The Systematic Investor #156 | feat. Moritz Seibert

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 [Music] welcome and welcome back to this week's edition of the systematic investor series with mort c but and i niels castrolassen where each week we take the pulse of the global market through the lens of a rules-based investor if you're new to the show let me start by saying welcome with the hope that today's episode will trigger your curiosity about systematic trend following as an investment strategy enough to check out the back catalogue and listen to the past episodes that you may have missed like last week's episode with richard where we discussed all things back testing and forward testing and everything in between i really think it was something that a lot of listeners think is important so if you missed it i invite you to go back and check it out let me also give a quick shout out to those of you who spent a few minutes this week leaving a rating and review i noticed some great ones from joey yaz hypnosis 8 and ernst mango omego just to name a few you should know that i do read all of them and i post many of them on the website and we're very grateful for the help that you give us in order to grow the podcast moritz you're back from holiday how are you yeah back from italy since i don't know 12 18 hours or something like that so got back last night just been great to be out of the country we've had a uh a very wet and cold summer here in bavaria um not sure if that's also true for for zuk where you live nils but it's been relatively cold here so it's been good to get to get away and uh enjoy some sunshine and warm weather in italy we were in rome it was 38 degrees centigrade there so that's actually too much but you know around tuscany florence it's just beautiful area good food nice people good beaches kids loved it so all good all good yeah i i went away for the summer which i normally do i will i go back north to to denmark my birth country and we had actually a pretty good summer but i will say that lots of our swiss friends when i came back have told me that it's probably been the worst summer they can imagine and actually just like in germany luckily not as bad as as in the eastern part of germany but or western part of germany sorry but there were certainly some issues with flooding and all of that as well so yes certainly an extreme weather-wise summer but i hope you managed to recharge your batteries more it's because we've got lots of great things to talk about today if i was going to do a quick market summary i don't know how much you actually have followed the markets at all have you been sort of completely away from it not completely but i make a habit of really you know trying to wind down a bit and uh read some books that have nothing to do with the markets don't log on to twitter all the time and not follow every move in the market so i i'm a little bit in uh a little bit off sure but catching up on things you're catching up on things that's great and i'm sure the market wrap that i'm gonna focus on this week will come as no surprise to you because of course the week finished with the quote-unquote all-important at least to the fed unemployment numbers which came in let's call it a lot weaker than expected and i love the way that one financial blogger put it he wrote is this bad news so bad that it's actually bad question mark and then the blogger went on to write now before the release everyone thought bad news would be good news because it would mean less tapering but this news is so bad it may actually be taken as bad news so kind of a a little bit of of mental gymnastics there and of course we're going to have to see what the fed will say next of course they may just come back and say well now we need more data because the data is not as clear as we thought at least as rules-based investors we don't really have to worry about how we should interpret these numbers of course since we are in the minority the immediate question for most investors is the number weak enough to convince the fomc to postpone the tapering of the open market purchases and given all the commentary of the various fed presidents and govern governors in the last few weeks you may conclude that the answer is a solid maybe there's no doubt that the 120 billion monthly purchases is distorting interest rates with rates across the yield curve very near at zero the fed can declare mission accomplished but part of the accomplishment is an arguably wildly overvalued and over confident stock market the question the fed members may be asking themselves is will tapering under mine record high stock indices it might but it's not part of the fed's dual mandate to keep unemployment low and inflation stable the unemployment rate by the way has fallen to 5.2 percent from 14.8 in april 2020 while the annual cpi is at 5.4 percent well above the fed's 2 target i'm sure the official response will be that we need more data to decide but of course we'll have to wait until we hear from them now more it's interesting to hear your kind of view on what's going on in your portfolio the markets given the fact that you haven't paid as close attention as you would normally do i guess that's the advantage of being rules based you don't have to spend too much time on it when you're on holiday so what does it look like now that you're back looks pretty good i mean you know i run my assistant every day even though i'm on on holiday but i'm not spending hours on it i'm up 26 here today august was up 1.10 july was up 60 basis points and slightly down in september but that's only kind of like one day i guess only friday now what has happened i think it's been a relatively quiet august for me i took some notes here to prepare for our conversation my portfolio is mainly long i only have a really just a handful of short short positions on and in august what i did exit is i exited a long position in cost b on the 19th of august that's a trade that i had on since december of last year i think i made a percent or so on that trade um the market had moved kind of sideways since april this year had to drop in august that scared everybody and that kicked me out i also exited the long euro currency position i exited the long silver position on 9th august i think i'm not the only one there catching up with you guys that was the day when gold and silver had this massive overnight sell-off and of course that kicked me out um it is what it is i would have loved it to be different uh because gold and silver are rising again but right now i don't have a long position and i also exited long platinum and that was it the short positions that i have on is i think yen and lumber number is a great short by the way i'm short libor in the september 2024 contract that's about it maybe five or six shorts and the meaningful long positions european wheat milling wheat uh it's kind of nice euro stocks dividends very nice the s p just uh looks very strong to the upside a long coffee long palm oil long natural gas natural gas is actually one of these things that i can yeah i can feel a little bit in my portfolio in in in nice ways and interesting ways as well i mean there's henry hub so us natural gas which is which is moving higher and i have a long position i also have a long position in dutch gas which is ttf and that contract is really