Transcript
Speaker: welcome to a new episode of the in silico terminal podcast my guest today i hope i can pronounce your name correctly kieran duff is that how you say your name that's exactly right exactly right how we doing dude i'm i'm good how are you Yeah, i'm very well.
Speaker: Very well. Actually, this podcast came... It's pleasure to be here. Thank you. Thank you. this This podcast came to be because you you were recommended to me in a very popular episode of of this podcast that I did last year with TraderXO.
Speaker: ah he He recommended okay yeah to get you on. Yeah, XO is a good friend of mine. Yeah, I've got a lot of respect for him. Yeah, his episode was really great. Like he he did a lot of preparation and um he's very enjoyable to talk to. I learned a lot from him.
Speaker: And it was also very, very popular. So since he recommended you, i was like, I have to get you on. You you you must be a good podcast guest then. Can you maybe maybe tell me a bit of all and but about what you are doing? Like how do you deal with with the markets?
Speaker: Yeah, okay. So, I mean, um first of all, I appreciate you for inviting me on and and listening to XO and i massively appreciate XO. He's a really close friend and, um you know, I learn a lot from him every time we sit down and have a conversation. um So, big shout out to TraderXO.
Speaker: and But I am a systematic portfolio manager. I manage other people's money through licensed venues such as DarwinX. I also work for DarwinX as head of UK growth and business development.
Speaker: So I see a lot of what happens on the back end in kind of the traditional finance space, less so kind of crypto, although I have some stories about DeFi back in 2020 and 2021 as well, because I was very active back then. But my approach is to build systematic portfolios, again, TradFi portfolios. So we're talking FX, metals, indices, and try and just...
Speaker: Try and outperform, you know, certain benchmarks such as metals, indices, perhaps crypto now that it's a little bit less volatile with a much higher risk adjusted return.
Speaker: So whilst I might not outperform the market in terms of raw return, you know, if you if you buy the S&P and it has 15 or 20% year. but then also has a 20% drawdown, um my risk adjusted in return generally is is far, far higher. And so my main goal and the way that I approach things is how can i squeeze as much return with as little volatility as possible? Because ultimately that's what professional allocators want. You know, they don't yeah care for 100% year or,
Speaker: you know hundred percent per month They want to know that they can part their capital somewhere long term with, I say, minimal risk, but minimal risk of the manager going in haywire and martingale in their way to to a significant return or something like that.
Speaker: So all my attention and focus goes into, yeah, how can I pull little bits of the market without taking stupid amounts of risk? And like I said, I do that on behalf of investors. I think my AUM currently is about two and a half million dollars. So i'm still I'm still a baby when it comes to professional asset management, but I'm i'm on the ladder at least and and that's kind of what counts.
Speaker: Yeah. how How did you get into this role? Do you have like a background in finance or...
Speaker: No, dude, I always say this to everybody who asks me this question is I, i tend to fall forward in life. Um, and I have a, I would say maybe some look on my side, but I try and get myself in the right circles of people. I try and get myself in the right conversations. And you know, again, I just fall forward. If I fail at something, i try and fall forward, not backwards. Um, and so I tried to find a circle of people who were doing what I wanted to do, which was manage capital on other people's behalf. Hence how I got in, in, you know, to, to DarwinX, um, which we can talk about in a little bit.
Speaker: Um, but I have no kind of background in, in finance. I have no degree that would, get me straight into Jane Street or you know UBS to go and raise capital for those guys but what i have is skill as a trader and now the network of that to be able to connect the two and be like okay I am good enough to manage capital on behalf of other other people now it's time to get my name out there so I put a lot of time into networking with the right people but yeah it Like I say, i i'm just I let life take me places where I never kind of imagined. And so when an opportunity presents, I'll just do as much as I can to to take advantage of it. But yeah, zero experience working in finance pre becoming a trader.
Speaker: So how ah did you acquire this, as you said, skill as a trader that you that you feel that you're also capable of managing other people's money?
Speaker: So I started trading crypto funnily enough in 2016. I remember the days when, you know, Bitcoin was hundreds and not tens of thousands of dollars or even hundreds of thousands of dollars for that matter.
Speaker: um And of course, everybody's a genius in a bull run, right? So I made... not Not crazy money. You know, i was I was just a young lad. I didn't have much money to play with, but I saw a good return on what I did invest and have a play with. And whilst it wasn't anything major, what it did was it stuck with me the fact that I could put money into the market and with with very little experience or knowledge at the time and be able to walk away with more money.
Speaker: right Because that's what we're all here for. no Nobody plays the market game for the fun of it. Like if if you do, you're a psychopath. But people play the market game because they put money in and they expect to walk away with more money. That's ultimately what we do this for.
Speaker: um' And I think when you have very little knowledge, actually, what of kind of the risk and you don't respect the risk, it's even more kind of enamoring. It just takes over and you're like, my God, I can't believe I did this with $1,000. Now I'm going to do it with $10,000 because it's the same thing. And this is so easy. Why doesn't everybody do it? And I think you very quickly realize that actually that's not the case. um and And most people lose. But then, um yeah, when the bull run kind of died out at the time, I, again, fell forward into FX trading and got pulled into various signal scams and and trying to, you know, have the cheap the cheat cut way where you buy a PDF from somebody on Instagram who has a Lamborghini and they would tell you how to trade and all this sort of stuff.
