TTU Podcast Episode #061. Peter Kambolin, Founder and CEO of Systematic Alpha Management. Show notes at:

Transcription

1 TTU Podcast Episode #061 Peter Kambolin, Founder and CEO of Systematic Alpha Management Show notes at: Peter: I think investors this time around will be a lot more careful from the standpoint of their allocations. I don't think it's smart to give all your money to similar CTA's that have 65%, 70% correlation for one another. I think this time around investors will have a lot more diversified CTA portfolios with some short term guys, with some mean reversion\ momentum guys, and it will be good for them and it will be good for the industry. Because it's unlikely that, that portfolio of diversified CTA's will have the same drawdown, as some of the lower term CTA's had in 2010, 2011, and This is Peter Kambolin, Founder and CEO of Systematic Alpha Management in New York, and you're listening to my year-end review on Top Traders Unplugged. Introduction: Imagine spending an hour with the world's greatest traders. Imagine learning from their experiences their successes and their failures, imagine no more. Welcome to Top Traders Unplugged. The place where you can learn from the best hedge fund managers in the world, so you can take your manager due diligence, or investment career to the next level. Here's your host, veteran hedge fund manager, Niels Kaastrup-Larsen. Niels: Welcome back Peter for this review of 2014, where we look at the big events from the point of view of your trading strategies. I want to explore both the ups and the downs as well as the big takeaway from what can only be described as an interesting year for systematic trading strategies, in general. But as we know, just because you're systematic in your trading, it doesn't mean that your strategy deals with market events in a similar way. So I want just to jump right in. Talk to me about 2014, from your perspective, and how the year evolved both from maybe your firm's perspective, but also of course from the trading strategies point of view. Peter: Well we have two strategies, two programs that we are offering our clients. One is a market neutral program that has a very long track record, but with ten years in fund. And that's the problem, that people know us. And we have another program that we launched about three years ago which we call the Multi-Strategy program. It combines, in a 50/50 blend approximately, Market Neutral spread trading that we do in the core program with directional trend following momentum models that exploit a totally different set of futures instruments. So if we look at 2014, it was a challenging year for our Market Neutral program. In fact in over ten years we only had our second losing year on track record. On the other hand the Multi-Strategy program This year it is close to flat, primarily because the directional component performed extremely well last year, somewhat in line with most CTA's that have Copyright 2015TopTradersUnplugged.com (CMC AG). All rights reserved. 1

2 seen very nice returns since May, June of last year. Similarly with us, our directional strategies performed extremely well. So as a result, in 2015 we are now offering our Directional program as a standalone for those investors that are interested in tapping into our skills, placing directional trend following using our systematic strategy. Niels: So what you're saying is that the Multi-Strategy could be completely replicated if you were to invest 50/50 - Alpha Futures and the short term directional program? Is it really that simple of a carveout, or are you doing something a little bit different when you have to make it a standalone program? Peter: Well, in the standalone program we won't do as much asset allocation or deleverage or re-leveraging that we are doing in the Market Strategy program. In the Market Strategy program, depending on which models perform well, we could potentially reduce allocation to certain models and increase the allocation towards the directional models. So it's not a simple 50/50 split in terms of the risk. So if one is interested in purely directional recurrence, short term fully systematic momentum trend following type of recurrence, then the directional standalone program could be a very interesting candidate. The good thing about that program is that it's correlation to other CTA's is still quite low. It's at the level of 15%-20% on a daily level. Even though we are trading directional trend following momentum models that are somewhat similar to what most other CTA's are doing. Niels: Sure. Okay, well, let's head back to 2014 a little bit now. Clearly a challenging year for mean reversion type strategies. Tell me a little bit maybe before we talk about why it was so challenging, talk to me about, just from a market point of view, which markets did well for you in the portfolio, and which markets turned out to be much more challenging to trade? Peter: Well our market neutral program trades so-called European spreads, where we trade US equity indices against European indices, hedged with local currencies. An example would be trading SMI Swiss market futures against S&P 500, hedged with the Swiss franc currency contract. We trade North American spreads, such as NCAP Index versus the Russell 2000, versus S&P 500. It's a very typical example of a North American relationship. Also we trade Canadian spreads, Canada, S&P 500, Canadian Dollar, or in the case of Canada, we also hedge with crude oil, copper, and gold, because the Canadian stock index is highly correlated to those commodities. We did not trade Asian spreads last year. We stayed away from Japan for approximately two years now. This is because the behavior of Japanese Nikkei 225 has been quite abnormal over the last three and a half, two years because the new economic policy that they introduced. And we stayed away from other Asian spreads as well. So basically, in the portfolio we were trading European spreads and North American spreads. And in Europe we also split Europe into three parts: Euro denominated countries, Swiss franc (in case of SMI), and British pound FTSE 100 trading. Copyright 2015TopTradersUnplugged.com (CMC AG). All rights reserved. 2

