“We do something that is very complex.” – Oliver Steinki (Tweet)
In the second part of our conversation with the cofounder of Evolutiq, we dive into the firm’s trading strategy, how they manage risk, and what kinds of investors they seek. He also discusses tips for budding entrepreneurs and firm managers, and how he deals with explaining the complexity of his trading models.
Thanks for listening and please welcome back Oliver Steinki.
In This Episode, You’ll Learn:
- The details of Oliver’s portfolio.
- What types of markets have more success for their models.
- Their unique allocation process.
- How to explain these scientific terms to potential investors.
“We don’t enter any position without an attached trading stop.” – Oliver Steinki (Tweet)
- The question of simplicity vs. complexity.
- How they managed intra-day forecasting.
- How he manages risk.
- The way that market correlations play into the portfolio.
“We are really uncorrelated because we have long and short positions on a lot of different markets.” – Oliver Steinki (Tweet)
- Oliver’s thoughts on drawdowns and the worst ones that he is expecting.
- How do you know when a model has stopped working.
- What Oliver’s experience has been as the world has become more divergent.
- What keeps him up at night.
“I probably worry to much. But hopefully over your whole career you worry too much but nothing happens, which is much better than being over-confident.” – Oliver Steinki (Tweet)
- What he is looking at in his research for the firm right now.
“Since we are research guys, we believe in constant improvement of the existing processes.” – Oliver Steinki (Tweet)
- How models can start decaying.
- Who he is targeting for his firm.
- How he manages the cash portion of his accounts.
- How setting up a company outside the big financial center affects his business.
- What questions investors should be asking him during due diligence meetings.
- The advice he would give to aspiring managers.
“One shouldn’t underestimate the time you need to spend to get the infrastructure and regulation right.” – Oliver Steinki (Tweet)
Resources & Links Mentioned in this Episode:
- Oliver recommends reading The Mathematics of Money Management.
- He also recommends Algorithmic Trading.
- Learn more about Harry Markowitz, mentioned in this episode.
This episode was sponsored by Eurex Exchange:
Connect with Evolutiq:
Visit the Website: www.Evolutiq.com
Call Evolutiq: +41 55 410 7373
E-Mail Evolutiq: firstname.lastname@example.org
Follow Oliver Steinki on Linkedin
“If you have a good business model, you need good endurance to go through the negative periods and always turn up again.” – Oliver Steinki (Tweet)
Oliver: I think it’s being honest to other people to be in a career because when you are in a career you will find something that is your passion. For me that’s really something nice that I have achieved with Evolutiq. Basically, I did the Ph.D. out of passion for the subject and now, out of passion for the subject I’m creating an investment management firm out of it. It’s just something really great if you can follow your passion because then it doesn’t feel like working.
Niels: I want to mention that today’s podcast episode is brought to you by the Eurex Exchange, which, of course, is the home of the European Yield Curve.
Oliver: Hi, this is Oliver Steinki, CEO and Founder of Evolutiq and you’re listening to Top Traders Unplugged.
Introduction: Welcome back to Top Traders Unplugged where the best traders in the world come to share their experiences, their successes and their failures. Let's rejoin the conversation with your host, veteran hedge fund manager, Niels Kaastrup-Larsen.
Niels: …The process regardless of what underlying it is, is the same.
Oliver: Yeah, yeah, so we always have the same investment process. Step one, we have the ensemble generation, so we generate a lot of levy process based on market models.
Step two is always independent of the underlying is the ensemble pruning. So basically we kick out the bad ones.
Step three is the integration of those that are left, how do you combine them to get one specific signal? Then we come to the element of portfolio construction, step number four, where we can discuss later in detail how we do it because I think we have quite a unique approach as well.
The last step is the implementation, which for us is fully automatic, but not just a technical detail because we will trade shortly more than 20 underlyings in a lot of different markets, so it just gets much more error prone if you were to do all this manually.
Niels: Sure, of course. Speaking of the expanded portfolio, is that going to be fully diversified, meaning financial markets and commodities? OK, do you have a view, just out of curiosity; are there some types of markets, meaning commodities versus financials that seem to have more stable or better predictions? Oliver: Yes, of course, we have different success. In the backtests, for example, silver was much better than gold, but we would not just trade silver because the beauty of our unique approach is that gold was not very strong, but it still made something like 10% over four or five years, which is not great, but still a good portfolio diversifier.
Niels: Tell me more about it, unless you feel you explained enough about the trading strategy itself but, feel free to describe more. I have another question as well sort of ramp up about the model itself, but is there anything else you want to add that you feel is important about maybe one of your USPs in terms of the model itself?