going you know through the roof right now because we have a gas shortage here in europe and that contract the market traded at around 1515 in march april of this year and it's now north of 50. the market is really scared that europe isn't going to be having enough gas uh you know the storage isn't filled up enough for the coming winter season so this this market is moving higher and what's interesting about this is that it's becoming substantially and increasingly correlated to emissions which i'm also alone and that's there's also a nice market to be long off but what's happening here and that's interesting is that because gas is getting so much more expensive it's actually lucrative for the power plants to fuel switch from gas to coal and coal is the dirty fuel that they shouldn't be burning anymore because of carbon dioxide emissions and when they do that when they burn coal to generate power they need to buy more emission certificates more emission allowances and this is what's driving the emissions higher right so you have a very strongly positively correlated market right now which which you know didn't used to be the case and that's what i'm saying like i can i can kind of like feel the gas positions in my portfolio henry hop ttf and also the correlation with emissions it's all very nice volatility because it's all to the upside and all very profitable right now but of course it does have a risk when that thing turns around and one of the things that i like about some of the spread systems that i have on is that it tends to get me onto different parts of the curve so for instance in ttf where i'm along the front contract my spread system also signaled me to have some short spreads on in deck jan and and also in in april may and that reduces my outright exposure to ttf at the short end quite a bit and makes the thing a bit more stable so gas is interesting let's see what happens there it's exciting and other than that easy as it goes yeah a couple of questions oh well first a comment i mean you're you're it's an interesting thing this thing about europe scratching their heads to find more gas and of course what better way to than to buy it from a reliable source like vladimir putin who's obviously always going to provide all the gas we need when when it fits into his uh schedule so but that's just you know my take on on this because of course there's this new uh pipeline being put in place uh into europe from uh from russia not stream two north stream too but let me ask you a question and i don't know if you know the answer to this but i'm just curious obviously you're having a great year so far how much of that would you say comes from classical your classical trend portfolio and how much is some of the new stuff that you've talked about that seems to be working really well like some of the spreads and so on and so forth yeah i'd say 75 to 80 this classic trend following okay i may actually you know look at becoming a little bit larger with these spreads yeah but i've only added them about a year or so ago so me being a scaredy cat i don't want to you know just uh switch that system on and you know go all in on that stuff try it out a little bit kick the tires see how it works in real time and in real life and it's i'm really happy about you know having put in all the work i mean it's took me quite some some time years actually to do that uh and then come up with that system but i like what i see and the way it trades and the way it behaves so um maybe by the end of the year or something like that i'd become a little bit larger with these spreads i mean mind you they're a little bit more expensive to trade of course because i'm trading on both sides of the markets right so it's twice the commissions but system diversification is is absolutely fantastic we've we've talked a lot about market diversification on that podcast and you know diversification across different time frames and and all of that is is super but if you can get system diversification on top of that that's great with systems that are still trend following type of systems even though it's a spread system systems that have trend falling properties positive skew these type of things you know no large downside tails to the left side and that's stuff that i don't like and my spread system just correlates with close to nothing to my outright trend following positions and therefore it's very valuable to me so yeah and the gas stuff i mean look it's it's it's politics i think to a large extent uh we hear news that russia themselves don't have enough gas this year i find that a little bit hard to believe but there was a story that it would take them from now another two months to fill up their gas storage i'm not sure if that's true but maybe they use that to say look we can't deliver more gas to europe because we're short ourselves i think they're i mean this is my my opinion personal opinion that might be absolutely wrong is um they're just using leverage to get nordstrom 2 going that's what they want and and then the gas starts flowing again well interesting point that you mentioned about system diversification that is another topic that we've discussed of course and i can almost sense jerry making notes as he's listening to you and i because he's on next week and of course he's going to try and i'm sure convince us that no no we should just do what we normally do so i think this is where rich and you and i maybe even rob are more sort of open to this thing about there might be other ways of doing trend following not saying that jerry isn't but in terms of actually implementing different types of trend following systems rather than just having multiple time frames so interesting point on my side in terms of our trend following strategies at done i mean you know whether the unemployment news is bad not so bad or even good our trend following strategies had a positive week where some of the long-standing trends in particular inequities and energies continue to march higher currencies also had a positive contribution this week but elsewhere in the portfolio we did see some smaller corrections in particular in the grains and the softs it is interesting though to me and of course my trend barometer is completely unrelated to what we do at done and and what my trend following model is doing but it is interesting to me that it's currently actually still showing pretty weak readings and of course the summer hasn't been super grateful for trend following returns so we'll see what happens next on the volatility side it was a very quite weak in u.s equity and volatility markets despite you could say the rather surprising economic data that we talked about friday in particular was as close to a non-trading day as one can imagine with the s p 500 moving by less i think than 0.05 percent while this is still you know it still led to a weekly gain for the s p and the vix remained more or less unchanged finishing around 16 and a quarter which was down ever so slightly since the last week the vix is approaching post covert lows again a level it has bounced off a couple of times during this year already and the positive change in fixed income strike volatility which almost offset the increase in the s p is therefore not entirely surprising downside protection remains heavily sought after and therefore it is quite expensive but