Speaker: um And very quickly, I realized that these guys were, you know, basically talking nonsense. However, I had seen again, going back to my point, I had seen that I put money into the market and walked away with more. So I knew there was something to it. And then I kind of I tried to find good mentors who I could, again, be in their circle and learn things from. And I very quickly found a trader called Tom Dante. I don't know if you've heard of him before, but he he kind of instilled. Yeah. So he's kind of like, yeah yes, he's a retail trader more so these days, but he has experience working at a prop desk and and, you know, kind of more the professional side of things. um And so that was somebody who I was like, well, if this guy has the knowledge of
Speaker: where I want to go, I'll listen to him as opposed to somebody with a Lamborghini, which is what I did. And, um, you know, I would say that was, that was probably 2019, maybe 2018, 2019. Um, and that kind of set me up for the next few years. And I mean, I've, I've traded discretionary for a long time after seven years or so, however long it was, I decided that it wasn't, it wasn't something that I wanted to do long-term um And I then become a systematic trader building algorithms and portfolios. And my life has changed ever since, um to to put it plainly, you know, and that's what I do now. That's what I do today. And I'm very good at it.
Speaker: What sort of stuff did you do when you were still trading discretionary and what made you move away from that?
Speaker: So i would I would just trade key levels in the market. So I would look at you know a chart of multiple assets. My watch list was huge.
Speaker: Again, I'd trade crypto, indices, metals, FX, you name it. If it if it was on TradingView and on my broker, I'd be trading it. um So I'd identify... areas that looked clean from a levels perspective perhaps there'd been some resistance at some point and then we broke through and it was pulling back into that area of resistance and I would look to go long um very very um low frequency you know i might take a few trades a month nothing nothing intense but it just it was something that never really who never really got the juices flowing for me um
Speaker: And therefore, like, that you know, I left it back in 2020. Well, it was the beginning of 2025 that I took my last ever, my last ever trade as a discretionary trader. And I have been a far happier man ever since.
Speaker: And them so what what kind of stuff did you switch to as a systematic trader? Like what kind of strategies? what What attracted you more about that than than the discretionary style?
Speaker: of For me, it was it was less about it was less about the strategies. It was more about the freedom. So everything that I do is algorithmic. So I don't have to worry about execution or watching charts or any of that sort of stuff. So I get a lot more time.
Speaker: I have a lot less emotional stress because I have two young kids. And when you're when you're pressing the button for buy or sell,
Speaker: as as As experienced as you can be, you there will always be an element of emotion there. you know You've made that decision to press the button. If it doesn't go your way, i don't care i don't care if you've been training for a year, 10 years, 100 years, there's always going to be that feeling of, oh, I pressed the button and it didn't work out. And I was finding that because i was trying to because I was trying to become more professional and I was trying to break that next level as a discretionary trader, these decisions weighed more on me than you know the the the whole time that I'd been trading because things felt more important. it was I was at a place where...
Speaker: I had a lot of connections who were like, okay, you know, come and trade with us. Depending on how you do, we'll back you with capital. And that was kind of like, okay, you're breaking into the next the next level here. And so every time I press the button, it was harder mentally. And so what I found was I was going downstairs to my kids, you know, as you can tell, I'm in my bedroom. um You know, I work from home, I trade from home, all that sort of stuff. um And so I would, I don't have a commute to work where I can debrief and forget about what's happened in the day, forget about the trades I've taken and the bad decisions, the good decisions.
Speaker: I would walk downstairs and my family would be there. And whatever mood I was in on the top floor is the mood that I'm in on the bottom floor. And so I just realized I was becoming a miserable person.
Speaker: And that was kind of what sparked me to make the change. And then my approach was like, um I guess i guess my my mind was open to all the possibilities that I didn't have as a discretionary trader.
Speaker: and And what I mean by that is you are just one person. You have one brain. You have a set amount of time. you You can't scan hundreds of markets. You can't approach these markets in many different ways. It's just too much for a human. And so I went down the rabbit hole of just investigating any any strategy you know on any different asset and seeing what would work and so I just started putting things together like I'd never traded breakouts for example um and so I was like well now I can trade breakouts because I've never tested it I've never seen it and and so I started building some breakout strategies and and they work phenomenally well over a long period of time Um, one of probably the most robust ways of trading in my opinion is trading breakouts. Then I started putting together some mean reversion strategies because that was more closely related to how I used to trade.
Speaker: um and then I would just trade random things or at least build random things like, I had a conversation with a friend of mine probably about a year and a half ago now, and he said to me on a call, he said, oh, have you ever noticed when the S&P closes red on a Friday, it always opens green and and ends green on ah on a Monday?
Speaker: And I was like, as a and that's a bold statement, right? Saying it always happens. And so I was like, well, I'm going to test that. I'm going to take your words and going build a strategy.
Speaker: And so I went away and I built a strategy around that thesis. Super simple, no technical analysis, no macro analysis, nothing. Is Friday's candle red? If so, you buy on Monday.
Speaker: And the strategy is phenomenal. ah over like 20 years, just phenomenal performing strategy. The problem is it has no capacity. So I don't trade it because you can only, you're executing in such an illiquid time of day, you know, Sunday evening that you just can't get any size down. So it's not tradable. But my point here is any data point, any variable, any conversation that I have, I try and build it into a into a strategy. Um, because there's really no limits into what you can build.