Oliver: I think the model itself is really a combination of two relatively unique approaches compared to using levy processes, which come out of the… In quantum finance, you often have the split between the P and the Q world. So the Q world has a risk-neutral derivatives processing that we usually do not forecast. Then you have the P world where you have the guys who do forecasting.
Basically what we have done as a transfer is using techniques from the derivatives processing world, and bring them into the forecasting world. There’s this one transfer or one USP of our strategy, and then the second one is applying these ensemble methods, and I believe that there are also other guys who do combine a lot of different models, but we are doing it in such depth with the three different steps I think we are quite on that one side.
Yeah, then the next step, which makes us a bit unique, is the portfolio construction process. After we come up now on a given underlying. We come up with something like we go long. We have a very unique allocation process.
Let’s say you would have a 1 million dollar portfolio and you would trade ten underlyings. A lot of people would just put 100,000, basically ten buckets of 100,000 and then underlying A you would trade with 100,000; underlying B you’d trade with 100,000 and so on. What we do is we start like that, but then we do two more steps.
The first one is what we call the volatility equalizing. So let’s say for your target volatility would be 10%. Then you have something like the DAX, which has roughly 20% annualized volatility. For that, we would half the exposure to DAX. So instead of trading 100,000 we would only trade 50,000. Whereas for something like the Aussie/dollar/yen, which is something like 5% annualized volatility, we would actually double it, so trading 200,000 thousand. What we want to achieve is ten equal risk buckets, not ten equal notional buckets.
Then in the next step we also did the weightings based on the Sortino ratio. As you realize Sortino keeps popping up all the time. So we would overweight them after doing this risk equalizing, or volatility equalizing as we call it, we would then come up with something that we call Sortino optimization where we do exactly the same process as before.
Let’s say our overall strategy is 1.5 target Sortino, we would then have a strategy that had a Sortino of 2 over the last one-year period. This strategy would be slightly over-weighted, a strategy that had a Sortino ratio of 1 over the last year. Then it would be slightly underweighted.
Niels: Now I wanted to ask you sort of a general question. If you think back on the CTA industry, and I don’t know exactly when you started following it, but what happened really a number of years ago was it came from being a very US dominated industry. A lot of large well-known firms had been around for 20 plus years, and a lot of the assets were there. Then comes along in sort of the late 90s early 2000s a number of I think perhaps more scientific, European managers and the investor base seemed to really like that approach and a lot of the assets really shifted and of course, in today’s world, we know that many of the powerhouses in this industry really reside in Europe at this stage. Not necessarily because performance has been better, but investors seem to have warmed up to this approach.
Now are you sometimes afraid that these very scientific terms that you use and that you describe your systems with may scare certain investors, unlike me. I’m happy to admit if I don’t understand it, but, of course, as you know, anything that we don’t understand we are more likely to say no thank you to. Trust me when I say that even basic trend following is sometimes very hard for many people to appreciate, so I can only imagine what it’s like describing what you guys are doing. Have you thought about that from a purely marketing angle, or when…?
Oliver: Yeah, I think actually it is definitely difficult for us because on the one hand we do something that is very complex. Being honest, it took me one year to understand levy processes fully during my doctorate studies. So I know they are very complex distributions.
If you were to have a five-year track record, basically we could show the track record and say that we have proprietary algorithms, and that’s it. Now we are now at the stage, and now we really have to try and convince people based on the academic arguments and that’s why the investors we can convince now at this early stage, they tend to be very well educated in this financial theory. They don’t have to know everything about ensemble methods, a relatively recent development, or they recently gained a lot of popularity, and also the levy processes. You really need to know quite a bit about all this theoretical or academic background in order to fully understand what we are doing.
We want to be super transparent, really lay down and disclose, that’s how we do it, so that someone who approaches us from the academic side, because right now we don’t have a very long track record. They can understand what these guys are doing and how it’s different from other CTAs.
Niels: Oliver, if you just take a step back and we think about simplicity versus complexity for example. The history of CTAs I would say probably come from the world of simplicity and many people have argued over the years and decades that using simple methods, in the long run, are more robust. You mentioned Harry Markowitz, obviously coming up with some important theories that took it a little bit more in the complex direction in terms of asset allocation.
Actually I believe that in a later interview he admitted, when it came to his own finances and how he would allocate his own money he would use the simple method. If it’s 10% in each bucket, that’s how I do it. So how do you, I’m just really interested in this, how do you as a person view simplicity versus complexity? I do understand your own product is complex. Because you maybe have seen it for a while it may not seem so complex to you, but in general, as a philosophy, do you believe that complex processes or complexity is stronger and more robust than simple stuff?
