sense to me even though I hadn't traded, it seemed correct. So yeah, that's how I did it.
Niels: Yeah, well done! Now just again staying on sort of the history, and this is something that I'm quite interested in… what was it when you read those books, the Market Wizard books, what was it about the books that fascinated you or caught your attention? Or… you know because they obviously had a very big impact on your career path, so can you sort of remember what it was that really sort of made you think, "Wow, this is something that I want to do!"?
Barnaby: Well obviously I think it's the combination of the fact that you can compete. You know that was prevalent throughout the book. And also money as well, I mean I like the fact that you can make a lot of money without having to essentially maybe build a business or things like that. If you could just get a hold of capital, you could make some serious money, and that appealed to me. I never really wanted to be a manager or anything like that. The idea that you can compete on your own against everybody else and make money at the same time, and it's your job. It just sounded brilliant!
Also, what I liked about the book was… Market Wizards, in particular, was the fact that there are so many different ways of making money as well. And essentially really all comes down to risk taking and psychology even if you're systematic, it's still about you know, finding points of the market where there's some sort of edge. It's all about cutting your losses when you're not making it. It's all about running your winners when you are making it, scaling down when you have a bad time, putting your size up when you're having a good time. All these things you know, that cross all areas of trading.
Niels: Sure, sure. No, that's very true, there're some good points there. You mentioned again sort of the point about the competition aspect that also sort of caught your attention. Besides trading which obviously is competitive, I mean you get your score every day I guess, have you done anything else in your life from a competitive standpoint? I mean be it sports, something else? Because it seems to be something that you thrive on.
Barnaby: Yeah, I mean anything… I mean quite a lot of athletics at school when I was young, I loved doing that… sprints, 100 meters, 200 meters, played rugby at school, quite a lot of rugby. In my later years, I started playing quite a lot of squash. That's quite an intense competition. Anything really, any game, any card game, whatever, it's fun!
Niels: Sure. Great, great! Now the first job, just staying with that a little bit, did they teach you anything in terms of trading or was it really something? Because I'm trying to see if there's any link back to kind of the "turtle experiment" in a roundabout way where people would be backed, and you would get some money, but you would actually also be given some guidelines or instructions or something like that. Was there any of that in your first job? Because you also mentioned that you actually backed some traders yourself later on, and I don't know whether there is a link since you've clearly read some books, and maybe you were even aware of the "turtle experiment" also back then?
Barnaby: Yeah I was aware of the "turtle experiment." But, to be honest, when I started there wasn't any sort of a graduate program or anything, it was just sort of… you initially started by sitting in front of a simulator for a few months and then you got capital to trade with. But one thing that's really helped is I sat next to a young guy who had… he had been there like a year or two years. He was doing phenomenally well. It was just really good to sort of see how he traded, to see that you could be fearless but also at the same time you know, you've got to cut your losses when it's not working. He was very clinical. He was disciplined. So I took a lot from sitting next to that guy, his name was actually Andrew Priston, and he actually went on to make a lot of money. I think he's still trading till this day. So that was pretty good, sitting next to someone to give you the confidence to do it yourself. That was quite important.
Niels: Yeah. No, it's interesting, many of the people I've spoken to when it comes to… and it doesn't have to be a formal mentor in that sense, but just having someone to as you say learn from in whatever way, just seems to be so important in the early stages of a lot of these very successful people's careers. Now if we fast forward to today, today you're running obviously Cardwell, and that's a big part of your life. What do you do when you're not working? What do you like spending time on?
Barnaby: Not too much time outside of work, I mean I've got two young children, so that takes a lot of time, but I like skiing. I don't go much, but I like doing that… Holidays, if you could fit holidays in there that's great. Visiting new places, I like doing that. Reading, reading I do a lot of. That's pretty much my life; it's not a bad life don't get me wrong!
Niels: Sounds pretty good, sounds pretty good! Now before we jump into the next topic, I just wanted to ask you a more, sort of a broad question. It's clear that the CTA space originates from sort of the longer term trend following strategies that you clearly were familiar with earlier on. You chose to be short term in your approach. So I'm interested in what the reasons were for you to choose to focus on this particular time frame, which frankly hasn't been that easy to master for many people. And we've also seen some firms exit that sort of time frame or the business as a whole because they didn't make it being a short term trader.