Speaker: The only thing that you can't put into your code, as you know, is your discretion. You can't, you can't get a, a coded strategy to go, ah actually, I wouldn't take this because of X, Y, Z. Um, but that's, that's, yeah that's, that's what I do. I just build anything that I can and, and see if it fits within the portfolio.
Speaker: Are the strategies that you're running more higher time frame or do also have like ah intraday things? Because I think those are quite like different ah scale things because for for the intraday stuff, you have to worry a lot more about fees and and execution infrastructure and stuff like that.
Speaker: Yeah, I used to trade a lot more high time frame stuff, but as I've become better at managing a portfolio, a lot more of my stuff these days is intraday.
Speaker: So ah the majority, I would say 90% plus of my strategies close at the end of every day anyway. So even though they might be trading based off higher timeframe um variables, they're still closing intraday.
Speaker: A few reasons why I do that. it It doesn't suit everybody, but it suits me. um You obviously get less fees because there's less overnight swaps, which... It's not a lot when you trade in small account, but as you start scaling, those swap fees become quite sizable.
Speaker: um There's less volatility. There's less chance of me waking up at 3 a.m. m and going, oh God, where's that trade? and Because I won't be in one. um And I kind of realized that even though you probably sacrificed a little bit of performance by doing this and by trading intraday, or at least I found that I sacrificed some performance,
Speaker: I have a far better life doing so. I have full night's sleep. i don't have um I don't have to worry about having triple swaps on a Wednesday, you know, if I'm holding a position. um and And it just suits me as ah as a manager to be able to trade that way. So yeah, majority of the stuff that I do is intraday, but I still look at higher timeframe variables, you know, breakouts on the daily bars and stuff like that. I still look at it, but it's managed intraday for the most part.
Speaker: So you're not trading any any crypto anymore, i assume? Just like traditional markets from open to close, I guess? Yeah, i would love to I would love to trade more crypto. um i still I invest into crypto all the time. So personally, I'm still very active um in in crypto, but I don't trade investor capital on on crypto, mainly because it's not available where I trade. um And I probably would. I did start to build some crypto strategies recently.
Speaker: um I would love to do a challenge where I just put, you know, a few hundred or ah or a thousand dollars into ah into a crypto account and just trade some breakouts, like super high risk, like, I don't know, 20% of the account just goes on on each trade or something like that. um But i just i just haven't got around to it. I've just been so busy over the past few months that when I had the idea, was like, okay, i'll ill I'll do it at some point and I've just never got back to it. Um, but I would love to trade more crypto, probably just Bitcoin and Ethereum. I think the, the, the, the lower down the alts you go and the thinner the liquidity gets, the less tradable it becomes from a professional perspective. Like you can trade it as kind of a retail guy trying to fill, um, you know, a thousand dollars of, um, size. But if you're trying to really size up, I think alts just aren't on the table.
Speaker: do Is there any strategy that you can maybe share that has like worked very well in the past but has decayed since?
Speaker: um ah Yeah, i can I can pull some stuff up. or do Do you want to see it visually or do you want me to just explain it? I guess you can show it as well if if you want. um I'm not completely sure it works here, but I think you can, like, share your screen itself. It would be cool to see. I'll just... ah ah i'll I'll explain it then because my files are all all over the place. I have quite a few strategies that the have decayed. of them...
Speaker: one of them One of them is less of a decay because of the edge and it's more because of market regime. um And it's a breakout strategy on dollar yen.
Speaker: And so USDJPY. And it worked phenomenally for you know the past 10 years, 15 years, 20 years. you know Year on year, it was it was just phenomenal.
Speaker: um But it was very skewed to the long side. And that's obviously because of interest rates by the BOJ. um And then more recently this year, the BOJ had been intervening and there's been some nasty wicks to the to the downside. And quite often, um at least i would say between April to August, I would be long dollar yen and there would be, your it looked like a crypto candle, right? There would be a fat candle to the downside. It would stop everybody out and then it would just completely recover for the rest of the day. And and it it became untradable for a period of time. And, you know, this is a strategy that has made me a lot of money trading it live. And the back test is also fantastic over so many years. And then all of a sudden, because the BOJ start intervening and in in the market to cap inflation, it's like it becomes untradable.
Speaker: Overnight like that. um And it's actually it's actually performed really well over the past two months because I'd never completely removed it from um from my portfolio. The only thing I did was um it used to take two trades occasionally and I stopped it from taking two trades.
Speaker: um Again, I gave up a load of performance doing that, but it was far more stable. um And that was purely just because the edge decayed because of market regime.
Speaker: um And so I tried to make a few different tweaks on that in terms of, like I say, reducing the risk slightly, not allowing it to take a second position, being a bit more aggressive with if we are right, can we lock some of that in so that we don't get those massive spikes to the downside, get stopped out and then it carries on.
Speaker: um I mean, i think I, ah do do let me see. Is there any way to share? ah should be able to um share this. Give me a moment.
Speaker: Okay, so this is a this is a chart of Dull EN, so USDJPY on the H1. Now you can see like anybody, anybody with a week's experience in trading, right. Can see that that is an uptrend.
