“If you can’t explain it in English to a reasonably intelligent person why it really works, then we don’t want it.” – Marc Malek (Tweet)
In the second part of our conversation with Marc Malek, we explore the strategies that he uses to build his models and how he explains them in simple terms. We go in depth about drawdowns and what investors should know about them. We also discuss what keeps Marc inspired, what he does for fun, and how he couldn’t imagine having any other job.
Welcome back to Part 2 of our conversation with Marc Malek.
In This Episode, You’ll Learn:
- Four strategies inside Conquest Capital’s Macro program.
“What we start with is what we believe is the blueprint of how markets move, and then we try to create different models that take advantage of the different parts of that movement.” – Marc Malek (Tweet)
- How the technological revolution has changed the market and trend following, now that everyone gets the same information at the same time.
- Marc’s approach to building trading models.
- Defining risk in terms of upside deviation vs. downside deviation in the portfolio.
- Why correlation is one of the most misunderstood stats that you can use.
- How Marc deals with drawdowns and why he thinks nothing new is happening now that did not happen before.
“Drawdowns are a very emotional issue and that’s really the one time where I think managers, including us, do something that feels better for the investor, but is probably not in the investor’s best interest.” – Marc Malek (Tweet)
- His view on backtesting.
- The challenges hedge fund owners face in the current business climate and why Marc is lucky to have investors that have stuck by him.
“We benefit from those sudden events that keep people awake at night; and the way we lose money is not from any fireworks, but from a lack thereof.” – Marc Malek (Tweet)
- Why Marc is motivated to keep pushing through this period and why he loves what he does.
“For each additional parameter that you’re putting into your model, you are lowering the number of the degrees of freedom that your model can have to react in different market conditions.” – Marc Malek (Tweet)
- The story of how Conquest Capital got its first investor, and how in the early days investors cared more about managers and interacting with them personally.
- Marc’s biggest failures, and how he overcame them.
“If I could do it all over again, I’m not sure I would choose to be in a risk averse strategy.” – Marc Malek (Tweet)
- The hobbies Marc has and why he likes seemingly dangerous sports.
Resources & Links Mentioned in this Episode:
This episode was sponsored by Swiss Financial Services:
Connect with Conquest Capital Group:
Visit the Website: www.ConquestCG.com
Call Conquest Capital Group: +01 212.759.8777
E-Mail Conquest Capital Group: firstname.lastname@example.org
Follow Marc Malek on Linkedin
“Even in its worst days – it’s still the best job in the world. It’s very dynamic, very exciting.” – Marc Malek (Tweet)
Niels: You're listening to Top Traders Unplugged, episode #032 where I continue my conversation with Marc Malek, the founder of Conquest Capital Group. This episode is sponsored by Swiss Financial Services.
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.
Marc: ...allocate within the different risk environments and so on.
Niels: So you have these four strategies within the Macro program, are you able to visualize and talk just briefly about how each of them implement what they do, just to make it simplified a little bit?
Marc: Well, like I said, in looking at our long vol component... all trades that we do are, in the case of Conquest Macro, we have about 35 different markets or so. In each one of those risk buckets, we're looking to extract a certain profile in the market. The way that I think of the way these strategies interact is essentially at the root... it's going to get us started into a different topic... but in a nutshell, I think historically trends happen because information comes out at different times and causes the market to move. If you go back in history, 20 years or so, information came out very slowly; different people got it at different times, and reacted to it at a different time, and created the trends that formed. I think as we..., and that was great for long term trend following. I think as we got to the end of the 90s/beginning of 2000, with the advent of the technological revolution, with the telecom revolution, with all of the advances that we have made on the internet and so on, the effect of that on the market is that it caused the way markets move to go from a much more continuous function, where it was somewhat of a smooth movement over time that was based on different people getting that information in different time and reacting to in at different times, to an environment where pretty much everyone that wants any kind of information gets it at the same time, and reacts to it at the same time.
Visually, if you want to think about it, think of it... let's say a market is going up, and it's going from point A to point B. In the old regime, it would go from point A to point B in a nice upwards sloping smooth curve. In the new risk regime, it replaced that with more like a step function where the market goes into a flat period, where nothing is going on, and news comes out. Everyone reacts to it at the same time. It causes a big jump, which is that up part of the step. By the time everybody reacts and the market absorbs that reaction, it flattens up again until you get the next news event, which again will cause it to be almost gapped, either up or down, and then stay there and continue. It's not really gapping in the type of markets that we trade, but a very short move. What we try to do is, we're still looking at the same... I mean the average holding period in Conquest Macro is about 6, 7 days.