Barnaby: Yeah so I mean, kind of two reasons, first is we spent a lot of time building long term trend following models. I mean I wouldn't say that they're easy to do but they're not that hard to do. The data is pretty clean to get, daily data is pretty clean, it's easy, you can use it. We found that it's very hard to distinguish ourselves in that timeframe, and I just knew it would be very hard raising assets if you're very correlated to everybody else, it's very hard to distinguish yourself.
There was kind of at one part there was a commercial decision at the beginning that, "Hang on if we do it short term we can definitely get more uncorrelated." The second reason is because I sat in front of those screens for so many, so many years. I know a lot of nuances and subtleties in creating the systems, these short term intraday systems… Because the problem is, you know the problem is everywhere.
Data is the biggest problem. It's not easy and continuous like daily data; it jumps around. You've got tick sized changes. You've got open and closed time changes. You've got problems with liquidity, volatility, and certain historical events. But because I sort of traded through everything, we can incorporate those problems into our backtesting, into the data sets. I think basically we just had the right ingredients. We've also got a good math background. It's systematized what we're doing, but also because we trade intraday, we can overcome the pitfalls. It's a tricky business, you know we've got one guy just on data, there are only eight of us.
Niels: Yeah, no absolutely, it's so important. I mean I'm no expert in the short term space per se, but I think maybe it's fair to say that at some level human behavior when it comes to greed and fear, or for lack of a better description… that's part of what happens in the short term. In particular maybe more so than in the longer term, and that goes back to human behavior really. But with less humans being involved in trading, we hear more and more about systems and robots and what have you, do you think that there will be an effect on the trading time frame that you've chosen, given the fact that maybe there is less human interaction in that time frame than there use to be back in 5, 10, 20 years ago?
Barnaby: Well, yeah I mean I'd definitely say that the markets are more efficient when I was trading discretionary. You know as years went by I noticed that it was harder to make money on a discretionary basis because the efficiencies and execution were doing so much better instead of people ripping the market with 1,000 orders they would now seed them in. So there's less market impact on orders executed.
Niels: But as you say, as markets become more and more efficient, maybe more and more being you know executed with very little human intervention whether the short term pattern structures, whatever we call it, are changing in a way that it makes it harder to be short term from a trading point of view?
Barnaby: Yeah, so I agree. So in the very, very short term, yeah, definitely. I think there's a lot of competition there over high-frequency trading, and I think these guys are finding it harder and harder to compete. We're lucky because our space is not really high frequency. We're sort of average holders three, four days so we're in that weird space where there's not actually that much competition. So it's been fine for us, we haven't noticed anything like that, and even if it does happen because we trade so many systems, it doesn't matter because the ones that degrade performance we reduce exposure to anyway. So they would eventually be taken out of the portfolio, and at the same time we're always constantly researching. So we hope that we'd always have enough things to do to create enough alpha, to plug the gaps.
Niels: Sure, no absolutely. Now you've clearly been on an entrepreneurial journey for quite a while. It's almost like sort of a quest to achieve something or strive towards something, but part of all quests is that from time to time you endure a certain amount of suffering or challenges, you know I guess that's normal. How do you in your mind, unless of course it has been smooth sailing all along? I don't know? But how do you frame the challenges and maybe even the suffering that you've had to go through in order to get to where you are today? How do you look at that?
Barnaby: That's a tough question. I mean I think as long as you sort of believe in yourself, and you stay true to yourself, you can… if the times are tough, it doesn't matter. If you know your trading is correct, your systems correct, when the market moves happen, you'll always make money. So you know, I think you have to accept that there's always going to be, in the trading world, there's always going to be tough times, tough years. You know it's not a job where, it happens daily or weekly or monthly, even yearly. You just have to accept that there're going to be rubbish times, tough markets, low volatility, etc., etc. I mean we've had, since we started the program in late 2008 it's been actually quite a tough time raising assets. We started at the peak of the CTA cycle, so once we got our three-year track record there's nothing left to get. But yeah you know, we knew we had a good product, and you know I think we've been proved right eventually. When the market comes back with volatility, we can make money.
