Is Trading Fewer Markets Actually Better? ft. Richard Liddle & Gareth Abbot

[00:00:06] Gareth

That is, essentially, the diversification space, the risk space that we are building the portfolio inside of. Looking at it that way, we can visualize every aspect of how we are building our heat; the risk that we’re taking. We can see the correlation. We can quantify every aspect of it. We have various ways that we control the heat inside that space.

[00:00:38] Intro

Imagine spending an hour with the world’s greatest traders. Imagine learning from their experiences, their successes and their failures. Imagine no more. Welcome to Top Traders Unplugged, the place where you can learn from the best hedge fund managers in the world so you can take your manager due diligence or investment career to the next level.

Before we begin today’s conversation, remember to keep two things in mind. All the discussion we will 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.

Here’s your host, veteran hedge fund manager, Niels Kaastrup-Larsen.

[00:01:33] Niels

Welcome to another episode in the Open Interest series on Top Traders Unplugged hosted by Moritz Seibert. In life, as well as in trading, maintaining a spirit of curiosity and open-mindedness is key, and this is precisely what the Open Interest series is all about. Join Moritz as he engages in candid conversations with seasoned professionals from around the globe to uncover their insights, successes, and failures, offering you a unique perspective on the investment landscape. So with no further ado, please enjoy the conversation.

[00:02:09] Moritz

Hi, everyone. I’m Moritz Seibert, and I’d like to welcome you to another episode of Open Interest on Top Traders Unplugged. Today, I’ll be speaking with Gareth Abbot and Richard Liddle from Bowmoor Capital, a UK-based trend following firm. I’ve really been looking forward to getting them on the podcast because truth be told, I had really never heard of Bowmoor Capital before last summer when a friend mentioned it to me for the first time.

I got really, and I mean big time, really intrigued about how they manage their portfolio, the small number of markets they trade, and of course, their very good performance during 2023, which, as we now all know, wasn’t an easy year for most trend following traders. So, Gareth, Richard, welcome to Open Interest. It’s great to have you guys on.

[00:02:56] Richard

Thank you. Thank you, Moritz.

[00:02:59] Moritz

I was just mentioning that 2023 was a tough year for trend following trading strategies. We didn’t record this, but when we just got on to the recording platform, we were speaking about the start of the year and I was saying there’s a bunch of funds that are up 3% to 4%, and some of the industry benchmark indices, they’re maybe slightly positive now.

I don’t want to steal your thunder, you just mentioned to me you had a fantastic start to the year. So maybe without too much of any specifics, tell us how your start to the year was.

[00:03:33] Gareth

We don’t need many positions in this compact portfolio to drive performance and are in a pretty good place with three fairly strong positions driving performance close to 9% so far. So yeah, we’re enjoying this month. On the other side, typically, we can have a majority of small losing positions and that’s smaller than usual. We don’t have many.

One position in particular, a small position in nat gas, a short position that’s offside but not particularly big either. So yeah, we’re benefiting really from three positions across the portfolio, one in the stock markets, one of the metals, and a soft as well. Good to put detail I suppose on the these, but yeah, some of our bigger positions, they’re going well.

[00:04:37] Richard

It’s a good start to the year because like you said, about 2023, we did have a decent year but finished, unfortunately, it was a hard December. So actually, it’s quite nice to bounce back with a really strong so far nearly 9% January following that December and start the ’24 year with a really positive month.

[00:04:57] Moritz

As I always like to say on the 23rd or 24th of January, we have a week still to go. The month is still very young. A lot of things can happen even in the last trading hour of that month. But hey, kudos to you. Well done. I think you mentioned short palladium, long cocoa, long the Nikkei. These are all good positions and they’ve been driving your performance, so congratulations on a good start.

Now, we immediately jumped into performance but what I really wanted to do to start out is your backgrounds, which I thought was fascinating because Gareth, you have a background as a quant and a mathematician, which in our industry isn’t all that unusual. And Richard, you have been, used to be, a Royal Air Force jet pilot, and that is the interesting piece.

Don’t tell me it hasn’t been exciting that your new role as CEO of Bowmoor Capital is even more exciting than that; maybe it’s safer. How does one move from flying jet airplanes in Iraq to running a UK-based trend following trading firm?

[00:05:58] Richard

There’s a lot of skills crossover which, on the surface, may not be obvious, but there are a lot of those discipline skills that I bring forward. To be perfectly honest, yeah, I spent 20 years as a pilot. I flew helicopters for 11 years and then jets for 10, ending as a flying instructor.

We have a very well-trodden path within the British military into either commercial aviation or into city jobs with the like of the large institutional banks. And I kind of made a decision, early on, that no civilian flying role is going to have the level of excitement that I had with those 20 years. I had a fantastic time, but it was time for a new chapter.

I wanted to make that transition into the city, into that traditional world, which we did have a well-trodden (what you call the golden handshake) golden handshake into the likes of Barclays, which was the one that I was looking into. Through a choice, shall we say, series of events, I was introduced to Gareth and Brendan, the trading team of now Bowmoor (they weren’t at the time), and a plan was formed.