We're still looking at the same movement from point A to point B, but we're trying to capture it through the different activities that cause that movement. So in the part that is flat, obviously there is nothing that you want to do. You want to try to be out of the market. If you think of it as a staircase, in the flat part of the staircase you really don't want to do much there, on the spike up, when the news event happens and causes the market to spike up, on the top part of the stair, what you want to do is you want your long volatility component to come in and capture that expansion in volatility. That's how we capture that. While that's happening, there are some counter trend moves that happen in the market sometimes. So we try to capture those with our counter trend strategies. In that case let's say that instead of the steps continuously going up, you can have three steps up, then two down, then another three up. We have again...it would be a lot easier to explain this with a board and some charts..
Niels: But you want to buy the dips in an uptrend, or you want to sell the spikes in a downtrend.
Marc: Both, that's essentially how we try to achieve the returns that we do. We have over 75 different models with decent, not high correlation to each other. Each one is trying to do a different things. Whenever we try to buy the dip in a market, we do it in a way with a very tight stop, because overall we don't want to take away from our profile of being long volatility. You can buy that dip in a combination of ways. Either through taking profit on existing position or establishing a new position with a very tight stop. If the dip is at an interesting point in a trend development and so on. Essentially we start with what we believe is the blueprint of how markets move, and then we try to create different models that take advantage of the different parts of that movement.
Niels: Let me ask you something, while maybe you get a sip of water, because you're doing most of the talking today, but let me ask you a slightly different question. When I hear you talk about this, clearly what you've built is a complex system, if I can call it that, with lots of models, lots of moving parts, and in my own personal experience we've certainly tried to build short term systems using many different market model combinations, but what we found in our research was actually that the more models... even though they were negatively correlated, the more of these things we threw into the overall program, it didn't actually help us a great deal, and that sometimes simplicity is not a bad thing. How do you view that? Or let me put it this way, how do you balance the complex versus the simple?
Marc: Very simple, each model by itself is actually not that complex. There are a couple of ways that you can build trading models: you can do what I call the trolling approach which is you test different algorithms and trading strategies and whatever works, works, and then you use it. You use it until it stops working and then you stop using it. A lot of people do this. There's absolutely nothing wrong with that approach. People look for pattern recognition all the time with the market and try to take advantage of those patterns. My problem with that approach is if I don't know why it's working, I will not know when it will stop working, and might end up really overstaying my welcome in that strategy. So we use a slightly different approach to strategy building or model building, which is that if you can't explain it in English, to a reasonably intelligent person and have them get the idea of why it works, then we don't really want it. For each model that we have we have what we've identified over 60 different statistical variables that we use as diagnostics for each model on a daily basis, and then we look at obviously the one week, the one month, and so on, to make sure that the model is still behaving the same way that we expect it to. However, like you very correctly said, if you come up with a hundred different models that do all these different things, and you throw them together, at best you are going to end up with zero returns a lot of the times. You end up really hurting yourself in some cases, for a couple of reasons: one is if you add a lot of complexity in a model, you're probably over fitting, and it doesn't work, and if you have too many different models you're just going to dilute your returns basically. You end up spending a lot of money making your broker's rich. The way that we've gotten around that problem is by not always allocating to all these different models. So what we have, is we have a lot of different models that do different things, but because of our reliance on the risk environment and various other factors, what we do is we're constantly turning on and off models based on our read of the market conditions in which they work or not work. So, yes you are correct, if you just throw a lot of very complex models in you are going to eat away your return and have negative expectations from that addition, but that's not what we do.
Niels: Let me just ask you a thought about this. I don't know whether you have an opinion about it, but I think it's important for the listeners to understand, because clearly a lot of people are really trying to grasp what you do but also, generally speaking about these type of systematic approaches, and here's the thing: you can come up with 50 different market model combinations and each of them look good and have a decent risk-return profile and for all things being equal, if you put them together you should get... on paper it looks stable and robust, but I think what people need to understand is that what can happen here is that you are talking about, in this case, 50 different return streams and 50 different streams of trades that are either going to be profitable or unprofitable, but what can happen, which we don't know, is that in the future the way these trades occur can happen so that you get a concentration of losing trades just happening at the same time, and that's where the drawdown comes from. It's not necessarily that the models themselves are bad or have changed; it is just that the way the returns or the trades come along and you get a concentration of losing trades and therefore that's where your drawdown comes from. Is that a fair way of describing how these things can...?
Marc: Oh yeah, absolutely, absolutely. Before we add any model to our portfolio it goes through very rigorous checks of simulating different market conditions, what trades, at the extremes, how big it's going to make, our positions visa vise the other models that we have, so we try do to as much of that work as we can beforehand, but even then markets have a way to surprise you, so even then what we do on a portfolio basis, then we have different sets of risk management tools that will make sure that we never go above the max position in any one sector or market, or whatever we want. So we try to capture it both from the construction, but also then from the overall portfolio side.
Niels: Is the Macro program... I guess it's an intraday program, in the sense that your short term trading does rely on continuous data and being able to react at any point in time, or are you able...