Niels: Sure, sure, sure. Just obviously, we're going to go into much more detail, but why don't you give me a very brief overview of the program you have today and where you're AUM is because I have a feeling you just crossed a big landmark or milestone, at least in my book, recently on that point.
Barnaby: Yes, we just crossed 100 million dollars!
Niels: Yeah, congratulations!
Barnaby: Yeah, which is great! It's been a long time coming, but we've made it, although now I think the new benchmark is 250 not 100 in any way!
Niels: I'm sure it is, yeah!
Barnaby: We'll see what happens. And then in terms of what we are and what we do, yeah, so we're a systematic managed futures program, that's what we do. We're short term, fully automated, fully auto-executed. So, in a nutshell, that's what we are; we trade over 50 systems. We trade across global futures markets, so multi-sector, multi-products, as many markets as we can get a hold of electronically basically. And we aim to create steady alpha and minimize our drawdowns, we like to keep our drawdowns to 10%, and work it from there.
Niels: So let's jump to the first setup question that I just want to talk to you about, and that's really about the organization. You already mentioned that you have a team of eight, but you're clearly in a space where infrastructure is important. You're trading relatively short term. So how have you gone about the challenge of setting up an infrastructure that, on one hand satisfies your own needs and requirements, but actually also maybe in the bigger picture satisfies the requirements of potential investors looking at you? How do you do that when you first start out and as you've grown?
Barnaby: Yes, when we first started out there was only four of us. So we've grown into a... How we've grown the business is very organic. You very quickly realize that you need to get up to some sort of institutional level, if you like, of business organization when you start raising money. So you need one guy solely charge your operations. Obviously being a systematic firm, we need at least one quant. When we first started off we had developed our signals. They were totally automated, but in terms of execution we weren't fully auto-executed. We had manual execution trade is putting us into the market, obviously. It's great in the sense that you've got a human overlay, but it's bad in the sense that you're limited to the amount of transactions you can do, especially if it gets busy. So over the years we've made more and more effort to become fully auto-executed. So we have complete straight through process now from signal generations to actually entering the orders into the market.
Niels: Maybe just for the listeners who are not so familiar with this part, and this is sort of digressing a little bit here, what are the challenges with going to auto-execution both in terms of actually getting it done, but also from a risk management point of view. I think sometimes when people hear a manager say, "Oh, but there's no human intervention and it just runs by itself, etc., etc." They can maybe become a little bit nervous. At the same time, I know that that is what a lot of investors would like you to be. So there's no human intervention in the process, but how do you balance this? How do you ensure that nothing goes wrong if I can put it that way?
Barnaby: Well, one of the first things to say is to admit that things always go wrong. Anyone who says things don't go wrong with auto-execution is lying. So, for instance, we never... We don't have any... running any black box running on the side. There's always someone, at least two people actually, in the office at all times overseeing, checking, making sure the execution is correct. Initially when we designed the software, we tested everything in the simulator environments. Then we go on to very small size in the real market. We try and write as many layers of risk as possible. Things for instance in case we haven't thought of something. We limit the amount of trades we can do per second, or minute, or market. We try and write all these limitations into failsafe in case we've forgotten to do something. So, yeah, it's done very incrementally, very slowly, very modularly. That's how we do it.
Niels: Maybe in your case as well I recently attended the conferences in Florida for our industry. You meet a lot of the brokers there, and they all talk about that they are unique because they have certain algos that they can offer. In terms of algos to execute, in the old days you would just have a stop order, and once it's hit you get things done as quickly as you can, so to speak, but today's world is a little bit more advanced. How much do you rely on the algo itself in order to get efficient execution do you think? if I can put it like that?
Barnaby: Yeah, that's a good question. At the moment, not too much. We do a little bit of TWAP VWAP. For us, it's more important to get our size through the market through doing more and more interactions of the existing systems. By that I mean base shifting parameters slightly, so you get different entry times and exit times. So you can get smaller and smaller amounts in the market. As it may be maybe our own kind of algorithm in a way. We tend to do more of that, than using these pre-canned exo-algo’s.