A plan was formed to take the strategy, which is Global Alpha, under a new umbrella, under a new fund structure, a new corporate entity, which became Bowmoor Capital. We put a very tight core team together, specialists in their individual fields with a very clear goal of making the strategy the sole strategy and allowing it to be open to far more investors than potentially it previously was.

I took that risk rather than going for the well-trodden path into the city and having a Barclay’s title next to my name, I thought, well, actually, my entrepreneurial spirit, which I’ve always had through many other ventures, I would rather lead this team. The power of the strategy is evident. The power of the team and the trading desk led by Gareth was evident.

And really, my role was very, I won’t use the word easy because I think distribution and raising funds (as we probably all know) is one of the hardest roles of getting a new fund off the ground. The strategy speaks for itself, but my role became very much driving distribution, driving the setting up of the fund structure. We were previously a Luxembourg structure, then moved it into a Guernsey fund structure with all of the complexities there.

So we were a new emerging fund even if the strategy was a legacy strategy, shall we say, and had been running for a number of years. So that challenge really appealed to me. I think it was in that challenge that I leaned upon the previous military experience, the handling of the pressure, the handling of those challenges. And it was using that experience in a completely different environment.

No longer was I sat in a cockpit at 30,000 feet, but I was actually managing the same level of stress and pressure, though, sat behind a desk; to get a new fund off the ground in a new structure and drive AUM, in it, to our lofty targets which we still have. So a lot of, shall we say, crossover skills applied.

The challenge, I would say, has been, again, on a par, and I thrive in it. I think what makes it exciting is not only the strategy. The strategy Gareth runs is phenomenal and really is market leading. But the team we have is small, highly focused, and highly dedicated. I think that really does make working with Bowmoor an absolute pleasure.

[00:09:43] Moritz

Fantastic, Richard. Now, Gareth, your experience doesn’t come from a 30,000-foot view from a cockpit. I think you are more the mathematician by background, at least this is my understanding. The question I have is a lot of quants, smart mathematicians like yourself, they go into all sorts of areas and finance like high-frequency trading and stat arb and derivatives trading, options, all that.

Why trend following? What piqued your interest? Why did it become trend following for you?

[00:10:16] Gareth

Not quite as exciting as Richard has just given as a background but, like you say, it’s quite familiar in our industry. My background is pure mathematics. It was a big theme in our house growing up. My granddad was a mathematician and I really got into it. My granddad, he was in the shipping industry back in the 1950s and ’60s. He was a naval architect, also a very enthusiastic mathematician, and often, that’s how we would spend our time.

Mathematics, solving puzzles, was a big theme in the house, that and sports. So yeah, growing up, it was always, I guess, a passion of mine. Between mathematics and sports, that’s really where my interests lay at that point. I think moving toward trend following, I was lucky enough to have the opportunity to just follow my curiosity. I didn’t really have much career ambition at university. I was just enjoying doing pure maths.

That was the thing I guess that I knew I did want to do. I hadn’t settled on an application, like a lot of my friends at the time, with ideas about doing something commercial, engineering it may have been. For me, it was just enjoying mathematics and that’s what I was studying, pure mathematics.

It really came out of the blue, the job offer in my final year, which took me on to a trading desk. It was a set-up. I was given a lot of freedom, a good budget, and discretion to set up a trading desk and set up strategies and really drive the performance. It didn’t take me long to find trend following and immediately really fall in love with it because I realized that it really is like mathematics.

It’s very pure. It’s price. That’s what we’re looking at. I guess, as well, that there’s then a lot of consistency between what I had been studying and the type of mathematics that I love, and trend following, this pure analysis, pure trend, were just price analysis, following the price, and all of the mathematics applied directly to the price series. So I felt, for me, that it was a really comfortable area that allowed me to continue thinking that way in the world of pure mathematics.

There are many great challenges for mathematics in the universe. Beating the market has been my favorite challenge. That’s not to say that I don’t have other interests. But beating the market, I think, is one of the greatest challenges for mathematics and I’ve had a lot of fun over the last, coming up, 25 years now facing that challenge.

[00:13:55] Moritz

When you say it’s a challenge for mathematics, it almost sounds like there is a solution. At the end of the equation, that’s the answer. That’s the solution. That is how you beat the markets. But I guess that’s not exactly like that, is it?

[00:14:10] Gareth

Well, I guess this brings in the whole concept of having a robust approach. There’s not the perfect answer; the idea of the holy grail. It’s having that approach, which can endure through thick and thin, that makes sense. It ought to work when the conditions are favorable, and it ought to manage unfavorable conditions. So yeah, for me, there is an answer.

In our system design, that theme of pure mathematics runs right through the system. The design around all aspects is functional rather than iterative. We avoid optimization, estimations, iteration. And yeah, I’ve approached it with my own strengths; pure mathematics and looking at everything from a functional analysis point of view and finding exact solutions.