Marc: Yes. We continuously trade throughout the day and the night.
Niels: OK. I want to jump to another section which is more on the topic of risk management, because one thing is to put the models together, the other thing is to manage the risk, and you just mentioned that you have some limits for each of the things, but let me ask you more broadly, how do you define risk? What is risk in your view, and what should the investors be looking at, because I think you've alluded to this, and that is that maybe a lot of people look at a certain risk number and they feel good about it because oh, the value at risk is such and such, but what they don't realize is that value of risk changes and when it really gets into hot water in the markets value of risk is not very useful, so how do you deal with that?
Marc: Probably I think value at risk is the worst things that happened to markets because it gives managers with investors a false sense of understanding of the risk.
Niels: So I didn't overstate that then?
Marc: No, absolutely not. I look at risk in a couple of ways. I look at risk in terms of upside deviation versus downside deviation in our portfolio. I look at it in terms of automatic expected drawdown, and I look at it in terms of what could happen given values, different movements in value based on the individual volatility of a given market and the correlation to the other markets that we have to the portfolio when we have max positions. So you want to look at individual position, and then you want to look at the correlation of these positions, because it's one thing to limit one market, but one thing that we know: correlation, again is one of, I think, the most misunderstood statistics out there. You have a lot of people that hate looking at correlation because they think it is unreliable, because correlations break all the time. But again, what we found is that actually correlation, when you look at it the right way, is one of the best sort of stats that you can measure in terms of fidelity.
The problem is that people look at correlation across a variety of risk environments. When you look at a 10 year correlation over a 10 year period of say a couple of markets to each other, over that 10 year period it's a constant ebb and flow between risk-seeking periods and risk averse periods. What happens is that correlation of those two markets that you're looking at is very different in a risk seeking period than it is in risk averse periods. To use the example we spoke about earlier, between strategies that rely a lot on liquidity, what happens to it when liquidity disappears, versus one that not as much, what we found is if you actually look at correlation - break it down, let's say you're looking at a 10 year period. Over that 10 year period instead of looking at the overall correlation, you look at the correlation in risk seeking periods and then you look at the correlation in risk averse periods, you get a much more accurate and a much more predictable, and much more true measure of correlation.
What we do is we not only look at an overall number of correlations, but we subject all these tests to again, further tests looking at them in different risk environments and so on. The only thing that we can control in that whole equation is how much risk we are willing to take. We can't control the return side of the equation because, depending on what the market gives us, is how much we can try to return based on the strategies that we have. We try to minimize the max drawdown, obviously to a point that, again, roughly a back of the envelope calculation, I think a reasonable expectation is what I call 15% return, 15% vol, 15% max drawdown. I think that is a good guideline over... if you look at different windows those numbers can vary, but looking at a long enough period, I think that would be a good sort of record to have.
Niels: That would be a terrific record. I think most trend followers would be envious if they could keep that...
Marc: Again, it's just too make sure that everybody understands, this is not for trend following, this is for Conquest Macro, which is sort of our 2 and 20 product, and not the trend follower, which is the 1 and 0. In the case of Conquest Macro, we really do try to provide some alpha that is very different than simple trend following.
Niels: Do you look at loss to stop as a measure? I'm asking this because I'm assuming you use stops for your trades, but I'm not even sure because I didn't ask you.
Marc: Oh yeah. For every trade that we do we have a stop associated with that trade, and a significant number of our trades we also take profits based on price level, we have time stops, we have stops based on volatility, so we use stops not just as losing sort of technique where you limit your losses, but we also use them a lot to take profits and vol stops and time stops and so on.
Niels: I think personally, I find that looking on a daily basis at what's the worst that could happen if everything got stopped out today, is a good measure of risk as well, and I also am not a big fan of the value at risk. I think it's worth pointing out to the audience that when looking at managers, and using your analogy about correlation and risk seeking and so on and so forth, I think what investors should really pay attention to is to actually look at the correlation in managers in a negative month, and in a positive month, and not just look at correlation in an overall month, because that doesn't really say anything in my opinion.
Marc: Absolutely. In looking at some of these stop losses, I mean yes, you want to look at what happens if every position that you have stopped out on that date; except that sometimes in cases where markets are negatively correlated. Let's take a simple example of your long bonds and long stocks. Those are trades that are working in a way most of the time end up working in different directions, so the probability of your getting stopped out on both on the same date is fairly remote.
Niels: Absolutely. I wasn't actually thinking about the probability as such, just people should be comfortable when... and I think investors should be comfortable with what could happen if my manager got stopped out of everything regardless of whether it was up or down.