despite all of that the s p realized volatility is at 3.9 for the last five days and 7.5 percent for the last two weeks so incredibly low all in all our volatility strategy had a positive week as well and a positive start to september for my trend following portfolio where as usual i can be a little bit more specific it was a down week it was a down week for the week but the month of september which we've started it's up 30 basis points it's up 9.57 percent yesterday performance breaks down group 1 so far this month in september so it's only a few trading days of course it's pretty much flat down four basis points group two up 28 basis points in group three almost flat as well at six basis points best sectors so far energies base metals and softs the worst sectors are currencies and bonds and if we drill down to single markets as you rightly pointed out more net gas is doing best nasdaq and australian spy makes up the last two top three markets so far at the bottom this month i have smi and the euro and the dax not doing so well and in terms of trading this week uh the week started off with the model buying some copper it closed and took a loss on a short position and then also it bought some coffee and some rbop and then on tuesday it was all about my fast reacting models because they had gone short the decks not long ago but then it just managed to get stopped out of those shorts and reversed to a long position as the dax tried to make a new all-time high it failed at the attempt so we'll still have to see how this these long position pans out but that's partly why it suffered some losses in in august was due to these short day or short-term models for sure what else did he do it closed out a long coffee trade and it closed a short euro position and lastly on friday it went out of a long cattle live cattle position in terms of the risk to stop level if everything gets stopped out on monday it would lose around 10.16 percent which is down from 12 and a half percent the week before all right so before we jump into the questions and we got quite a few of them which is great uh joe derek james sebastian and mickel send in some questions but i wanted to ask you just to go back a week on the topic that rich and i discuss namely back testing and i'm just curious marc maybe you could just share a few of your thoughts in terms of how you your framework is for for doing a back test uh what you look for maybe other certain things you feel is really important to uh to ensure so that you um don't read the wrong things into uh to the back test results so to speak the back testing question um well what do i look for it starts with the data right so make sure you have good clean data you can get data you can use daily bars you can use weekly bars some people use monthly bars even i like daily bars my system would work as well on weekly bars by the way i've tested that and that is a good sign so what i look for is a large sample size of trades with can i start to interrupt you here marks that was one thing that came up in the discussion as well is there such such a thing as a minimum number of trades that you need to have before you feel yeah that's it because we we've used this sample size you we've talked we talked about it almost every single episode but i don't know if people really know what do we mean by that does that mean a thousand trades does that mean ten thousand trades do you have any uh view on yeah it's difficult to get to ten thousand trades unless you do um you know you have a very short-term system or you use a lot of data going back uh decades and decades but um in in like our trend following world if you have a few thousand trades historically that's kind of like what it is but think about it that way right if you had a trade a sample size of um of 50 i mean what does that really mean statistically you can roll the dice 50 times it doesn't mean that it's or you can flip a coin 50 times if you do it 50 times it doesn't mean that it's 25 times head and 25 times tails there's going to be noise in that distribution and so what it is that we'd like to do is we want to remove that noise as much as we can and the larger our sample size the more we'll reduce it and the more we'll kind of like crystallize our edge in that sample so a few thousand trades and then be mindful that if what you see is not what you like because the p l is too rough and too volatile and the drawdowns are too large i mean that is likely an outcome of your back test then you'll probably feel that urge to um to do something that's better because you think that you can do better you can sit in front of the computer and you can find rules that make that p l curve even smoother or much smoother so that you'll have a better time with it and that's that's exactly the point when you start entering the optimization environment and you have to be very very careful there i think because it is so easy to find filters and additional parameters and all sorts of things additional entry rules additional exit roles scaling in scaling out volatility control yada yada we've spoken about that many times all of those will probably have the effect of making your historical experience a better one because you'll have less drawdown even higher returns more winning trades all these type of things but you're essentially curve fitting something to the to the past that is unlikely to occur again in the future and therefore that's the wrong thing to do and it also reduces your sample size because every additional condition that you put on top of your system is kind of like more specific to something to a specific event to a specific you know price action that has occurred in the past and the more specific you make that the less the likelihood that that exact same pattern is going to occur again going forward so accept the fact that a trend following a classic cavement and following p l return stream is rough loose fans pi fit all people right um and and rough is good find other ways to reduce that roughness if you must i mean i've kind of like i've mentioned that before i i stripped a lot of things off of my trend following system over time you know becoming simpler again and then and less complex in the way that i trade actually embrace that volatility waiting for that outlier and then have these you know massive moves i guess in prior years i wasn't really primed already yet to um to trade in that way um it caused too much pain made me nervous whatever it was i i'm kind of like now more steadfast in that that i can just accept it and take it on the chin and i know that what i'm doing is the right thing to do still that doesn't mean that i'm that i'm not looking to trade better i always want to trade better and why not you can add markets to your system that helps jerry has added the single stocks i'm sure that helps i've added spreads that helps me that's a form of system diversification the larger the account size the more you can diversify across time frames the more you can trade markets that have a relatively large notional contract size bitcoin for instance etc etc so so this this is what i would focus on is is not to um not become too specific in your roles but rather broaden your market set broaden your time frames and broaden your system diversification but stay true to trend following principles you know don't put system on