Niels: Sure, sure, now you're clearly in the growing phase of your business. I'm sure it will become much bigger over time and so on and so forth. If you take your trader hat off and you put you're building a business hat or the entrepreneurial hat on, how do you build a strong culture do you think in an organization?
Barnaby: Yeah, well it's a lot easier if it's kept smallish. That helps, we're all in one room, all together. I don't sit off anywhere in another office. We're in it together. I try and make sure that people's conversations are in line with their abilities, makes sure everyone gets a slice of the pie. We're also at the same time making sure that things are locked in so; you dangle enough of a carrot to spread it out over time as well. Things like that, it's quite a tough question actually. Maybe I should spend more time thinking about that.
Niels: Well, you can say it's a tough question in a sense and I think it's tough for, because many of us when we come into this world, we don't necessarily come because we've studied business organization or anything like that. We usually come in because we love the markets, and we love trading and so on and so forth. Suddenly we're forced to also become managers of people, managers of organizations, so you're right, it may not come natural to us to think about these things, but obviously they are important if you look at the long term picture and you may end up with a firm of 25 or 30 people, who knows. So yeah. It's interesting.
Barnaby: Yeah, you're definitely right there. That is a work in progress for me as a person.
Niels: Now you came from the discretionary side. You clearly have some experienced people alongside you. How do you combine the qualitative overlay with your quantitative approach? Do you know what I mean? How do you get the best of what you knew as a discretionary trader, or from an Experian executing trader from some of your staff, into the systematized, fully automated process?
Barnaby: Yes. That comes in sort of various ways. Like I said earlier because we know about the nuances of the intraday markets, the intraday prices. We can apply that to say a backtest. We can look at assumptions made by the quant team. If we make a lot of money over 1:30 we can come in and say, "Hang on guys there's usually an economic figure of 1:30 UK time. Is it realistic we're getting those prices?" There're qualitative inputs in terms of where we go in system design. We can look at maybe our existing systems. We can see where they're failing and try and hypothesize about how we can create a system to make money in the areas the existing systems aren't making, so there's... We can have a qualitative directional input into the research process.
Niels: I guess also, I guess, that goes also for the fact when it comes to execution. The experience that you've had and your team have had over the years having to do the execution yourself, I guess that helps when you start building the automation into it.
Barnaby: Yeah, massively. So you start with assumptions on your own experience about what kind of slippage you can expect per market. You build that into the model. You then reassess it as time goes by. As you trade, we constantly reassess the slippage. Are we under, over, scoring it? Yeah, loads of different things.
Niels: Sure. Now I want to jump to the next point that is about track record. Because obviously your track record is a little bit shorter than some of my other guests who have been around for 30 years. So they might have evolved even more so. When people look at your track record, how should they read it? Meaning, how has the program in the last six, seven years, how has it evolved over time?
Barnaby: So in terms of evolving, I wouldn't say it's actually evolved too much. Yeah, I'd say the duration in the short termness has calmed down a little bit from when we first started. That's because we gotten better on the quantitative side, the execution side, so we go a little bit shorter. In general we've tried to keep as many of the original models as possible. We just add more and more as we go by. There's really not much style drift anymore. We'll continue to stay in the timeframe we're in. We'll just try and add more and more competing models that help with the equity curve.
Niels: Ok, Ok, and you mentioned yourself that you managed to start trading pretty much at the peak of the CTA cycle. It has been, indeed, very challenging, but I'll give you certainly the applause for having posted positive returns every year since your inception. I think that's very impressive given the environment. Why do you think it was so difficult for the traditional CTA strategies to perform when you look at it from your point of view?
Barnaby: Um, one thing that helps is volatility, especially for us short term guys. You need some sort of continued movement. Even if you're a long term trend follower, you need continued movement. We haven't seen too much of that. This and last year we did, but before then we hadn't seen much of that. There's a lot of sort of risk on, risk off trading. So everyone became very correlated. So markets weren't hedging themselves off like they used to historically though that seems to be coming back now. So that was another reason.