Marc: In so far as your tolerance for a stop is going to affect your position size and your risk taking, you want to be very conservative in how you measure the risk, but there is such a thing as being overly, or unrealistically conservative in the sense that, again, one of the reasons why I hate value at risk is that it automatically assumes a blind approach of shocking the portfolio on a 3% move or a 4% move or so on. For very long term strategies maybe that works. In our case, our stops are within 1/2 a standard deviation, or 1 standard deviation, so by the time the market moves 3% or 4%, or 3, 4 standard deviations, we are long gone out of our position. So it's a completely irrelevant number. So when somebody looks at a snapshot of our portfolio, and brings me a vol number based on a 3 standard deviation move, it's completely ridiculous. It's meaningless, because we trade only the most liquid markets in the world, so the probability of a gap in any of our markets is almost nonexistent. As a matter of fact, I don't remember it ever happening since we traded. Given that our stops start within a 1/2 standard deviation of where we are, and probably end at 1 standard deviation, why am I looking at the losses on the full position from the 1 to the 3 standard deviation Marc? You want to be very conscious of your risks, and of position size, and of everything getting stopped out, but you also want to do it with somewhat of a common sense approach.
Niels: I want to jump to... I've got a couple of sections left that I want to talk to you about, and we're not going to go into every single question because we want to keep time to a certain level at least, but I want to talk a little bit about drawdowns. Often investors get very uncomfortable during drawdowns, and it often causes them to redeem, probably very close to the worse drawdown of any manager. I want to ask you about how do you,.. as a manager, when you look at your drawdowns, how do you know if something more fundamentally has changed and therefore you need to adopt, or adapt maybe is the right word, your strategy? Because what we have seen, certainly in the trend following space, is that since 2009,even strategies that have been doing well with a certain drawdown profile for 10, 20, even 30 years, suddenly the drawdown profile has changed in the last few years, and many people have recorded larger drawdowns than they have prior to that. How do you view that?
Marc: A couple of things, first there really is nothing new about what is happening now, in the big picture, than what has happened before. Of course, the way a max drawdown happens is because you made a worse drawdown than the last one. Well, yeah, that's sort of, at any point, if you go back to a lot of these strategies prior to 2009, you will see a point in which they made a drawdown, then they rallied and made a lot of money, and then they probably made a worse drawdown, so they hit a new low in their drawdown, but they came back and made more money, and now they hit another lower low in their drawdown and, mind you, they will come back and make money, because fundamentally, I don't think anything has ...I don't think anything has fundamentally shifted in the way markets work.
We have seen a temporary shift, which is in this overly active intervention of the Fed and the markets, but this is, in a way, what they are doing is not new, because the Fed always intervenes in the market, all the way from the open intervention when they tried to manipulate currency markets by intervening in the currency market, to changes in interest rate, which really is part of their mandate, but it's still an intervention. What they did this time is they took it to a new extreme. It's not unexpected that that new extreme caused a corresponding new extreme in strategies that don't benefit from these interventions. At least in the US we are seeing the beginning of the unwind of that sort of level of activity on the side of the Fed. I have every expectation that ...my own personal view is we've really, rather than try to solve the problem that caused 2008 here and in Europe we've tried to throw a lot of money at the problem. We haven't fixed any of the problem, but now we're short all that money that we threw away at it, and we're still left with a bigger problem.
So my own expectation - I think a trend follower is going to come back and actually have probably a much better year than they've ever had before to be commensurate with having the worse drawdown than they've ever had before. Maybe that's part of the question that you asked, but let me answer of how we deal with this in Conquest Macro. For every model that we build, and then that model becomes part of a sub-strategy, and then for every sub-strategy that we have, and then that sub-strategy becomes part of the portfolio then for our portfolio. What we do... what I mentioned earlier is we've identified over 60 different statistics that we use as almost like diagnostics. So what we do is we take a number of winning trades or losing trades, any one of those stats, let's say. What you do is you have a running total of these stats. You build the distribution for these stats, then on each day you go and calculate that number for that day and you see if it fits within the distribution. In addition to that, what we do is for every model that we build, and again for the sub-strategy, the strategy, the portfolio, we build a beta version of it.
So what's a beta version? For example, let's say I have a short term trend following model in Conquest Macro that has a 7 day average duration. I know that something as simple as a 5 day breakout strategy, roughly you have the formula between the holding period and day breakout is that for an end day breakout it's going to have a holding period of 1.5N, so roughly a 5 day breakout is going to have a 7 1/2 day holding period. Really, the beta of my Macro short term trend following model is going to be a 5 day breakout on exactly the same market in the same proportion. So let's go back now and say OK, when I look at, on a daily basis, I look at these diagnostics for every model that we have and sub-strategy and portfolio and so on. When you are looking at over 60 different stats, the chances are that any given day you're going to get a few of these stats that are falling outside of the distribution, which is really nothing unusual because that is how distributions get built.