Speaker: But if you look at what's happening is you have these stupidly big candles to the downside and then it just completely recovers for the rest of the day. It does it here, completely recovers. Does it here, completely recovers. If I go further back, um,
Speaker: They were a little bit more aggressive. So we have this over here. We have these over here. i was actually short there, so it benefited me. um But quite often, like, it just becomes untradable because you can see it's constantly grinding to the upside. But then when you have a candle like this, you know, where you might get long, and then all of a sudden it... it you know, shits the bed and then completely recovers. how how do you How do you trade for that if your strategy can't adjust in real time? Like it's it's really, really difficult. um
Speaker: And that was a bit of a, who it was a bit of a humbler for me because i had relied so much on the strategy for such a long period of time and then for it all all of a sudden to kind of crumble beneath you and you're like oh my god what am i what am i going to do without this system like how am i going to am i going to move forward um but you know you just have to adapt and and make some changes and hope it works out for the best you know
Speaker: How many adjustments do usually make to strategies? Or even but what I was asking myself earlier, how how does a typical day look for you now? Because you mostly have these intraday automated strategies is running. You said before that you're like um not so dependent on on what is happening on the screen anymore. You don't have to like to thatmo you don't have to put in that much manual input anymore. so Like, what what what do you actually do? Are you like mostly researching? Are you like fine tuning stuff? Are you backtesting things?
Speaker: So so there's ah there's a few things that I do day to day for the for my portfolios. The first thing is I constantly have an incubator. So it's a second portfolio that runs on live data. um And that's where all of my strategies come out of the development phase, go into the incubator, and I get live data for three months minimum.
Speaker: um So nothing ever gets built and then goes straight into my live book. It always goes through the incubator. Because I want to know are the fills that i get live similar to what I see in the back test.
Speaker: um If it is and everything works from a code perspective, then I can migrate it live down the line if there's place in the live book. um So i I do still do strategy development.
Speaker: I still build strategies, I mine strategies, and then I put them through the incubator. But the majority of my focus and attention, at least for the past six months, has been on the portfolio infrastructure.
Speaker: So I have kind of the the strategy layer, which is all of the individual strategies. um And then I have the portfolio layer that sits above it and it kind of governs all of the strategies. And I was writing an article actually this morning for my sub stack on this. And I was talking about how that works. And I try and think of it like, imagine a prop desk or, you know, a hedge fund trading floor. You have each individual manager.
Speaker: And that's the same as my strategies, right? But then you have the governance layer, which is, you know, it might be your risk manager or your desk manager or your, you know, head of, head of bund, head of FX desk, all of these different people who sit above the managers and govern what they do.
Speaker: I you're only allocated this amount of risk today, or you're stopped out for the day. So you need to, you need to stop trading. And I tried to rebuild that with code into my governance layer, which sits at the portfolio portfolio level. So every time one of my strategies fires a trade, it pulls information from the governor layer and kind of goes, okay, yeah, I can trade or no, I can't because I'm already, you know, the portfolio is already long on gold. I can't double up that position.
Speaker: Because again, I'm all about the the the low volatility returns. i I don't just want to double or triple up on a position just because three strategies have fired off in the past, um which I had in in in April this year. ah Trump said something about the Strait of Homo.
Speaker: About seven of my strategies went short on gold and dollar yen. I got stopped out on all of them. And then about an hour later, they all went long on gold and dollar yen. And I got stopped out of them all again.
Speaker: um And so I asked myself, how do I make sure that never happens ever again? Because that was some form of hidden correlation that I didn't know was in the book. um And so as part of building the governance layer was how do I stop things I can't foresee?
Speaker: you know It's very easy to say, oh well, the strategy has a max drawdown of 1% or it only loses $50 on a trade or $100 on a trade. But how do you plan for things that you can't foresee, i.e. shock events, hidden correlation, hidden tail risk? How do how do we plan for that?
Speaker: um And that was those were really the questions that were going into into my development of my governor governor layer or portfolio layer. And that is where 90 percent of my attention goes now.
Speaker: Very rarely is it on individual strategies and tweaking the strategies. It's usually how do I get the strategies to play better together as a team? And that all all is within the portfolio layer.
Speaker: I must have said governor and portfolio layer about 400 times in that. so
Speaker: So it's basically um how I can imagine it. It's like you're a a one person hedge fund kind of like you you don't have a to a team, you do like everything by yourself, right?
Speaker: Yeah, everything, everything is built and and managed by me. Yeah. Did you have do you have like a coding background? Or did you like teach yourself how to how to code? Or did you use say I
Speaker: So again, i don't have a background in coding or development. However, about three or four years ago, i started learning how to build, it started with basic HTML, CSS and and Java, um, uh, JS and I kind of learned basics. I can read code. I can understand it, but I'm by no means a developer or at least not a competent developer. So the majority of my coding goes through Claude or Codex. So I build a lot with with AI agents.
Speaker: But luckily, I'm in a position where I can read the outputs and understand actually what is happening um and what needs to be included in certain bits of code. So I'm not a developer, but I and I used to work at a tech company before I before I so even started trading. um So I used to be included in sprints and stuff like that. So I understand the developmental process, um all of which I bring across now, even though I'm working with AI agents, which I think is is probably one of the better, most efficient workflows anyway, because the speed of which that you can build, as long as you understand where the flaws are within that, you know, having an AI agent build something for you, um, is just unprecedented. You can have something built within minutes and it's incredible.
