Artwork for podcast Top Traders Unplugged
SI333: Convexity - The Trend Following Game Changer! ft. Richard Brennan
1st February 2025 • Top Traders Unplugged • Niels Kaastrup-Larsen
00:00:00 01:12:54

Share Episode

Shownotes

Most investors focus on returns, but what about the shape of those returns? Today, Rich and I explore the often overlooked concept of convexity - a game-changing principle in trend following that helps us thrive in unpredictable markets. Inspired by insights from our recent guest, Dave Dredge, we break down why mastering convexity is key to cutting losses early, riding big winners, and staying ahead of market turbulence.

If you’re serious about protecting and growing your portfoli - while making smarter, more systematic decisions - this episode is a must-listen. Join us for a deep dive into the mechanics of risk, reward, and the powerful edge convexity can give your portfolio!

-----

50 YEARS OF TREND FOLLOWING BOOK AND BEHIND-THE-SCENES VIDEO FOR ACCREDITED INVESTORS - CLICK HERE

-----


Follow Niels on Twitter, LinkedIn, YouTube or via the TTU website.

IT’s TRUE ? – most CIO’s read 50+ books each year – get your FREE copy of the Ultimate Guide to the Best Investment Books ever written here.

And you can get a free copy of my latest book “Ten Reasons to Add Trend Following to Your Portfoliohere.

Learn more about the Trend Barometer here.

Send your questions to info@toptradersunplugged.com

And please share this episode with a like-minded friend and leave an honest Rating & Review on iTunes or Spotify so more people can discover the podcast.

Follow Rich on Twitter.

Episode TimeStamps:

01:49 - What has caught our attention recently?

06:20 - Industry performance update

11:10 - Why does convexity matter for trend following?

36:13 - What do we actually mean by convexity?

42:04 - Sharpe world - the pitfalls of today's economy

52:30 - Why risk and volatility can feel like the same thing

56:08 - The core of trend following

58:22 - Why open trade equity makes no sense for managers

01:00:03 - The business of trend following

01:01:37 - The exponential potential of trend following

01:08:05 - The times have changed - why is trend following more important than ever?

01:11:22 - What is up for next week?

Copyright © 2024 – CMC AG – All Rights Reserved

----

PLUS: Whenever you're ready... here are 3 ways I can help you in your investment Journey:

1. eBooks that cover key topics that you need to know about

In my eBooks, I put together some key discoveries and things I have learnt during the more than 3 decades I have worked in the Trend Following industry, which I hope you will find useful. Click Here

2. Daily Trend Barometer and Market Score

One of the things I’m really proud of, is the fact that I have managed to published the Trend Barometer and Market Score each day for more than a decade...as these tools are really good at describing the environment for trend following managers as well as giving insights into the general positioning of a trend following strategy! Click Here

3. Other Resources that can help you

And if you are hungry for more useful resources from the trend following world...check out some precious resources that I have found over the years to be really valuable. Click Here

Privacy Policy

Disclaimer

Transcripts

Intro:

You're about to join Niels Kaastrup-Larsen on a raw and honest journey into the world of systematic investing and learn about the most dependable and consistent yet often overlooked investment strategy. Welcome to the Systematic Investor Series.

Niels:

Welcome and welcome back to this week's edition of the Systematic Investor series where Richard Brennan and I, Niels Kaastrup-Larsen, where each week we take the pulse of the global markets through the lens of a rules-based investor.

And I also want to say a warm welcome if today is the first time you're joining and if someone who cares about you and your portfolio recommended that you tune into the podcast, I want to say a big thank you for sharing the episode with your friends and colleagues. It really means a lot to us.

Rich, it is so great to be back with you this week. How are you doing down under?

Rich:

Very good. Niels, I know you've been busy with your traveling, but I've been stuck here in the heat down in Australia. So, we've got the monsoon coming in into Queensland and lots of tropical lows about, so, impending cyclones coming. And it's miserably hot. But apart from that, Niels, everything's good.

Niels:

Okay, that's good to hear. You know, I was thinking about whether sort of big the big news from your part of the world. Of course, the big news from my part of the world is still this thing about sort of Greenland, and Denmark. Even in the airport, as I was flying back from Miami, the guy who was checking me in said, so, what's happening with Greenland? So, it's kind of funny. And I will say one thing, if nothing else, the PR for a small country like that is quite extraordinary. So anyways, maybe we can have a positive spin on that here, over here.