3 So in terms of how each sub-group performed, I would say the worst performer was the Euro denominated spreads. The best one out of the three in Europe was the FTSE 100 relationship, and SMI was somewhat in-between. In US/North America, last year we actually only started trading MidCAP index. Before that we traded a lot in Russell 2000 against S&P, so we did not trade MidCAP in prior years. But we realized last year that the liquidity in the MidCAP index is quite good, somewhat overlooked by many traders, on the other hand. And with all the spreads that we traded last year: MidCAP (S&P 400 it's called), index spreads performed the best last year. Niels: Sure, sure. Now you mentioned something that I think is important, and, of course, the listeners can go back to hear much more detail about what it really means in the longer conversation we had last year. But you essentially have three components in a position, you have the relationship between the two markets, and you have the hedge. If you look at the and I don't know whether you look at this, you probably do But if you look at the performance distribution, or if we call it that, was it the relationships between the markets that you were trading that was difficult, or was it really the hedge of the relationship? Peter: Well, we look at all three legs of the spread through equity markets and let's say currency hedge or equity hedge. They are all the same to us We look at the P&L for the spread itself; we don't care if we're losing on a currency position, for example while we're making on the equity indices or vice versa. Because sometimes the movement of currency could cause some very unusual price action in the equity indices because there's a high correlation between currency moves and equity markets moves. So, we are let's say, willing to take a small hit on the currency while we will be making a lot bigger gains on the equity trades or vice versa. So to answer your question, we do not look at separate P&L's coming from each leg of the spread. We look at the whole position, and for that reason it's hard to answer your question. Niels: No, no, I understand. No, that's fine; that's fine. Now, clearly there were some big moves last year, and I imagine for the trend following models, they enjoyed that, and for the mean reversion models, maybe that was more tricky to navigate. But, my question is a little bit different. When you look at the year as a whole, and you look at what the markets are there any other the things that you saw last year that make you feel that maybe we need to look into this a bit more because we didn't do as well as you would have expected? Now we all know we can all lose money in a bad environment, that's fine. I'm actually more looking for the question to be, was there anything that you said, "That's surprised me a little bit, that we couldn't make more of that environment?" Peter: Right. First let me describe the environment. Last year was quite unprecedented from the standpoint of volatilities in different markets. During the summer month's volatility in equity indices, and in some currencies were at an historic low level. When volatility is so low, it's very, very difficult to make money, especially in the mean reverting type of strategies. Because a market that's slowly drifting, let's say upwards, and that's what was happening with equity indices. You know, a trend follower could catch that equity move, while we are Copyright 2015TopTradersUnplugged.com (CMC AG). All rights reserved. 3

4 looking for intra-day volatility spikes that cause markets to be dislocated over a short period of time and that could potentially generate good positive returns for us. So, an extremely low volatility environment in 2014 was the main driver of a difficult time for our Market Neutral program. And again, if we look back, the last time we saw such slow the environment was end of '06, We know what happened in '08, and '09. So, it's quite possible that we are at the point where there could be big potential shifts in sentiment and in direction of the markets. And this year we've seen already volatility spiking up to very healthy levels, equity indices selling off, etc. Niels: Sure. No, absolutely. Now of course 2014 will be remembered for a few particular themes. I think it's fair to say that people will look back and, in particular, at least they will be remembered for Ukraine, Russia, and oil in the latter part of the year. I don't know if you can, but it would be great if you could, try to visualize a little bit how you navigate themes like that if you can. I know you're short term and so on and so forth, but the reason I ask this question is because I think a lot of people will always think that themes like that are negative. And that's probably because they come from the long-only world where oil stocks collapse because oil goes down, or whatever it might be. So, I want to try and just paint a slightly different picture that describes how themes like this do not necessarily mean that there are no opportunities for other types of strategies. So I know that there were some big moves, and obviously in your trend following program, clearly a component clearly the latter half of 2014 was very, very strong. I note that in the mean reversion program the spike down in the S&P in October may have been one of the challenges you have. So talk to me a little bit about those kinds of three or four themes that we saw last year. Peter: Well, yeah, actually in October we implemented a minor adjustment, let's call it. When we were re-backtesting our models, and typically we would re-backtest all the models once every two months. We force our model to take more trades. We implemented certain filters, certain parameters during our backtesting procedure where we wanted to increase the number of trades that we picked. Even though we are short term, and compared to many other CTAs we are one of the shortest term CTAs out there, we wanted to keep our trades even shorter than what we had before. Niels: Why? Peter: That's related to these drifts or directional moves in the markets that caused us pain and difficulty. We would take a position. The position would slowly drift against us, and instead of closing the trade within let's say one day (that's a typical holding pattern for us - 6, 8, 10 trading hours). Some of the trades that we kept last year lasted for several days that, for us, is very unusual. If they last a long period of time, that means that there's no mean reversion. It means that the trade is going against us. Copyright 2015TopTradersUnplugged.com (CMC AG). All rights reserved. 4