[00:15:27] Moritz

I guess keeping losing trades small and letting the winners run is part of the answer to that equation of solving the market puzzle. But, Richard, maybe let me bring it back to you. You’ve mentioned your fund and the launch in Guernsey or the migration from Luxembourg to Guernsey.

Maybe tell us a little bit more about that Global Alpha Fund, I think it’s called, which launched in the summer of last year. What does it trade? How does it trade? Is it pure trend? Just the high-level summary of that product?

[00:15:57] Richard

Absolutely. Again, we are a single strategy fund, so it is exactly as you say. It’s the Global Alpha Fund which is run out of Guernsey under Northern Trust. We trade 21 markets. The intricacies and the whys and the wheres of the 21 very much I’ll leave for Gareth, but we trade only 21 markets of pure trend.

Unlike our competitors and our peers who may trade hundreds of markets, we are the anomaly, shall we say, with only 21 and I’m quite sure that’s where this conversation will lead because it is a bit of a USP of ours. Once Gareth explains the why we generally find investors and those knowledgeable in the space, actually, their eyes generally open to think, well, actually, yes, I can now understand why you’ve got so few markets versus the hundreds others may trade.

So yes, we trade 21 markets. We cover all sectors of the global economy from the metals to stock markets to softs, etc. We could go into the specifics, if you wish, of those 21. I’m more than happy to list them off. But ultimately, we work at around what we refer to as a MECE portfolio construction where we’re mutually exclusive but collectively exhaustive, hence maximizing diversification.

Simply put, it’s those carefully selected markets in an optimal number of markets to cover, like I say, every sector across the global economy. They’re highly liquid markets and they’re selected exactly for that. So they are the 21, shall we call them, most liquid markets, meaning that we’re never going to be left in a position where liquidity for the fund may become an issue.

To briefly cover what we’ve got, we cover indices. So we’ve got Euro stocks, Dow, and the Nikkei. We cover interest rates, so we’ve got the SOFR, the Euro bond, and the US long bond. We’ve got currencies. We cover British pound, Canadian dollar, Aussie dollar. Metals: gold, copper palladium. Energy: crude, heating oil, and nat gas.

We cover grains, corn, and soybean, softs, cocoa, and coffee. And finally, livestock, live cattle, and lean hog, and then obviously, the portion in cash. So they are the 21 giving us, like I say, that MECE portfolio construction.

[00:18:31] Moritz

You teed the golf ball up for me very nicely. Thanks for that, Richard. By the way, I just listened to the markets that you’ve mentioned there. Wheat is missing. There is no wheat contract you’re trading. Or did I miss that? I’m not sure. Look, this is the thing. This is exactly where that conversation is going to go and hope it’s going to be interesting.

Blackjack 21. Why 21? Why not 22? Why not 20? Why not 30? Why not 40? How did you come up with this MECE, mutually exclusive, collectively exhaustive (I think you called it) group of markets?

[00:19:09] Gareth

There’s asymmetry in the design and a good balance. With the 21 markets that Rich has summarized, we cover all the sectors. We only do the two grains. We do corn and soy. We left wheat on the subs bench. We brought the two softs since. We do coffee and cocoa and two livestock, cattle and hogs.

Adding a third in each of those sectors sure can be done and there is liquidity. But we feel that we have the balance with the nine out and out financial markets, nine out and out commodity markets, and the three metals which make up the overall 21 proxy both because we’ve got a blend of precious and industrial. So we have that even balance, this unequal allocation to financial and commodity.

Touching on the MECE theme, this is really powerful for us because it’s not just the portfolio. The strategy is a MECE trend following strategy and, no doubt, one of our strengths. The strategy is driven by MECE system and portfolio construction. So it’s not just a portfolio designed to minimize correlation and maximize diversification, which is quite simply what that means in this case; minimum correlation, maximum diversification.

We have a broad and comprehensive approach to trend following beyond just going out in the markets. Our system-driven process is portfolio construction. We combine a highly efficient trend capture, which I think we’re going to talk about, with the traditional conservation of capital to produce our edge. This is what drives the mathematical expectancy which any strategy requires to be profitable. Positive mathematical expectancy is required.

Quite simply put, profits must frequently outweigh losses. And in the case of trend following, that’s typically a low win rate and a high profit rate. The greater the efficiency of the trend capture (again, we’ll hopefully come on to this), the greater the mathematical expectancy. That’s one of the cornerstones.

One of the foundations of the system is the efficiency of the trend capture which, as I say, drives the edge; the mathematical expectancy, which is key. That is ultimately the key. With that in place, it comes down to how to back your edge and how to manage risk, essentially.

On the portfolio side, first of all, the idea of the MECE portfolio we’ve touched on there, it’s a highly liquid construction as Rich describes. The portfolio constituents are the highest global liquidity major macroeconomic markets. It’s a carefully selected minimum number of markets for the portfolio. Why minimum?

Well, in order to value the impact of every trade and avoid dilution of the system, we want to value the system, value what we do, value our trend capture. We cover every sector, every region around the global economy and, like I say, it’s equally weighted across financial and commodity markets.