Where I start to worry is if a cluster of the stats start to break. If a very large number of these stats start breaking. Mind you, right now I'm not looking at any return, whether it's making money or losing money, we're not even looking at any of these, we're just looking at the stats and these stats can break both because it is making too much money or losing too much money, or not making any money. Once we've determined that model is behaving differently than our expectation based on those diagnostics, the first thing that we do is we go to the beta version of that model and see what the beta is doing. If the beta version of that model is displaying the same breaks in these diagnostics, then what that tells us is that this break is much more likely being caused by a change in the way markets are behaving rather than the way our model is behaving. So that gives us a comfort level that the model is not broken, but then it makes us go and take a look at the market to see if there is anything that has changed in the market. Now, 99 times out of 100 nothing has changed in the market, and again these distributions happen because different things happen at different times, and they get to build these distributions, then they end up coming back and correcting themselves. However, we've had cases where we've decided to shut down a model or shut down a market because of this.
For example, looking at... giving you a very obvious example, but, trading Euro/Yen futures in Japan is not really a good idea in a short term trading model (laugh), because something had fundamentally changed in the way that market trades, or for example sometimes in a given model if we start seeing breaks from the beta where the beta is behaving a certain way that is still in our expectation of how it should behave given the market condition, but our model is not, then it goes on the operating table and we try to dig deeper, and we have no problem killing a model if we think it's not behaving as we thought it should. To counter for that what we do is whenever we introduce a model, we never introduce it fully into the portfolio. So even after the model makes it from the research phase into being ready to go into the portfolio, it goes three months of paper trading without any money; if it survives that, then another three months in the portfolio at a 1/4 weight; if it survives that then another three months at 1/2 weight, and so on. So it's not until over a year's time that it will go into the portfolio at full weight and even then it's one of 60 plus different... we do a very incremental approach to this stuff.
Niels: Sure, makes sense. Now, there's an intellectual side to drawdowns that we've just discussed, but there's also an emotional side. How do you deal with drawdowns from an emotional side? How have you learned to ...
Marc: If your strategy... if nothing fundamentally changes in the market, and your models are not broken, then a drawdown is the perfect place where you should double down. Now, obviously, we don't do that. In being a money manager, or hedge fund manager, or CTA, 99% of the time your interests are aligned with your investor’s interests. You do something that is good for you for your business and for your investor. A drawdown, as you correctly mentioned, is a very emotional issue and that's really the one time where I think managers, including us, do something that feels better for the investor, but probably not in the investors best interest which is cutting positions in a drawdown and reducing risk in a drawdown. Investors demand that you have some drawdown mitigation technique and within our fund we have hurdles where if we had X drawdown then we cut the position by this much, and if we hit Y then, we cut them even more, which is... I'll be the first one to say it, not the best portfolio decision, but it's a business decision and it's something that investors ask for. I'll also tell you that we have very large managed accounts where investors choose that they don't want to do that. Either they come and add themselves, or they ask us not to cut positions in a drawdown. It takes a very astute, and an investor who really understands your strategy not to want that.
Niels: Just a final question about drawdowns and risk and so on, and so forth. Is there anything that keeps you awake at night - something that you worry about when you look at the way your program is designed and your strategy and the markets and the way the world is today? Is there something that you say, mmm, I don't really want this thing to happen? I'm not saying that you have sleepless nights...
Marc: By the nature of what we do the bad periods are, for us, are a bleed over time, and the good periods are the big shocks. So if I was managing a hedge fund that is your typical long risk hedge fund, I would certainly be very worried every night that one of a million different exogenous events can happen, and I can wake up to a 20% loss. That type of event can never happen in our strategies because, by definition, any exogenous type of event like that would be hugely beneficial for us. We benefit from the sudden event that keeps people awake at night and the way we lose our money is really not with any fireworks, but the lack of. So it's sort of death by a thousand cuts, and not explosions. I might stay up at night hoping...(laugh)
Niels: (laugh) Maybe that's what's keeping you awake, the fact that nothing is happening. That's interesting. That's a good way of looking at it. One other topic I want to touch upon. I know we've already talked about it. It's a little bit about research. Investors, they want us to do research, they want us to innovate, but they don't really want us to change. We have a certain profile, and you've explained that when you do implement things it doesn't change the overall profile, so that I understand. But I want to ask you a little bit about backtests, because when you do a backtest and you finally come up with something that looks promising, it obviously looks good, that's the nature of the beast, but things do not always work out as the backtests suggest, so how do you balance, or how much weight do you put on a backtest, and what other things, it could be intuition - I don't know, but are the things that are important for you when you make the final decision in saying, yeah, this model I really like and we should use that?