5 Niels: Remind me again, Peter, why can the trade length be so different? I don't remember that from our first conversation. Peter: Well, again, we are looking for a mean reversion backward move. So let's say the spread is touching our upper band, and we are going to go short. We expect a reversal, so we would take a short position and then for 3, 4, 5, 6 days the spread slowly drifts upwards - continues upwards. At some point, it will hit our stop and we would lose our stops at 2%, 2.5%, 3% depending on the models. This is not the best way to stop-out of a bad trade. We work to make our models more flexible and even on small reversals. It is not like a big reversal with a gain, but there's always zigs and zags. So these reversals, we would rather get out of that trade and wait for another opportunity. What we did in late October when we re-backtested our models, we realized that all of a sudden in 2014 we were holding our trades for a longer period of time than we used to in the past. We forced our backtesting to produce such parameters where we would basically shorten the time that we spent in a bad trade or a good trade; it doesn't really matter. So that's how we decided to fight with these drifting markets that were low volatile markets that weren't a theme last year. Ever since we did this, we've had a positive response. In November was a small positive, December was strongly positive, but it's also related to a better environment we believe that we're seeing now with the volatility spiking up. Niels: Do you think... I mean spread changes in terms of what you trade, do you put a lot of weight on the sort of human behavior? Meaning that we as humans cause these spreads to go in and out one way or the other? Peter: Well, yes but when we play these trades these are fully systematic trades so sometimes we cannot even explain why certain markets are going up. Niels: No, the reason I asked that question is simply because I guess more and more trades now-a-days, more and more trading around the world is not done by humans. So the human element, so to speak, in some regards is diminishing. I was just wondering whether the different market environment that you've now observed, and you clearly reacted to it. I wonder whether that's caused by the general environment such as the theme: oil, Russia, whatever, or it's also slightly caused by the fact that there are less and less people making buy and sell decisions in the markets. More and more of that is done by some kind of algorithmic trading strategy. Peter: Well, that's why in our main marketing program we stayed away from trading currencies, spreads; stayed away from trading commodities and fixed income markets even. Those markets are primarily driven by institutional investors or these computer-driven programs. Equity indices are still interesting for us for the reason that you just actually brought up. A lot of retail non-professional participation is in stocks and equities. On the one hand, of course, the percent of algorithmic executions is growing in equities as well, on one Copyright 2015TopTradersUnplugged.com (CMC AG). All rights reserved. 5