top of system and you have like one trend following system and you combine that with a quote-unquote crappy mean reversion system that will really kill your trend following experience um i i don't like these type of things so it needs to be in a trend following framework trend following mindset and then i i like i like the way jerry has said it many times on this podcast is don't fall in love with these historical curves i mean that is that is just a chart it's like a picture look at some of the key statistics uh what's the average win what's your average loss win loss ratios these type of things i look at the maximum drawdown of course but you don't really need a ton more than that at least i believe that i'm not really interested in a historical valued risk figure for my system like what would the var have been in you know i don't know 2011 during the uh during the europe euro crisis i i don't know that that that doesn't add anything to my uh to my thinking so i look at the historical volatility i find that interesting like what has realized i look at some monthly returns one thing i do look at is margin utilization and i want to be conservative with these things i don't want to run a system where i'm getting kicked out of positions because i'm just trading i'm trading too you know too large a size i have too much risk in terms of margin and you need to be cognizant of the fact that margins can increase and do increase every time volatility in the market is is rising so that's one of the things i look at so that i can make sure that i can stay with my positions other than that not much more sure i used to look at probably a million things that's why i said like you know when i left peaks complexity type of but all of these statistics i mean you can come up with mar you this that sortino sharp blah blah yeah fine but why i mean what it wouldn't have had an impact it wouldn't have changed my decision making process with that trend following system so i don't really need that anymore so so let me this is actually i'm going to jump straight into one of the questions we got from derrick because he he's asking what are the most important risk metrics you monitor and then he goes on to say that from what i'm saying you know every every week it's it's the risk to stop so i'm going to comment on that in a second jerry weighed into this on twitter saying you know he looks at every trade average win average loss percentage win risk of ruin and and as you've said you know some of these are the same of course and what you've said also is that you really try to cut down on the number of things you look at and i think that's absolutely true and i think that's something that comes with probably just experience i mean in the beginning we were tempted to look at so many things that we think would be interesting and meaningful but in reality it comes down to very few things so let me ask you a slightly different question and that is what metrics would you look at to get a clue where you say no this system i wouldn't trade i mean where i said you know what what are you looking for to make the judgment saying yeah this is fine or no this is not fine is there anything you found that kind of triggers you to make a decision one way or the other yeah i mean of course i mean imagine imagine you run that back test on a system and it has a few thousand trades um that's just sample size and it comes up with an average trade that's negative well i'm not really interested in trading that system i guess right because what it statistically means is that if i put that trade on then on average i'm going to lose money so no that's not for me i also want to stay away from systems where my average loss is greater than my average gain you know that speaks to the skew of the distribution with our trend following systems we keep our losses very small we keep them you know to a certain number of atr's and in my case hopefully in the absence of the market gapping but we keep the upside open and so that means that we have that asymmetric distribution of gains versus losses it's something that i'd like to see the consequence of that is drawdowns and volatility and roughness of p and l but the positive aspect is robustness of the system and having a great time and a reliable system having a great time that you know sometimes when that thing works it really shoots for the stars so like that that's about it i'd say i mean i don't want a system where the historical drawdown i mean probably there is a limit where i say like if if the historical drawdown is say 70 or so then yeah i probably wouldn't trade that system but even there you could say well reduce the size reduce your risk right if you only traded half the risk then that 70 drawdown becomes a 35 percent drawdown or else being equal so you can adjust things but the drawdown in and by itself is probably not the single metric upon which i throw a system out or switch one on or off okay no that that's cool so i would say derek it's true that i talk about every week about the risk to stop it's not really a it's i wouldn't say it's the the main risk metric for me i think it's a really useful statistics for a number of reasons it obviously tells me a little bit about the riskiness in the portfolio but i think more importantly it tells me or it helps me monitor if something is out of whack so for example if you certainly don't say that that this has happened but if you were to have an error in your code and you suddenly ended up with a massive position then that would show up because you your your wrist to stop would blow out and so that's a that's a good way for you to quickly see that something might be wrong so i think for me it's just kind of a monitoring tool that that i uh that i use and of course we can we can do this because we have or everyone can use this as long as you use kind of hard stop-loss exits on all of your positions when looking at a back test i agree with you moritz i agree with jerry i think it it is about keeping things relatively simple and of course these you know having more average win compared to average loss all of that makes complete sense i think though what i found is that one of the things that i believe will potentially break your confidence in your system it's how deep and how long your drawdown is so for me drawdowns are actually quite important and i think there are different ways you can look at drawdowns there obviously you could just look at the max drawdown you could look at that you could look at the length of your drawdowns and and so on and so forth and i think one of the kind of other metrics is that people also look at so this is not unique is this ulcer index which measures and identifies the average drawdown that you can expect uh should that you should be able to stomach if you trade a particular strategy i think that's important i think sharp and tortino i don't really pay much attention if any to those things i think actually sharp could be classified as a risky statistics to look at not a risk statistic so so things like that i think for me it's and i also think that one thing that i want to do some more work on and and have already started doing this with rich and will come out with something in in the fall and that's more in in this sort of path dependent way of looking at drawdowns because you can have very different