Speaker: Um, but there are drawbacks to doing that, yeah you know, and you just have to be able to be able to manage those drawbacks. How do you manage um execution for for your strategies? do you have any more complex execution algorithms? Are you like T-wopping into positions are you just like markets ordering in and out of stuff?
Speaker: How does that work?
Speaker: I usually, for for for the majority of my systems, it's all market execution. um A lot of my strategies were built as pending orders. So it would it would recognize a high or low for a breakout, for example, and it would just set a pending order there. The problem with that is it's very difficult to manage spread um when you when you have resting orders in the market.
Speaker: um For example, you know, there could be some there could be some news that gets released and spread could blow out. You could tag your pending order. Price never actually reaches there. just It's just there because of spread. And then you get stopped out straight away. And that happened to me a couple of times. I also had um an occasion where I got tagged into a silver position way above all time highs because it was a broker in efficiency. And it it just tagged my... um it created an order about, I don't know, 30 points above all time highs. And again, just stopped me out straight away.
Speaker: um And so what I did was I changed everything. I rewrote all of the strategies to be instant execution or market execution. with spread filters, with slippage filters, um all built at the portfolio layout. So the strategy sends the order to the to the governor and the governor goes out, well, spread is outside of what we would usually do, which is kind of ah a rolling or dynamic spread check. So it checks the average of the past five minutes, make sure it's not like double that. And then also has a max spread that it won't allow ah to execute if the spread is bigger than ah X, Y, z
Speaker: um I have slippage checks on it as well. So if it it will calculate as the orders being pushed through, are we going to get slipped on this based on how fast the the market is moving based on the liquidity behind the book? um And if so, it won't execute again.
Speaker: So I'm always looking for how do I get the best execution? How do I do that? And I believe you can only you can only get granular with that and ensure you have best execution if you're using market execution orders. um Because it's easier to, I always say it's it's it's better to be proactive as opposed to being reactive. There's nothing worse than and being tagged in a trade you shouldn't be in because of spread or or what have you and then wanting to get out of that position.
Speaker: But you might be offside massively and there's nothing you can really do about it. Whereas if you're proactive and you can stop that order before it hits the market, not only have you saved yourself from the potential stopout or or being offside, but you've saved... the execution costs and fees, which, you know, again, unsizable amounts is is a lot of money. And I truly believe you can only do that by being proactive and executing the market.
Speaker: How does ah the structure work with the company that you are working with? And also, how do you, um what kind of clients do you have that, like, who are the people that invest into your, don't know if fund this like the the correct word, but like whose money you manage?
Speaker: And how do you how do get more of those people?
Speaker: the The structure of DarwinX is a regulated asset manager here in the UK. So a Darwin, which is the the the term we use, is basically an index of your strategy.
Speaker: So the DarwinX has its own internal risk engine, which again is just an algorithm that makes sure that a trader doesn't go haywire basically. And the the Darwin or the index of your strategy is basically your trades along with the DarwinX risk manager.
Speaker: um And that allows you under kind of a signal provider agreement to trade other people's money because the regulatory burden is sat with DarwinX.
Speaker: So investors can go allocate capital to strategies knowing that it's safe-ish because we're a regulated asset manager and we also have our risk manager that's present on every single strategy.
Speaker: um A lot of the capital that comes through the platform is our own. um We also have a master fund called the Index. which we've raised a lot of capital for as well. um And then you just have retail investors who ah you know like the platform and like allocating to you know managers like myself.
Speaker: So I currently have, I think it's $1.8 million dollars through DarwinX. The rest of my AUM is through boutique fund agreements um with other regulated funds outside of DarwinX. where I will just trade an account for them. And again, the regulatory burden sits with them and I just have a signal provider agreement.
Speaker: um In terms of getting more investors, it's It's really difficult if you're not regulated. Like that's that's the one thing. Being a fund manager is really difficult if you're not regulated, which is why you have to pay a premium and trade through funds that charge for that service, right?
Speaker: You know, it's not a free lunch. DarwinX takes a percentage of the fees um because they're providing you the the licensed fund route where you eat. With DarwinX, you don't even have to find investors because there's already investors on the platform.
Speaker: um But in terms of growing it, like my goal is just write about what I do You know, I have I have a YouTube where more recently I've been live streaming. I have a sub stack where I write about all my stuff in the hopes of, OK, if somebody if somebody likes what I do and, you know, wants to potentially have exposure, they have a path because they can they can find me on DiwanX and they can allocate capital to me. the The problem is if I ever induce an investment, you know, that then comes down on me because I'm not a regulated financial advisor or asset manager. um So you have to be super, super, super careful.
Speaker: um But yeah, that's that's that's the way I manage capital. um I also have some close friends who are in ah institutional cap raising. A friend of mine, Claudia Quintela, she used to work for UBS and Morgan Stanley, and she now runs her own company, Vibe Advisors.
Speaker: And she raises capital for institutional managers. So we're talking tickets of upwards of $100 million dollars and and and more. um And I had a conversation with her before and she just she called me a baby manager and it really hurt my ego because I was like, what do you mean? I'm a baby. Like um I was like, I've got i've got a year long track record. You know, i'm I'm managing like a million dollars. I can manage up to 20 million dollars. And she was are you just a baby? She was like the allocators I work with, they won't they won't get out of bed for under 100 million. And so I put a lot of time again, going back to the execution, I put a lot of time into, it well, how do I raise my capacity? Because most traders will just trade a system that they found and that works with zero regard for how do I scale this, which is a beast in its own right.