Oliver: I think that it’s a difficult answer to give here because I think you can see it from both sides. I also strongly believe in the concept of Occam's razor, so you should also aim for the simple solution to solve a given problem. If you have a very complex problem, you might not have one simple answer. That’s why we try to keep it as simple as possible. And that is the simplest solution we have come up with. Of course, it’s still very complex, but basically that’s our approach to it. If it’s right or wrong, we can probably have another chat in five years and then we will see if it was good or if something very simple would have been much better.
Niels: Just as a last question on this topic, because I do think it’s really interesting. In your design, you’re obviously dealing with complex issues and you’re using complex processes, but in the actual design are you trying to simplify some of this in the way you apply it, or do you just come up with the algorithms that actually just handles everything, even though it’s a bit complex, but you can program it, so it deals with it? Do you try for simplifying it, or do you just say happy to take all the complexity on?
Oliver: No, we really try to simplify it as much as possible, but for example coming back to the question of portfolio construction, we could allocate just 10% to each bucket, but I really believe that this is not the correct approach if you want to have equal risk buckets. It’s just not, for me, it’s counter-intuitive to have $100,000 in an underlying which moves 50% a year like VIX, and the same amount in something like Aussie/dollar/yen, which moves 5% a year. We always strive for simplicity. We don’t want to make it super complex, but we also want to still be precise and doing what we believe is right.
Niels: Sure, absolutely. I think most people now-a-days realize that you do need to volatility adjust your positions to have a chance of not shooting yourself in the foot. You mentioned that you run the system every day. Now a new forecast will come along and it will tell you what to do in terms of positions. What about intra-forecast, the 23 hours and 59 minutes that goes in-between the two forecasts, do you in anyway shape or form have any stop loss or exit that may be triggered before the next forecast?
Oliver: Yeah, yeah. We don’t enter any position without an attached trading stop. So it’s directly from the start when we enter the position we have always a trading stop attached, but we do not have take profit orders. So we stay in a position long until the signal changes to short.
Niels: Absolutely. I’m just curious in terms of the model itself, as you’ve implemented it, for how long has it been like this in your research? Has it been fixed more or less for a few months, or for a year or so, meaning are you constantly making changes to it even as we go along or is it just now a matter of having more markets but essentially the model itself has been stable for a while?
Oliver: Yeah, the model has been very stable for a while, and it’s more a question of adding more markets.
Niels: Let’s shift gear then to another important topic, I think, and that is risk management. We touched upon it a little bit, but I would love to ask you just how do you define risk in general? What is the risk matrix or risk framework that you look at?
Oliver: What we look at is downside volatility and maximum drawdown and maximum daily loss. We are still relatively small and only trade liquid markets, we believe that by always attaching a stop loss, we can, to a certain degree, limit the downside risk in any given day.
Niels: Sure, what kind of targets are they and how do you actually try and achieve them or be within them in practical terms?
Oliver: So what we have, the question is you enter a position given the underlying and then the question is where do you put your stop loss? Do you put it 2% below, or do you put it 10% below? If you put it 2% below you have the problem that you might be stopped out quite often, where over the long run your prediction was correct but you have been stopped out. If you put it at 10% below, you won’t be executed that often, but when you are executed you will really have lost a lot of money already.
So here we try to strike a balance because we target 10% return. Why do we do that? Because it’s what we do in order to build a track record. We believe that that’s the kind of style of volatility and return is what most institutional investors would like to see, somewhere around 10% to 12% annual return and 6%, 7%, 8% of volatility.
Because our constant is very flexible, we also have some clients saying, no I can stand a much higher volatility. For us, it’s very easy to offer double or half of it. It’s really very easy for us to customize it to any desire to risk-return profile a client would want to have. So we have this target Sortino ratio but it is 7% volatility with 10% return or 2% volatility with 3.5% return, or if it’s 20% return with roughly 15% volatility, that’s what the client has to tell us.
It’s the same with underlyings. There might be a client who has, because of some constraint he would not want to trade commodities, for example. It’s very easy for us to just trade all the other markets for him. So it’s a very customizable strategy we have to offer in my opinion. Niels: The stop loss, is that the same regardless of what kind of market you trade?
Oliver: No, no, it’s a function of the volatility and our maximum daily loss is below 1%. We don’t want to lose more than 1% a day, so that’s the maximum constraint we have, and it’s automatically, the distance is updated based on the last year return. So we constantly check if over the last year on VIX the ideal return would have been X%, with a stop loss of X% below, that’s how we update the stop loss systems.
Niels: Now tell me a little bit about correlation. You’re obviously trading in a somewhat… You could say it’s a short term strategy because it could change every 24 hours essentially. How do market correlations play into the portfolio if at all? Meaning is it more the model correlation that’s relevant, or is it perhaps the market correlation that needs more attention?
