In partnership with:
CME Group

47 How to Start Managing Client Capital with Bastian Bolesta of Deep Field Capital – 1of2

"With hindsight, I would probably argue that we were naive." - Bastian Bolesta (Tweet)

The story of Deep Field Capital's Founder and CEO is an interesting one. Bastian Bolesta met his future partners whilst spending a semester abroad in China before returning to Germany and then moving to Switzerland to join them. The team also moved from the discretionary trading space to a unique way of systematic trading. Deep Field is a relative newcomer to the industry as they don't have a 20+ year track record, so our conversation has timely insights for those looking to start a firm and begin managing external client capital.

Thanks for joining the discussion and please welcome our guest Bastian Bolesta.

Subscribe on:

Subscribe in iTunes Subscribe on Stitcher Subscribe on Soundcloud

In This Episode, You'll Learn:

  • Bastian’s background and how he got interested in finance.
  • How he grew up in Frankfurt and went to university there.
  • What he learned from a semester abroad in China and why he returned there after graduating.

    "It was a very exciting environment in terms of the country and how fast it was developing." - Bastian Bolesta (Tweet)

  • How he met his future business parters and started their first business.
  • How they went from the discretionary trading mindset to a systematic trading firm.

    "We started to develop smaller pieces of programs to support our discretionary trading." - Bastian Bolesta (Tweet)

  • How they came up with their initial trading ideas.
  • The difference between discretionary and systematic traders.
  • What he does when he’s not running his business.
  • Pros and cons of trading external capital and how they started trading other people’s money.

    "The manager actually has to be very careful about what investors he brings in as well." - Bastian Bolesta (Tweet)

  • How Bastian looks at the industry as a young and vibrant firm (started after 2008) and offers a different perspective then those who have 20+ years in the industry.
  • How he built the business from the ground up and how to grow a business smartly.

    "Life is way too short to work with people you don’t like." - Bastian Bolesta (Tweet)

  • How to convince investors that you can compete with larger managers.
  • How regulation affects the business.

    "We were just fast enough to develop in a less regulated environment." - Bastian Bolesta (Tweet)

  • The track record of their strategy and how investors should look at it.
  • The details of their trading program, and why they work in “themes”.
  • The allocation process for his program.

Resources & Links Mentioned in this Episode:

This episode was sponsored by Swiss Financial Services:

Connect with Deep Field Capital:

Visit the Website:

Call Deep Field Capital: +41 41 511 5588

E-Mail Deep Field Capital:

"We only employ trend following strategies, but we don’t show any correlation to the trend following space." - Bastian Bolesta (Tweet)

Full Transcript

The following is a full detailed transcript of this conversion. Click here to subscribe to our mailing list, and get full access to our library of downloadable eBook transcripts!


Welcome to another episode of 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 as the most honest and transparent account that I can make of what goes on inside the minds of some of the best investors in the world delivered to you via a one-on-one conversation. Today you're listening to episode 47. If this is the first episode you've head, you may want to go back and listen to all the earlier conversations. Before we go any further, let's find out who's on today's show.  


Hi, I'm Bastian Bolesta, the founder and CEO of Deep Field Capital, in Switzerland, and you are listening to Top Traders Unplugged. 


Thanks for doing that Bastian, and by the way, if you want to read the full transcript of today's episode, just visit the TOPTRADERSUNPLUGGED.COM website where you can find great details about today's guest. Now let's get started with part 1 of my conversation. I hope you will enjoy it. 

Bastian thank you so much for being with us today, I really appreciate your time. 


Thank you very much for having me today, Niels. 


My pleasure. Bastian, as I prepared for our conversation today, I found a couple of very different approaches. Not just to trading but also to the company's history and the way it's evolved. The big question on the trading side begins: How do we set the best team of systems and models on an ongoing basis? And the big question that you and your partners also had to deal with is: the advantages or disadvantages of managing external client assets. So I think there're some great stories and topics to dive into today that I think the audience will appreciate. Before we get to your story, I wanted to ask you a slightly different question, a question that I found that many of these successful guests that I have on the podcasts tend to also find it a little bit challenging at times, despite it being quite a straight forward question. So it goes something like this: If someone that you don't know comes up to you in a social gathering, not as a business pitch or situation, and they basically ask you, "Bastian, tell me what you do?" How do you respond, how do you explain what you do? 


I generally say that I run my own financial business together with three friends of mine and that we develop algorithmic trading programs based on our own trading and market experience. Of course, it always depends on the setting. If people are more interested then fine, and some not. But generally I start with the "having your own business," part. Because that's one of the key, key drivers for us doing what we're doing. 


Sure, great. Now let's stay with you for a little while longer. Tell me your story, how you got into this business in the first place and perhaps in order to put some extra color on, you can go back as far as you like and how you were as a young man growing up perhaps? 


Well, now-a-days we basically run Systematic as a management company called Deep Field Capital. But how did we get here? In terms of my background, I was born in Germany and raised near Frankfort, so naturally an interest in finance, the financial hub of Germany, was one of the first things to think about when actually living there, because you saw the skyscrapers all the time, and heard impressive stories of people working there. So I was quite interested in finance. But I didn't really know how to start, what kind of angle, what being in finance actually meant.  

So when I finished my A-levels, I decided to study business administration to focus on finance at Frankfort School of Finance Management. That particular University was strongly supported by the big financial institutions in Germany, where they have the idea that when you study there, you can work at the same time. So basically I studies at the University, but at the same time I started to work for Deutsche Bank as well, all in all for five years, beginning in the late 1990's. I spent a larger part of my time in the beginning in risk controlling, particularly then on market risk controlling, which was quite exciting. I first started in Frankfort. Then I could spend some time in London as well, and later on in New York. So this was a really exciting part in terms of working and on the academic side. I realized quite fast that whatever you're studying there doesn't necessarily directly connect what you have to do on the banking side. It was still quite interesting.  

The second part I went then to China on a semester abroad, which was a very interesting time, because it was a totally different setting. I had to study there a little bit, and in Mandarin as well, which I could speak slightly, but not as good. So I needed friends supporting me, translating whatever the slide said. It was the merger and acquisition course, for example. I could follow, driven by pictures, but it was a really interesting and difficult time, but a very exciting environment I would say, in terms of the country, how fast it was developing, and what kind of opportunities were available to pretty much everybody venturing there.  