Speaker: So I put a lot of ah time, energy and effort into ah raising my capacity from an executional standpoint so that I can take more size. And I think i think now my estimated capacity based on my recent market orders is about $300 million. dollars But I raised that from about 15, 20 million before I started optimizing for capacity. So even at $15 million AUM,
Speaker: Sounds like a lot of money, but if you're taking performance fees of 15%, 20%, it might change your life. It's never going to put you in a Lamborghini and allow you to go and, you know, do do whatever you want. When when you're talking 50 million, 100 million, 200 million, which I am nowhere near, by the way. um That's when that's when you know you go, oh, that dude, yeah, he's ah you know he's got houses in every country, every corner around the earth and all that sort of stuff, any car that they can want. So there really is a different level um when when it when it comes to institutional institutional trading.
Speaker: How did you manage to increase the capacity like that? like What have you done to... how does one go about increasing the capacity of the strategy, sort like their portfolio in general?
Speaker: That's a great question. um the The obvious one is trade futures. So i i my portfolio is CFDs, which I know people are probably listening and go, oh like CFDs, gross. um But that's where I started and it scaled and worked very well. So i'm i I'm a firm believer of if it's not broken, don't fix it. I am in the process of rebuilding the portfolio optimized for futures because if if you have an institutional allocator who's ready to drop 100 million, they're not putting 100 million into a CFD broker. they They will put that into a futures a futures broker owned by themselves and you will access that capital that way.
Speaker: So I need to rebuild for for futures. That's the first and most obvious one. Second one is trading more liquid assets. So if you're trying to trade um cross FX pairs like the pound against like the great British pound against the Aussie dollar, your spreads going to be wild, your capacity is going to be low. it's It's not good for execution. Even crypto to a degree. Crypto is like very, very illiquid in comparison to traditional markets like gold.
Speaker: um You know, having your holding tines slightly longer. So scalpers are always going to have less capacity because they've got more market impact. They're more reliant on getting the best price. yeah um The things that I did was I removed strategies that weren't optimized for capacity.
Speaker: So like the S&P strategy that I mentioned earlier, I traded that live and it butchered my ah capacity because I couldn't take more capital on because all the capital flows through all the strategies that I have. And that was holding the whole portfolio back.
Speaker: um the The other thing that I did which made the biggest difference was I implemented an order splitter. So it basically TWAPs into the market. So it will send my first order. It will then delay the second split by three seconds and then it will delay the third split by three seconds. So if I'm trying to offer a contract, it will split that size into three separate positions and it will delay by three seconds or so. I found three seconds is is three by three. So three trades by three seconds is the perfect, most optimal contract.
Speaker: um balance in terms of raising your capacity to a level where you can really, really scale. Because three seconds doesn't sound like a lot, but on an on a ah an order book at a broker, that is a lot of time for um the book to fill back up. um But also you don't get hammered on your execution because if you're trading a breakout,
Speaker: and price breaks above a massive 100 day bar high, let's say, that's going to move very quickly. If you're delaying your trades by 10 seconds, you might miss 50% of that trade as it breaks out. So really, my execution is over six seconds.
Speaker: you will miss some you won't miss a lot of the move generally but i get three positions uh with three seconds delay so you have position one three seconds position two three seconds position three so it's six seconds across the three um but ah but you know again liquidity of the asset is is massive and the strategies that you trade as well like people People don't know this because they haven't looked into it, but generally a lot of strategies that you find online, their capacity is awful anyway, because if it's a popular strategy, there's a lot of people trying to fill size and often you're going to get squeezed and it's not going to fill at the price that you want. um But again, like if you're trading with, let's say your stop is 10 points,
Speaker: okay and you get slipped by two two points, you've been slipped 20% of your position. Now, if your stop is 100 points and you still get slipped by the same amount, so you still get slipped that that two points, it's only 2% of the position, which is a massive difference, right? So if if both of those traders get stopped out, trader A, who offered a 10 pip stop, has an extra 20% loss because they got slipped by two. Whereas trader ah B, who only got slipped by 2%, only has an additional 2% of their loss.
Speaker: So there's there's little things like that. Generally, you know, the the higher the timeframe you trade and the less that you're getting in and out of the market on highly liquid assets, the higher your capacity is going to be.
Speaker: But... Everybody should look at this as early as possible because the last thing that you want to do is build a strategy that has a track record. And then two or three years down the line, when somebody comes and goes, I want to give you 10 million to trade with and you go, ah, I can only handle 200 grand. It's like, well, that's not a good position to be in. So I think optimizing as early on for capacity, if you're looking to scale,
Speaker: you know If you're just looking to sling some money on on crypto or you're just looking to trade prop firms and not really you know scale and make a career, then it doesn't really matter. But if your goal is to become a better trader and scale, capacity is so, so important.
Speaker: Can you briefly explain to to a crypto noob like me and probably most of our listeners what the difference between CFDs and futures are?
Speaker: So futures futures, you're hitting direct to the market, right? So you're actually trading the contract that you want to trade. um Whereas CFD brokerage is kind of a synthetic, it's called contract for contract for difference. um And it's basically a set ah a synthetic contract built by that broker.