Oliver: So we have done some backtests correlation matrix which is quite positive in my opinion because we have a correlation of -0.03 with MC World, for example, -0.2 with the Edge CTA Index, so we are really uncorrelated because we have long and short positions on a lot of different underlyings.
For example, let’s say we would only trade gold and there’s also another guy who trades gold or you take gold as a benchmark itself, we don’t necessarily have to be highly correlated because we can be long, short, and neutral. We have physically three possible outcomes compared to OK we are just long gold.
Niels: Absolutely, now of course with returns comes drawdowns and they affect us differently, and I appreciate that the track record may be not so long, but you’ve done a lot of testing, so you’ve probably had some expectations. Just remind me what your expectations in terms of your worst drawdown would be, for example. Also, what are your thoughts about if you get to that level and maybe go beyond that? How do you frame that?
I know this might be a little bit difficult to answer because you’re not quite there yet, but it’s something that is so important because I think one of the strengths of some of the managers who have been around for 20, 30, 40, years is the fact that they have gone through a lot of drawdowns and they have survived them, so it is a very important part of what we do as managers. Tell me about what your expectations are and how you want to deal with it if you get to that level.
Oliver: So we have in backtests only had a maximum drawdown of 5%, but I think long term what I would consider an attractive strategy is when the maximum drawdown and percent is equal or lower than the annualized return. So for us we target 10% annualized return. If you would stay in terms of maximum drawdown below 10% of the long one, I would be happy. What we would do if you would ever get there, already when we come to 75% of that, we are constantly checking our models, checking the assumptions. Could there be any tweak relative to the backtest that we have not taken into consideration. So far we haven’t found any kind of mistake or error, but it’s a big question I think you’re aiming for, or the answer to that question is when do you stop trading?
Niels: Yeah, I mean there are two questions. One is when do you stop trading? I thing that I find is an interesting question. The other thing that I personally think is an interesting question is that when you use these complex systems, how do you know if a model has stopped working?
Oliver: Yeah, we have, to a certain degree, we have some semi-automatic approach to that because we have the Sortino rate, what I told you about earlier, so a strategy that is not working well will have a pretty low Sortino ratio over the last year, and therefore we would underweight it automatically and those strategies that are working well would have a high Sortino ratio and therefore be over-weighted.
Niels: That is true, but I will challenge you a little bit on this, even though you’re the one with a Ph.D., so I need to be careful here, but one thing is to underweight things that may not have worked, but the fact that you’re picking something that has worked it could be just at the time when it stops working, and therefore you end up first disallowing something that may start working and then you keep on investing through a model that stops working, so this is the compound of false signals, if we can call it that.
So this is the challenge and sometimes when you just have a simple, or simpler model where you can see each trade and you can see everything that goes on, sometimes it can be quite easy to say, actually there’s something that’s not quite right here, and I need to reinvestigate that. Once you get into artificial intelligence and predictions and so on, I guess, at least in my mind, it becomes perhaps a little bit harder to find the root cause of a potential underperformance period, but I’m sure you have ways of dealing with that, but that’s just my own sort of simpler observation I guess.
Oliver: No, no, I agree with you. The more complex the strategy, the more difficult it is to come to some form of let’s call it performance attribution, because in the end it’s like you want to know what have been the non-performing underlyings that’s still pretty easy on a portfolio level, and then you want to know, OK for those underlyings where your performance has been weak over the recent time period, why has it been like that?
That is the question because we have roughly a bit more than one million models per underlying which then leads to the forecast. We store the P&L of each of those one million models over time. So we have a huge database, as you can imagine. We can really go into the details and really pick what is going on here because with every statistical learning approach, if you have a total regime shift, and you will have some period of adjustment until the outer learning works again. You can have, and you will have periods where you basically, your models are not working. The art is you cannot work, but it has to be limited and you have to make it until the time when they work again.
Oliver: Speaking of regime changes, in my mind you could say that there has been a little bit of a regime change recently, meaning that after 2008, 2009, a lot of central banks decided that they were going to work together and have a very coordinated approach that clearly meant that convergent risk-taking became in favor. The world was reasonably stable in some respect and the other types of strategies, where the CTAs are usually namely in the divergent part of the risk-taking spectrum were struggling.
Last year the central banks decided to stop coordinating their policies completely in the US. They didn’t want to continue the QE. Japan decided to accelerate QE and Europe said, “Well, maybe we’ll do QE.” Of course, we know now that they started doing it. So in many ways, from my point of view, the world has become much more divergent than it was only a year or so ago. Now to me that’s a little bit of a regime change, and I wanted to ask you, in your research and your test, what’s been your experience in the period 2007 through 2014? Could you see these things happening in the performance data as well?