Niels: Yeah, I'm just curious about this... your view and your answer on this question and that is, if you were given the choice of being able to look at a twenty-year track record of someone, or clearly knowing that the program had changed and evolved over time; or a simulated track record on the current system that the manager was running, which one do you think is most meaningful to look at?
Barnaby: Mmm, that's a good question. If it was on a current program and it was backtested using exactly the same parameters they were using now, I don't know if that would really be possible if it were dynamic, bit if that was the case then I think that has as much credibility as the real track record, if it's not fudged.
Niels: You know I agree, I think it's an interesting question because we do face... I say we, but investors face the conundrum about track records I think. I don't know whether they think about this in that way because people are very quick to disregard a backtest. On the other hand, if you're dealing with someone who's clearly experienced. Who knows how to do a backtest, then I think, personally at least, I think there is a lot of meaningful information in it. Although I'm not sure that many of the larger managers today would actually give some of that research data outside - maybe they would, maybe they wouldn't. It's interesting. Anyway, you don't need that, you've got a real track record so people can look at that instead.
Now let's move on to the heart of the strategy itself, namely the program. If you're just looking at it from the 30,000 foot point of view, what is it that you're trying to deliver with your program?
Barnaby: So yeah, what we're trying to deliver is... We're trying to deliver market movement, stretching alpha for market movement on shorter timeframes, non-trend timeframes, basically, not long-term timeframes. We're looking basically; we're looking to find points in the market where there is a higher probability than normal of the market moving away from that price point. We don't necessarily know which direction it's going to go, we just know for that given price, or that given point in time the market's going to move away more than it would normally do.
Niels: Can you explain that a little further for me? I'm not entirely sure that I understand what that means, whether that means that you're trying to revert back to some mean, or whether you're trying to actually follow a short term trend.
Barnaby: Yes, I guess essentially, you could classify as finding some sort of short term trend or short-term momentum in the market. So what we do, if you want to profit, if you think you've got a point in the market where the market might do that, if you structure a trade with a fixed stop-loss and you give it time to run, so you give it some sort of time based exit, if you structure the trade around that, you can make money.
Niels: So that's interesting. So you talk about obviously having a fixed stop. I think that makes perfect sense and maybe we'll talk about that a bit more when we get to the risk management side, but then you mention the word time based exit, which many people probably, I'm not sure that they are fully aware of what it means, but I guess in short it means that you're exiting the market based on a certain amount of time. It could be days, it could be hours that you've been into the trade. So my question is, why do you think that time based exists are the way to go? Because I don't think many people use that. I've come across it before, but it's not often.
Barnaby: Yeah, so traditionally if you're a trend follower you probably do some sort of hugging volatility stop. Something which follows the trend. The problem is if you try and do that intraday on a very short term basis, the market's way more volatile than the daily market is. If you start using these hugging volatility stops, it doesn't work. There's too much whipsaw. There's too much erratic behavior. The edges are still there in the market for it to continue in the direction you hope it's going to go, but the only way to capture it, we think, is via time-based exits, letting it run but actually having some sort of trailing stop.
Niels: It's funny, because when you describe it like that, you use the word trend following, and you use the word time-based stop, many would probably say well there's an inherent contradiction between the two. Because trend following is all about letting the trend run for indefinite if the trend was still up or down. You can't predict anything about time. How do you, I'm interested in how you would respond to that, but I'm also interested in how do you find out how long should a trend... how long should a trade be in order to capture the trend that you think there is.
Barnaby: Yes, essentially, you can do a time-based exit on a long term trend following. What you do is just... you would backtest and look at the parameters and have time as a parameter and stop and see where the P&L basically goes positive as time increases. Obviously on a backtest you could always find some sort of mathematical optimum on whatever... on the parameter of time. What we do is we don't really tend to pick optimal parameters, we try and pick a set of parameters. So we don't assume there is an optimal, we just say, if a trend following system is profitable with a time-based parameter exiting at two months and beyond, we can go and exit it at two months, three months, four months, five months, six months and that will still make money. If you don't have a volatility stop, it will probably still make money.