real-time experiences depending on how how that path is once you get to a drawdown so those are the things but i do think that we all agree that things have to be generally pretty simple in order to be robust and of course you i'm sure we all have some kind of rule of thumb in terms of what we expect the drawdowns would be based on our volatility and things like that and and those things are worth keeping an eye on and see if the back test makes sense in that respect but i want to stay with derek because derek's sending another question um and i can't wait to hear your answer on this uh morris so he sent in another question and that is if you had to start the business again from scratch what would you do differently or keep the same how would you try to be unique if at all so i don't i don't run a well i i do run a business but it's a business within dominic group so it's kind of like a it's definitely a different setup than say chesapeake right but you've also been involved in starting a business right so yeah you know exactly what it what it takes i mean i can from what what i can say and uh to to let you guys in on that i mean it's it has to do a lot with uh with people and trust so when when i started a quantum together with a bunch of friends you know this is now almost 10 years ago of course i was 10 years younger i think i was uh to a certain extent naive and blue-eyed and in the way that i thought about the future success of that firm i was doing that together with a friend from winton he came from winton capital brand name cta um we got some really senior guys on board great names uh perry kaufman on there it's like you know it felt like if if this doesn't work then nothing in the world can't work this just needs to work and people will throw their their money at us and of course nothing could be farther from the truth right it was very difficult to get seed money we eventually did get seed money you know it's it's now a profitable firm the firm is still around but it's been difficult nevertheless because the way you have been the majority owner of that firm the way we've built that up is we've kind of like put in all the everything had to be super everything was expensive there was an expensive office you know super expensive i.t infrastructure market data people on the payroll regulatory costs but no revenue right and i thought well you know it can't take long and then we'll make a ton of money performance fees management fees uh the aum will be in the hundreds of millions and none of that happened so you have that business risk where it becomes really difficult to you know go through that phase because you have to cut down expenses nobody's feeling good about that nobody's really earning money and then i mean in our case it's all about the people then if you have between five and ten people and not everybody can work through these periods in the same way as everybody else can there are some people that just cannot live in that with that uncertainty and and you know for understandable reasons i mean if you don't make an income it's it's very difficult with family and all these type of things right and then you also in our case we had people of different um in different age groups i was certainly the youngest at the pack there i had a very long-term view of uh building that business kind of like okay keep the risk down you know build that up and then here we go for the next 30 40 years of course if you have a you know partners who are 20 say 15 to close to 20 years older than you are they have a different perspective on what they want to do with that business you know for them it's not a 40-year journey it's more like you know that thing needs to make money within the next 10 to 15 years because then i you know i'd like to retire and that impacts risk-taking so what i want to say is i don't think i'd make anything too different in terms of um trading style trend following all of this i kind of like i had already figured out back then probably do it less complex the way that i do it now but really if you want to start a business be so careful about costs keep them down down down down make sure that you get some track record under your belt some friends and family money in there don't hire too many people off the bat don't be naive don't think that you'll be the one that can make it um it's very difficult yeah no i i completely agree and i think it has become a lot more difficult actually since you started your last firm and and and i've certainly also been involved in a few startups in my career so so i i want to start from the reverse actually because derek ends the question by saying how would you try to be unique if at all and that kind of reminds me of this quote that in order to stand out we need to know what we stand for and i think you know having a really really clear definition of what it is you want to do um with your with with your new firm i think is incredibly important i i will have i will have to say that i think the barriers to entry in our industry has gone up significantly not and it's not even just about the aum you start with there's so many other things that you now have to so to speak tick in order to attract investors not least a certain length of actual track record you know everyone can show up with a simulation but it really means nothing so in terms of actual track records so all of those things you uh you need to uh you know you as you rightly said more it's a really long term view you have to take so the way i think about it today is there are a couple of things you could do if you have a some sort of a track record that people can check and see and you might have some some good experience already in the industry and you want to kind of branch out one way you could do it is of course you could try and find a strategic partner to be part of you know i i guess also in in my youth i thought it was better to be you know a big slice of of of of a firm rather than being a smaller slice of an even bigger firm so if you can find a strategic partner even if you have to give up some of the equity or profits whatever it might be i think that's a good idea but i will also say that i think it's very hard to find today i think there's less people than 10 20 years ago who are actually willing to do these seeding or early acceleration deals and the conditions you get offered nowadays are really not that great so you have to be careful but you could be lucky and find someone who who wants to support you so that's one thing the other thing you could do is just to try and say well why do i what what do i really want to achieve with my trading you know why do i want to manage a lot of other people's money and i think you have to be clear on that because i do think and i do see some firms where you could say the aum is a lot smaller but if you don't have too many people as you rightly said you know don't maybe start out with too too big an infrastructure but if you have enough people to to do what you want to do and you have enough aum to pay the bills so to speak but you still had that upside in terms of performance fees and so on and so forth i think you can actually build really good businesses even at a much smaller size if you're if you're also careful about the expense side of of of that so um yeah i'm not so sure that the way you know most of the people