Oliver: Actually, it’s a very good question and Peter and I we discussed it on Friday because the question is, for me the definition of a regime shift. When you look at, as you mentioned correctly, the central bank intervention in the financial markets since 2008, it has led to record low yields on fixed income instruments. Basically, it’s a range bond process because; OK now here in Switzerland we have some rates go negatively, in Germany, Scandinavian countries. We see countries with negative rates, but it’s still limited. I wouldn’t expect like a negative 10%. That would really hurt.
It’s very difficult for me, or I find it a very difficult questions and I don’t have the answer how to predict a regime shift or even to recognize one. For me a regime shift is where something changes but you would have to define and to be very accurate from an empirical point of view you would have to define beforehand, OK if all of that happens, that’s what we would consider a regime shift.
Niels: How long to you go back in your backtest? How long do you test the model back?
Oliver: Early 2008.
Niels: Ah, OK, OK, that’s interesting isn’t it, because you’re really dealing with that period that everyone felt both on the upside and the downside. Anyway, last question on risk management and it’s very simple, what keeps you up at night? Is there anything when you think about risk, do you think about your model, do you think about the markets you trade, you think about any other thing that could really disrupt what you do, it could be an IT issue, I don’t know, is there anything that keeps you up at night, Oliver?
Oliver: Yeah, of course because we are a start-up. The not so great performance over the last month keeps us up, and we are constantly challenging ourselves, is it for us OK in line with expectations, but how can we improve further. From an IT perspective, we have triple backup so I’m still pretty confident, however, we still monitor the trades all the time although it is working without any backup.
So, basically, to be a manager you have to be a bit paranoid and constantly worry probably too much and hopefully over your whole career you’ve always worried too much but nothing happens. That way it’s much better than the other way around where you are overconfident, and you spend too much time on the golf course, but then you have some blowup, and that’s what no one wants to see. People trust you with their money, so it’s your responsibility to do the best that you can 24/7 for them.
Niels: Sure, absolutely. Now let’s jump to the topic that is also something we’ve already touched upon which is research. I just want to ask you a couple of questions in this regard. First of all, investors to some degree they want us to keep innovating and doing research and so on and so forth but they don’t want big changes in the profile of the strategy. When you sit down, the four PhDs within your firm and you discuss research, and you have your brainstorming sessions, what’s the conversation like right now at the moment? What are the topics that you discuss?
Oliver: We are working in the same direction on our research because we have basically the whole strategy is based on the initial idea of my Ph.D. to use ensembles of levy process models. Now we’ve already come quite a bit down the road towards transforming it into a fully automatic strategy. Plus we are still in the research. Portfolio construct is an ongoing process. How can we improve it? The optimization for Sortino is an ongoing process. The ensemble method by itself, how do you over or underweight a strong performing or weak performing models is a constant issue. Which market do you add as the next one is an ongoing challenge?
So of course we have a lot of different topics where we are challenged and focus our research on now. I think what we have all in common is this big wall map, OK that’s where we want to be, and now it’s just a question of how do we get there. Sometimes we turn left, and then we realize, OK we should have turned right, but it’s nothing that we are departing too much from the road to where we want to go. We know there are still certain things to achieve, for example until we have all twenty underlyings traded but we know how to get there and now it’s a question of basically man power and research hours to implement it.
Niels: I wanted to clarify something with you, and I’m not even sure if I fully understood that. Clearly you're applying some kind of an artificial intelligence in your machine in terms of learning. So if part of the improvement is the machine becoming better at learning, if I can put it that way, where can you add value to that learning process as part of your research? Is it more models so that there is an even more votes, if I can visualize that for people that each model has a vote and that’s what the prediction becomes is it in that area that you can add value to machine learning, or how does it actually work?
Oliver: I think, as I mentioned, I think we could extend the number of models. We could extend the calibration techniques we apply to them. We could improve our pruning methods, so how do we decide which models to kick out? We can also improve on the ensemble integration step, so how do we combine those models that are left over? And we can also improve on the portfolio construction point and once we are a bigger firm, managing more money, we can also probably in the trade implementation right now we just do a limit over a certain price and we buy the underlying future. If we would manage a couple of billion, we would probably have to think about things like iceberg oils or so on so that we don’t leave a big footprint in the market.
Right now we are very far away from that problem, but if you are Winton trading several billions you will have to take these problems into account. We have the general outline, the general investment process. We have come up with procedures for each of these steps that we consider good, but as we are research guys, we believe in constant improvement of the existing processes. Also for the strategy, it might not sound like much but if you’re fixed 56% correct instead of 54%, it’s a huge different financially in terms of returns.
