The Systematic Investor #137 | feat. Mark Rzepczynski

you're about to join jerry parker maritz siebert and niels costrip larson on their raw and honest journey into the world of systematic investing and learn about the most dependable and consistent yet often overlooked investment strategy welcome to the systematic investor podcast series [Applause] [Music] welcome welcome back to this week's edition of the systematic investor series with mark residzinski and i niels castellassen where each week we take the pulse of the global markets through the lens of a rules-based investor now for those of you who are regular listeners our conversations are intended to motivate and inspire you to continue your rules-based investment journey and if you're new to the show we hope that today's episode will trigger your appetite to learn more by diving into the back catalogue and listen to all of the past episodes that you may have missed like last week's episode with jerry where we discussed whether we are right in using historical volatility for position sizing and how much of a portfolio should be allocated to trend following strategies just to name a few of the topics we covered but mark it's uh fantastic to be back with you this week how are you doing how things were you are in the u.s i'm living well in a k-shaped recovery actually this week i got my second vaccination so i'm uh i'm fully vaccinated and ready to do whatever i have to do to travel and under current restrictions but uh i've got my shots so i'm feeling good can i ask which one you got i got the pfizer okay some of the vaccines have been under some level of scrutiny as they have found that at least with the estracenica one apparently it develops its own disease that they've never seen before the country that i come from originally denmark have completely stopped using it which is quite interesting anyways this is certainly not a podcast about vaccines on the other hand we are going to talk about quite a few interesting topics i have to say i'm excited about today's conversation but before we dive into that i normally do my little market wrap and i do find that there are some interesting things and narratives happening in the market right now for example we've seen this relentless will and effort from the fed and other central banks to pop up asset prices especially the financial assets and that has gone hand in hand with the mantra don't fight the fed there's no alternative to buying stocks while we have interest rates that are being suppressed so to speak and therefore most investors don't fight the fed so to speak but there is another message coming from the fed that i think investors are maybe happily ignoring because the fed has recently at least in my mind been saying very loudly we're going to do everything in our power to create inflation if investors are going to really adopt the don't fight the fed mantra then shouldn't we take them at their word and therefore we should be biasing towards higher inflation which means financial assets may not be the place to be i don't know the answer of course i just think it's worth having an opinion about it and there's another interesting thing that happened this week while we saw treasury yields actually drift a little bit lower from the kind of sell-off we've seen in bond prices earlier this year we've seen a little bit of relief in the last few weeks and the thing that stood out was the optic in the reverse repo the rrp because at the march fomc meeting the committee announced that they would expand the rrp availability to 80 billion per counter party up from about previous 30 billion per counter party now the presumption at the time was that the fed wanted to expand it programmed to soak up so much of the excess liquidity in the system the rrp is essentially a de facto overnight security issued by the fed to select counterparties including banks and certain of the money market funds and prior to this week this facility was only doing a total volume around 20 to 30 billion a day but then on thursday this week the rrp surged to over a hundred billion dollars interest rates payable on the rrp is zero and so you may be wondering why investors would buy a security that didn't pay any interest i think the answer is that a money market fund would rather and zero percent and maintain its government only profile than by a negative yielding t bill in and of itself this optic in rrp is something that may not uh give us any clue of the of the fed's next action and of course it suggests that the fed has just flooded the system with the liquidity however i think it might be worth just keeping an eye on these volume numbers further clues in terms of that liquidity indeed now mark i always want to hear your 30 000 feet view of what may have caught your attention overall before we dive into the details and so on and so forth so since we were last on a few weeks ago about a month ago what what have you been noticing well the one thing i'd like to go back on your comment and reverse repo because i think that a lot of people when they think about trend followers or a lot of macro traders they think that we're looking at very high level issues and what you find out is that there's a lot of value in what i call the plumbing of the system this is that that when you think about market efficiency a an academic will always think about market efficiency is that can i come up with a grand theory that unifies everything and people who are involved in the markets have to constantly be involved in the plumbing of markets and if you're in futures markets you'd say what are the contract specifications what is that going to mean to how markets move even in a sense is it as a focus more on global macro i feel i spend more time in the plumbing than i do and and some of the macro issues because i say this is where all the activity and this is where crises are going to occur and uh i will say that and this will be a theme that maybe we could talk about a little bit later is this is that i wrote about something earlier in the month that is that i think it is i'm constantly facing i'm ex-anti-ignorant about a lot of events that have occurred and then x post i become very knowledgeable and what i mean about is that did i spend any time at all or prior to the gamestop issues thinking about gamestop and the answer is no i was ex-anti-ignorant when you think about our our friend in asia and his levered trades in prime brokerage i i think i understand the prime brokerage model but i probably would say that i was ex-anti-ignorant on exactly what was going on with this particular case or what are some of the problems associated and then after the fact i become very well educated quickly but it's almost too late it's so one of the big problems that you're always faced as an analyst and a manager this is that you read about some sort of explosion or some issue and you said gee i gotta now get up to speed on this issue i've got to do it really quickly the next thing you know you feverishly try to learn everything you can but then you find out you you you take a step back you say it's too late that horse has already left the barn and so how do i prepare myself for the events that i'm ignorant about that i got to deal with later on and you know what that is a great segue for one of the points that i wanted to talk about just after the kind of update on performance as such but another thing you mentioned two things here we could just talk about covet and how many people knew or talked about covet prior to january of last year and how many people know a lot about covet now that's how it always works right the question is it's almost like how as a investor do you battle this phenomena of if of moving from ignorance to knowledge but there's a time factor when you have this switch when you as you acquire knowledge there's a time of uh timing effect the timing effect is is that you're always late true and then of course assumes that you need knowledge to make a be a successful investor so i might come back on that point in in in just a second before i do so let me just give my quick run up in terms of performance on our side at don it was another solid week in terms of performance interestingly enough performance really came from only one area of the portfolio really and that was commodities and i think it serves as a good reminder what the c in cta stands for and you can even narrow it down further to the grange which had a really great week from a trend following point of view but there were a few other of the seas so to speak that did well currencies copper coffee cotton were a few other that perform well in our portfolio and elsewhere in the portfolio we really saw very small give backs in things like energies fixed income and equities but all in all a positive week strong month so far the trend barometer actually also saw an optic on my side although it finished the week at 41 which is kind of a neutral reading but it was up for the week and that in itself tends to suggest that the environment for trend following is improving in terms of our volatility strategy this week was also a profitable one and from my colleagues running the volatility strategy they pointed out a few interesting things such as