Then I came back, and this was kind of difficult when you come from this really fast development environment, and you come back to Frankfort, and go back into Deutsche Bank. Being German myself, I don't want to be too hard on being German. Deutsche Bank is a particularly special financial institution I would say and keeps itself like that as well in comparison to the others. So, this was a really difficult transition, basically, coming from China and going back there. So I finished my degree at the University, and I actually changed the department as well. I worked then in relationship management for financial institutions in the Middle East, which was a little bit more dynamic I would say than the market risk controlling. But all in all, I somehow did not feel as passionate about being part of a bigger institution as I was before actually going to China.  

I somehow had the idea to develop my own business, and make my own experience what it means to set up a business and try things out, not being part at this particular time of a large institution. So I quit, in 2003 and went back to China, and tried all kinds of different businesses in this "wild west environment" I would say. It was really a "wild west environment" in the early 2000's in China. I set up together, with a friend, a headhunting company to recruit talent for financial institutions. So, trying to combine my understanding of the financial markets and my experience from back in Frankfort, and somehow use the environment we actually found in China to build up a business 

Most of the businesses we built up throughout the years there were not as successful, but we learned a lot, with some of them we could make some money. This was quite interesting. But then probably the biggest change in terms of my future development was when I met my today's partners. Two of them were traveling in China in 2005 to visit a friend of mine. They knew each other from University. So, I met these two guys, and they were day traders. Self-funded day traders actually. Not working for any kind of company, but actually starting really small, being probably really lucky, being at the right time at the right place, in the late 1990's or early 2000's. And now (they) basically had all this experience and understanding about markets and how to make money from this day to day encounter with market sitting twenty, twenty-four hours, five days a week or something like that in front of the Bloomberg screens. And this was really exciting for me, where we thought, well, that we could do something together. And the Idea was to take their trading knowledge and understanding, and the passion about trading and developing trading ideas, and combine it with my passion for building up businesses, entrepreneurship and that's basically when we partnered up and started ISATYS basically, which later developed into ISATYS Advisory, back here in Switzerland. That's probably how the entire thing started.  

What could be interesting for the listeners is that, how did we get from these prop trading into actually doing something with external investors? This was first driven by the idea to take trading ideas and somehow offer investors a part of it, of basically an opportunity to participate in these ideas. We first did that by developing products, particularly structured notes for institutional investors in Switzerland, mostly pension funds, which were running these large portfolios and were interested in getting exposure to certain trading ideas. But the entire organization structure did not really allow for reacting fast to certain fast rising market opportunities. So, we structured products, which allowed them basically to get these exposures. They then bought these products from banks who issued them, and this was a really interesting time because we could take our ideas and actually support external investors on their portfolio management with ideas we developed from our prop trading. 


Do you have an example of that Bastian? What would be an idea back from that time that you thought, " This could actually be something an institution would be interested in?" 


Probably, there are probably two dimensions. One dimension is just exposure to something. So, you can develop what is now days a "bread and butter business."- Everybody can do it basically. You develop a basket of certain companies for example: the water industry and you say, water is the new big thing. Or actually, we did a lot of products on China as well. China was not as hot in the early 2000's. It wasn't the focus of the majority of institution investors here. So we developed a basket of single stocks or indices on China, and this is probably the first part of such a product. The second part is that you give it some twist that you have mechanisms using different kinds of options to lock in returns to add a capital guarantee as well. Of course, always dependent on the issuer. That's something we all learned a little bit later in 2008. That 100% capital guarantee only means something if your issuer is still around. 


This is an easy example. So, mechanisms to lock in returns, exposure to things they could not directly invest. That's probably an easy example of what we did there. 


Sure. I wanted to... there're two topics that I mentioned, and you've also alluded to it already. And I want to explore them a little bit further before we dive in too deep. The first thing is really that one of your partners really comes from the discretionary trading side. And I'd like you to share the story of how you went from the discretionary trading mindset to becoming, essentially a systematic trader. That's not something that many people do. If you start off in one direction, you tend to sort of stay with that. Tell me about how that came about? 


Very interesting question. It's probably the first start that could be that we, in the beginning while we were actually venturing into the systematic space, we ourselves were not thinking about it like that: now we go systematic, and we actually move away from our discretionary roots. We first started to develop smaller pieces of program to support our discretionary trading. These programs were basically trading ideas put into a systematic framework and traded in an automated fashion. This was driven probably by two major factors. The first thing is, if you do things like that, things become more precise. If the heat hits the market and a lot of things are going on, if you do it automatically in an automated fashion, you take away the emotional aspects of trading.  

There's certain trades where emotional aspects might be helpful for you, but many, many trades are actually much better executed if you take the ideas you had once you were not under stress, and you somehow put them into a strict framework, which is then automatically executed by a program. And this was one factor why we started doing that, because it supported us on the discretionary side and made things much better, in terms of execution particularly. The ideas were still the same, but execution became more precise taking away the emotional aspect. The second thing is that you can do much more at the same time. A small team... a human being in front of the Bloomberg terminal of the trade station can only follow a certain number of trading ideas. If you start to do it in a more systematic automated fashion, you can actually do more things at the same time. So, these were probably the drivers when we started to explore that particular field.  

Maybe a different, additional driver could be that Ralph, who was basically trading on the discretionary side and Heiko who was in the early days a discretionary trader, but always very chart, analysis driven had a very interesting exchange over the years and developed the idea of working closer together. Ralph was the guy basically first looking for the program supporting us on the discretionary side. So, we realized over time that this is something worth spending more time on. It actually was quite beneficial in terms of making money, so it was quite profitable employing these programs. At the same time, it was a very exciting and interesting area we hadn't explored earlier. But, as I said in the beginning, we didn't think about ourselves as being discretionary traders, and now turning systematic. We basically continued what we were doing the entire time. We exchanged the human being against a computer which had to follow through certain rules, which was still defined by a human being. But not when heat hits the market, but when you're calm, and everything is well thought through. 