But with this compact diversification in the markets and such a strong internal diversification with the system strategy, the reliance on even just a single market trend to drive portfolio performance is very important to us. One strong trend can generate all of our returns while we simply deal with the background noise.

Positions are weighted to offset and normalize daily movement across non-trending markets like a noise-canceling effect in the background. So we build our positions in these normalized units so that a good day in a stock market can offset a bad day in a soft market. What that means is that we achieve this noise-canceling effect.

Non-trending positions will cancel one another out while performance is driven by our strong positions or the trends that we capture. Like I say, just one can drive the entire portfolio performance, which is something that we value greatly.

[00:24:17] Moritz

Let’s maybe take this in two steps. We’ll definitely get into the efficient trend capture and the things you do there, Gareth. Thank you for the explanation so far. Now, you’ve mentioned, or I think you’ve mentioned (correct me if I’m saying that wrong) that maybe more markets might be diluting your returns.

Now, you as the mathematician, we understand that there is a marginal diversification benefit to be had by adding additional markets to the portfolio. So if you increased it from 21, to 22, to 30, to 40, to 50, all the way to 100 or more, you would have an additional diversification benefit which is there for you to grab. You can have that.

It would theoretically improve the risk-adjusted returns, which means you can trade your portfolio in a different way, size positions differently. You decided to not do that. This is kind of difficult. You said it’s a USP, but it’s difficult to square that up mathematically.

So maybe explain to us why trading 21 markets is better than trading 100 markets. Or maybe start with, I presume you have run your system unchanged on more markets, just add them to the system, add them to the code and let the system run. What happens? Doesn’t it perform better?

[00:25:49] Gareth

No. We can improve Sharpe if that’s important. What’s most important for us is the return profile of the system. We’re looking to double capital at every economic cycle in a robust way. So every three to five years, we want that to be a double rate. We want to double capital every three to five years and are quite happy to see a maximum drawdown of around about 20% through each cycle.

So I’m not all that concerned with Sharpe ratio. In many ways, it doesn’t make sense to me, this combination of risk and reward. I like to look at them separately. I like to look at return and risk in many ways and separately. There are very few metrics that I feel combine both these concepts well. We do look at some Z scores ourselves, but I don’t really value Sharpe ratio.

That’s really the only thing that I can see improve, at the point where we go for that scale of diversification in the market. It’s not the only way to diversify. For me, it’s not the most valuable way to diversify either. We value the internal system diversification more than just adding markets. This is what we call smart pyramiding or whole trend. I could perhaps try and explain it in a couple of ways.

[00:27:34] Moritz

Oh, yeah. Especially the smart pyramiding. That sounds very intriguing.

[00:27:38] Gareth

Yeah. Pure trend is out there. It’s pretty well understood. There are different interpretations of it. For us, it expresses our 100% commitment to trend following. We know that trend following price trends are a clear reputable source of alpha for a robust system. Price trends, the trend alpha is systemic. Its base of trend following works because it’s based on how the economy works. There are few solid sources of alpha other than trend alpha. Like I say, it’s systemic. It comes from our economy works. The idea of whole trend, this expresses the knees deep commitment to maximum trend capture.

To maximize the mathematical expectancy as far as possible with the most efficient trend capture that we can achieve, we’re looking for maximum price movement and profit across all markets. We’ve committed more to that area, to that efficiency of achieving the maximum price movement, the capture and profit across the markets we trade, before we just add more markets with a multi-strategy, multi-frequency system. This is the idea of the smart pyramiding.

So we’re casting as broad and net as we can across the markets that we trade. We trade multi-strategy, multi-frequency; four strategies across four frequencies. These are established industry methodologies that we’ve upgraded. They’re based on simple mathematics, all trend following, all in different ways.

This composes an internal diversification space within the system to diversify and pyramid every trade. So. trailing stops is a good example. We use stop losses but not stop profits. The exits are consistent with the entries and are determined by this smart pyramid.

So the diversification that we can achieve this way with the smart pyramid, we have the four strategies on four frequencies, gives us 16 individual strategies all trend following, all in different ways. There we have the diversification to capture different price structures, the many different types of trends in all these markets.

[00:30:31] Moritz

That is probably what you mean with the smart pyramiding. As you have said, it is this combination of systems. You have I think a total of four model types, if you will, and then four timeframes. The total of that is 16 systems. When you combine them, you essentially have this layered approach which we could say pyramids you into position.

But it does not say, according to the original definition of a pyramid, whereby you would add to a winning position and just become larger and larger and larger, maybe adjust the stop, you would simply trade another timeframe or have another model come in. I’d like to really speak to you about that, and to you Richard as well, we’ll tie you in a sec.

Here’s just a final thing on the markets and prediction around markets. You’ve made the decision to trade 21. You’ve made a decision to trade, say, the Nikkei index I think you’ve mentioned at the beginning of the show, and not the DAX index and not the FTSE index. Why is that? Did you look at the performance?