that the momentum of the s p 500 is still pretty strong with 97 of the stocks trading above their 200-day moving average which is the highest level since 2009 historically and that tends to give one to three months positive returns but there are some signs of potential exhaustion in the index both the russell 2000 and the nasdaq 100 already had some corrections this year and there are several indicators like the increased margin speculation and some extreme sentiment readings that could be seen as warning signs i guess also equity indices are already above most sell side analysts and their price targets for the entire year and they are trading at p e ratios only seen back in 1999 2000 and finally another statistics i got from them and that is the past year seen more plus one percent gains versus losses than at any time before in the last two decades so maybe it's time to think about getting some exposure to trend following a volatility if some of these signs are to be taken seriously now for my own trend following model portfolio where i can of course go into somewhat more detail it was flat this week it leaves it up 3.58 for the month up 12.93 for the year this month so far the performance is coming from group one and two models so classical trend but also trend following quote unquote with a flavor of how a discretionary trader would trade they were up respectively 2.2 percent group 1 and 2.06 percent so far this month while the faster reacting group models are down about 0.68 percent are being whipped around a little bit in equities and bonds this month but not much in terms of sector attribution this month equities are doing best followed by base metals and then the grains and the worst sectors this month is really the currencies and the bonds in terms of single markets if we drill down there australian spy and aluminum are doing best with copper coming in in in the third place and at the bottom this month we see the euro we have the yen and we have the us 10 year in terms of trading activity the week started out with the system getting out of some short yen positions and also buying some mexican peso as well as palladium and the model also tried for some of its fast reacting models to go short the dax when it sold off during the week it also bought a little bit of grains and it also added some copper as those markets moved higher and in terms of the risk in the portfolio right now i use this risk to stop meaning if all positions got stopped out on monday it is expected then to lose thirteen point eight sorry thirteen point eight two percent yeah that is the number which is up about two percent from eleven point eight six percent last week and that's obviously partly because markets have been moving in in the trending direction so to speak so as long as the stops don't react as fast as the markets are moving then this risk open risk goes up a little bit and there were about 15 trades this week so average nothing extraordinary now mark i have been looking forward to discussing some of these points the first one actually is based on what's happened during this week so this is brand new interesting stuff that we can talk about but we do have some questions from raymond and dirk that we're gonna dive into a little later but what i'm talking about is really the news we got from a very large swedish systematic global microphone called ipm they decided to close shop and return all of their assets to all the remaining assets to their clients the aum in the last few years i don't mean many years two or three years max had dropped from about 8 billion to less than a billion this year and i wanted to first maybe ask you a little bit how you see the difference between systematic global macro and what we do as trend followers because i think that many people are not quite sure what the difference is and often prefer at least that's my experience they often prefer the systematic global macro because it seems easier to understand what are your what are your thoughts i think that when you describe systematic global macro you would say that there is an increase in the number of signals beyond price that will determine what will be by and self in the portfolio and in some senses that what you're doing is you're increasing the set of information that can be used by which to make uh investment decisions i think that there's a interest from investors for this for the simple reason is this is that it can tell better narratives and it makes them feel comfortable you say that if someone said i look at the global pmi trends if i look at what's happening to consumer sentiment surveys if i look at certain other macro information that gives them a sense of comfort that you're including all of the set of information by which to make your decisions and then you think about the trend follower generally and we'll say i'm very focused is that i'm i'm saying that all of information is going to be incorporated in price and those prices may have some lag to what may happen in the future there's an under reaction so consequently we'll say that the trend followers said i'm not going to look at a lot of this other information because all of the information is already embedded to some degree in price and that the information that i may be seeing that's fundamental could be lagged because it could it's what happened in the past it also is information that could be subject to revisions what i might see is that unemployment this month may be different than what i receive after they revise it next month and so there's a fundamental view on how you use information and what's the set of information that you should use by which to make decisions and i think that most of the investors will always say more is better than less so anyone who uses more information by definition has to be better than those who use less information now i'm not saying that's true but i say that's the assumption that people are going to be making yeah so a couple of thoughts from my side first of all i was very fortunate that early on in my quote-unquote podcasting career i actually interviewed the founder of ipm annas lindell so there is in the very very early part of the top traders unblocked catalog of podcast a two-part interview with with anas lindell so if people want to understand what the strategy was at least back then those would be two good episodes to spend a little bit of time it's very informative it's very interesting we talked about your ignorance ex-ante and your huge amount of knowledge exposed and this is interesting because systematic global macro i kind of understand why it's popular because if people say yes but our models are telling us that in inflation rises it has these consequences and people can say oh yeah yeah i've seen that before and and that makes total sense and so i think they can quickly get quite comfortable with the idea but i do think that some of the things that ipm cited or at least maybe the press cited when they were talking about their decision was the fact that they felt that markets actually weren't reacting as they should so to speak based on the economic data and we've all talked about this before we've seen it we saw the worst crisis in history in our history at least last year from a gdp point of view from an unemployment point of view etcetera etcetera yet the markets rocket to all-time highs at least in terms of the equities and and so on and so forth correlations broke down between gold and bonds and all that good stuff so to me and i don't know how to phrase it in an elegant way but i think one of the key differences and i think this is a hugely underappreciated attribute that trend following gives investors so if you on one hand you could say that systematic global macro they built their models to deal with risks and patterns that we've seen before in the economic data in this case for the most part but what trend following really does even though we use historical data i think we built our models to deal with surprise and this is why i said to you that maybe you don't need to know anything and still be a profitable investor because i think that's what we're trying to do by not being too clever about how markets should react uh but just saying okay if they react this way this is how we're gonna deal with that and therefore we don't really worry too much about what's gonna happen uh in the future so those are some of the things that that i've noticed but i have to say firms like ipm have been incredibly successful in attracting investors over here in europe and they were really the talk of the town and it was very difficult to persuade people that a pure price-based strategy that knows nothing about the future would be a better bet than a sophisticated economic model that analyzed all sorts of economic data there's this all views that old school versus new school and the new school that that uses uh new computing and data science techniques are going to be better than than the old school and it's unfortunate when any firm goes under and i could say that running firms is that it is really difficult because as you lose assets you face the the difficult task of understanding the economies of scale or diseconomies of scale of running a business and so you go from a billion