How did Ralph and Heiko, how did they initially come up with some of these trading ideas, which you later on managed to a large extent turn fully systematic? What was the original thinking, the original idea from trading in that way? Because it sounds to me like, although you say it was discretionary, it sounds to me that they did actually have certain rules at least that they were looking at or looking for, or maybe I'm wrong here. But tell me a little bit about sort of the... just before we jump into the systematic side of things, how that initial side from the trading aspect looked like? 


That's a very fair question. It probably goes down to the terminology: what is discretionary, and what is systematic? Of course, every single trader follows certain kinds of rules. There are hardly any people just waking up in the morning, turning on the Bloomberg and saying, "Oh, that's a hot buy and I don't know why, but I just think it is." This feeling you have that you should just do something, it's probably not the main driver of why you're actually acting on the market. So, everybody has certain kinds of rules. I think the major difference between a discretionary trader and a systematic trader is that the discretionary trader can always overrule his rules. He actually thinks, "Well, that's how I want to do it, it's well thought through," and then he sits in front of the screens, there's a lot of action on the market, and he, I wouldn't say he gets doubts or second thoughts, it's more like, "Oh no! that's not the right point, I'll wait a little bit." Particularly on the losing side, "Oh no no no no! This will come back!"  

You have this strong exit signal, based on your rules, but you just don't follow it through. In some cases, this gut feeling makes people successful. But I would probably argue that in the majority of cases it's a problem for them. So, if you turn systematic, that probably means in contrast to the discretionary part that you have to follow through whatever you thought through in the beginning because it's probably done by computer or you outsource it to a different trader who is just supposed to hit the button if the green light goes red or green. The trader themselves don't really know why it's going red and green. He should not think about it; he should just execute whatever the signal says. That's probably the difference between the discretionary part and the systematic part. I would agree that of course the ideas which are now reflected in trading in our Singularity program, they have always been part of the thinking of the entire team, but not traded in such an automated fashion I would say. 


Sure, absolutely. Before we get to the next question I wanted to ask you, in terms of the initial transition, I just want to ask you kind of a slightly personal question. You are busy building, running, Deep Field Capital, what do you do when you're not busy building and running Deep Field Capital? What do you like to do when you're not in the office? 


Thinking about Deep Field Capital (laugh) no, family: I have a lovely family. I have two children, two and a half years old and six months. This takes a major part of my time, together with my wife. We live in Switzerland; it's such a beautiful place. I'm currently looking out the window, and I can see the mountains. Some of them already having snow. So, you can do so many things out here - going mountain biking, hiking. Prior to actually taking external assets and investors’ money, I did ice climbing as well and things like that. Rock climbing - still do rock climbing, but it's safe, don't worry investors. So, there's plenty of things to do here. It's a really beautiful place to live together with a family, and building up a business as well. And that's what I spend most of my time on. 


You alluded to it just now again, and I wanted to try and get into this mindset, because this is quite a unique situation in my view. So, I wanted to ask you a little bit about the debate that you had internally and everything that you really had to come to terms with in terms of pros and cons of managing external capital versus doing what you did initially, which was just to manage your own capital as prop traders essentially. What was that thought process that you went through in order to get to the conclusion where you are now, where you actually run external capital? 


With hindsight I would probably argue that we were naive. It's really tough for us as a team, but we had an idea of what it means to manage capital for external investors. We had built these relationships with institutional investors here in Switzerland. And I had my experience with institutional investors in the Middle East when still working for Deutsche Bank. So, we knew what it actually means to deal with investors, but the large difference to our experiences in the past and what basically happened once we actually accepted external assets for trading in the Singularity program is that now we're actually actively managing money. The structured products we developed, these were one-time products, they were not actively managed, you did something good or you didn't do something good. But you couldn't do anything about it. The only thing the investor could sell it with a loss or he was really happy because it was a successful trading idea put into a product.  

If you actually accept assets, and you're basically trading it even when the program is doing on a single day you're still responsible for whatever's happening. These investors have invested in the product based on certain information provided to them; on relationship you have built with some prior to that, maybe they're not caring about what you're doing there because they liked you and know you from different kinds of businesses. This was a totally different thing. It put a lot of pressure on us, in different types of situations. Which we didn't anticipate when we actually sat down and said, "Well the Singularity program we started trading it live in July 2010, and it was quite successful. Well, it's successful; it has capacity. We don't see any issues that we would run into or any limitations on the capacity side. So why don't we take additional assets from external investors?" We had some good experiences working with external people in the past, why not do that as well on the active asset management side? So we started the Singularity fund one year later, in September 2011. We haven't really covered all the aspects that would actually change once you start trading external assets. 


Can you give me an example of how managing external capital challenged the way you think about what is best from a trading point of view? 


In terms of that you basically have to spend additional time, you can't just spend time on research for developing your program, but you have to spend more time on operational aspects as well, and on constant communication. Is that the direction you wanted to go with this question? 


Yes, but also in a sense that sometimes you know as a trader that this is the best way, or this is the right way to trade. But sometimes you have to make commercial decisions, where it could be about, I imagine that some people have a certain risk tolerance themselves, and they know that the program is solid, and they trade it like this. And then you see you start to taking external capital and then you find out well, actually maybe investors don't have that risk tolerance, so I can't trade it this way. There are some examples out there in our industry where some firms that have been incredibly successful run with very high volatility. That's simply because they do know that over time it is the right thing to do. But at the same time, it may be very hard for them to raise assets, because not everyone is willing to have that kind of volatility. So, it's more that side of things as well that I thought would be interesting to hear your thoughts about. 


In the beginning there wasn't such an issue in terms of this aspect. We did not sit down and decide to become an asset manager. At the center of our project, used to be and still is the proprietary trading. So basically putting our own assets at risk. This goes for testing new trading ideas and the development of new trading ideas, as well as the actual trading. Around this... let's call it traders approach, we have built an institutional set up to enable external investors to invest alongside us into our systematic strategies. This aspect of being alongside us was probably one reason why we started the external asset management business as well. Because we didn't see it as the center point of what we're doing, but actually it's something we attach to our prop trading 

This was a really good idea, and it still drives us the majority of the time. However, once you start taking external assets, you feel much more responsible as well. You feel new pressures you haven't really experienced before. Particularly in times when it's getting difficult, when you have drawdowns. If you lose money, and it's your own money, it's your own problem. If it's the teams money, it's the team's problem, and we would have to discuss in the team. For example, if the calculation was wrong or we had an operational hiccup, or if the program was down for half an hour, which, fortunately, doesn't happen. But in the early days when we started to explore in 2009 and 2010, there was still the operational side. It wasn't as institutionalized as today. It didn't really matter so much because we knew that it was our problem. If you're really unlucky and you can't trade on a particular day, which would have turned into the best trading day of your life, or of the year, it's your problem.  