Niels: Yeah, that much I understood from my other conversation. As I mentioned I really go the understanding of the importance of this number that is very interesting. Now when you look at models and all the work that you’ve done, do you believe that models have a certain lifetime, that there is a decay in models where they just simply stop working.
Oliver: I believe when too many market players are playing according to the same model then they could stop working or if you have a future regime shift with totally new drivers of price. For example, the entrance of the central banks is clearly the entrance of a very big market player with very deep pockets and that clearly had some effect on the prices where they’ve been interfering in those markets.
Niels: Let’s talk a little bit about the business side of your company. Moving away from the trading and the research, obviously you’re a very experienced entrepreneur as we learned earlier on. When you look at the landscape of investors today, who are you targeting and who do you think will be a well-suited client for the strategy? It goes both ways, who do you think might be interested in you, but are you also, and obviously you can’t be too critical in the beginning, it’s about getting assets in, but is that something that is part of your long-term profile?
Because I know a very, very successful short term trade company based in Zug, who were interviewed earlier on the podcast. Their CEO clearly described that it was not about that all clients are the right clients and part of their learning process was really to say no, and just focus on a smaller number of investors with whom they could really get into the detailed research because it allowed for them to have much longer relationships even through the bad times.
Oliver: Yeah, I believe the same. For example in terms of minimum account size, there has to be some minimum. You cannot first when you want to trade a diversified portfolio of twenty underlyings you need to have a certain number of minimum, you need to put up some margin, so that’s one constraint.
You also… What we focus on right now are these emerging manager investors because you know we don’t have a very long track record, so those investors they then basically invest based on their academic merits or the originality of your idea and we believe we have an interesting idea and we also focus on a couple of wealthy individuals that have a certain level of education and entrepreneurial success they have a certain type of risk where they really… similar to comparing it to a start-up investment because you know there are still ways to improve it and it’s early, but if it goes well, it will go very well.
So this is why right now we focus on emerging managers and wealthy individuals, because, for example, if you look at the ideal kind of ticket in terms of asset size, it would clearly be of a big pension fund. They could invest 50 million in one go, but they would have criteria such as, OK you need to manage at least 100 million and our ticket is not allowed to be more than X percent of your total AUM. As we are still very small, it doesn’t make sense to contact these kinds of investors. I think it’s over the lifecycle of an investment management firm that at each stage you have different ideal kinds of investors.
Niels: Sure, absolutely. Tell me Oliver, when you run a strategy like yours, what’s your average margin to equity and what’s the range of margin to equity that you see on a daily basis?
Oliver: It’s between 2% and 7% or 8%. I would say the average is around 4% or 5%, something like that.
Niels: OK. Well that leads me to the next question that I have, and that is, in a world where, certainly in your case you have a lot of cash, especially if you had a fund and people were investing through a fund, then you would have about 90% to 95% of the money sitting in cash. How do you, in a world with zero interest rates, and as you mentioned, in some countries even negative interest rates, in a world where there is doubt about the safety of the banking system still, I guess, I don’t know whether you have the problem right now or not, but how are you going to handle that issue that’s called just the cash portion of an investors account or fund?
Oliver: We have managed accounts, if you have a fund it’s separate and you’re not exposed to the risk of MF Global kinds of cases, but we have to make sure that you deal with brokers where you have the managed account which have very solid financial power.
Niels: Yes, that’s one thing. To me, that’s more about where you put the margin and how you do that. Actually I learned quite a lot from the sponsors of our podcast, the Eurex Exchange because they actually educated me when I saw them last month that there are some new rules now where essentially you can get much more safety for your clients of the positions that you have if one, you’re clearing a European counterpart and two, for the markets that you trade on European exchanges. So there are new things happening.
I was more interested in the none trading margin. If you have 90% of a fund and it’s sitting in cash. You probably unlikely would have that sitting at the broker. You need to do something else with them and I was just curious whether you have thought about what you would do in an instance like that because when I hear about your strategy it reminds me again of other short-term strategies where the most efficient vehicle for you, really would be to get as many investors as possible investing to you via a fund. Even though you say you could manage 100 accounts, I understand that, but actually just the way you trade and sort of the frequency, it could mean at least having a fund vehicle at some stage would be relevant. So I was just curious to know whether you thought about how to manage the other part of the portfolio, the cash in the current environment?
Oliver: Yes, I talked to some of our investors and they would like to have this portion invested in some money market type of securities to at least get the minimum, but most of them said OK, we are fine with it sitting in cash because we are interested in… we give you 100 we care about what happens to the 100 and where do we stand at the end of year one, year two, and year three, but I agree with you, long term it would certainly be on our target list. We would certainly like to raise a fund long term. This problem you mentioned, it’s a realistic one. If you want to make too much return on this 90%, probably you have to take some risk, then it’s a question of how much risk do you want to take on the 90%.