to eight billion dollars is that you start staffing up you start building you know structure around around a eight billion dollar institution or you as a bigger institution you started losing assets this is that how do you deconstruct the firm how do you reduce head count how do you start to cut costs what do you do for for in a changing margin world it's it's difficult to deal with now i will say that one of the challenges whether it's a trend follower or a discretionary global macro or systematic global macro is that you're earning your fees because you're supposed to learn how to adapt and understand the changing markets in some sense people pay you to say that if markets change you're supposed to figure it out and that's why we're paying you a management incentive fee and and it it's a tall task but that's what you're that's what you're being paid for yeah it's funny if we digress a little bit from our normal conversation and just dive into the topic of managing these businesses i've run large cta businesses you have as well so we have some experience with that and i wonder a little bit they were closing their fund at 800 million dollars under management still now i know of course you're absolutely right that from 8 billion to less than a billion that's and we don't know the full story they could have had many more redemptions already knowing about it and say okay but still i it goes back a little bit to getting new talent into the industry also because it seems to me that there is this how should i put it change where suddenly you almost can't be a successful manager unless you have 500 million plus under management which i don't i really don't think that's true or necessary especially in a world where you have so much technology that can work for you especially when you have so much so many super external service providers that you can utilize as well and i think it's a shame if we end up in a situation where you have to manage almost a billion dollars for people to even want to do it and this is not to make it a biased conversation but as most people know it done we don't charge a management fee so whether we are at a billion or 2 billion or 500 million we need to make money to our clients and we've done that for 47 years so that's another pressure you get but it also shows i think confidence in what you do and the strategy and showing that you are willing to sit on the same side as the investor so i think personally it's a very fair way of doing business but then of course as i said you have other businesses that even if they're got one or two percent management fee unless they make 10 million dollars plus in fixed fees every year they don't even want to do it i think that's a shame now there's a the economics of running a firm is a very interesting topic because i say that given given for example it's not if you want to start just a small cta and you're doing friends and family office but if you want to be institutional ready the cost of running a cta to get started is fairly high it's much higher than a discretionary trader because a discretionary trader doesn't need as much quote unquote infrastructure so when you think about it the cost of starting us a small cta is very high then the economies of scale kick in and you get to half a billion two billion and you could actually run a lot of money and and the margins are very strong then you get to the point where now you start to see some liquidity constraints now you have to build more models now you have to add more infrastructure how do you deal with the fact that some markets start to be closed because they're less liquid all of a sudden the cost structure goes up and the impact on return becomes very high after you get to beyond a certain number and it could be 2 billion 3 billion but i think that it's for example if you look at a cta you can trade anywhere from 50 to 100 markets but i will tell you this is that if you have a meaningful allocation to something like coffee sugar cocoa and you're running a two billion dollar cta and then you want to sort of get good execution costs on that you'll be surprised at what happens to you when you start to try to get good execution for some of these markets especially in the commodity side it is really hard to reduce that cost and control the cost of execution and this is where and again this is just my personal opinion but i do think this is where unfortunately we do see perhaps a little bit too much greed not just in cta land but in in anything to do with managing money even in you could maybe talk about things like arc invest right now at whatever 50 60 billion on a management at some point finding the level where this is the best level of aum for my clients and saying okay this is where we we stop and we don't take any more assets i see so few firms doing that and i see so many people later see their aum drop like a stone because maybe performance it just starts to lag and and so on and so forth and aum is very often the cause whether people will admit it or not and you see this for example in the largest ctas this is that as you after you get to a certain point and i'm not going to say what that point is it could vary but i have an idea what that point a point is is that you're going to have to start dropping a lot of commodity markets so you can't trade 50 liquid commodity futures markets if you're a multi-billion dollar cta so by nature you're going to have to now start to focus in on financials and right away that starts limiting your diversification it limits what is the set of choices you have even for trend following and then what happens is that you really got to be very good at getting those those decisions and i think that the i want to say it's a dirty secret i think everybody will talk about this is that if you go an annual performance for a lot of ctas is that they may be trading let's say 50 markets but we'll say the majority of your returns are probably going to uh come from less than 10 probably about five so there are five markets gonna make you a lot of money there may be another set that are just a little bit and then there's something that you're going to scratch and you're gonna and then they're another set you're gonna lose on but i think that when people look at the return attribution for a cta what makes them sometimes nervous is the fact that they'll say whoa you made all your money in that year from crude oil trading and you say yeah that was a great trend he said but i thought i bought this diversified portfolio you say no i you bought a diversified set of opportunities of which crude oil became the best one and that's where we generated most of our return and i i think that's these are little things that you know and you got to get under the hood you start finding out about how ctas operate it's neither good nor bad it just it's just the way the the the markets behave and it's the way the return profiles of of the strategy behaves and i think that's a good point and of course we often hear jerry talks about the this example i think is from 1991 or 1990 i can't remember where he made 30 in i'm just saying unleaded gas or something like that i can't remember which market something like one of the energies and he made it all in december but that was also the whole return for the year for the whole portfolio so you're exactly right we have to talk about diversified opportunities not necessarily that the returns themselves will be diversified from many sources they don't have to and it kind of ties in quite nicely you said something early which unfortunately i forgot that fits nicely into the next topic that that i wanted to touch on this was prompted by an article i was sent by a good friend in the industry i'm not gonna i'm not gonna name any names here but it's important for me to stress at the beginning that i'm using this article as an example i have actually only a lot of admiration for the firm they've been around since 1971 so for 50 years in our industry that requires a lot of discipline and success and it commands a lot of respect there's nothing negative in what i'm about to bring up as a point but i do think the topic is interesting because it is about this change of regime and whether the old style of trend following still works we know that's often up for debate or whether we need to move on to some new types of techniques in this particular case it's machine learning and the firm i'm talking about is milburn middleburn were in some articles a couple of years ago i think in these exactly two years ago in the hedge fund journal where they talked about their journey from being the classical trend follower they started three years prior to to don uh did so they're probably maybe the oldest still in business uh or maybe they're not really a cta anymore i don't know but they talk about their journey where they started moving towards quote unquote statistical learning based framework let's call it machine learning away from the classical trend following and i think and i might be wrong a little bit about the timing of all of this so i apologize in advance but i do think that a couple of years ago they it said it says here over the last day so this is the article from march 2019 over the last decade it became