that are on this podcast will will you know we've obviously grown up in a certain way of doing things and and we see all these legacy legacy firms with great stories great track records great aum and all of that i'm just not so sure that that is something you can easily achieve today unless you have something that is absolutely so unique that you can demonstrate it very quickly and people can see it really quickly that you have found something that nobody else is offering but that in itself is also quite a tall order i would say so but it's a great question derek and so we appreciate that and and maybe jerry next week we'll have some news on what he might do if he was to start all over today [Music] let's leave derek's questions and move on to joe joe rise in first off i'm addicted to your podcast i look forward to new episodes much like i used to wait for new cartoons on saturday mornings when i was a kid so that tells you about my age bracket i trained follow us and foreign stocks i also tried to trade as many etfs as possible for diversification medals commodity currency etc and i wanted to ask your opinion on one difficult subject for me personally my question is when i look at the charts that make it through my screeners i inevitably have to decide between stocks at all-time highs and ones that were higher in the recent past sometimes much higher but are now recovering and trending upwards again psychologically speaking it's difficult for me to see past the top of a chart with new all-time highs when it's so easy to tell myself that since the recovering stocks have been higher before there is a precedent and they can reach that level again so is there any justification for my bias against all-time highs and fears of buying the top or do i need to just quit overthinking it and as jerry often says take all the trades thank you so much for your time and your podcast from joe in pennsylvania thank you joe for that and for your kind comments moritz what is your what is your advice for joe i absolutely love buying all-time highs so i i'd say don't don't overthink it if your system tells you to buy that high and it's the all-time high then you buy it if the market has corrected from and high you got kicked out of your position it hit the exit and the market starts rising again you get an entry signal you take that entry signal and you go long again it is really i wouldn't overthink that and i definitely don't want to penalize an all-time high against another type of a high more like a local high because the all-time highs even though you know they look like well the market has never been here before it's never been higher sure that's the definition of an all-time high it's still a good a good system to buy all-time highs so come to terms with that i'd say and just just go along so of course joe not surprising i will agree with moritz but i'll tell you i'll tell you why i agree um two things first of all just staying on the topic of all-time high i mean did you know that the s p has made 53 all-time highs just this year alone we still have five months left i mean that's just one reason why all-time highs are not is not a bad thing and if you look at since 2009 for example i mean how many times has the equity markets made all-time high so as mart says don't be afraid of the all-time high but the reason i would not overthink it is if you do you cannot in as far as i can tell from your question you can't be 100 rules based right because you are now deciding discretionarily which trade to take and i think for me that's the cardinal flaw in your in what if if you if you do that compared to what we do where we don't have an opinion whether it's a good trade or it's a bad trade we simply just take the trade so i think that's the advantage you get it's the discipline it's the systematic and rules-based approach and as we've talked about in the last few weeks in particular with rich and that is what we do is we hunt for the outlier and there is just no way we can tell in advance what the outlier is going to be so by definition you have to take all the trades i'm afraid joe so hopefully that works out for you in the future but i love the question and thanks for your kind words next question is from james james writes hope you're well just wondering if you could recommend a software program for the retail investor that has limited programming skills that can back test the classic classic trend following strategy discussed on your show more stu are you familiar with any um software where you don't have to do the heavy lifting in terms of programming i think you and i will kind of think about the same one yeah i mean there's quite a few that i've used i mean there's there's one that i'm i still have installed on my computer uh it's it's trading blocks i think that's very straightforward to use i mean limited programming skills is what you need it's you know based on the scripting language chart based but easy to use pretty much in the same way that tradestation or trade signal ninja trader those are other names you know you can use them in the same way but trading blocks is really really really well built for trend following type of trading systems and there's uh some systems in there that are pre-built which you can use like you know i think to the turtle system is one system is in there right so you can you can use that as a base start from there or just use that yeah but i will say and and i think you and i will both agree that we're not endorsing any of these systems we're just mentioning systems that we have heard about or familiar with but i will say on top of what uh march has said i would just say james that you should still i mean even though you buy something that's off the shelf and may or may not be you know have you know a lot of people testing it and running it you should always be critical about what you know what you see and what you do um there can be box there can be other things so but it's certainly something that might help you and speed up things instead of trying to build it yourself of course so uh so i agree with what moritz just said there another question mart that you might be able to answer here is for sebastian sebastian rights interactive brokers just introduced negative interest rates for trading accounts above 50 000 at least in europe starting from tomorrow so that's a few days ago do you have to deal with similar restrictions in any kind and how do you deal with that i i am affected by negative interest rates um in the injector brokers account in my rjo account and essentially all of the accounts except for my checking account which i have here at the local savings bank but trading wise it is it is a it is an issue if you are a euro denominated investor or swiss francs denominated investor or danish kroner denominated investor it's less of an issue if you're in the us and your accounts are in the u.s dollars so what can you do well the the i think the objective is to hold as little cash as you as you can you know you can't hold zero cash you always need to have some cash on hand to operate as a trader but you don't need nobody forces you to keep all your residual non-margin cash in euro cash and pay negative interest on that right i mean you can buy other securities those could be you know in the u.s you could buy treasury bills and and notes and stuff but here