Speaker: So there will be central banks pricing the assets. That's what you get access to on the futures market. Then you get ah the primer primes or the liquidity providers who will price ah their feeds based on what the true asset price is. So they'll kind of compare that to a futures price.
Speaker: They will then charge the broker for their flow. So the broker then takes in traders, takes in investors, whoever, and they they offer their own feed. So they will take the feed given by the liquidity provider. They will mark up the spread so that they will add additional costs onto what the liquidity provider or the bank is saying, and they will charge that to you as the trader.
Speaker: The problem with that is you're bound to how much size that broker has. ah If, for example, um ah maybe I should explain a book versus B book. So there's there's two main ways that a broker deals with flow.
Speaker: Okay, and flow is just the the volume coming in from the traders. um There is a a booking mechanism where they send the real flow to the market. So if you're placing trades, generally that what they will do is they will step aside, they will charge and they will move it to the the live market.
Speaker: um the other the other what The other way that they deal with flow, that's generally for winning traders, right? And people who have proven themselves consistent um and become a problem if you're taking the other side. What they will do with the majority of traders on their books is they will be book them.
Speaker: um And B booking basically means I'll take the other side of the position. So if you go along with 0.1 contracts or 0.1 lots um on on an asset, the broker will fill you at that price and they will just hold the other side of the trade, knowing that eventually it's like a casino you're going to lose. um And that brokers make a lot of money doing that because, you know, most traders don't make money. um The third option, which is slightly a bit of a hybrid, is when firms internalize the flow. So let's say let's say me and you both go long and short on the same asset with similar size. They'll match our flow.
Speaker: So they'll put my trade against yours and they'll capture the spread in between. um So there's very little very little risk for them because if you lose, you pay me. If I lose, pay hi you. um That's a very simple way of looking at it. There's more complex mechanisms behind the scenes, but the broker doesn't care because they captured that charge or that spread in between. um And then they also make money from swap fees and stuff like that as well. You need different licenses per different ones. um But this is why people tend to pull their face at CFDs because you're very dependent on the broker's feed um and um the the quality of you know the the bars that they produce. Whereas the futures market, you can trade and you know you're getting the same price as everybody else. You know that you're not being manipulated in any any which way by the broker. um And not that the brokers are always out to get you, but you know you're adding another middleman who's taking a cut of what you make.
Speaker: Like I say, I still trade through a broker. I still trade CFDs because it works and I've raised capital doing it. But eventually I'm probably going to get to the point where to to break the next layer and to go into the next level of asset management, I will have to trade futures. Because if you walk into if you walked into Jane Street and said, ah i'm I'm here for a job, I trade CFDs, they'd probably ah boot you out of the top floor window. um So that's kind of the that's kind of the main differences. Me as a trader, I would rather be B-booked, meaning that the the broker takes you the other side of your positions because you tend to get better execution because they will fill you at the price because no order is going to the market.
Speaker: So if they quote you 120 for euro dollar, you're likely going to get 120 for euro dollar. Whereas if you're A booked, you get that real market impact because you've been passed through to the market.
Speaker: So I would rather be B booked, but I wouldn't be B booked because... um I have a track record so you can see that I'm not good to take the other side. um And generally what they will do is they will take a period of time where they B book you, they will manage if they're losing money or making money. And on the very small amount of traders that they lose money on, they will then move them to book.
Speaker: You don't know this from the front end. You don't know whether you're being B booked or A booked, but you can generally tell sometimes based on your execution.
Speaker: I see it. Interesting. how How many strategies are you currently running concurrently? And how do you make sure that they that they're like uncorrelated in a way because you're focused on minimum drawdown? What kind of drawdown do you have on your portfolio and how do you what do you do to achieve that?
Speaker: So I currently run 13
Speaker: or 14, one of the two, 13 or 14, but some of those strategies trade multi-assets. So some of them trade S&P and the NASDAQ, some of them trade FX pairs and indices.
Speaker: um But the way that I manage correlation between them all is there's ah there's a few different mechanisms. So the first one is i have asset limits. So I can only trade X amount of size on a specific asset, meaning that I can't do what happened in April when I said about the the hidden correlation shock that I had. If I've filled up the amount of available size that I give myself on an asset, I can't then put more size down.
Speaker: I have bi-directional limits. So if I'm long on euro dollar and another strategy fires long, I won't take that trade because, and again, this is all baked into the governor layer. Because if I take that trade, I'm not trading new alpha. I'm just doubling up the size on what I'm already trading.
Speaker: um I also implemented something which has sparked quite a few debates with people I've told about, but I give myself one bullet in the chamber um is is what I call it. ah And effectively what that does is that manages, if I've taken a losing trade in the day, won't then take another position. And the reason that I implemented that is because I want to know that my bad days are minimized. And over the past two years, analyzing all of my live performance, what I noticed um and the data told me was if I start a day bad, it very, very, very rarely recovers and becomes a winning day.
Speaker: So I decided to give myself one loss. If I take that one loss, I don't take another trade for the day. It took a little bit of a hit on the performance.
Speaker: So when I run all my tests and stuff, yes, the performance is slightly lower from just a raw return perspective, but the volatility reduces by like 100%.