Marc: You can't really blame the backtest. The backtest is just what you put into it. Obviously garbage in, garbage out. So, 99% of the work happens before you go into a backtest. What I mean by that is look, again, it's a continuum meaning if I take a very simple strategy that's a 50 day breakout strategy: buy when you break on the upside, and reverse and sell when you break on the downside, and I put this into a backtest, it's going to give me what I believe is the highest confidence level backtest I can find. It's as pure a beta as you can find. Even in that backtest it's still not going to cover every market condition that can happen. So think of it very simply, again, going back to 2008, if you run that backtest up to 2007, you're going to have a certain risk-return profile in it, but then you add 2008 to it and all of your stats are going to go right through the roof, although it would have been very positive in making money, but again it's not going to be true to the back test because it's going to make different extremes.
If I think of this simple 50 day breakout strategy as one end of the spectrum and then the other end would be the completely over fitted strategy that will give you the 4 sharp, but from day 1 it will never work. Where you want to be is obviously a lot closer to the first one than the second one, so how do we do this? One, is you want to think of the number of variables and parameters that go into your model as each one being inversely proportional to the number of degrees of freedom that you are putting into your model. With each additional parameter that you are putting you are lowering the number of degrees of freedom that your model can have to react in different market conditions. Again, taking it to the limit, put a thousand different parameters on, you can get a model that never loses on a backtest, but will have zero chance of winning. That's the first layer that you look at. Then, even then, lowering some of the parameters you can always try to solve for the perfect parameter, and sometimes that perfect parameter is actually a singularity, or close to being a singularity, where if you start deviating a bit to the left or to the right it completely breaks down. You can easily fix for that by, for each parameter that you are looking at, making sure that any marginal change in that parameter value up or down by 25%, 40%, or 50% is not going to have any significant effect on your portfolio, meaning that you're not really back fitting to that exact one parameter that made things work for this. There is some level of comfort. There are different techniques that you can do that you have to do before you get anywhere close to a backtest, that again, will ensure not that the backtest is going to be the holy grail, but that it gives you a fairly good estimation of what your risk return profile should be. It's subject to what the markets give you.
Niels: Do you ever feel restricted in your research because you have this objective of creating a program that does well - kind of a portfolio protection program, if I can call it that, and that obviously puts in some limitations, I would imagine. Do you ever feel restricted in your research, saying I'd love to put that in, but it just doesn't fit...?
Marc: No, because we don't. We don't think like that. What we do is, in research we go wherever research takes us and again, we try to achieve our negative correlation in Conquest Macro by doing a lot of asset allocation. However, we also have another program called Conquest STAR. Conquest STAR, which stands for "short term absolute return" is essentially a version of Conquest Macro, with only the absolute return mandate. It's a short term strategy that doesn't have a performance and risk aversion mandate. There are certain models that we come up with that might go in Conquest Macro where we'd only turn them on in one environment or the other, but can go into Conquest STAR as a full-fledged model. So we go where the research takes us. Again, going back to the concept of the Lego pieces. You keep building those Lego pieces as best as you can, and then you use them in making the different portfolios afterwards.
Niels: Yeah...I want to shift gears on you again, Marc, I want to go to a few things on what I can the business side of things, and then at the end we'll finish off with some more general and fun stuff. I just want to hear your view and opinion: on one side the markets have been challenging and therefore many strategies have had a challenging time in the last few years, but there's also a business side to what you do as a manager, because you are an entrepreneur, so you're running a business at the same time. What has been your challenge as a business, do you think and how is business?
Marc: Well look, obviously it could have been a lot better. We are used to being in the kind of space that we're in, which is the trading world space. I would say that even though we have strategies that do... we have funds that are ready to be launched and strategies that do well in risk seeking periods as well, really our bread and butter is still coming from the Conquest Macro side of the world. So when you're starting off with a strategy that really looks great only about 30% of the time, and the other 70% of the time you're playing defensive, you're already used to some of these ups and downs. What makes this particular stretch more painful is just the length of it, because in these six years, not six but maybe since 2009 - five years, we should have had at least a couple of really nice moves into risk aversion, which we didn't have because of the global central bank action, so it's been a particularly long period of drought and, let's be frank, a lot of funds like us have gone out of business, and some other ones decided to change their strategy and do other things. We're lucky enough to have investors stick by us. We're still around, obviously, we're not making the living that we made in 2008, but it's the nature of the beast, sometimes you go through good periods and you go through bad periods. What I'm hopeful for is that, I think really this massive compression and selloff of vol and intervention, and everything is setting us up for something that will rival 2008 if not worse. So think whoever is left standing, when we finally turn the corner on this, will end up being in a very, very good position.
Niels: You mention something which I think is interesting, and the whole explanation about the businesses and it goes in cycles, and so on, and so forth. I don't want to put any words in your mouth, but you probably could have retired a few years ago and not be going through all of these phases, so I'm interested in what motivates you? What keeps you going? What makes you want to come in every single day and do what you do?