There’s a background to my question. So, the background is, if you analyzed your return stream and you found that there is a set of markets that have better and more persistent, statistically significant and persistent trending properties, then you could arguably say that you only want to trend follow these markets.

Say the Nikkei is 2x or 3x better in terms of return output than the DAX index, then you could say that you don’t want to be trading the DAX index at all and you will be committing all your risk budget to the Nikkei index as a result of that.

There are some market participants, trend following traders in our industry, who do that in the alternative market space, deliberately removing the traditional markets from their portfolios because they say they have inferior trending properties and they replace them with these alternative markets where they say that they have higher Sharpes or higher risk-adjusted trend following returns.

I respect that decision, but that is also kind of like making a prediction that that persistency will continue in the future, that what you have observed there will also be true going forward. That, to me, is always a difficulty with these predictions in general because I really don’t know which of these fantastic markets that we trade are going to be throwing us off these great trading opportunities. It might be cocoa, it might be sugar, it might be the DAX, it might be, for once in a while, the FTSE index. Who knows?

So to what extent did that play a role when you looked at these markets? How did you decide which ones to exclude? Is there kind of like the thinking there that the markets that you trade, they just had maybe produced the best returns in the past and therefore you select them for your portfolio and you threw the other ones out? How did you form that portfolio when you view it through that lens?

[00:33:47] Gareth

We didn’t in any way optimize any aspect of the strategy, the system, or the portfolio. There’s no question we could select 21 markets which would massively outperform what we have here, but that’s just not the way that the system has been designed. Through testing, we’ve always required that the system would perform given a reasonable sample of markets.

The system will perform well. Going back to the Sharpe ratio, that might not hold up so well if that’s important, but given eight markets, just a proxy from each sector. So that’s always been a requirement through all the system design and testing is the system must perform given a reasonable sample of markets, and will do.

If we run the system on one through sector, eight markets, performance is good. Even fewer markets than that if we get some sort of diversification between financial and commodity. We wanted to avoid the portfolio looking like we had cherry-picked as well, which is why there are no surprises in there.

For example, in the stock markets, we’ve gone for the multi-reaching major markets. So we’ve gone for Euro stocks for Europe. We’ve gone for the DOW in the States. That could have been S&P. NASDAQ would likely outperform them both. We’ve gone for Nikkei in Asia, and typically Hang Seng would outperform there. So we wanted to demonstrate the power of the system, which we feel this portfolio does very well.

There’s no question, like you say, that there are bigger returns up for grabs in some of the smaller, more obscure markets. A good example would be Bitcoin. There’s no question that the system likes Bitcoin. If we throw Bitcoin in there instead of something like Canadian dollar, yeah, that has a big performance impact. Maybe that’s for a later product. Maybe we will take that view and have a more risk-seeking product in the future.

But the purpose of Global Alpha is to have a very solid product, it’s all major market, and demonstrate the system capability. Like I say, the system will perform given a reasonable sample of markets. It doesn’t need to look for that extra alpha that is out there in many of these less liquid markets.

[00:37:05] Moritz

Sure. We can never be sure, and neither can you be sure that you have the best set of markets because you very likely will never have the best set of markets. So, in your case, a simple test would be if you replace the Nikkei or the Euro stocks or the Dow, I think you’ve mentioned, with whatever, the Australian stock index, the FTSE, and the Canadian stock index, and you replaced your palladium with platinum and your heating oil with gasoline. If you have another set of 21 markets, performance will be different. But what you’re saying is no matter how you slice and dice it, every single one of these portfolio combinations, over time, will have a positive expectancy.

[00:37:46] Gareth

All I feel strongly about, in that regard, is that we give some consideration to the diversification. So again, just sticking with the stock markets, the Euro stocks is in there as a proxy for Europe, the DOW for America, and the Nikkei for Asia. So like for like changes just to maintain the diversification.

[00:38:11] Richard

We do monitor and track your well over 100 markets. To swap in and out is not something that is taken lightly. But we have looked at it in the past, and we are currently at present as well, where we look intricately at one specific market and decide whether it’s better placed to fill that slot in the portfolio.

What Gareth mentioned is that, on my aspiration certainly, we are currently testing what we refer to as a wildcard; so again, 21 less liquid, more volatile markets that could offer something different to an investor.

But core Global Alpha, like I’ve mentioned, that we are a single strategy fund. I would like, in time, potentially this year, to launch a higher volatility, a higher vol product on less liquid markets, to offer the investor (like Gary said), those risk-seeking investors (potentially people who’ve already been invested in our core offering) an opportunity to step into something that offers a little bit more potential reward but also is a little bit more volatile.

That’s on the whiteboard. We’re looking at the markets that we would use in that. That’s quite an exciting project that we’re working on. We’ve carried out a lot of testing on it. Hopefully, we’ll be looking to launch an internal managed account shortly to put that live, and then who knows? Bowmoor might have that as an offering within the next 12 months.

[00:39:45] Moritz

Well, interesting. Looking forward to that.