Speaker: You know, the the volatility massively reduces. And there's been times since I implemented that change where I've woke up in the morning, I'll check my account, I'll see that I've been stopped out overnight. And I'm like, well, that's the day done. I'm not going to be in another trade today. so And it's a really nice feeling knowing that the worst your day can ever get really is one loss. And that, you know, you're not going to keep getting stopped out and keep getting stopped out and keep getting stopped out. um I do run my strategies at really low risk. So on my master account, my max drawdown is 1.4%. On my risk adjusted account, which is what investors see, my max drawdown is 5.83% with return over the past year and a half.
Speaker: um So it's nothing crazy, but for the level of drawdown that's there, like it's a pretty, pretty good risk adjusted return. um I'm actually, i'm I'm currently, I wrote a post about it yesterday. I'm the number one UK based trader at DarwinX who's included in our index, which is the master fund. based on annualized return and risk adjusted return which to me is like a massive deal right because not only do i trade darwin x capital i also work there and my job is to head up growth and as i started to scale and grow with darwin x i was like this makes my job 10 times easier because i don't have to convince somebody to go and trade at darwin x i am the trader at darwin x um
Speaker: So it was like a massive, massive achievement for me. And like I say, I get quite sentimental about about the work that I do because I put a lot of time and energy into building good things and focusing on moving forward and writing about it as well. You know, there's there's nobody who i really had to read their work or talk to when I first started. And it feels good to see people coming to my my DMs and ask me questions about some of the stuff that I've done. Um, you know, everybody likes to be recognized for the work that they do. And I, I feel like I'm finding a place where, where that's happening to me, which, you know, feels really good.
Speaker: Yeah. ah Sorry for going off on tangents, by the way, I tend to do that a lot. you um Don't worry, about that that's what the podcast is for. made me want to ask, like what do you suggest ah for people that are just um starting out now and want to like go down a similar path as you? And also, what are the the most common mistakes that people make when when they do something similar?
Speaker: it's I'm presuming when you say starting out and going down a similar path, you're talking about systematic trading and building algorithms and not just in general trading. um bit Yeah. if if it's just If you're starting out I would manage, try and manage your expectations as best as you can. You're not going to, you're not going to scale in the first year. You're not going maybe not even going to scale in a couple of years. But the the curve that you see with what,
Speaker: trading in general, but algorithmic trading specifically is it's this exponential curve that, you know, just seems like nothing's happened. and then all of a sudden the scaling happens and it happens really quick.
Speaker: Um, so is to manage your expectations and learn as much as you possibly can and implement it into test accounts, incubators, move it live. Um, you know, data is king and this is why AI is so fantastic. I've said quite a few times now ah to go all in on AI agents when it comes to building ah building strategies and portfolios, just because of the efficiency.
Speaker: um I did write a handbook. It's on my website, which is, you know, it's like an 80 page book on... starting in systematic trading. So selfishly, i'd I'd plug that and say that's probably the best place to start. But it is very specific for the path that I went down with the end goal of managing investor capital. Not everybody wants to do that.
Speaker: So you kind of have to figure out why you're doing in this. What are your long-term goals? Do you just want to be a trader or do you really want to scale and become a professional or institutional trader? Because it's two completely different paths. And if you want to be an institutional trader or ah I don't like the word institutional. I'm going use professional. If you want to become a professional trader, you need to be putting in habits and and routines very, very early on, as opposed to retroactively adding them, you know, two years down the line. um
Speaker: But I would just say, just learn as much as you possibly can. Read as much as you can about strategies. Have an open mind. So don't block anything from day one. um you know, and go, oh, well, I know that doesn't work because of X, Y, Z, because I'm going to be blunt and honest. Most traders don't have a clue. And most traders, at least that I've worked with, hold preconceived notions that are completely false.
Speaker: You have to relearn everything because you don't realize how much you don't know. um i I think with... In life in general, but specifically in trading, people build a thesis before they see the data and then they fit the data to to prove the thesis, um which we call in in in my world overfitting.
Speaker: um So you would say, you know, ah I want to trade this because I know it's profitable. You're then going to find a way to make a backtest profitable.
Speaker: You're going to tweak it and you're going to change things. You'll remove yeah days of the week where it underperformed. You'll double size on the days of the week where it did perform. You'll do anything that you can to prove that thesis right.
Speaker: And like I say, it's a people do it in day-to-day life. um ah But it has to be the other way around. You think of a thesis and, um oh, sorry, you gather data on a variable and then you build the thesis around that data.
Speaker: i did it I did it the other way around when I first started and hence why I was building rubbish systems. um But you just have to take yourself out the equation, forget everything that you think that you know and let the data do the do the work for you.
Speaker: Yeah, I think that's that's very good advice, especially for systematic trading, but also just like trading, maybe maybe even life in general, that you have to like drop your preconceived notions a little bit differently.
Speaker: I think we've we've covered a lot of ah topics here today. It was a little bit different than the usual episode because it's not as crypto-focused as usual, but I think it was very interesting. So I want to thank you very much for coming on today.
Speaker: and um yeah, is there anything else you would like to share?
Speaker: No, I mean, i i appreciate the invite. I always love um talking about what I do and hopefully providing some form of value to people. The only thing I would say is if, um you know, if people want to learn more about that, I do have a sub stack where i I put a lot of time and effort into writing articles. I release two per week. It's completely free. um You know, so if you do want to if you do want to hear me ramble and go on more tangents,
Speaker: that's the best place to find me um but dude it's been an absolute pleasure and you maybe we can rock it again sometime yeah for sure thank you very much