But, once you have external investors’ money, it's a totally different game. Because they will ask you, "Why didn't you trade yesterday? I picked up the broker statements, and I can't see any trades or much less trades than normal. What happened? Didn't you miss some of the performance?" And this is totally different, because if something like that happens, you can always work on it and explain everybody tries their best to have the best operational setup and everything, and we're really lucky that we're working with good service providers, and that our set up today is very solid. But, before something like that happens you already feel pressure. You're already afraid that something like that might happen. Or in the drawdown, you're afraid that the drawdown will continue. Whereas a prop trader, or just trading your own money, you're not as afraid, because you believe in your system.  

You constantly do reconciliation checks that the system is doing what it's supposed to do. If it's continuing in a drawdown, and when you continue to lose money over a certain period of time and you're happy with it because it's doing what it's supposed to do, that's about it. You can continue focusing on new research or just go outside hiking. If this happens with external assets, you're much deeper in communication with investors. They start questioning if what you're doing is correct or if you could change it a little bit and it would be much better. If the risk is not too high. This is totally different and could overwhelm the team, particularly in the beginning. We have learned to deal with that. We had our larger drawdown in Autumn last year, and learned a lot about that, and did the mistakes back there as well, and I think that we are much, much stronger now, and much more relaxed as well.  

You could probably argue as an investor, you have to be careful in selecting your managers, in terms of what your portfolio looks like, but also what kind of managers you want to work with and invest with. I would argue that exactly the same goes for a manager. The manager has to be very careful about his investors as well. We were very hungry in the beginning and accepted all kinds of different assets. If you want to grow your fund, and someone comes by and says, "it sounds really interesting what you're doing, and you have such a great performance." You would probably accept his hundred thousand, two-hundred thousand, five-hundred thousand Euros, because it will bring your fund further. But, with this particular type of investor, I would call it "fast money" or investors chasing past returns. You have to be really lucky in terms of timing. They're good people, and it's a valid point that you look at investors and managers, and if they do performance, you want to have a share of it. It's perfectly fine.  

From the manager perspective, you have to be aware that this money will disappear as fast as it just appeared. You have to be lucky in terms of timing because if you're at a beginning of a business cycle where the market environment plays into your favor, then it's perfectly fine. This particular investor will make money, he will see one all time high after another all time high, and maybe he will have made enough money once you actually go into a drawdown period, which will happen, over an entire business cycle, you will have these periods. And maybe he will even stick with you. But if this investor is investing in a rough time, or just in a transition period where markets get more and more difficult for you, and you as a manager know that that's a certain period that you have to navigate these rough waters. But they most likely will turn into something which is profitable for you. Taken that your models are not broken, or you will not come up with new ideas in case the environment continues in that fashion. Then this particular investor will leave you again. And your funds will get smaller again. This can be really tricky because you're first so happy that you have rising assets, and then you're so disappointed that you're losing them again. This has been an interesting and important learning, but I would not argue that I wouldn't do it again. If you're a young manager and you set up your own business, I would still probably still go for it for the money. Being aware of these different types of investors is a really important thing. 


Sure, the reality is really is that I think you made some very important and some great points, Bastian. And at the end of the day, I think it's important to realize that this is the way the journey is. That you don't start out as a new manager, and suddenly you get only long-term institutional clients who will never redeem and so on and so forth. You have to take what comes along. The other thing I think that is very important is to just be open about the fact that we're not perfect. Not even the big companies today are perfect. We may not see the flaws, but they're there. And we're all human beings; we all put our trousers on in the morning in the same way and so I think that's very important 

Before we jump into the more business and strategy sections, I wanted to have a fresh perspective from your point of view. Because many of the guests that I've had on, and have been very privileged to have on, have twenty, thirty, forty years of experience in the hedge fund or managed futures industry. That obviously gives them a certain perspective as to how they look at the industry. In your case, you're coming from it as still relatively new, but you've actually ventured into this area at quite an important time: just after 2008, and so on and so forth. How does this fresh perspective, without all the back history from the 90's and so on and so forth, how do you look at the industry today, and all it's been through in the last couple of years? How do you see it? 


That's a good question, because, in the beginning we didn't know much about the industry. As I said earlier, we were not aware that we were venturing into the systematic space, or didn't perceive ourselves as doing that. Once we were in the systematic space and providing a product as well in the Singularity program, We didn't really know how to describe it. So, what are we doing here? So what we had to do was reach out to different managers, and to go to all kinds of different fairs, and listen to all kinds of different presentations in order to get an understanding of how the big guys in the industry are talking about their products. How do they perceive the current status of the industry? And of course, we developed an understanding that managed futures programs, particularly trend following programs were going through a really rough period. They had strong difficulties in the past 2008, 2009 environment.  

But we ourselves didn't really face these struggles. Parts of our programs were already trading in the 2008 environment. Of course, the entire program is built on data sets, including 2008 as well. Referring to your question regarding the 1980's or the 1990's, we didn't look as far in terms of data, or how the program was working, because we were only trading futures. We depend on highly liquid electronic markets. If you look at the liquidity, and how liquidity has evolved over time,  in the majority of futures you can't really go much beyond the 2000's. For the majority of futures we are trading the liquidity became sufficient for the way we are trading in 2002, 2003, and in some cases 2004 and 2005. The equity side for example S&P probably earlier, but if you think about grains or copper, it's certainly later. So from this natural setting what we were doing there, we were not looking much further than what was going on before 2000.  