clear to us that the explosion and data had the potential to have a major impact on understanding markets this meant simple rules based approaches to using momentum were perhaps not going to work as well going forward sounds very similar and i don't really want to put words in his mouth but it sounds a little bit similar to what david harding did around the same time moving away from trend following in a sense and then i think they went on to in february of of of that year maybe or there around yeah maybe in the month before the article they talked about as of february it had phased out traditional trend following entirely in favor of its contextual statistical learning framework work across the main long short diversified multi-market and commodity programs completing the firm's transition from a rural space to a pure data-driven approach so that's the context that's the framework so they obviously were early adopters of trend following they were early adopters of more statistical based approach and let's just call it machine learning to to keep it simple and then last year there was an article actually i think this is from their own website so it's probably a paper they wrote where they talk very openly and i think this is the beauty of what they've done is they've been very open about this journey and and their lessons and i think this is what i wanted to talk to you about mark because i do think you've brought this up and then you think you've actually worked with kind of machine learning artificial intelligence so this is why i thought it would be interesting for our conversation today but they talked about what they learned from using machine learning strategies in march of 2020 and so they write we run a number of programs at the firm but the common thread is that they that the active risk taking the signals that we generate that tells us whether to go long or short or whether to take more opportunistic or defensive stance in the market are all driven 100 by machine learning technology that's really just to confirm what i said earlier and then they go on to say this unfortunately didn't help those investors who were looking to ask for to potentially provide some relief from losses and they were seeing elsewhere in the portfolios we felt for those losses too as substantial investors in our programs ourselves so clearly what they're setting up here is the fact that march the february march period last year turned out to be incredibly difficult for them despite having switched from to these machine learning strategies and i'm just finding a few paragraphs here that that just to give you a little bit more context mark before you give me your thoughts they write they if you follow the equity markets closely as the pandemic accelerated we started to see some very unusual behavior in many actually almost all of the equity markets that we trade as an example at one point in march we saw the s p drop more than 28 percent in a matter of 13 trading days this was something truly without precedent using the s p as an example again we also saw materials sustained volatility with daily price moves exceeding 10 percent both up and down again this was behavior we had never seen before so again going back to this are we building models based on things we've seen before maybe again systematic global macro does the same i don't know or are we building models like i think trend following really are we're building them for surprise for constant change i think again this is an interesting thing they also talk about and i'm just quoting from this paper the accuracy of the models forecast in the equity sector was certainly worse than we had hoped okay that's fair and maybe we need to not go into too much further details about what happened but i think the whole point is that they and their models were surprised i think they intervened manually i think that somewhere in the article they write that they had to intervene manually because again they felt that this was an environment that had not been seen before by their model so given your experience and kind of tying in what we heard earlier about systematic global macro now it's machine learning what are your thoughts what do you think what do you think investors should take away from some of these from some of this evidence i guess we could say it is playing out in front of our eyes i can start with that the systematic space is a big tent there's a lot of room for for interpretation and differences of opinion on how you can make money and i probably am a little bit more of an agnostic and say this is that i think that if you think you can be able to use some of some tools and you could create an edge and you uh feel comfortable with the relationships that you found that's great and more power to you probably jerry if he was on this ask the same question he'd be his hair would be on fire if with the philosophical differences and and i think that in itself is shows that there's a wide tent in the systematic space this is a couple things that always jump out is that from a from running a firm i think that the important point is to have transparency so if you make a transition or you do things that changes your model and approach as a systematic manager are you are you disclosing that to your clients are you being able to tell them what you're doing and where what is your philosophy and i think from the transparency when i use the word philosophy is very important so a philosophy for markets that says can you be able to create an edge do you believe that markets are efficient or inefficient how inefficient and do you believe that there are relationships between fundamental factors or other factors and returns on a go forward basis and i think that doing due diligence on managers which i've done at some point when i was running a fund of funds and even for some of the work that i've done in an advisory basis is that you always start with the idea what is the philosophy that the manager has on how markets operate and given that philosophy now how do you use the tools and you could say machine learning is a tool how do you use the tools that are available to exploit the philosophy or the view that you have towards market behavior and the question comes in is always is that first are you comfortable with their philosophy second are you comfortable with the tools that they plan on using and third can you be able to appreciate and can they be able to convince you is that that the use of their tools can gen and given their philosophy can then be converted into returns and in some senses is that philosophically it means is that they say that those who use a lot of machine learning other techniques says there are relationships that have existed in the past that will continue to exist in the future what they're saying is that those relationships may be hard to find that those relationships are maybe what we call non-linear so non-linear in the sentence is that they may need conditions by which d you would they would that you can exploit them but they are repeatable they're accountable and you can be able to then exploit them in the future if you believe that then you're gonna you you can use machine learning and it makes perfect sense if you have a philosophy that says that there's a tremendous amount of uncertainty in the world and uncertainty means is that relationships change and in some sense those relationships because they change i will get false signals if i look at the past and try to use those those relationships in the past to try to exploit something in future and and i think that for example is this is that a classic trend follower would say is is that well given the uncertainty across all these relationships i have to only use a primal number of relationships and the primal number is that what happens to prices in the past is likely to continue into the future it may not be clear exactly how long but i could say i can use or i can exploit price relationships to be able to find trends and in some sense the trend follower is a passive investor in a sense is that a trend follower would say i try to find trends and then i will make returns if the trends actually exist so that doesn't mean that they're passive and they're not doing anything because they're trading they're following their models they're looking at these rel trends but they're saying if there are no trends i will not make money if i'm a trend follower the machine learner or the we'll call it the systematic global macro person says is that i'm not a passive investor i will go out and find relationships and i'm going to when you think in an active sense i'm going to exploit relationships i'm an activist in the market i find new data i manipulate data and then i exploit data so it's a very active engagement with data and relationships between markets yeah it's interesting to me and of course i'm incredibly biased in this discussion but i find it interesting that still in my mind so relative few investors have a meaningful allocation a lot most investors have no allocation but there are some and they have a relatively small one but there's still very few who have a meaningful allocation to trend following but when i think about it and that is when we look back