Niels: Actually just to clarify I wasn’t even thinking about making a return on the money, I was just wanting to make sure you could keep it safe. Having money in the banking system today, at least in my mind doesn’t necessarily mean that your money is safe. Anyway, that’s a different discussion. Just one more question on the business side of things. You’ve chosen to be setting your company outside the financial centers, if I can put it that way, how do you think that affects you? Does it make it harder for you to get people to stop by the office or vice versa? Is it better for you? Is it a more creative, calm environment to be removed from the noise of London or Wall Street? How do you phrase that and why did you decide to set up your company where you did?
Oliver: Basically, for me, I find the Swiss financial regulation quite attractive, so we wanted to set up in Switzerland. As I’m German, for me it was easier to do it in the German speaking part than the French speaking part, where I used to live before. But I think now-a-days we live in a global… the financial markets are global. I travel a lot, so I’m in London roughly once a month and in New York three or four times a year. I’m also teaching in Hong Kong, so anyhow, I can combine it with also meeting some investors there, so of course you need to meet those investors in person, but where you do your research, most of them they don’t really, the investors don’t really care. For them we are like, we are in a village outside of Zurich which is considered a financial hub. I think you have quite a few CTAs and family offices in that area. You have Mann Group and Favicon and so on. So it’s not that we are totally remotely somewhere. I agree we are not in the center of London or in the center of New York, but we’re not totally away from the financial centers.
Niels: Absolutely. Before we go to the last section, I wanted to ask you a question that I think is relevant, probably for all managers, but I can imagine for a group like yours where you’re going out and having to do a lot of due diligence meetings or due diligence phone calls, and you get all these questions all the time. What do you find that investors, maybe because of your slightly unique strategy for sure, but what are the questions that investors should be asking you? What should they really dive into which they may not be diving into today when they talk to you about your strategy?
Oliver: As an investor, I think what’s more and more important are the regulation and infrastructure requirements. If you look at classical DDQs, you have a lot of questions: How many? You have the key man at risk. You have this kind of stuff. You have what happens if your computers breakdown. How are you hedged against this kind of risk because I think most investors understand that there will be months of time periods where you just believe that the market goes up, but it goes down. I mean that’s life, it’s just about basically eliminating all of the other blow-up risks.
If by offering managed accounts, as we do, investors are pretty safe. We cannot run away with the money, nor do a Madoff kind of case. I think since Madoff we have seen a huge increase in the demand for managed accounts. Also if you have a fund, you cannot do it anymore very cheap structure with unknown auditor zones and all that is needed to have one of the big four. So I think that’s what people are really focusing on because this is the risk… Basically exact career risk as an investor. If you say OK, we go with these small guys from Evolutiq and then in two years there would be some huge issue because we wouldn’t have a proper backup structure, then it would be that career risk. If adjusted performance doesn’t go according to expectations, they can pull the plug.
Niels: Let’s jump to the last section as mentioned. This is sort of more; I call it general and fun. So it’s a little bit all over the place really. So I wanted to start by asking you. As a new manager, but again as an experienced entrepreneur, what advice would you give to other new managers or aspiring managers? What has really been the key takeaway of your learning starting out in the systematic trading space?
Oliver: I think one shouldn’t underestimate the time you need to spend to get the infrastructure and regulation right. This is a key point, and I think now, with all this new regulation it’s becoming more and more important. It’s clearly a red flag for any investor if you do not fulfill the highest standards in these fields. Therefore, I would really recommend that everyone focus and become informed about these kinds of things before you start your own asset management firm.
Niels: Sure. When you, in leading up to this, I don’t know if there was one or maybe a number of firms that you looked at, and that inspired you and you aspire to get to where they are today. Are there any firms out there where you said, “Yeah, this is such a great story and such a great firm that if I one day could be something along those lines, I would be very happy and very proud?”
Oliver: I know a couple of people at Winton and at Mann Group, and I think those are really like excellent places. They have managed to create this awesome research environment but still have a very profitable business. I would like to do something similar because if you want to hire intelligent Ph.D. guy, you have to make it a nice place for them to work so that they can really focus on the research. By focusing on the research, this should improve your performance.
Niels: Yeah, those are certainly some great names you mentioned there. Now you’ve study a lot. You’ve read a lot of books. Is there any particular book, maybe two that if you look at it purely from a trading point of view, that you would recommend to other people to study. Maybe not something that is too deep into Levy processes because I think that is a little bit unique. But are there any trading books that you’ve come across that you felt were really good?