Now going out and talking to all these different players in the industry, we understood that the industry was going through some kind of crisis, and experiencing a lot of money outflows. The interesting thing is I was just talking about certain types of investors chasing past returns, and was referring to smaller tickets like 500K or 100K or something like that. But actually I would argue that the industry experienced that on a much larger scale as well. Managed futures was so successful in the 2008 environment. So when crisis hit the market, they could actually show and play to their strengths. What happened afterwards is they got a lot of money. A lot of institutional and professional investors, which should know better, not necessarily just chasing the returns, invested into this space and were quite disappointed over the last couple of years when the space didn't deliver. For us, being a really young and fast developing and merging manager, we observed the entire thing but were not as affected by the struggles the majority of players went through. If you look at our overall development we were somehow able to make a decent amounts of money, and deliver good performance with the program, even in that difficult environment, but couldn't escape the deteriorating effects of a more difficult market. It became even more difficult for us. We were more like a free spirit in the slightly darkened and sad environment. This space was facing over the last couple of years. 


Now let's jump to the first topic relating more to your business today. We've heard the story and as a small manager, we don't need to spend too much time on how you've structured your infrastructure and so on and so forth. But, do feel free to share some of that. But what I'm more interested in is more the mindset behind the journey. Meaning as you have to build an organization and have to build a business, how do you decide where to allocate new resources? And what to add to your organization first, and how do you balance the risk of adding people too soon before the assets come and adding them too late because people might not invest with you because there's only X number of people? I think a lot of people would really appreciate to understand a real life, real time example of how you build a successful asset management company from scratch? 


For us the starting point, as I said earlier, wasn't necessarily, "Let's sit down and build a business." We already had our trading business going on. The cornerstone of what we were doing back then was that we were doing it as friends. So I sometimes refer to it as "family style." So we did it "family style." We sometimes still say that we want to keep it family style, which puts some limits on how many people you want to hire. If we incorporate new people in the team, there's generally always a thought, "Can these people stay with us for longer?" We're not just looking for their talent or the new ideas that they bring in, but also their personality. If you can actually imagine spending time with them on a remote island for more than an hour. And considering that work takes a major part of your time, life is way too short to work with people you don't like. That is one major cornerstone or major pillar of the business.  

Then in our case, you have to look at what the Singularity program is actually doing. It's a really fast, highly reactive program. We have a lot of trading. That puts a lot of emphasis on building up a top institutionalized infrastructure once you start accepting external assets. Particularly if you want to attract institutional assets. We were basically forced, right from the beginning when we said we would open it up for external investors, that we'd have to spend a significant amount of resources and time on building up this operational side. In terms of having top tear brokers, in terms of having all these different recovery aspects of a business, different other places. Doing it in a virtual space, so using cloud computing for the servers, but also the calculations.  

This was an important thing that we had to focus on right from the beginning. If you can't deliver that, particularly when you're trading so much, we will not attract large assets, particularly institutional assets. A lot of resources on that, this was kind of up to some point unexpected from the team as well, because it took away some resources that we normally would spend on creating new ideas. But fortunately we started to realize that quite fast. We started to outsource many of the things probably larger investors would do internally. So, middle and back office functions, for example, have been outsourced to a top service provider in the industry. So we get the data, and do the reconciliation inside, but they basically gather all the data.  

Everything has been automated on our side in terms of the actual trading and in terms of how the reporting systems are working. So if you start using very good service providers, you can actually keep it family style. It takes a while to build up these relationships until everything is running smoothly, but once it's in place you have so much more time and resources to spend on your program. And as we are trading fully automated, we don't have to spend much time looking after the program in terms of execution. The major resources went into operations, and to continue our research. This is still the case, we probably see more and more resources on research. Particularly in the last eighteen months, because how the operation is set up, is pretty much finalized. We just onboarded another top-tier brokerage relationship, in order to be a bit more flexible on what we can offer to our external investors. But we're happy with the service providers we already have on board for the last couple of years. I would say if you're a small manager and you start to think about what can be outsourced, and what do I necessarily have to do in house, and what kind of resources can I use to do your work in house. Then you can actually keep it family style, but grow substantially in assets.  


I appreciate that. One question though and it's probably to do more with perception than reality. How do you convince a potential investor that you're able to compete and deliver the same operational excellence as a firm with fifty people do? Because at the end of the day, that's a major challenge, I think, when it comes to attracting these investors. Frankly, a lot of the people who have to make the decision about allocating to a small manager, probably feel some pressure in terms of the way they look at their due diligence. And they would feel more comfortable with a bigger firm, even though there may not be any factual difference in terms of operational excellence. How have you gone about convincing people that it's just as good to go with you as it is to go with a much bigger firm? 


It somewhat goes back to what I just said earlier. If you try to build up everything yourself, you're certainly under much more pressure in terms of, are they really able to do that and keep the high quality? So you really have to think about what you want to keep in-house and what you can outsource to the top tier service providers. A top tier service provider helps investors to tick a box, particularly on the operational side. All our trading programs are developed in-house. The trading side as well as the research side, the back testing side, are all in-house development software. People have to trust us that we know what we're doing there and that we have the recourses to keep that up to date and progress over time as well.  

The additional aspects of middle and back office functions, reporting things, these can be outsourced now-a-days. It did not work that way ten or fifteen years ago. There was no offering to smaller managers; things were way too expensive. But now-a-days even top tier service providers have offerings for smaller managers. You don't need the big, big operational package by one of these top tiers. You can just take part of it, and build up over time. If you need an additional reporting function, you can get this additional reporting function. If you need a regulatory reporting function, they can offer it to you. It always takes time to implement these things. I think that is really important thing, helping investors to trust you on these aspects of the operational side of the business, and not be afraid that you are a small manager in comparison to the larger ones. 


I just had one more question, I was just curious now that we're talking about the challenges of building something up, what about regulation? Regulation has changed a lot, even since you started. It certainly hasn't made it easier to be a small, in my view. What do you think about what's happening from a regulatory point of view, and how does that affect your business? 


That's a really important aspect of the business now a days. You're perfectly right; it changed dramatically. We probably were lucky that we were just fast enough to develop in a less regulated environment, where we could explore in all different directions, getting an understanding of what we really wanted to do. Now, we're well prepared in order to face these changing regulations. If you're a manager or you actually want to become one, and you have a trading idea or program, and you want to start your business today, it's definitely more difficult. For us, having a running business, spending time on developing an understanding of what kinds of new regulations are coming up, for example. Thinking about the AIFMD in Europe or the equivalent in Switzerland, or all these different changes in the US as well. It takes so much time to get an understanding, and you don't necessarily have the financial resources in order to have external consignments doing all aspects of that for you.  