and we look back at history what what is history history is usually an account of surprise we don't put mundane events into history books we put all the big surprises and events into history books and so you would think having a strategy that's designed to cope with the next surprise makes sense and i think maybe we still have to do a better job in explaining why people should have a meaningful allocation to to trend in their portfolio or to surprise so to speak you know that i'm always good about circling back to what we originally talked about so what we talked about our ex-ante ignorance and exposed knowledge is that i think that the difference between risk and uncertainty is no ability okay risk is when i can be able to understand all the different choices i may not know what will happen in the future but i can understand all the relationships and say there are relationships that exist there's probabilities for different events occurring i could look at all the different cases i don't know what case will happen in the future but i can understand all the different choices or alternatives or scenarios uncertainty is that we don't know what is possible okay and if you believe that we live in a very uncertain world then you are going to go down one path which is to say i gotta i in some cases want to use some more simplistic views if i believe that i i have a large degree of no ability and i can then condense this into a risk problem then i could say is that i want to use machine learning or i want to use more fundamental information and a lot of this has to do is that you look at naturalistic decision making and this is from the psychologist gary klein and then the head of the max planck institute in berlin is the professor geigenger they talk about heuristics now in general is that when we talk about kingdom and and behavioral finances is that the use of heuristics has always been viewed as a that'll be make that's irrational that people who use rules of thumb are ignorant that but in reality is this is that using heuristics is the way we deal with a lot of different uncertainty and when you think about trend following is a form of heuristics it's a rule of thumb that say that when i'm faced with a lot of uncertainty what do i do i fall back upon simple rules and the simple rule is that if the prices are going up they probably may continue to go up if prices are going down then they're probably going to continue to go down the hyper rationalists would say who don't believe that there's a lot of uncertainty and believe that the world is knowable and it can be assessed as risk would say that's a that's an irrational that's a behavioral bias that you shouldn't use for those who are saying that in a real world context which are faced a changing uncertain world that this is really important and gary klein often talks about what is called uh recognition prime decision making there's a recognition prime decision making is is that and he studied this with for example a battalion chief an on a fire brigade they see a fire now every fire has some similarities but the every time every fire is always different now he has to determine am i going to send people into the burning building and is it going to crash down on them or they're going to be able to put the fire out from going inside this is that so what he's going to look for is going to be use his he's going to be looking for key signs or rules of thumb by which to make that decision and he has to make that in a split second he has to make it under intense stress and then some of those is that can you be able to do an optimization under those conditions meaning that i can figure out all the different probabilistic choices weigh them and then use an optimizer to find out what i should do and i'm not against optimization okay and i think optimization is very powerful when you have enough time and you have enough information but under you know stressful conditions using a recognition prime decision model is better using rules of thumb may be better when you face maximum uncertainty and so a trend follower saying this is that when i face a high level of uncertainty and markets are constantly changing which they are because it's there's a behavioral component the components of who trades is constantly different then i'm going to i'm gonna i'm gonna have to be using these rules of thumb and we find this already i think we were talking a little bit about machine learning or you mentioned machine learning this is that it's if i'm doing facial recognition of trying to say can i be able to take photos and be able to find out of thousands of photo what is a cat or a dog you could say that the features of a cat or dog may be different but i could say there are sort of characteristics of what it means to be dog or doggedness and i could then be able to learn from different photos what it means to be a dog and i can then be able to reinforce learning because i could say i get penalized if i find a cat when i'm supposed to look for a dog like i can use that technology now what happens if let's say the features of a dog are constantly changing so that the dog today is different than the dog yesterday that becomes a much more difficult problem i'm not saying it can't be now it can't be solved it just takes more work and at the same time as is that you're going to get more error and that's what you find with some of these firms some of them have been done very well and i think that they've exploited different opportunities in the marketplace but at the same time there's also a place for those who are more recognition primed rules based heuristic based investors such as trend followers yeah now i go way off off track here in in my own little mind but you can even make the same argument you don't even have to go for a different kind of dog or different kind of a cat i think it's well known that in the last few years facial recognition is something that is used heavily in for example airports so we we feel safe because all these cameras they will recognize our face and they will alert someone if they see someone that they don't like so to speak on some kind of watch list but suddenly what changes is the fact that suddenly we have a pandemic and we always wear face masks and actually then face recognition isn't very effective when everyone is wearing a face mask legally in the airport so circumstances can change and i think that's the point that i do think people just need to recognize because you could have a simple rule that if someone in an airport is running around with a machine gun then that should alert you that's that rule doesn't change right you don't need face recognition for that purpose but maybe i should stop here before i go completely off track the point i'm saying is that i still think the strength of trend following lies in the fact that we are not trying to pretend that we know what the next event or the next prize move is going to be and they it doesn't have to be something that's happened before as we saw last year because things will always change and there's always going to be a new crisis right now now when you think about this that most people who have been trained in science have read the the work of thomas kuhn on scientific revolutions and what what you find out is that it's not science is not very linear that it actually has big jumps is that when conventional wisdom is that you found that the data no longer supports or you get alternative you have failure of existing theories with empirical data then there's usually a paradigm shift and i think that before you get those paradigm shift shifts what you'll have is this situation is that that we've had in the post gfc period this is that trend following did well during the great financial crisis 2008 2009 who came out of that crisis is that something did well but then you say in the 2010-15 in that that post gfc period trend following didn't do as well as as it historically is done there have been other periods of failure for trend following but this was not a very good period and when you think about it this is that it's natural that then people would say i need to rethink my philosophy i need to say are there other techniques that i need to do should i switch my view of the world and my fl my underpinnings of how i look at my model some firms did that and and now they're either successful or they're finding out if the world moves back to more normal that this is that then it's going to be more difficult others have taken a view is is that i've seen that there's booms and busts in in momentum trading trend following and my job is to say how do i use my risk management to ensure i can ride through this period of time until i get to the other side and in some senses that you can see that some are saying is that i gotta live with this because i know that over a century trend following is going to do very well okay so i just got to batten down the hatches try to make sure that they ensure that i stay in business and keep moving forward others say i want to make that switch i'm going to go into a different change my philosophy and use different techniques and we'll say that the marketplace is is interesting because they could say we'll reward those who are