Anyways, we’ve got an amazing lineup of topics that I think people will really enjoy today. But before we do that, I'm obviously always curious in terms of what sort of been on your radar since we last spoke, which would have been before year end. So, anything in particular that you've been following along that's outside of today's topics?

Rich:

Look, I suppose I've been captivated by the Trump elections as everyone else has. I've been captivated by the Australian Open tennis, which I've been watching eagerly. I've captivated by the cricket.

But apart from that, Niels, I suppose I'm a good trend follower in that I've sort of kept my head out of the news in general, but very busy with my models. I have been spending a lot of time listening to a lot of Dave Dredge videos. I know you had him, with Cem, just recently on a great discussion and he's someone who really resonates with me. So, the topics today are going to be drawing on Dave Dredge-isms. So, it's a Dredge-ism hour and how we apply that to trend following.

So, looking forward to getting into those topics.

Niels:

Yeah, indeed, indeed. A great conversation and I know these topics actually today will be really fun as well.

Now, that being said, I think for me what's been on my mind, I couldn't help notice that there was a certain tech company that started out pretty shaky this week. Nvidia of course is the one I'm talking about here.

And all of this created a little bit of kind of a talk main topic when I was attending the Global Alts Week in Miami this week. On one side, on one hand, I think it reminds me how easy market sentiment can change these days and how valuations really are a moving target.

I think it lost like $600 billion in a day. That's meaningful.

It also reminds sends me of the difference, really, between being a short-term manager where your models may, you know, react to a big sell-off like we saw on Monday and then also the longer-term time frames that we focus on where we try to ignore some of that noise. Now, of course, sometimes it's good to react quickly this week not so much since a lot of this kind of disappeared again the following days.

Then the other thing, of course, that I will want to mention this week is just that, you know, attending the Global Alts Week in Miami this week really showed me how far this industry has come. First of all, let me just thank all the people who came over to the Dunn Capital booth to express their appreciation for what we're doing on the podcast. That was so nice, and it really means a lot to hear feedback like that.

And, of course, I also saw, as I mentioned to you Rich, two of our wonderful co-hosts, namely Katy and Alan, who were also there. And that's obviously also a great treat to see them in person from time to time. And also, there were lots of our guests and peers in Miami. So, it was quite special.

And then the other thing that it made me think about, so I attended the first of these conferences 32 years ago and this was back when it was called Managed Futures Association and, funnily enough, I actually ran into someone in Miami that I met 32 years ago at the first conference, in London I think it was held. We were maybe 100 people back then, very informal in many ways.

And fast forward 30 plus years, we had 5,000 people attending Miami and live coverage from CNBC and celebrities from the financial industry, not to say the entertainment industry as well. So, I think the industry has come a long way. Obviously, it is much more than managed futures today than it was back then. But I'm super excited to see where the industry will go in the next 25 years and I hope that that we’ll all still be part of it, talking about it then. So, quite a fun week in many ways. Yeah.

Now speaking of trend following, it is month end, it's actually February 1st today. So, let's talk a little bit about January for a moment. I mean it was a mixed month, I think, according to sort of the early data, although with a little bit of a bias, positive bias I would say.

It strikes me that there is an increasing divergence in the market that also leads to, you know, a pronounced dispersion in returns between managers. I wonder if this is just the beginning and that the divergence in markets, which is most likely caused by central banks now having their own agenda instead of this very super coordinated agenda we saw a few years ago. Now they're doing their own thing, BOJ is hiking, Europe is lowering rates, and the US is kind of at a standstill right now. But it's pretty interesting and I think it could be quite useful for models such as ours, from my vantage point at least. I'd love to hear your thoughts afterwards.

Maybe a little bit of a surprise in terms of what did well. Not surprising necessarily. The equities did well. There's still a lot of effect from the election. Maybe surprising that something like German DAX. When you hear about the German economy, you might not think that the DAX would be a great source of performance, but that's what I'm seeing this month.

Coffee has kind of helped out nicely now that cocoa is taking a little bit of a breather, and gold hitting some highs - doing well, but also some of the smaller markets. I mean, live cattle is not a market that many people trade. Trend followers do, and it had some good momentum recently.

Of course, there's also a little bit of challenging things going on in our portfolios nowadays. Again, from my vantage point, this could be different manager by manager, but sync was a little bit of a challenge this month.

So, anything that stood out to you in January as such?

Rich:

No. Look, I tend to agree with you though and, for instance, our firm, East Coast Capital Management, we've had a pretty good month. I know your firm has had a good month as well.