This can be overwhelming for somebody starting a new business. In our case it has resulted in the situation that we are most dominantly focusing on the US as a market. Our entire regulation efforts, and for the expanding of a business, we currently focus on the US. While we're actually sitting in Switzerland, we do not actively approach Swiss investors or European continental investors. We have all these existing relationships, which is fine, and there's additional potential for us to grow with these relationships. In terms of a regulatory environment it has become much more difficult for a new manager, but also for a manager of our size at the moment. You always have to face the question, "Do you want to be AIFMD regulated or not?"  

This will takes away a lot of time for you for the next six or twelve months if you go down that road. You need sufficient assets as well, in order to pay for the running costs. I'm not quite sure if that really helps the end investor here in Europe. I understand that people who develop these regulations probably have the best intentions in mind to protect investors. That's is one of the key reasons, at least what you read and hear about. What's actually happening is that it's getting really difficult for smaller managers to offer their investment ideas to investors out there. I'm not just talking about retail investors; it's even institutional investors are limited in the options that they have in investing based on these new regulations. They have to tick the box.  

If they invest into someone who isn't regulated with a certain standard, and something goes wrong, they face a career risk. If they are right, no one will say to them, "You made a really good choice, you picked a really good manager despite that he's not regulated to standard A, but just to standard B." So there's a big career risk for decision makers on the institutional side. They most likely will not take this risk, unless they know you really well. I strongly believe that this regulation will limit the ability of the European alternative investment space to develop new trading ideas and bring them to the market. I would argue that it's not necessary the really big managers that come up with new ideas. Of course, they have all these large resources, and hundreds of PhDs developing certain things. They all have really interesting products and have been around for a long time, and I have a lot of respect for what they're doing. But, a lot of interesting ideas are coming from smaller managers as well. Some of them fail, others are successful. And if you take away the opportunity for them to grow because investors provide them with capital, being aware of the risk of providing capital to small managers. Please people, investors, keep in mind that not everyone has to be protected. If you have a regulation like that, it forces investors to act in a certain way, and it limits the ability of the space here in Europe to develop. I think that the US is going in a totally different direction. I think that the business environment there becomes more interesting. And in certain other geographical areas for example Asia, where you see a growing community, particularly in our managed futures space. So, I think that it will be difficult for Europe, to keep that edge that it might have had over the last ten years. 


Sure, absolutely. I think you said it very diplomatically. Another way of saying it, of course, is that it's interesting that many investors are not allowed to invest in a very diversified portfolio of markets with a high level of risk controls. Yet, on the other hand, they're allowed to go out and invest all their money in Enron, and they could go with any other stock for that matter, which they could lose overnight. Regulation is certainly an interesting factor in all of this.  

But anyway, let's shift gear and talk about something which is really important, I think, when you look at any manager. And that's the track record of a strategy, because as we know, many strategies start out in one way that builds a long track record, but over time it evolves. And, therefore, it can be very difficult for any investor to look at a track record and get a meaningful idea of what performance risk might actually look like, going forward. And of course, we know that we always have to say that past performance is not indicative of future results, which is very true. But, looking at your track record, how would you say people should read it? Are there any things along the way since 2010, when you started, where you said, yeah, this was one period of what we did, we then it evolved a little bit into this, and how would you advise people to read your track record when they look at it? 


Particularly when referring to the chasing past return aspect of what we discussed earlier, I think that there are two dimensions. The first dimension is: has the manager done something differently over time? Has the program evolved towards a certain direction? And the second dimension is: have the markets evolved? Has the market environment been the same the entire time? Because of course, the results are dependent on the market environment as well. And if you want to look at both dimensions, you have to take a lot of time, particularly when you're sitting down with a manager to get an understanding of what you're actually doing, what kind of results you expect in a certain market environment, then refer back, "Okay, 2010 had such and such a market environment, 2011, 2012, 2013. This has changed; this is one aspect why your P&L went up or down in certain periods.  

At the same time, they have to cross check, have they done anything differently? I would argue that the managed futures space if you work on the systematic side, you constantly work on the research side to cross check if your earlier assumptions are correct. If you've done something wrong, if everything is working in line. You shouldn't only do that if you're in a drawdown, but you should constantly check. Also, when you have higher than expected returns, you better make sure that you understand why this actually has happened. At the same time, you come up with new ideas, and you want to introduce these new ideas to the trading program. Make sure that it doesn't result in what people refer to as style drift. Because if you have investors just rating returns, they are fine if you have returns. As a manager, you would like to head more in the direction that investors have you in their portfolio because you deliver a certain risk-return profile.  

These investors are really the exciting ones, because you can have really deep discussions with them, and they stick with you as well when you have a drawdown, and you can learn a lot from these discussions as well. In our case we are more interested in these kinds of investors: the institutional, professional investors, who put together portfolios with you being a building block of that portfolio, in the expectation that you deliver something, and don't have any style drift there. So, you have to make sure if you come up with new ideas, that these new ideas are well thought through, tested from all kinds of different angles in order to reduce or mitigate the risk that they change your risk-return profile.  

At the same time, you have changing market environments which could make this picture a bit blurry. You might think that you're still doing the same, but you're not because the market environment changed, and it just looks the same. Or the opposite, you're doing the same, but the results tell you you're not, but it's driven by the market environment and not by the new ideas you have. That's a thin line you have to walk being a manager, by further evolving your program, or developing new ideas. This is something that takes a lot of time when sitting down with an investor. They have to have the patience and the interest to develop the understanding of what kind of market environments you are expected to deliver, and what kind of performance.  

If you look at what happened in the Singularity program, the core of our program has remained the same since July 2010. So the idea of how Singularity looks at incoming price data and analyzes this price data, and generates signals, has remained the same. Our strong focus on trend following signals, despite not being a trend follower, but we'll probably go into that a little bit later in our communication, it has remained the same. We have an aspect where we say that if something is not automated, you should automate it over time. So whenever you have a manual process in a systematic way, but manual, you should try to automate it to exclude human error, and at the same time to have the potential of increasing complexity to look at a higher number of variations that a human being might not be able to cover in that manual process.  