successful regardless of what is the transition or what is the philosophical change towards markets yeah no absolutely [Music] now mark we're already almost at the one hour mark and we haven't even started on any of the topics that you uh kind kindly send over so why don't we just if you look at your list and why don't you pick a couple of those that you like that you thought would be really good and we'll keep the other ones for another time we also have a couple of questions uh that i would like to get to so when you look at your list of topics what what stands out to you right now what are the things you think we should be talking about today it's interesting and this all goes back to uh and just a quick anecdote when we could probably come back to this is this is that uh i was going back over some research and about 40 years ago michael jensen who is one of the real top people in finance in in finance for the you know 20th century said we're done market efficiency has been one of the most studied areas we've seen and this is almost settled science we know markets are efficient we don't have to worry about this topic anymore let's move forward there's nothing more to see there's nothing more to see do here and i'm looking at this 40 years later and i look at first trend followers still exist in been successful uh successful behavioral finances come into existence we've talked about the the limits to arbitrage which offers opportunities and inefficiencies this is that surprisingly is it what some people viewed as settled science 40 years ago where we've been able to get new information we've come up with new theories that we found out is is that in some senses is that there are people who were trend followers back then nurse and some are still in business and so i think that if you look at i think you've your friends had done is that they were fighting the same battle about whether they could make money in an efficient market 40 years ago and they're still probably having that same discussion yet they're still here 40 years later yeah it's funny and we we as you say we've been around for 47 years and our current strategy at least in its kind of form has been around for 37 years so we have a pretty long track record through you could say different market regimes and one of the statistics that i like to look at is actually how has the performance evolved over three different periods of time since inception in 84 of the strategy since we made the first major upgrade in 2006 and when we made the second major upgrade to the model in 2013 and what's really interesting and i think people would be surprised to know because they have this notion about recent performance but what is and i have to be a little bit careful for regulatory purposes about talking about specific numbers so what i will say all three periods double digit returns net of all fees but more interestingly the longest period of time you would expect that to be the highest annualized average return and it is but what you also have to account for there is that up until 2007 at least at our firm we included interest income like most people still do into the composite performance we have to prepare so there was some of that up until 2007 since 07 we just decided that's really not any return we generate so let's not include it in our track record of course the clients get it in their accounts but that's different so of course you would expect the returns to be a little bit different but actually the returns and they are they're a little bit lower a couple of percent low on average since 2000 so between the long period and since 2006 and the period from 2006 and the period from 2013 till now are almost identical and i think people would be surprised about that actually and and of course you can always find rolling three-year five-year periods that were lower or higher of course you can but over that period of time so over around 15-year period and over around an eight-year period those two periods as an annual average return are almost identical so you could also argue the other way around and that is as you said yeah of course we're still here trying to find out ways of dealing with all of this and efficiency and so on and so forth but actually trend following still can still find ways of dealing with this of course you have to adapt and you have to evolve and that's what these two major upgrades reflect but it certainly doesn't suggest to us that trend following is less efficient in extracting profits from from the market and this gets back to a whole other topic which you could always save in your our topic to do to-do list it's getting longer by the time by the minute is that when you buy a trend follower you're asking asking for something that's positively skewed you want to have a positively skewed return and in fact that's what a trend follower is trying to create is that you say you're long straddles but in some senses that you're making money when market you know moves to extremes divergent trading is as i call it and the problem with divergent trading or positively skewed return profiles is that you don't know when those positive skew events are going to occur you hope that they occur when you know other markets are falling so that you have this negative correlation or crisis alpha but you don't know for sure but the problem comes in is that over you look over a longer time period returns could look very good at the same time is is that if you as you look with your telescope or your microscope and you say you go from let's say a 10-year period to a five-year period to three two year period which you find out is this is that as you look into shorter periods of time the performance of trend followers usually do worse as you go in as you magnify into shorter and shorter time periods to the point where it ends because you say look you could have a five year period and you have one year that's has a huge return and others are just modest or might be negative and and in some sense the battle that is often faced both for the investor who chooses to trade like a trend follower and the investor who invests in other firms is that that you don't like positive skew even though they say they they may want it they don't really like it this is it you actually want to have something that is is that makes your turn and then when you have that negative skew event when it does poorly well on some sense human nature tells us this is that if that negative skew event happens when all the other managers are doing poorly you can shrug your shoulders and say everybody got slammed in march of 2002 what can you do no one was able to miss that and so people actually have this they say they don't but they have the preference for a negative skewed type of return profile and what you could talk to a trend follower you get the opposite return profile where you say okay it's possible this is it i'm just doing okay okay negative year of this and then all of a sudden you get this huge outsize return and what's maddening for investors is that okay if i get that outside return because there are a lot of mean return chasers that they say well now i got to put my money into a trend follower and now i'm going to go back into a period where i might be quote unquote more dormant so it's it trend following requires you to throw out a lot of conventional wisdom so that and you have to be able to accept a different view of how markets behave and return profiles for managers actually exist yeah with all investing certainly it's not just a trend following track records but with all investments unlike the weather forecast where you will look at the weather today and and we can be pretty good at forecasting what the weather is going to be tomorrow but if we ask if we're asked what's the weather going to be in 10 years we have no idea but it's actually the opposite with track records or annualized performances where you could say we have no idea what the performance going to be tomorrow or the next year but once you get out kind of 10 15 20 years you have a pretty good idea of what the annualized return is likely to look like so i think that's quite important the other thing i just want to bring up it reminds me of a conversation i had this week with a with a potential investor and we were discussing about the kind of this maybe there's a need for more protection since a lot of markets are a little bit stretched and interest rates are going up and so and so forth and they were saying that yeah they had a little bit of volatility in their portfolio as kind of protection and they're considering the time when they would need more and i'm just thinking it's very hard and obviously the true answer is it's impossible to know when you need more when you need it as you rightly said because these strategies are what people say they want but they don't really like them they tend to try and time when they need this kind of these strategies in the portfolio and i'm not so sure that's a great approach to this as frustrating as it can be to have it all the time i think it's a better approach than trying to time it at least we've never found a way to do it i i once wrote a comical piece is that you remember that movie uh a few good men but tom so they