6 hand. On the other hand five, six years of very strong bull markets produced another generation of on-line traders. People in the regular... non-professional people now believe that they if put their buy order in, in the morning, by the end of the day they will be up, you know, a 1/2 of a percent, or whatever. That's how markets behaved over the last two, three, four years. So the amount of these speculators - non-professional speculators usually grows after bull runs. That's what we saw back in '06, '07. At that time there was a big number of on-line accounts and on-line trading by regular people and that's why it was so much easier for us, I would say, to produce very strong returns in '08, '09. We were basically trading against nonprofessional people who were very emotional about their decisions, and they cannot react as quickly as we can to certain events. Hopefully, we think that, going forward in 2015 and beyond, this type of environment will come back. Basically, people need to lose a lot of money. I'm talking about retail people before they close their accounts. So it will take some time for them to get out of the markets altogether. This year even... earlier this year markets were at all time highs one more time, although things are different today and over the last week or so. My point is that most likely the environment for us, for our type of trading, should only get better going forward because of the reasons I've... Niels: Sure, makes sense. When you look back on the year as a whole, what's your highlight when you think of 2014? Peter: Well, definitely the events in the Ukraine and Russia impacted our trading. There was some very unusual activity, I remember, in the FTSE 100 market right before the Crimea events, and right after. From the rumors that we heard, people... basically they knew that Russia was going to take over Crimea and that, that would very definitely be a negative event for the markets. So there was a very unusual sell-up in the FTSE market, in particular, right before the Crimea events. Then there was a rebound later on, a few weeks later. While other markets behaved a lot more normally. For us it was a negative event because the FTSE market became extremely abnormal, most likely due to a lot of initial shorts - covering shorts were done by possibly Russian banks or some other insiders that knew what would happen. So definitely that was the event to remember. The second event to remember, even though the impact of that event will be realized over time, is the fact that US stopped its monetary QE policy. On the short scale of 1 to 3 weeks or even months, it's difficult to feel that. It happened, but going forward maybe, partially, the reason, that volatility is rising nowa-days, is partially related to the fact, you know US Fed no longer supports the market directly. Niels: Sure, so in a sense that's kind of one of the things that I also want to touch upon. Clearly there's been some prominent people out in the last couple of weeks really arguing the case that the world is becoming much more divergent,. Clearly this is not just from a market point of view but also from a policy point of view. You mentioned one of them is now that the central banks are doing different things rather than going in tandem. Divergence in general obviously could lead to higher volatility and is that why you are excited about the Copyright 2015TopTradersUnplugged.com (CMC AG). All rights reserved. 6

7 future that you would welcome divergence rather than this sort of more controlled convergent environment we've been in? Peter: Well, generally, the higher the volatility, the better for us. What causes it to go up? Sometimes it's hard to know or predict. Also, what's important is that central banks now a days are a lot more transparent about their future actions. So markets now a days... prices in future events, I would say in a better way. The news is out. For our particular program, what is not good is some unusual all of a sudden news announcement that is totally unexpected. That most likely will produce some very abnormal moves in different markets. Clearly European authorities and the US Fed are quite transparent trying to describe exactly what they are planning to do in the future, and people and traders and markets have enough time to participate and to price that in. Niels: Sure, it makes sense. You mentioned that you launched a new program, you made a couple of small tweaks to the mean reversion models, are there any other sort of changes at your end in 2014 that you want to highlight? Peter: Well, we improved the quality of our backtesting when it comes to directional programs. Partially we believe the reason for the very strong returns is due to that. We now generate backtests that don't rely on... we're doing walk forward out of sample backtests. This means that we would produce them, let's say going back three years. We would generate certain parameters from back three years ago. We would trade these models out of sample for a period of one month, let's say. We would record out of sample performance. We would then re-backtest the models and then use a new set of parameters for another month of trading, also out of sample. Basically we are generating, during our backtest, truly out of sample performance that allows us to see which models are good and which models are out of... you know over-fitted. Niels: Are you able... this is certainly a theme that I've heard from a number of people. It seems to be the way research is going. I imagine you run your simulation environment side by side with your real environment and look at how the two stack up, and have they been running in line? Peter: Yeah, the matching is very, very good. If you do a backtest with out and out of sample back test, you could potentially produce a beautiful equity curve pattern that's impossible to generate in real trading. What we are doing, we are basically seeing which model survived over in the production code by real market conditions and environments and which models we need to keep in the portfolio and which ones are just... So that type of walk forward out of sample backtesting we started doing in the summer of last year fully for our momentum and directional models. Clearly the results are extremely positive. Niels: Does that mean, Peter, that over time you're going to get to a point where you want to almost recalibrate the program on a monthly basis with potentially new models or a new lineup for the next say 30 days (I'm just picking a number here)? Copyright 2015TopTradersUnplugged.com (CMC AG). All rights reserved. 7