Final point on this, to either Richard or Gareth, but when you think about the equity markets, you’ve mentioned the Euro stocks, and you trade the Euro stocks as the proxy for the European equity markets. Say you risk X on the Euro stocks, now you could trade the DAX index, the CAC 40, the IBEX, the SMI, and say the FTSE, these are five stock indices from Europe.

You could now risk one-fifth of X, your original risk budget of X, risk one-fifth on each of them. That would give you the same exposure but at slightly different points in time, because there is this bit of a diversification benefit that does exist between these markets. So IBEX is, I guess, performing a bit stronger than whatever the Euro stocks this year or the last couple of months, I don’t know.

That to me is something that’s out there that you could take that wouldn’t be creating a dilution to your returns because you’d still be getting on to these big trends and you’d still be risking the full amount, eventually, the full X. You see the point?

[00:40:59] Gareth

Sure. As we consistently take that approach across the portfolio and trade more markets in each sector, then we do start to dilute the impact of an individual market trade. That’s not to say that approach is not effective. Trading hundreds of markets can work, can have a good risk-return profile. It’s not for us. It’s not for us in this product.

Like I say, we really value the internal diversification more than just adding markets and we value the system, what we do, and the impact of an individual market trade is very important.

[00:41:51] Moritz

That is the other thing. I think you have put a lot of value, and I do agree with that, on system diversification. You have diversification that stems from the market selection, and you have the diversification that you get from trading across different timeframes and different system types. And maybe, Richard, we’ve already touched on the 16, maybe you can give us a bit more of a high-level overview of the timeframes and the model types that you run there.

[00:42:21] Richard

I’ll probably leave the model types to Gareth, but the timeframes we look at are not optimized in any way. They are simply based on Dow Theory being a month, a quarter, half-yearly, and yearly being the timeframes. The reason we don’t optimize them quite simply is we may optimize it, but it’s only optimized until the time of a further optimal may be present.

Therefore, sticking to the Dow Theory timeframes is something that, for Gareth, has been very key. Now, those four time frames are then coupled with four models. They are the models which are used for the entry and the exit signals. Gareth, high level, do you want to touch on the four models that you use, or the four indicators that you use coupled with those four timeframes?

[00:43:12] Gareth

Sure, yes. Again, we’re on this idea of whole trend and smart pyramiding. We’ve spoken about the systemic source of alpha; the trend alpha. Most of the original pure trend CTAs were driving out by panic reaction to market conditions, like 2010, and were victims of their own huge success in many cases. Trend following was easy money in the early days for over 40 years. All of the classic single strategies had proven robust, individually.

In fact, every single global alpha strategy had outperformed on its own. This is very important to combine the two. When conditions get tough for a pure trend approach is when whole trend gets going and the combination will deliver the alpha, the clear repeatable alpha, the edge that we need. Like we say, the diversification isn’t just across markets. We have a 3D dynamic diversification.

So there’s a smart, precise position weighting. This is based on real mathematics rigor. This is where it comes in. We touched on the market strategy a moment ago. These are the three key components of the system: the portfolio allocation, the risk management, and the market strategy.

The market strategy is based on pretty simple mathematics. The portfolio allocation we’ll talk about just now, and the risk management is where the rigor comes in. That’s where some real grown-up mathematic comes in. So the 3D dynamic diversification, this is dynamic allocation across a three-dimensional diversification space, which is like a prism.

The dimensions are multi-strategy, multi-frequency, and multi-market, which we’ve touched on. The multi-strategy theme is very different established trend following methodology. I’ll explain about these. That’s really, I guess, what the question is at the moment, as well as multi-frequency. This is the Dow Theory, primary, secondary, and minor trends.

So across the multi-strategy diversification, we’re running four main trend following strategies, all, like I say, trend following, all in different ways. Two of them are breakout event-driven strategies which are very price sensitive and will react at a signal; will trigger based on one significant price movement. The other two are evolving autoregressive strategies based on lags; so less sensitive to one individual price sample based on the analysis of lags.

In combination, this allows us to capture many different price and trend structures. Because, as we know, they’re always different. Even within the same market, one trend to the next can be very, very different. They can be similar. You can see similar patterns across markets, but you have to be prepared for one trend to the next being very, very different. It could be very aggressive, very volatile.

Most of that movement could take place in a very volatile way over a short period of time. In other cases, these trends can be very stable and slow and evolve over many years. With the four very different strategies, we can capture many different types of trend structures. These four strategies are then applied over four frequencies.

In line with Dow Theory, that’s just a consistent application of Dow Theory. We apply the strategies over a primary trend timeframe, twice over secondary trend timeframes, and then on a minor trend lookback as well. None of these lookback periods are optimized in any way.

I spoke earlier about the avoidance of optimizing in any aspect of the system, and this is a good example. We haven’t optimized any of our lookbacks. We’re looking at the lookback periods which are consistent with the economy that we work and live by. So we’re looking at a year to multi-year, we’re looking at six months, we’re looking at three months, and we’re looking at a month; quite simply, calendar lookback periods.