Niels: Ok, and how far out do you actually go? Clearly you mentioned three to four days average holding, but how far do you actually go out in your trading, how long can a trend be for you in order to capture it?
Barnaby: Yeah, so we've got a very small actual trend following component. That goes out for a maximum of 30 days though, to keep us within our remit, so there are very, very few trades at 30 days. There's a lot of skewed to one to two days, then three, four, five, six... It's like a decreasing curve on the frequency of exits.
Niels: So tell me a bit more about the kind of models that you run. You say that trend following is not necessarily what you do for the most part, but tell us a little bit about the different type of models you have inside the program and the reason why they're there. What do you like about them? What are you trying to achieve by using them?
Barnaby: Yeah, so one of my favorite ones are momentum anomalies intraday. So intraday momentum. So what we do is we look for a pickup in speed of the movement of the market intraday. That in itself is OK. You can make a little bit of money with that, but it doesn't... It's not good enough to create a program on. What you need to do is, we combine it with some sort of consolidation. So you have a pickup in momentum; the market starts to pause... We've got proprietary ways of measuring that or picking it up, and then what we find is that's a much better way of making money with intraday momentum. The fact that the market pauses again gives you a better reference point.
Niels: To get in or to get out?
Barnaby: To get into the position. Then you can have your time-based exits. You can vary the time-based parameters and run the trade with some sort of fixed stop that we would have backtested. I like that particular set of models because they're very fractal. They'll work on lots of different timeframes. You can use then on 30 second chart, bar chart. You can use it on a five minute, 30 minute, daily. It's very versatile and very low on parameter sets. So we do a lot of that. We do a lot of work looking at the open. The open...
Niels: Yes, that's what I picked up when I early in my introduction referred to Toby Crabel, and I have no idea whether that's true or not, but obviously that was one of the things that he got very good at and famous for, so tell me about that, that's interesting.
Barnaby: I'll tell you a story there, funny enough. I heard about Toby Crabel, and I managed to buy one of his books on Ebay. It cost me $1,000. It was quite a good read, and it gave us a few starting points or starting places to research. There's not many of those books around. I got a few ideas from Toby Crabel there. When we started backtesting, you do find that the market moves differently on the open than it does in any other parts of the day. If you can set up trades in certain ways, you can definitely make money, also from gaps in the market as well. We find opening gaps from a psychological point of view means that the market's done something different. Now the question is will it psychologically accept this price or not. So you can profit from how the market reacts after a gap as well.
Niels: Tell me more about when you say that markets behave differently at the open. Tell me more about what the difference is and why you think they are behaving differently - if there's a good logical reason why they should be behaving differently.
Barnaby: Well I guess one simple way of explaining is, if, especially with a gap, for instance. If the market gaps... Say if the market gaps up, short term participants are going to panic quickly. So from a psychological point they are going to be erratic in how they get out. So that tends to overplay or overpush the market. That's just an example, I think, of how you could perceive it from a psychological point of view. Also, I think just from a practical point of view as well, I think a lot of people just trade on the open. If they have to get in or out. I think they design a lot of the longer term models to trade on the open or on the close, for instance.
Niels: The open is interesting isn't it? I guess from... I'm sure you've been thinking about this for awhile and that is with electronic markets being open almost 24 hours a day, the difference in time at least, between the close and the open is becoming much smaller, so you would think that the need or the reaction from people might also change because maybe there's not such a big difference between the close and the open because they happen within an hour or so. Is that true that things actually are changing just because the exchanges are going more or less 24 hour trading?
Barnaby: Yes and no, we've sort of seen a little bit of degradation maybe in some of the markets, especially the grains. They've been chopping and changing closing times back and forth for the last five, six years. In general, even if the futures markets trades longer hours, the tendency being an underlying cash market still seems to be fixed - has a fixed open and close times. So that time gap put into the market on the basis of the cash as well. So the effects still seem to be there.
Niels: Interesting. For my own personal observation, I guess it could be somewhat harder to backtest these things because of the structural changes that the exchanges are making to this thing about the open.