But when I look at S3 Trend index I see it's down for the month. So, I do think we are sort of seeing this dispersion coming through and trying to understand why. Perhaps they're not exposed to the commodities we are like coffee and some of the ones that have really been driving our growth, or perhaps there's model dispersion there. I'm not quite sure, Niels, but it's perplexing to me. So, it’s clear that not all trend following is alike by their definitions.

So yes, it's going to be a month of dispersion, I think.

Niels:

Yeah, yeah, true, true. So as of last night, the 31st of January, my trend barometer finished at 41. So, that's kind of a neutral stance on performance. I think that's probably also what we're seeing in the early numbers as you mentioned, Rich with the trend, various indices, CTA indices.

We only have the data up until Thursday and I do expect that yesterday was probably a little bit of an up day, so, the numbers could be a little bit better than expected, than what I saw on Thursday.

So BTOP50 looks like it's going to finish around 70 basis points, up for the month and therefore the year. SocGen CTA index up 20 basis points as of Thursday. SocGen Trend, as you mentioned, surprise, down 52 basis points as a Thursday, it'll probably narrow that gap a little bit Friday. The Short-Term traders index up 35 basis points. Then we have the Bridge Alternatives which is just flat fee trend following down 78 basis points, that's also surprising to some extent.

But, in the traditional world maybe not so surprising. MSCI World still powering away up 3.47% in January. The US Treasury Bond Index from S&P, of 20 years and more, was up 31 basis points but it is actually down more than 5% for the last 12 months. Then the S&P 500 Total Return up 2.78% so far this year. So, interesting start to the year I would say, and we'll see how it all plays out.

I think we teased it enough that we're going to go through some really interesting topics. So, once again, it'll be like going to college here or university when we have you on, Rich. You're going to educate us in a really fun way. And, certainly, a lot of this and a part of this will be inspired by our friend Dave Dredge and his wonderful writing.

All right, let's dive in. One of the things that certainly is talked about more and more in our industry is convexity. Now, some firms actually specifically position themselves to get investors positive convexity, in this case. But let's talk about what it is and why it matters for trend following. Because a lot of times it's talked about and written about in relation to other strategies, not so much about sort of trend following. So, let's dive in with that perspective.

Rich:

Yes, Neils. So, a big heads up to Dave Dredge for this one. I've been inspired by a lot of his recent videos he's been doing. He strongly resonates with my philosophy of being an outlier hunter. I think this is because we're both sort of striking the nail of convexity on the head with our models.

Dave, obviously, is in the world of tail protection, which is fairly well exclusively looking at how to protect traditional portfolios from adverse volatility events. And you know, something like Nassim Taleb’s fund, Dave Dredge, someone like Wayne Himelsein, they'll be looking at integrating tail protection strategies as allocations into traditional portfolios to provide this degree of protection for uncertainty.

Now, in our trend following world, we embrace convexity at all levels, not just in protection, but also in looking for opportunities through diversification and by attacking what we call these outliers. The principles that Dave discusses and convexity, which is a term that was effectively developed in options world, in options speak, is certainly embraced by trend following.

So, these three topics, the first one, I'll be discussing is what is convexity, so that we understand what we're dealing with. The second topic that we'll be looking at is the traditional investor and how they view risk and how that's not the way to view risk. It's what we call, from Dave Dredge, Sharpe world thinking. So that will be the second topic. And then the third topic is how we integrate convexity into our trend following models.

So, we'll start on this first topic. So, let's start by looking at what convexity is.

So, in the world of finance, traders and investors, they often seek stability, predictable returns, smooth equity curves, and strategies. They claim to navigate uncertainty without turbulence. But markets aren't predictable, as we all know, and the real world isn't linear. It's wiggly. Financial markets are defined by convexity where small changes can lead to disproportionately large outcomes. This is nonlinearity.

So, if we can imagine any financial market, let's say the S&P 500, there will be regimes where it is fairly muted in its performance, but then there'll be long periods of regimes where it explodes in its performance. There'll be periods where over five years we have annual returns of say, 25% plus over those five years.

acted in the tech collapse in:

Now if we are a traditional investor looking at traditional risk measures, their assumption is that the overall, or the average volatility, can be explained. And what we're saying is that is not the way to deal with risk.

Risk is being able to adapt to the market that's currently in play. Whether it's explosive or whether it's going into a drawdown, we need a model that can adapt to that environment and address the acceleration that exists in that model.