Marc: I can't think of anything else I would rather do (laugh). My wife would kill me if I spent my days at home, so I kind of have to leave the house (laugh). No, but, on a more serious note, even in its worse days, it's still the best job in the world. It's very dynamic. It's exciting. Things are moving all the time, from research to trading ideas, to looking at the markets and what you have done before is working. It fulfills you on so many different levels. In a way, most jobs should do that. I think what this does is it has much higher highs and lower lows. It's very addictive, but in a good way, I would say.
Niels: Do you remember how you got your first investor?
Marc: Yeah, you know, when I left I told you I partnered off with a friend of mine and I had gone from running this global group at UBS and having a huge team all around the world and I remember me and my friends were sitting... we rented a one room office on Park Avenue that we just carved out a small conference room, just big enough for the conference room table, but it was really a one room office. On a Friday night, I think it was; we were sitting and building desks that we bought from Staples. I didn't know that they don't come assembled. At that point, sitting there, I'm saying, "am I sure that I am doing the right thing?" That's when it hits you, but no, there was no doubt really about that step. We started... I put in my own money. My partner at the time put in some of his money, and that's how we started. I think when we started it was like 3 million dollars. Just through contacts... I miss those days because back then, even in that one room office, we started trading with our own money, but pretty much people at the time - this was not so long ago - in 1999, you met with the people who's money it was and typically these people didn't care what our office looked like, or how many people you had, or what art you had on the wall, they came and they spent 2, 3, 4 hours with you and if they believed in you and what you're saying, they gave you money, and if they didn't than they passed, but there was no bullshit. It was really about the strategy and about what you're doing. Unfortunately, that doesn't exist anymore. Now there are so many layers in-between you and the ultimate investor. Whether there are many different professional, like investment due diligence people, and some know their stuff and a larger majority of them don't. They come with a list of boxes that they need to check without really fully understanding what it is that you do. So, in a way, it was more fun before and in a way, easier. Right now that has sort of shifted from investors really looking and delving into the strategy, to right now, more of the herd mentality where either the big funds get bigger, because no one gets fired for buying IBM, or you have to put in really outsized type returns for a year or two, then they come and they buy the return, again, without really understanding what caused those returns.
Niels: That's a perfect segue to my last question in this section and that is, you've obviously been part of hundreds, I'm sure, of due diligence meetings and answered thousands of due diligence questions, even though probably most of them are the same, but I want to ask you, what are the investors forgetting to ask? What are they not asking you that they really should be?
Marc: It's tough to answer that question, because due diligence meetings now have stretched to that 4, 5 hours, and then not only investment due diligence, but you have operational, you have legal, and all of that, so I think most of the relevant questions are being asked, what I'm not sure of though, is what people are doing with those answers. I think the questions are there, but I'm not as sure that someone is processing all of the answers in the right way.
Niels: Yeah, my own impression is, and that's one thing that I, I don't know if I've ever been asked by someone coming and doing due diligence, and that's the why. They ask a lot of what and how we do things, but they never really ask why we do it. I think that's missing, because I think if people understood the why they would have a much easier way to one, communicate to the investment committee, because usually the investment committee members are not really present anymore, but just also generally to understand the business as such, because at the end of the day the only thing that differentiates one manager from the other is the manager themselves.
Marc: Absolutely, whether it's systematic or whether it's discretionary, it really boils down to the manager. In my mind, I don't see a difference between these things. Ultimately it's the person.
Niels: Yeah, which is also why I end up with this section called general and fun, because this is where, Marc, we really get to know you (laugh).
Marc: (laugh) I'm scared now.
Niels: (laugh) Scared now; that's right...no there's no need for that. I just want to ask your opinion. We've touched upon it in different ways. There are many people listening today who maybe want to find out what it takes to become a great trader and so on, and so forth. From someone who has done it over a long period of time, what would you say it takes today to become a great trader? What are the personal traits that you need to persist in order to achieve that?
Marc: First, thank you for the compliment, but I wouldn't presume that I am there yet, so I am still looking with the rest of them. I'll tell you; discipline - discipline is extremely important. You need to have not blind faith in yourself, one should always question what they're doing, but you really have to have the ability of tuning out the rest of the world around you to the point where you're comfortable saying, yes, 90% of the people, or 95% of the people are wrong and I'm right. It's not very easily done. I don't know whether it's hubris or what you want to call it - thinking too much of yourself, but it takes a real sort of combination in-between having the right discipline to question certain aspects of what you are doing and how you're doing it, but also the fortitude to stick to your guns. You get so much unsolicited advice at the top and the bottom and all points in-between, and you have to be able to really process the information without the noise. You need to start with some level of talent, but very importantly, you need to be able to compliment it by having the right people around you. Successful traders are successful because they're good at what they do, but also, to a large extend, they've surrounded themselves with people who maybe are even better than them at what they do. I have no problem hiring people that I think are better than me. It just makes the whole thing more enjoyable.