Oliver: Yeah, what I recommend in my class and also in the resources at the end of all of the different slides, I think, if I would have to choose two books, one is from Vince, The Mathematics of Money Management and Risk Analysis Techniques for Traders, which I found very useful for the question of how do you size your positions. Then from a guy named Chan, Algorithmic Trading Winning Strategies and Directional, where he also has a second book where he basically really tells you how to build your algorithmic trading firm. Those two I found very useful.
Niels: OK, in terms of other books, just, in general, I’m a little bit curious, other books that you’ve read that may have impacted you as a person if it’s a little bit outside sort of the trading space, is there any?
Oliver: My favorite author is Jack Bassoon, he was a French intellectual and he writes a lot of very interesting books. I also really like to read about history because I find that you can learn a lot about it. So another very good book is 500 Years of Western European Culture. It’s one of those books where basically every page you can learn a lot of new stuff.
Niels: Sure. Now, I think it’s clear for everyone who has listened to our conversation today that you’ve been starting a few businesses along the way. As an entrepreneur, if you just take that hat on, what’s been the biggest failure so far in doing that, and what did you learn from it?
Oliver: The biggest failure so far, all the companies are going OK, but sometimes they took longer to… basically until the revenue came in. So I think if you have a good business model you need to have the endurance to really go through negative periods and always come back, always stand up again and really continue to improve your business model until it finally comes into place.
Niels: Now Oliver, I’m not entirely sure how old you are, so I’m going to have to ask you whether you have children?
Oliver: No, I don’t have.
Niels: OK, so I’m going to rephrase my question and say, when you do have children, if you do have children, if you could take just one of your skills that you have today, and you could pass it on to your child, what would that be and why?
Oliver: I think being honest to other people and being curious. When you are curious, you will find something that is your passion. For me, that’s really something nice and what I have achieved with Evolutiq. I did the Ph.D. out of passion for the subject, and now out of passion for the subject I’m creating an investment management firm out of it. It’s really something really great if you can follow your passion because then it doesn’t feel like working.
Niels: No, absolutely, that’s true. What about a fun fact, Oliver. Can you share a fun fact about yourself? Something that even the people who know you – could be your colleagues, could be your family may not know about you?
Oliver: Most of them know that I’m pretty clumsy so I’m unable to fix anything, unable to cook, so I’m really, as a cliché a mathematician guy.
Niels: Right, so the nerd factor does actually exist.
Oliver: Right, the nerd factor.
Niels: Excellent. Now I said earlier, and I asked you earlier about what investors should be asking you when they talk to you and I want to make sure that I’ve done you justice as well, so is there anything that you can think of as we’re wrapping up that I’ve missed in our conversation today? Something that you want to bring up towards the end here?
Oliver: No, I think it was very interesting for me. Thank you very much for your time. I am just summarize again, I think we have a couple of USPs compared to other CTAs. We are a very well educated team. We have this unique focus on Sortino ratios. We have this unique academic approach of applying levy processes through ensemble methods. Also, we have the USP that we believe that we know how to run a company.
Niels: Absolutely. As we wrap up, I do want to just mention one thing, and that is to thank our sponsor, the Eurex Exchange for sponsoring today’s episode. As many of the listeners know, of course is that the Eurex Exchange is the place where you can go and hedge your portfolio risk. Before we finish, I want to ask you, Oliver, if you can share where the best place is for listeners who want to reach out to you and get to know more about your firm, where is the best place to go?
Oliver: I think it’s easiest to go through our website Evolutiq.com and there you can send me an email, give me a call. You have all the contact details over there. I would be happy to share my experience with interested people who are listening to this podcast.
Niels: Absolutely. This has been a great conversation. I really appreciate that. You have introduced us to your class of algorithmic trading. That was great. The transparency and the willingness from your side to share your views on your strategy and your firm I also really appreciate that. Of course, as most people know, all the details of today’s conversation can be found in the show notes for this episode on TOPTRADERSUNPLUGGED.COM. So I hope we can connect at a later stage, Oliver, and find out how things are progressing. In the meantime, I wish you and your colleagues all the best.
Oliver: Thank you very much. It was a pleasure to talk to you, Niels.
Niels: Thank you. You too. Bye.
Ending: Thanks for listening to Top Traders Unplugged. If you feel you learned something of value from today's episode, the best way to stay updated is to go on over to iTunes and subscribe to the show so that you'll be sure to get all the new episodes as they're released. We have some amazing guests lined up for you, and to ensure our show continues to grow, please leave us an honest rating and review on iTunes. It only takes a minute, and it's the best way to show us you love the podcast. We'll see you next time on 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: 08 Jun 2015no comments