There have been two things where we automated aspects in our trading program. In 2011 we automated our allocation process from a manual process by our portfolio manager in 2010 to an automated process where the program is putting together the portfolio based on the same rules that the portfolio manager did in the past. This also allowed us increasing complexity, in terms of how you can actually put together the portfolio. The second thing I would argue is the allocation process in terms of markets: What kinds of markets do you have in your portfolio? This has been something we've been researching for quite some time as well. Which has been introduced in 2014, where now the number of markets in your portfolio on a monthly basis can be higher or lower depending on certain drivers or market conditions, particularly volatility in markets. This has been something that was a manual process in the past and has been automated. That's probably it in terms of how the program evolved. 


Now we're going to jump into the nitty-gritty of the program. I want to make a couple of observations, because I think this is where it's really interesting, and I very much look forward to your explanation. I think one way of looking at what you do is to try and make the distinction between the fact that many managers when we research certain things, we put together a model that has certain parameters. We tend to run that model over a number of markets, and these things are constant. But what you're doing is very different in a sense that you're trying to set the team, both in terms of models, systems parameters, and markets, every single month. So, take us all the way from the top, and all the way down, and explain what the Singularity program really does and what it's all about. 


So basically today we have the Singularity program which is a futures trading program. And we have a second program that we call Singularity Pure Equity, which is basically taking the Singularity trading philosophy and just applying it to the equity markets,. These are basically our two programs. They both have the same original thought as to how we look at markets and how we trade the markets. Singularity is purely systematic and fully automated. It trades only futures across liquid future markets, currently up to twenty-three markets across all asset classes. So that is a starting point when you want to categorize it.  

We have designed it to deliver what we call a multidirectional alpha stream. Basically, independently from the direction of whatever market, trying to capture different return sources, or return drivers and generate an alpha in comparison to traditional approaches, particularly in the alternative investment space, and consequentially showing a low correlation. We do that by employing a quantitative analysis of real time price data. The interesting part is that while we, I said it earlier while we only employ trend following strategies, we do not show any correlation to the trend following space.  

This was a very difficult aspect for us as well, when developing a better understanding of what we were doing when communicating to the investors. As we did not sit down and develop a systematic program, and no one on our team came from academia in terms of systematic trading, or used to work for a different systematic asset manager. We haven't had the terminology in order to describe what we were doing. And as we were only trading trend following signals, we went out to investors and said that we have a trend following program. We made money when they didn't make money, and we lost money when they made money. So no correlation, so we had to learn over time by communicating with investors that while using trend following signals, we probably are not a classical trend follower. Particularly in the way we are trading.  

We now-a-days strongly believe that this idiosyncratic profile is driven by what we refer to as intra-market diversification which is probably a term that is just used by us. We came up with this term by going out to investors... one day, for example, in the really early days we were trading eight markets at the time. Our performance was quite good, and we sat down and said, "Well, we have a trend following program, we currently trade eight markets." And the guys stopped us immediately and said, "Whoa, wait a moment, you're only trading eight markets? You're not diversified. How can you only trade eight markets?" This was the big question, why were we not diversified? He said, "Go back home and do your homework!" or something like that. So we went back home, and we said, well let's do our homework.  

First we only used a trend following program for signals; we deliver performance here, but he said we should do our own work, so why are we not diversified? What does it mean to be diversified? So we looked at different programs. I would argue that a classical trend following program has a smaller number of well researched and robust strategies or models, and they are employed across a large number of markets - as large as possible, fifty, eighty, one hundred markets. Some people tell you that they're even trading more markets. That's probably the aspect that this particular investor was referring to when he gave us homework.  

We said, "well, but we are trading eight markets and we actually do not feel that we are not diversified in terms of if you look at our risk and how everything is evolving, we have quite some diversification here." Then we realized that diversifying across a high number of markets is something we could refer to as cross-market diversification. Which is an important thing, and has been around for a really long time in investment. Everybody knows that if you diversify across different asset classes and markets, there's something worthwhile there. Then we looked at what we were doing, and we employed a really high number of strategies or strategy variations in every single market we are trading. When we went to this investor and said that we are trading eight markets, what we forgot to tell him was that we were trading eight markets, but we had around one thousand variations of strategies being active in that market while we were speaking.  

We came up with a terminology that we refer to as a strategy element. These are not just thousands of strategies, but they are variations of strategies. If you look at Singularity now-a-days, for example: if you have a portfolio of ten million, you would expect one thousand five hundred strategy elements in up to eight or twelve markets, maybe fourteen markets. That means that every single market trades, let's take ten; it makes the calculation easier and has one hundred and fifty strategy elements. So what does it actually mean? We have to look at what is a strategy element? This is a term we came up with. As I said earlier, we developed our trading software and testing software all in-house. If you're a programmer, you have these objects, which are hard coded and defined. A strategy element is an object in the programming language where we define certain rules, and we define it on three different levels. Then you take this object and put it into the trading program and let it trade for the entire month. 


I'm going to stop you there, and just try to explain again with using a terminology also that everyone can understand. Let's make it really basic here, in terms of what variations could that be? 


Sure. So, you have a strategy element, and it has three layers. On the first layer, we define what kind of trend following strategy this little fellow is using. You can think about it as individuals, thousands of individuals with different characteristics who look differently at markets. The first level is trend following. Let's take moving average. The second level is what we refer to as trading parameters. So one trading parameter could be time, one hundred and fifty days on average. On the third level, you have risk parameters, stop loss, and you can calculate stop loss in different ways. So, this little fellow gets a certain type of stop loss. Or take profit as another thing, so he does takes profit based on certain rules. This is hard coded, so this little guy will always follow his rules, based on these three layers.  

Let's say this is little guy, number fifty, and now we have his friend, number fifty-one, he is also using a moving average, but he doesn't have one hundred and fifty days. He has one hundred and sixty days, but the same stop. Then you have number fifty-three, fifty-four, fifty-five, so you have a high number of variations once you start slightly changing or giving different values to the three different layers. So potentially you can create a universe of thousands of combinations of strategy elements. And that's what we're doing when we're defining the potentially tradable universe on a monthly basis.  