had jack nicholson playing colonel jessup and they said like you'd need me on that wall you however repugnant you think i am you you want me there you need me there and i sort of the comical part was you say that manage futures and trend following is the colonel jessup of your portfolio you you'd need me there you want me there this is a you find me repugnant in my recognition heuristics but at the end of the day as i said if there's a if there's a massive divergent event you'll be happy you had me the funny part is mark the funny part of all of this is that for those who fully embrace the concept make a meaningful allocation to say trend following or it could be a long short volatility strategy that is also an absolute return strategy right it actually allows the investors to take more risk in their portfolio allows them to own more equities in their portfolio really and i think this is often missed that it gives them a lot of freedom elsewhere in the portfolio to know that they have this in there we had this deep discussion with a a former employer which i'll remain nameless and he said look we get our volatile trend following program he said i need to have a certain amount of my portfolio and treasury bills because i want to offset you know some of my risk and i was arguing and said no you want to do just the opposite is that if you have this risky trend following strategy that you know is going to do well if stocks go down in or or and you really believe that this is that we should say let's put all the chips on the table and put more equity exposure exposures as i said given and the more volatile the the trend follower the more risk you could be able to take because for a dollar of invested in trend following is going to give you that much more diversification benefit so you'll turn around and put in more in your risk the risk side of the equation and this is why the topic of the 6040 portfolio is interesting right now because and i was watching a webinar yesterday about the point and there are some interesting points made but to me at least this issue where we've seen some very few but pension funds embracing risk mitigation and creating risk mitigation components comprising of long-dated treasuries and trend following but given the fact that now treasuries have been yielding such a low return and in fact if you were long in the last nine months you would have lost 17 percent also so quite a nasty sell-off in long-dated treasuries but you are absolutely right that having a large exposure to trend following does allow you to take more risk quote-unquote equity risk elsewhere in the portfolio and i think people completely miss that point but hopefully this is something that people will start to reconsider mark again just being conscious of time is there another point you want to uh or you want to mention today or do you want to go into a couple of questions let's go into some questions this is that we seem like we've had this stream of consciousness we don't need to go to other topics we just just get started and bam we can use up an hour pretty quickly so let's go i know let's go to some questions okay okay cool so this is what i really love about the podcast so this question actually comes all the way from china so i love this i love the fact that raymond who is based in shanghai writes in with his question of course the question was inspired by the conversation i had a couple of weeks ago with moritz about chinese commodities raymond asks i'm a big oh he writes i should say i'm a big fan of your podcast i've learned a lot from you and the other guests very glad that last week you talked about trading chinese commodities as far as i know winston and aspect and i'm sure there are more by the way have been trading and operating in china for many years and did much better than their and i was think the quote-unquote previous cta programs meaning those without the chinese commodities he makes some reference to aspects returns with or without these commodities because i think they published a piece they must have been from where he got the data and he writes so i want to know if you can share some particular methods or criteria for select for market selection other than just trade all the markets available based on liquidity and market participation so from your time if you put on your sort of cta hat mark what were some of the things that you liked or your research team like to look for in in adding markets and i will also say that i think a lot of firms have different views on on on this we at our shop we have not embraced this philosophy of trading as many markets as possible for sure so we fall in one category but i know and i think raymond is right then more and more managers have decided to add significantly more markets and therefore also embracing things like chinese trading more and more markets by itself is always problematic because that means is that the average exposure that you have at any one market declines so by definition you're going to dilute those successful markets and when we talked before is that you find out that most of your exposure comes from just a few markets in any given year if you increase the number of markets and that means you could have those four you could have said i exploited all of those markets that had great trends i just didn't have enough of it so that's the pr that's your worst fear is is that i got everything that were the big moves right i just didn't have enough of it so when you say do i want to trade chinese markets futures markets what i'm looking for is uniqueness and difference in in some senses that you can exploit opportunities for example in copper or gold but that becomes a 24-hour market and so what you're really not being able to get any uniqueness or anything special not unless you look at the very localized short-term trading systems on a term basis is that the gold market is going to be affected by london new york and let's say maybe shanghai so adding chinese gold futures is not going to add a lot of value what you do find is that there's a whole set of other markets which are unique and different that aren't traded in other parts of the world and those are the ones that i would always focus on because by by definition their return pattern can't be exploited in the other 18 hours of markets and second of all is is that it may offer unique diversification opportunities yeah no i completely agree with all of that so i hope raymond that that is helpful the last question today is from dirk he writes in you probably discussed this on your podcast but listening to must be last week's episode number 136 of the systematic investor i was wondering did you back test vol position sizing against simple one over n allocation cures if you found significant differences in portfolio of all and returns now you heard you mentioned earlier you listened to jerry and me last week of course one of the key discussions we had or conversations we had was from a great question we received and that is have we completely misunderstood how to use volatility inside the cta model meaning should we use historical volatility as we do now or is it some kind of prediction when we do that and therefore we are a little bit not living up to the full philosophy of trend following where we don't know what's going to happen so to speak and i think this comes back to just just sorry this comes back to the fact that then the other discussion that we always end up having is this the continuously volatility adjuster type methodology versus what jerry does what maura does what i do in my trend following model portfolio where we just take the risk based on the entry point we size our position and then we don't change that position until we exit right and this is a very complex issue and i think that i think researchers who have shown that volatility adjustment does well for for trading they often i think they miss some of the we'll call it the plumbing of how this all works and some of the issues that are are based with doing the adjustment because what you could have is a situation is that you have a good trend and if volatility goes up and you're going to rebalance on volatility you could actually be taking money away from good profit making opportunities for the simple reason is because the volatility is going up but you had good volatility you have deviations above the mean which you're trying to exploit and at the same time you're taking money off the table from the big problem i always found with volatility trading is that especially if you have crises where you have a spike in volatility i have low volatility and if i'm a volatility adjustment means i'm going to have high positions in low volt markets okay then i have a volatility spike or a volatility event and i'm not saying it's a trend event but of all a volatility event and then what happens is they could say now i'm going to sort of have to cut all of my positions and usually if let's say that you have the market going up i can have a situation where i'm cutting my positions on on on a trend that's going higher so i'm cutting my profit opportunity then all of a sudden this is that now i'm at the top of the volatility market starts to reverse i'm going to take small