in europe even that's coming with a negative interest rate if you put your money into boo bills or shots or bubble or bond or even the books allowed to 30 years it's all negative right so that really doesn't help so it forces people out on the risk curve where they either go into credit risk mortgages these type of things i i don't do that because i don't have i have no edge in picking credit risk and i don't want to spend time on fundamental analysis on on any of these type of things but i've mentioned before that i'm interested in these cash and carry trades where i can get a a positive yield and these cash and carry trades they could come in the form of holding a single stock long and being short a single stock futures contract against that which trade on your ex and you're taking that single stock off of somebody's balance sheet for instance they bank and they pay you a premium for that and that moves you into positive interest rate territory you can do cash and carry trades with spot emissions here in europe and they pay about 100 basis points and that's a very very nice trade because you're essentially holding an instrument that's backed by the european commission and all the governments here and your shorter futures contract against that well we do trade futures so no problem there you could do that very very lucratively in the digital asset space those uh basis trades are becoming a bit more interesting again right now as we speak they've obviously collapsed when the bitcoin collapsed to thirty thousand but you can do them there just just be careful on how you do them like where you hold your digital assets and uh and how you structure these type of things but as far as i'm concerned there are quite a few opportunities for you to move out of the negative interest rate environment as a proprietary trader or even as a even as a cta that's you know open-minded and willing to engage in these traits of course you have to understand them research them make sure you execute them properly and not just shoot from the hip but it's possible to do that yeah no i completely agree with you more it's and and i think generally speaking uh sebastian you you just have to accept that if if if there is a currency or a number of currencies that has negative interest rates there's not much you can do about it other than accept it and because every time you're going to try and avoid that you're going to just take on some other kinds of risk and frankly you know with say half a percent negative interest rates per year your systems should hopefully still be able to produce a a a positive outcome it's not going to be devastating to your uh to your trading i would say last question is from michael i have a feeling that michael is a fellow dane so hello to you michael great show keep up the good work i'm a data scientist getting getting into trend following and i got curious about your back testing method after listening to episode 155 you mentioned that a back test of say the last 50 years of data doesn't necessarily make accurate predictions about the next 10 years as markets always change this makes perfect sense my question is then would this also have applied 10 years ago in that case can you make the back testing up to 10 years ago and then keep the last 10 years as unknown for the model those 10 years would then indicate how the model had performed 10 years into the unknown future simulating that you did the back testing 10 years ago and then ran the model live for 10 years this is similar to splitting data into training and validation for machine learning models to take this a step further you can split the last 50 years of data into five segments of 10 years and then fit your model on four of the five segments five times that way you will end up with five model predictions of 10 unknown years each which you can use to calculate both the average return and standard deviation next you can fit slash train your model until the average return of those unknown validation data is as close to the return as the training data as possible this is a common method for a ver for avoiding over slash under fitting in machine learning and i'm curious if you apply a similar technique in your back testing so we're back to back testing where we started markets yeah no a very good question uh mikel um a lot of traders do what you've just described which is in sample and out of sample testing and your out of sample period could be as you've suggested the past 10 years or the past three years the past five years it could also be at the beginning of your data sample right you could say i'm leaving the first 10 years out and i'm testing my system on all the remaining data and then i see if it would have worked back then you know moving back in time 30 years 40 years whatever at the very beginning where you haven't seen the data yet so i think that is that is interesting and um and a nice thing to look at what it also means and i want to come back to that sample size discussion that we have if you have 30 years of data and you leave 10 outs then you're not looking at one third of your sample size so it does come with consequences and so i've done what you did in the past which is i've used in sample and then i kept some data the current data out of sample on which i did not train my system and then i had a look at you know how that would have worked i now feeling confident enough that with my trend following trading style where i keep the losses small and i let the windows run i'm okay to look at the entire data set and test on the whole thing to exactly get what is the average trade the average win the average loss and all these type of things and i don't think that my type of trading system because it's i've removed a lot of the complexities over time is is that sensitive to um to an in out of sample type of analysis and of course that is completely different to what you've just mentioned in your question nils when it comes to machine learning and you know these type of systems nlp etc etc there you have a lot of dependencies on the data with how your systems work so you probably are forced to do what you say but with my trend following training style i don't really see the need to be honest at least i don't see it right now i don't see it any longer i used to do that don't do it right now yeah and i think that's uh that's a great answer uh and to a great question by the way amigos i appreciate that so in a funny way i've kind of done some of what you've just said with my trend following model because we did the back test from 1990 up until say 2007 each then we went live and traded that and then around 2013 ish we probably updated some of the models parameters based on further tests and then since 2013 i haven't changed a thing so the numbers i have quote every week is based on something that hasn't been touched for almost 10 years at all and so if i look back at that and i talked about it a little bit i think with rich last week it's certainly most people would never do what i've done which is you end up with something where you do a forward test without trading it live for for 10 years but that's obviously because i work for another firm but what's interesting about the experiment is that the question would i change anything today knowing what it's now done in in out of sample and the answer is probably no