Then you have a process where you systematically select these strategy elements across a given number of markets. It's a process which slowly builds up the portfolio framed by strong risk criteria. In the end, you have these one thousand five hundred strategy elements. For example: for an account of ten million, and then these elements trade independently for an entire month. This is a really interesting and important aspect. For example, let's take two little fellows in gold. Gold is trading; something's happening on the market, and one strategy element gets a signal to go long. So it buys one gold contract. Another gold strategy element, two minutes later gets a signal to go short because it's based on different rules based on these three different layers. So it's a different entry point for both of them. We let them trade freely. The portfolio is the net result of all the precisions and actions of these strategy elements. This results in a higher activity because they all trade independently, they don't look right or left like all the other guys are doing. You do that for an entire month, and that's why we argue that there's an intra-market diversification. Because these strategy elements, even when being employed in the same market, try to capture different drivers of return. They do not depend on the actions of the others. They all do it independently. And some of them are successful in those months, and some are not successful. Your trading result is a result of all these independent actions of these strategy elements. 


Why did you choose one month, why not two months or three months? Why does the team need to be reset every month? 


That's a really good question, because one month is not a natural given point. In fact, of course once you develop something, we as a team always argue you have to make sure that you mitigate the risk of curve fitting. So if one month is just the best you can do based on the data set that you are developing that's a really dangerous thing to do. So you basically should test if it also works with for every three weeks or five week, or even much longer periods. We realize when testing it that the reality is probably somewhere between three and six weeks. That's all fine and doesn't have to be one month. But if you come from an asset manager perspective and you're dealing with outside investors, generally you have incoming assets and out-flowing assets on a monthly basis. So it naturally makes sense to allocate the portfolio on a monthly basis if it's not putting you in a much worse position in comparison to five or six weeks. So when we tested that we basically got what we refer to as a plateau result. Four weeks was pretty as much good as three weeks or five, or six weeks. So we could be sure that we haven't just pinpointed and were lucky and just found exactly the right spot for the data we developed the system on. It's actually quite robust and stable because we are sitting on a plateau when actually doing it on a monthly basis, so that's the reason. 


So you have all these different strategy elements in your portfolio that you could choose from and we need to talk about how the selection process takes place, I guess as well, but tell me before we go down that path, tell me a little bit about these small fellows, these strategy elements. What do they look like? What kind of strategies... because trend following is many things to many people. You mentioned moving averages, maybe there's some other things, but also I'm thinking in terms of time. How short is the shortest fellow and how long is the longest fellow? 


Actually we sometimes refer to them really as fellows, because this makes things... I like the graphic aspects of it. They all use trend following, so as you perfectly rightly said there is more than moving averages. We have range breakout as well which probably is important. We did not come up with a new idea or concept of how to look at trend following or what trend following actually means. We were actually not even bothered to be a trend follower. We said we like these basic strategies and they have been well researched, and so we don't have to put much time on that and we can actually focus on how to actually define these fellows based on level two and three.  

The idea was basically to have a real high number of variations, and that's probably where our proprietary trading on the discretionary side comes into play; where we have all kinds of different ideas what can actually happen in a market and how these fellows actually could and should react to it, but not in order to just pick the ones you think are the best, you actually offer the entire process a high number of variations. You create a large potentially tradable universe of strategy elements, and at the same time you spend a little bit more thought on how they're actually selected. That's probably an additional very important part if you think about optimization, particularly over-optimization as well. Probably it makes sense to look at the allocation process at this point, which is a two-step process in our case, where the universe is first defined in terms of which markets I'm actually potentially ready to trade for these particular months. In the second step, all strategy elements in all these markets be ready for trading for the particular month, are basically competing for a spot to be picked. It reminds me sometimes of the movie Shrek, you know where the donkey is jumping, and he says, "pick me, pick me!" It's a little bit like that. 

In the early days, basically what happened is if you took a strategy element you could calculate the profitability of this strategy element based on what is called a Forward Walk Optimization Process. You basically can create heat maps. It gets really tricky when you have a higher number of parameters with an even higher number of variations for these parameters because then the potential tradable universe becomes really multi-dimensional and it's not possible for a human being to get an understanding of where profitability lies in that particular universe and where are the areas which are not as profitable. In the early days when this process was not automated, the portfolio manager sat down and actually systematically selected strategy elements based on different criteria, among others profitability.  

If you do it in a color, colored heat map, profitability is green: dark green is highly profitable, and red is highly loss making for that particular period you are looking at. Then you build up the portfolio. The portfolio manager, naturally as human beings says "well, take the green ones, they are high profitable, so this is a really good choice." Unfortunately, he somehow got it right. This is referring to 2010, but fortunately in 2011, as said earlier, we introduced the automated selection process and then we realized quite fast that actually, if you do it in an automated computerized way, the allocation process actually picks a lot of red ones as well, because red ones are maybe not profitable over that particular time window, but they're diversifying. As an important aspect, as I said earlier, the entire allocation process is framed by really strict risk criteria. In our case there are several risk criteria. One of them, probably the most dominant one is downside volatility. So the portfolio is built up by selecting these little strategy elements, step by step, putting together a portfolio, but it's constantly checking that the combinational strategy element that it is selecting is meeting the risk criteria. So there is no incident in the past, based on the data set that you are using, which has caused downside volatility in excess of 10% in a 1X version. If there's something that is a highly profitable combination, but it does not meet this risk criteria it cannot be part of the portfolio. So the little guy can jump as high as he wants and say, "pick me, pick me," if he... he might be highly profitable, but if he in combination with the others already being picked is causing a risk incident, he will not be picked. It's a really sad thing for him because by himself he is really profitable, so he probably will not understand not being picked, but the allocation process is framed by these risk criteria and it cannot be guided by just profitability. 


Sure, sure, I'm glad it's all done from an electronic point of view otherwise there'd be a lot of disappointed fellows in your office every month. 


That's correct, if they're standing in line waiting to be picked, that would be a really sad picture because the majority of them are actually not picked, so they all have to go back home. One has to say they have a new chance next month. This goes for the strategy elements and this also goes as I said earlier it's a two-step process, and the first-step markets are selected. I just the other day I explained it to an investor how the market selection process actually works and I said, well, imagine all markets - we currently have 23 potentially tradable markets, and maybe a short detour here... 


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.