In combination, this is how we achieve the smart pyramid. We can pyramid in and out of positions with this combination of multi-strategy, multi-frequency trading. The exits are consistent with the entries. We don’t just trail stops. We only use stops to stop loss not to stop profits. So we limit the downside, limit the upside, and exit, which always seems to be an interesting talking point about the system.

The exits are consistent with the entry. So if we have entered a trade based on a Donchian channel or a moving average cross, then that’s how we’ll exit the trade. The stops will be placed simply to limit the entry risk, the loss of the trade, not to take profit. In this way, we achieve the outcome of the smart pyramid, and we can capture the many different types of trend structures which simply trailing stops doesn’t do very well; which doesn’t allow the same freedom, which logically is consistent with how the entry decisions were made. If you enter based on a linear slope, it’s sensible to exit based on that same slope. Like I say, we leave the stops just to limit loss; the entry risk.

[00:50:15] Moritz

Coming back to the smart pyramiding, I guess when you say run each model, or each system, on one-sixteenth of risk, you would have that result, because every model has a one-sixteenth footprint, if you will. Some trade faster, some trade slower, and then over time, they would add to an established position, and they would also remove from an existing position given the fact that they’re trading across different timeframes.

But you’ve also mentioned, Gareth, the grown-up mathematics, they come in within that prism of markets diversification, timeframe diversification, and model diversification. So that suggests that is not a simple linear addition of one-sixteenth of risk but there’s something else going on that you do with a portfolio as it’s being built or as you’re holding these positions. What’s that about?

[00:51:13] Gareth

Yeah, we spoke about this before, didn’t we. I think this is perhaps a different way about looking at risk which we can get right into. Let’s give it a go and see how far we go with this. Our view here is that global alpha can stand the heat. So our distributed risk, our heat, is managed inside a 3D risk diversification space like a prism; the risk brick, which we’ve already just been touching on there where we have that the space is formed in the three dimensions.

We have four strategies, and four frequencies. So that gives us a surface of 16 points of diversification and those points are the units which we build the position with. That’s how we build our heat. That surface is then applied across 21 markets, so it creates this prism. If we are looking at the front surface: four strategies, four frequencies, the 16 units there; and then the depth of 21 markets gives us a diversification space with a volume of 336 units.

These units are the building blocks. The portfolio is constructed to precision scale within the three dimensions of strategy, frequency, and market, containing heat; a risk space within strict limits of risk, equity, and margin, and with infinite capacity (in theory anyway) as the position unit size contend towards your contracts.

So we apply the heat risk controls, in this risk diversification space, to improve risk-adjusted return. Like I say, we construct the portfolio to precision scale within this three-dimensional space. So I suppose it’s a slightly different way of looking at it. We look at the efficiency of the space. Like I say, with the dimensions of four strategies, four frequencies across 21 markets, there’s 336 units in there. That’s the volume in which we can build the risk.

The surface area is 368 units. So we’ve got a very high efficiency of the space to begin with. There’s a 91% heat efficiency there, when we compare the volume to surface area, which would be much lower in cases where a system has less internal diversification.

If it’s a single strategy approach, trading one type of trend following strategy on one frequency (whether many CTAs would take that approach or not), but that would be a very inefficient space; risk space with a much lower vol, down to closer to a 25% efficiency number comparing the volume to the surface area of that diversification space.

So, it then becomes about how we control the heat inside that space. That is, essentially, the diversification space, the risk space that we are building the portfolio inside of. Looking at it that way, we can visualize every aspect of how we are building our heat; the risk that we’re taking. We can see the correlation. We can quantify every aspect of it. We have various ways (and I’m quite happy to go into these ways) that we control the heat inside that space.

[00:55:35] Moritz

Let’s do this a little bit.

Before we do that, Gareth, we need to define what heat is. I think it comes from Secoda and Druse. Heat is, essentially, the risk or the equity at risk initially, and then also (depending on how you view it) as the trade develops, so the risk to stop, if you will.

Is this your definition of it as well? What do you mean when you speak about heat in the portfolio?

[00:56:04] Gareth

It’s the total distributed risk. So for us, the positions, the total distributed risk in strict 3D limits, the positions build accurately, and the risk to scale in market generic normalized units. So it’s how we distribute our risks, in terms of building in the units.

I guess an example of that would be with the 16 units provided by the four strategies across the four frequencies. At the point where all 16 units build the position, we call that a fully loaded position. We have one at the moment, the position we have in palladium, we have a short palladium position which is fully loaded. So all 16 units are short. In that case, the four strategies across the four frequencies are all short.

In that sense, that’s how we view the heat within this diversification prism. We look at how we’re building units across the portfolio. In the case of palladium, that would be a full layer.

[00:57:28] Moritz

So I presume palladium now has a lot of heat in your portfolio relative to say, I don’t know, nat gas I think you’ve mentioned initially, where you have a little bit of an offside position, I think you’ve used that word. Is there like a dynamic or ongoing adjustment of that heat going on between markets in the sense that, say, the palladium heat or the palladium equity at risk to stop becomes too large, you would reduce that position to control that heat, reduce it and move it somewhere else, or is that something that you would refuse to do?