and the reason i say that is when i look at the profile of the return since 2013 or so since we made the last change it's pretty much what i would expect from a trend following system you know it's had its good years it's it had its drawdown going into the end of 2018 is where it had its worst drawdown and then it's had a great run you know last year was incredibly strong this year is okay so far and so on and so forth so it actually is doing what i would expect a trend following uh strategy trading those markets of course if i had included bitcoin and ethereum and stuff like that maybe it's done would have done even better but given the market universe that we selected all those years ago i would say it's doing exactly what i expect and therefore probably you could always oh you know try and optimize but it for me it's been a really helpful exercise because i think it's confirmed the way we thought about building a trend following system all those years ago and actually i think we've achieved exactly that so anyways but great question michael so i appreciate that now mark those were the questions i have one question left for you which has nothing to do really with trend following before we wrap up and it's just that there are some things happening in germany this month such as a an election and since uh since you are based in germany i thought you might have some interesting thoughts comments germany's as we know very very important part of the european union and so yeah so what should we expect from uh from the germans this month i i don't know i mean um but what has been very interesting is that the social democratic party the spd yeah has made a tremendous comeback um they've been in the trough i mean even a couple of months back they were kind of like a party not to be reckoned with and you know some people say well maybe we just don't need that party anymore they're like getting close to 10 i mean it's kind of like very very very very low numbers compared to where they uh where they used to be and the conservatives the cdu and csu they were way ahead as were the greens and this has completely changed it's it's really a head-to-head race between uh the social democratic party and the cdu it seems right now but the social democratic party is even five points ahead of the ceu as of the latest polling and he's the current finance minister right the guy who's finance minister um the vice chancellor that's olaf schultz so a real turnaround of things what's so interesting is that the even the greens and the liberals they're all kind of like on the same level now so what that means is you have at least um i think five or six possible coalition combinations and that i'm not sure if that's a good thing or a bad thing because uh okay let's let's just take two of those out because they would include the far left or the far right so probably that's not going to be a starter but that still leaves you four combinations four possible combinations of parties and so how how long will it take that country to actually find the coalition to run the government and you know this can take months i'm sure if that's good but yeah that's that's what it is like right now interesting certainly worth i mean again as people know who listen to us every week they know that it pay it makes no difference to uh how how we trade that's right and what our systems will do but it might uh just as you rightly say it could increase some the levels of uncertainty if they can't figure out how to work together or whatever it might mean yeah so it's always interesting to see some some markets may even get impacted by by results i mean if say the greens sure becoming part of the government and let's say they uh they have a very high vote there then uh what does that do to the emissions market what does that do to the energies markets if they say you know what forget about 2050 we're doing everything by 2030. what does that do to power prices and so that that could have an effect but other than that it's normally um not much of an issue other than taxes and these type of things you know if uh some of those parties they're calling for substantially higher taxes and uh that of course means that i would be paying or we would all be paying higher taxes on our trading profits so yeah you can always move a couple of hours uh to uh to the south and you'll be uh in switzerland so uh it's always welcome pay my taxes in bitcoin which is uh something that you can do you can ex that's exactly right and uh we didn't even get to talk about bitcoin but of course as most people will have noticed uh bitcoin in ethereum and all the other cryptos are big on on on a tear again uh we'll see if it makes new all-time highs but it's certainly been an interesting six weeks are you have your systems kicked in on the long side again in everything oh yeah yeah i see a big smile at you uh so yes i'm sure that's uh that's uh that's the case yeah yeah cool we'll see uh we'll see i think jerry also managed to stay long some of the bitcoins as well so that's good okay well speaking about performance it's only been a few days of course in september so far and and these numbers are as a thursday it's only really after a couple of trading days in september nevertheless beta 50 is up 56 basis points up eight and a half percent for the year suggest ct index up 34 basis points up almost seven percent for the year the trend index up 39 basis points up almost nine percent for the year and the soccer and short-term traders index up 24 basis points for the month up 42 basis points or 0.42 percent for the year the trend barometer i'm into that already it's down at 30 which is weak so the time frame i'm using for that clearly is not the same as the longer term systems but we've talked about that prior msci also still making new highs up another 72 basis points so far in september up 17.62 for the year bonds are giving up a little bit down seven basis points in the world government bond index so far this month this was fun great to have you back morita glad you enjoyed your holiday as i mentioned next week it's uh jerry who's on hopefully we can pick up some of the things we talked about today system diversification could be one jerry if you're making notes but in any event what makes it even more fun is when you write into us with questions and you can of course do that to info toptradersunplug.com and we do our very best to give you our thoughts on the topics on the questions and of course as mentioned at the beginning we would be so grateful if you could just go and leave a rating and review uh in itunes for the podcast because it makes a huge difference for the algos that essentially promote the podcast within that environment anyways from moritz and me thank you ever so much for listening and we look forward to being back with you next week in the meantime 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 at 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 investor [Applause] [Music] you

Channel: Top Traders Unplugged

Views: 215

Rating: 5 out of 5

Keywords: niels kaastrup larsen, moritz seibert, 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, diversification, all-time highs, classical trend following, backtest, risk levels, monitoring, negative interest, retail

Id: ZOlvXrQQjHA

Length: 65min 18sec (3918 seconds)

Published: Tue Sep 07 2021