8 Peter: We backtest once every two months: out of sample for two months and then we would rerun the parameters. Models we keep more or less the same, it's just certain details are more defined. Niels: But you would then fine-tune the live trading systems or parameters according to your findings on a bi-monthly basis? Peter: Yeah, that's what we do, correct. Niels: Very interesting. I've actually had one guest on last year, where they completely set a new... or at least they allowed the models to pick a new team every month. Obviously it could be picking the same team, but it's kind of going in that direction I sense. It has actually, in a sense been, to me the big challenge for many CTA strategies: do you stick with the fixed parameters that you've been researching that have stayed with you and that have worked in the past, or do you actually allow the models to adapt over time and maybe...? Peter: Well yeah, it depends on your lookback window. If your lookback window is, let's say 6 months only, or 1 year, then your models will adjust and change a lot every re-backtesting that you do. In our case, our lookback window is four or five years. So two months of new data should not generate a difference of parameters all together. It should produce a slight adjustment, slight modification to the new environment that is. Niels: Yeah, I wanted to just ask you one thing, just as a general industry question before we round up this short review. Clearly 2014 was a great year for systematic traders, for CTAs in particular. We know what happened last time that occurred which was in 2008, there was a big inflow of money into this space in 2009 it started, and Then there was a big outflow when investors either became bored or disappointed, or maybe they realized that they didn't know exactly what they bought. That's very destabilizing for the managers and the businesses as a whole. You've been around for a long time. How would you go about managing the situation so that we don't get a repeat of what I call the inflow/outflow or hot money, or whatever we call it that we saw only a few years ago? Peter: Well, first of all I'm quite happy and excited that CTAs are back in the game. There were people speaking about that. Investors are a lot more interested now-a-days then they were compared to a year ago, let's say. So that definitely is a positive development. Second, I think investors this time around will be a lot more careful from the standpoint of their allocations. I don't think it's smart to give all your money to similar CTAs that have 65%, 75% correlation to one another. I think this time around investors will have a lot more diversified CTA portfolios with some short term guys, with some mean reversion\ momentum guys, and it will be good for them and it will be good for the industry. Because it's unlikely that, that portfolio of diversified CTA's will have the same drawdown, as some of the lower term CTA's had in 2010, 2011, and Copyright 2015TopTradersUnplugged.com (CMC AG). All rights reserved. 8

9 Niels: Are you also going to be more critical, do you think, in the money that you will be offered from investors? Are you going to look at what kind of investors they are? Peter: Well, our job is to produce good returns - generate pure alpha returns and to offer investors the best liquidity terms possible, that's out job. We will not have any restrictions or holding periods. Basically, if we do our job well, we as a firm in Systematic Alpha and we as an industry of CTAs and managed futures, money will stay. So no, we will try to attract capital from all sorts of investors. If we get a long term investor of course, it's preferable but we would take basically anyone. Niels: Sure. I've only got one question left, really Peter, today. Before I ask you that, is there anything that you want to bring up that I haven't listed in my topics I wanted to cover today that you feel that would be beneficial for the listeners to be made aware of? Peter: Well, most investors know Systematic Alpha as the Market Neutral spread trading firm, and we are very proud of our core program. It has won four major CTA awards in the past. In 2015, we would like investors also to know that we are now a lot more diversified as a firm. We do have a very interesting Directional Momentum program that is doing extremely well. We would welcome any increase or interest, and we would gladly describe and explain why we think our directional models are different and why they performed so well over the last few years. Niels: Sure, that sounds good. Just my final question, Peter. That really is when you look into 2015, what would you wish for in the coming year if you had a magic wand? Peter: Well, typically the worse it is for the markets, the better it is for us and for many CTAs. So I wish the bull run is over for the equity markets. I wish for a lot of volatility and a lot of good healthy trading out there. Hopefully the governments around the world will intervene less and less into the workings of the markets and let the traders settle the score. Let them decide what is the true value of a particular market or position. I think that would be the best for everyone. Niels: Sure, I think that probably in an economic world, that would make a lot of sense. Now, unfortunately, as I mentioned it's a short episode so we're going to leave it here. For those who want to hear much more from Peter, please go and listen to our previous conversation on Top Traders Unplugged. Peter, before you go I want to thank you for being on the podcast again, for sharing your insights. I want to congratulate you on the launch of your new product which is interesting, and hopefully we can talk about that at a later stage. I want to wish you and your firm the very best for 2015, and I look forward to catching up during the year. Peter: Thank you very much, Niels. Also, I want to congratulate you on your success in your amazing production. Copyright 2015TopTradersUnplugged.com (CMC AG). All rights reserved. 9

10 Niels: All the best, Peter, take care. 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. Copyright 2015TopTradersUnplugged.com (CMC AG). All rights reserved. 10