positions if i was on the wrong side of that trade what happens is that i'm not going to make back my money because at the high volatility event i'm now going to have all of my positions smaller so you could see this happening if for those people who are wrong-footed in march of 2020 what happened is is that they cut all their positions now i'm in a high volatility period i have a great opportunity to see a reversal and stocks going higher at the end of march but i'm going to be taking all small positions because i'm in a low volatility world and i can't make back the losses i had that occurred earlier in march if i was wrong-footed going into that crisis yeah not to give you too much pushback because i don't disagree per se but having lived through march last year and working with a product that actually has some level of adjustment within frameworks of of risk and therefore where volatility does play a role it could also have worked out reasonably well for you at least during parts of the crisis because what happened really was that a lot of the risk was caught as you rightly said in the early part of the crisis when volatility exploded but that also meant that if you just take a simple sector like equities where most trend followers came in completely long given the returns or the moves in in equity markets at the prior period so as the market sold off positions were caught but when the signals changed too short volatility had increased so much that position size were actually relatively small they were short but that but not so big and so when the v-shape recovery occurred we didn't suffer so much because position sizes were small but in fairness and i think that's the point you make very well it also meant that the recovery the subsequent recovery once signals turned back into the where the trend originally had come from inequities to the long side yes then you're right then position size were small and therefore it was harder to make back any meaningful profits in many of these positions so i don't know whether you can say evened out but it certainly fall from an investor point of view those who did that it certainly meant that volatility in the performance was not an issue there was no sleepless nights with because of massive volatility and returns it was actually very well managed and so to me last year has always stood out as a year of risk management that's really what i think we got stress tested in the real world because as we talked about earlier when we talked about machine learning no one had seen this environment in their historical data which of course we use to build these models going back to i mean the question was in reference to your last podcast with jerry and jerry said i've gone back and forth on this issue i could say i've seen both sides of this issue i probably would say that i do vol adjustments and i think fall adjustments is an important part of the model and looking at correlation because that tells me something about what the contribution of each market will be towards risk that those are important components that add to what i call it portfolio craftsmanship so you can be a trend follower and that's your core philosophy now how you size positions how do you account for volatility how you account for correlation how do you rebalance portfolios of markets those are all a part of what we call the craftsmanship of portfolio construction and there's a lot of room for value added for portfolio craftsmanship beyond trend following and it's i think that's the important part of what you pay for a trend follower or any systematic manager this craftsmanship issue so it shouldn't be taken lightly and at the same time this is that you say you do have to spend a lot of time with that i could say two people we could have the exact same trend following view and if we differ on how our we construct our portfolio and our craftsmanship we will get widely different returns and i think from a due diligence perspective when investors look at trend followers and systematic managers and i'm not saying that this all i'll make a generalization they spent a lot of time on the philosophy okay you're a trend follower tell me how you find trends and they may spend less time on the craftsmanship of how the portfolio is built when in some sense i could say that might be the defining difference in characteristics of the manager i think that's very great that you brought that up and a really important point i'm sure we're going to get back to it maybe one day you and jerry will come on at the same time and now that he's had the discussion with rob maybe it's time for him to have the discussion with you for or against volatility adjustment in the usual gentleman-like style that we have here on the podcast anyways let's leave the questions like that then we have some more for next time of course in terms of topics and i'm sure we'll come up with some new ones i want to just as usual just let everyone know that performance so far this month beat up 50 up 1.33 in april as of thursday night up 4.04 percent for the year such a nct index up one and a half percent up 4.11 for the year socked in trend little bit lower returns up 0.85 percent for the month maybe more in line with my trend barometer actually and still up four point eight six percent for the uh year and the short-term traders index interesting enough is down 41 basis points as of thursday in april and up 1.56 for the month as i mentioned the trend barometer is 41 at the moment you can check that out every single day on the website and equity is still doing well msci world up four point seven eight percent as of last night up nine and a half percent year-to-date mark any particular other than your your own great blog post that you write that people should go and check out anything else that you've seen in terms of interesting content convert context conversation anything you've so well noticed yeah i think that it's just amazing the amount of good writing that you see on on blogs in general and and i think that on a lot of these topics that i'm not going to go into specifics but i guess i'd say is this is that the writing and reporting and especially some of these events that we've seen either spax bitcoin uh gamestop prime brokerage and inflation is that when i always talk about this exposed knowledge i'm amazed at how quickly if you can use different sources to get that wisdom quickly and and on a lot of unusual events so i'd say that to keep up your avid reading of twitter and other blogs because i think there's a lot of good information out from experts yeah no absolutely there is indeed and i am myself a big user of podcast other people's podcast so i completely agree with you there i probably listen more than i read post simply because i can listen when i'm doing other things like walking the dog which my wife likes me to do so yeah no i think there's an amazing amount of great content out there on that note we're going to wrap up this week's conversation if you thought we did a good job in providing you with some content that you liked why not support us by going over to itunes and leave a rating and review so more people can find the podcast we would greatly appreciate that next week i'm joined by rob as usual that's going to be a super fun and educational conversation so make sure you keep your questions coming in you can email them to me at info toptradersunplug.com and we'll do our best to get them on and you can of course follow mark follow me on twitter linkedin and where else we appear from mark and me thanks so much for listening and we look forward to being back with you next week until such time take care of yourself and take care of each other thanks for listening to the systematic investor podcast series if you enjoy this series go on over to itunes and leave an honest rating and review and be sure to listen to all the other episodes from top traders unplugged if you have questions about systematic investing send us an email with the word question in the subject line to info toptradersunplug.com and we'll try to get it on the show and remember all the discussion that we have about investment performance is about the past and past performance does not guarantee or even infer anything about future performance also understand that there's a significant risk of financial loss with all investment strategies and you need to request and understand the specific risks from the investment manager about their products before you make investment decisions thanks for spending some of your valuable time with us and we'll see you on the next episode of the systematic [Music] investor you

Channel: Top Traders Unplugged

Views: 362

Rating: 5 out of 5

Keywords: niels kaastrup larsen, systematic investor, top traders unplugged, investing, hedgefund, trading, trend following, managed futures, risk management, popular traders, investing money, investing for beginners, trading news, investment news, investment advice, trend following updates, news for traders, top investors, investing podcast, Mark Rzepczynski, Systematic Global Macro, classical, modern, Trend Strength, Dunn Capital, AUM, ARK, position-sizing, volatility

Id: AMRp9L5Sl68

Length: 81min 24sec (4884 seconds)

Published: Tue Apr 27 2021