[00:58:11] Gareth

There’s a combination of both aspects there. So whilst the position in palladium itself is built by the smart pyramid, and in this case, all 16 units are short, at the same time the risk per unit is controlled in a simple way by the bet size itself. So there’s a bet size within every unit.

The unit equates to a determined bet size. This calibrates the risk per unit. One of the aspects there is the very familiar methodology of volatility normalizing. So if, in the case of palladium, the market became more volatile, then the unit size, the position size of that unit would decrease to uphold the target return.

So it’s the same idea, and it’s ATR based. These units are volatility normalized so that the outcome, or the target outcome in any event, is that a unit in palladium will move the same each day, on average, in dollar terms, as a unit in gas. Albeit those positions are different sizes at the moment because, like I said, in palladium we’re fully loaded short 16 units, and the short position in gas is only 5 units. But those units are volatility normalized in a very similar way to the methodology that many people will be familiar with, just ATR-based volatility normalization.

[01:00:08] Moritz

Sure. It normalizes expected losses through the initial position sizing. What we don’t know is the outcome to the upside because we leave that open. An interesting pointer is, when you initially size the position versus the ATR so that they have that same loss expectation, that is one bet. That is one hand that I’m playing at that blackjack table that the markets give us.

I kind of like encapsulate that bet and leave it alone, which means it’s either going to be stopped out or it will find its exit at some point in time. But an interesting question is always, is there added value or an edge that you can add to that initial position sizing by dynamically adjusting that initial position size over time, as the trade progresses, as a function of changing ATR, changing volatility, changing correlation, changing distributed portfolio, all that type of stuff.

It is an interesting point because these methods, these overlays usually aren’t rested on a sound trend following footing. They have, essentially, a different signal generation process that doesn’t stem from trend. It comes from second derivative price dynamics. So is that something that you do? Do you see edge and alpha there? Is that important? Could you just run your 16 systems in a linear fashion, stack them on top of each other and also get great results?

[0]1:01:52 Gareth

I get the idea of the balance between accuracy, system accuracy, and micro tinkering. I guess the more you try and factor in the more micro-tinkering, the more of the risk of getting whipsawed, essentially. The robust approach is to avoid micro-tinkering too much and trying to be too accurate versus having a robust level of system accuracy.

We do, dynamically, adjust the position. The ATR is a good example. Based on the volatility of the market, we dynamically adjust the position size, the size of the unit just, again, to target that outcome. The units are generic, market generic normalized units moving approximately the same in risk terms, even in dollar terms, each day so that we achieve the noise canceling effect across the portfolio.

A good day in one market will offset a bad day in the other, and performance over time is driven by the trends. Like I say, just one strong trend can drive performance while the rest of the positions can cancel each other out if this normalization is accurate. In the case of the ATR dynamic adjustment, quite simply, if the ATR is to go up, the position size, the size of the unit is reduced so that we stay on target for that daily movement of a unit.

That’s reflected in the stops as well. So, in the case of the ATR expanding, the stop would expand, and the position would be reduced. It works the other way around. As the ATR reduces or contracts, stops come in, position size increases. Again, it’s to achieve the outcome of normalizing volatility across all the markets, which, for me, is a very powerful aspect of risk management.

Now, if we didn’t do that, would the system still work? Absolutely. This is more for me a point on system accuracy, improving accuracy, improving efficiency and the performance profile rather than it being a necessary aspect of the system for performance. It’s more of a performance driver.

[01:05:02] Moritz

Fantastic. Gareth, Richard, I think we should leave it there. It has been a lot. I found it really interesting. Listeners, I hope you found it interesting also. Quite some different takeaways and nuggets that we’ve heard from Bowmoor Capital today. So thank you very much Gareth and Richard for joining me today. It’s been really great to have you on the podcast. I really enjoyed that.

Listeners, as usual, you can find Gareth and Richard’s contact and social media information in our show notes. And please send us an email, info@toptradersunplugged.com if you have any questions about the things we discussed today.

So, Gareth, Richard maybe in a year’s time or sooner or a bit later, we’ll reconvene on that show and review 2024 and how you guys have fared. I definitely do wish you all the best also with the new fund project that you have, the new higher volatility, Bowmoor on steroids fund, that you’re working on. Again, thanks very much for being on today.

[01:06:11] Richard

Thank you.

[01:06:13] Ending

Thanks for listening to Top Traders Unplugged. If you feel you learned something of value from today’s episode, the best way to stay updated is to go on over to iTunes and subscribe to the show so that you’ll be sure to get all the new episodes as they’re released.

We have some amazing guests lined up for you and to ensure our show continues to grow, please leave us an honest rating and review on iTunes. It only takes a minute and it’s the best way to show us you love the podcast. We’ll see you next time on Top Traders Unplugged.