Niels: To keep the noise out, as you say, do you have any personal habits that you do that helps you do that on a daily, weekly, monthly, whatever, basis and that you would say has contributed to your own success?
Marc: Well look, I'm really a trader at heart. I use quantitative techniques to implement my views. Whenever those noises are really loud, it's going to sound very strange, but I find it very therapeutic and much more calming to go back and look at the markets, look at the charts, look at our stats, look at the models, how we're doing, it's a way for me to remember how much work we've put into what we're doing and what we're doing to make sure that we're still... we believe that the models are still performing the way that we expect them to. It's more diving deeper into the things that are causing that pain at that moment, but if you've done your homework right from the beginning, it gives you a certain level of comfort.
Niels: In a long career there's always going to be highs and lows, and there's always going to be things where you said, yeah, I did that well, and there's going to be things where you said, no I didn't do that well, I failed at this. Is there anything where you can look back and say what your biggest failure has been and how you ended up overcoming it?
Marc: I can go on for a long... but I'll choose the two big ones. If I could go back and do it all over again, I'm not sure that I would choose to be in a risk-averse strategy. I think it's very important in a portfolio, but you inevitably end up being the person late at night in a party going around and telling people to stop drinking because they're going to get a hangover. That's a boring person; no one wants to hang out with them (laugh). It makes it a much tougher sell to do what you are doing. I have the same skill set that allowed me to do what we do here, I could have used it on the long side of things, and I think it would have been a much easier path and probably, financially, a more successful one because what ends up happening is when you're in a long risk strategy you're making money when everyone is making, but you're also losing money when everyone is losing, so you share the highs with a lot of people, but you also share the lows with a lot of people. In what we do, when things are great for us is when they really are horrible for everyone else, so you can't even celebrate, but also when you do your best - in a year like 2008 you still end up suffering significantly because you become the ATM machine. We're up over 50% in both of our programs in 2008, but still saw our assets being cut in half. So that's one. On a portfolio basis, we've done a lot of things that, in retrospect, were not the best thing to do, and we've undone them. That's part of why, I think, today we have the best portfolio we've ever had. It's because of all of the mistakes that we have made in the past and what we've learned from them.
Niels: I've only got two more questions I'd like to ask you, Marc, and that is, is there a fun fact that even people who know you don't really know about you that you can share? I've had very varied answers, so anything goes here I think.
Marc: Fun fact? I have to think about this one. My kids think I'm a lot of fun (laugh). I love to do dangerous things.
Niels: Give me an example of a dangerous thing?
Marc: Rock climbing, flying a plane, jumping from a plane, that kind of stuff...
Niels: There is a danger side to you.
Marc: It's more sort of... I wouldn't do anything that I think truly is dangerous. I tend to be very risk-averse when it comes to that. I think, just looking at the probabilities, why people might perceive danger is actually a fairly safe thing.
Niels: My last question, Marc, I asked you earlier about investors not asking the right questions, or failing to ask important questions, so I want to put myself on the spot as well and ask you what I failed to cover today. I know that we can't cover everything, but is there anything that you think I missed and that you want to bring up before we end, to do justice to yourself and to Conquest?
Marc: I think that you asked all of the right questions in the time that we had. As long as it has been, I'm sure we could have taken another 4, 5 hours and talked about the rest of the stuff. As you can see, I'm fairly passionate about this, so when I start talking I can't stop in some cases, but I don't think there is anything glaring, but just by nature of how wide-ranging the topics we discussed, each one of those you could spend another hour or two on, but that's not for a lack of you asking the right question, it's more of just timing.
Niels: Sure, absolutely. Well, before we finish our conversation can you tell our listeners where they can best reach out to you and learn more about Conquest please?
Marc: Of course, our office is in New York City, we're at mid-town, sort of hedge fund central. Our phone number is (212) 759-8777, my email address is email@example.com. If anyone has any questions, comments, they can feel free to send me an email, and I'll either answer it myself or route it to the right person.
Niels: Fantastic, and of course, all of these details will be in the show notes on the TOPTRADERSUNPLUGGED.COM website as well, Marc. I also want to say one other thing to the listeners, and that is, in the email that we send out, there is a place where you can click to thank Marc for sharing his story and his expertise and I encourage everyone to do so, and I'll certainly be the first one to say, Marc, thank you ever so much. I think it's been a great learning experience. It's been a great conversation, and I appreciate your transparency and your willingness to share these insights about your strategy and your firm and the industry as a whole, so I really do appreciate that.
Marc: Thank you very much. I appreciate your time and the listeners time.
Niels: I hope that we can connect at a later stage and see how things are progressing at your end, and maybe dive into some of the topics in even more detail than we did today.
Marc: Absolutely, thank you Niels.
Niels: Thank you. Take Care.
Ending: Thanks for listening to Top Traders Unplugged. If you feel 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: 18 Sep 20141 comment