Barnaby: This is where our background expertise come in because we lived it and breathed in from of the screens. We know all about the nuances there. That's all incorporated into our data set, our historical data set.
Niels: Absolutely. How many markets, Barny, do you trade?
Barnaby: We trade about 60 at the moment.
Niels: Those over 50 models, are they allowed to trade all 60 markets or are they more... Some models only trade grain, some models only trade foreign exchange, or how do you do it?
Barnaby: No, in general, our models have to trade, have to work on all markets. There're a few exceptions like short end doesn't move enough intraday to warrant it being in the bucket, so there're a few exceptions to the rule. Yeah, in general, we want to trade across as many markets as possible. That's how we know our system is robust. It works on all markets.
Niels: Maybe not a question specifically to you then, because you're smaller than the biggest firms in the business, but there is this debate, if I can call it that, between being fully diversified or being overweight financials and clearly the bigger you get, the most likely the more you will overweight financials where you will have better liquidity. Do you have a view on this? Is it your intention to say, "the day I can't be fully diversified, I'm going to cap my AUM," or do you think that it actually doesn't really matter? You can continue to find ways of making money even if you have to do it more in the financial sectors?
Barnaby: Yeah, that's a tough question as well. I'd probably say, for us, we would start capping it possibly. If I was so big that I had to start going more into financials... Yeah, I don't know, it's different for long term trend followers, especially in the bond markets. You get more carry effect as well. So that's not really in the majority of futures markets. It's more in the bond futures. So maybe they get an extra double whammy, as far as the trend there.
Niels: Are there any particular sectors or markets that your approach works better at?
Barnaby: Not really, no, no.
Niels: Interesting, interesting. Anyway, I was interrupting you when you were talking about the different types of models. You clearly like the momentum. You talked about models trading around the open, what else are you doing in your program?
Barnaby: Yeah, so we've got, as I said, a small trend component. There's probably no need to go into that. Everyone who's been interviewed probably already knows about that. We have quite a few different patterns we've sort of discovered over the years through our observations, which we backtested, and they seem to work going forward. So we've got a selection there we incorporate. We also look at time effects. So things could be calendar month effects, different things like that. At the beginning of a month there seems to be an effect in the market, so we try and capture that. So yeah, a variety of different things.
Niels: Sure, sure, and again, to the broader audience, but probably also to people who are specifically interested in your strategy and what you do. Are you able to visualize for us how you create a strategy? What actually happens from the idea generation to implementation in your program? What stages do you go into in order for it to meet all the... Tick the boxes and so on and so forth?
Barnaby: So how we work... What we don't do is go and data mine for something. We usually start with an observation of some sort, an hypothesis, and then work it from there. So, like I said, because I've traded discretionary intraday for so long, I've got a backlog of ideas to test. We go through those.
Also another starting place is, like I said before, if we've got an existing system and we see it's failing in certain parts of the market, we try and hypothesize a way of how we might be able to make money from the area where existing systems are failing. So they're good starting points. Then what we do, we'll roughly backtest it in a commercial piece of software just to sort of see if it's doing what we think it should be doing.
Then what we do is give it to one of our quants, and he'll go and actually properly program it in and we'll backtest extensively over all of our markets. We'll go back through it for debugging, etc., etc. Then what we do, we then follow it hypothetically. We usually wait for it to go back through the watermark before we even consider putting it in the real portfolio, just to give it an extra bit of confirmation, maybe like a forward test you could call it, in a way. So that's kind of how we operate.
Niels: Sure, and you talked about certain models that you like. I think certainly part of the audience that are listening...
Ending: Ready to learn more about the world's top traders? Go to TOPTRADERSUNPLUGGED.COM and sign up to receive the full transcripts of the first 10 episodes of the show and visit the show notes where you can find useful links to other amazing resources. Thanks for listening and we'll see you on the next episode of Top Traders Unplugged.
Become An Insider
Subscribe for free and be the first to receive new and exclusive interviews with the world's top traders. As an insider we'll also send insightful bonus content direct to your inbox.Free Instant Access »
You might also like:
Date posted: 02 Mar 2015no comments