Now what I mean there is that if we look at a financial market, and let's take a model such as trend following, in periods of time where there are no trends about, we don't expect that model to perform. And so, our equity curve should expect drawdowns or what we call stagnation over those periods of time.

But then in other periods of time, that market might have a significant material trend. We would expect our equity curve to respond to that trend with a significantly rising equity curve.

The fact is that over the course of 10, 20 years or whatever, we find that the returns that are generated from our models, applied to those markets, are not linear. There's not a linear, progressive, graded, performance level at all times across that market. There are periods of time where we have explosive opportunity and there are periods of time we stagnate and have drawdowns.

Now, when we look at lots of different markets, as we do under a diversified portfolio, we have a lot of different wiggly lines which are equity curves developed from responding to those financial markets that are nonlinear in nature. Now those wiggly lines are nonlinear representations for equity performance mapped to those returns of the market.

There's a nonlinear relationship between them. So, when the market Explodes with an exponential trend, we get an exponential rise in our equity curve. There's a nonlinear relationship between that.

Now what convexity does is it explains the relationship between the returns and our performance in terms of curved lines. And there are two forms of curved line that express any model that is applied to the financial market over the long term. They will either be convex or they'll be concave. And this looks at the relationship between the payout, between the risk and reward relationship.

If we have a low risk and a very high reward, it's going to express what we call a convex curve. It's a curve expressed like acceleration, where you get progressively higher, and higher, and higher, faster, and faster, and faster -

an exponentially rising curve. It's not a linear straight line.

And when we are applying something like a mean reversion strategy or strategy with what we call negative skew, we get a concave relationship. We get lots of small wins - win, win, win, win. Which is sort of the rising linear progression of an equity curve. But then when tragedy strikes and the markets enter a new regime that isn't predicted in advance, it falls off a cliff. So, we get a massive drop in equity, a significant increase in acceleration, but it is a concave acceleration as opposed to a convex acceleration.

So, what we're focusing on here is convexity. So, what we're saying is that we're dealing with asymmetry here, where the relationship between market movements and portfolio returns are nonlinear. It ensures, under convexity (which is effectively positive skew), it ensures that favorable volatility accelerates gains while we mitigate losses through effective risk management.

So, it's unlike a straight line relationship. Convexity produces a curve allowing portfolios to amplify upside potential while limiting downside risk. We get this convex upside.

This is crucial in a world where markets are inherently nonlinear, unpredictable, and prone to fat-tailed events. So, we'll look at the difference between convexity and linearity and why it matters.

So most traditional portfolios assume linearity. They're looking for the straightest possible equity curve, the most linear equity curve, the curve with the highest Sharpe ratio, the straight line. So, this is where the input is proportional to the output. The linear representation is where we get an amount or an input, which is proportional to the output we get.

So, imagine walking at a constant pace. Every step covers an equal distance, representing a linear system where changes in market conditions result in a predictable proportional relationship. So, let's now contrast this with convexity.

So, convexity behaves more like driving a car, as Dave Dredge would put it, pressing the accelerator doesn't just move the car forward, it speeds up exponentially when we hit that accelerator. We get the curve when we hit the accelerator. But acceleration is a funny thing.

We not only get acceleration to the upside, we can also get acceleration to the downside, and that's called braking. So, imagine you're in a car, driving at speed and you push on your brakes. There is not a linear relationship with the pressure you exert on your brakes and the speed at which your car stops.

You'll find it also follows an exponential convex curve where immediately when you apply the brakes, you get the biggest impact on your reduction in speed, and then it slowly tapers off. So, braking at high speeds requires disparately proportionately more force.

And convexity is about acceleration when conditions are favorable and braking effectively when conditions are not favorable. And it's therefore a dynamic process. It's not like a linear process that assumes constant linearity at all times, the same conditions, the same expected returns over times. What this does is it responds dynamically to the nonlinearity of the markets.

So, when markets are expressing this nonlinear exponential uptrend, we are using convex models to amplify that beneficial volatility. But when we are in an adverse regime where things are unfavorable to our trend following, we are applying the brakes.

And what that does… So, if we don't have brakes on our cars, and our brakes in our trend following world, of course, is our stops, our small bets. They're the effective brakes that ensure that we never let the adverse volatility become a tail event.

What we're doing with this asymmetric convexity, having brakes and being able to take opportunities, exponentially amplify our returns in favorable environments, we are creating asymmetry in our portfolio returns and exploiting beneficial volatility. So, the reason this is so important is that convexity allows us to achieve the greatest compounded returns over the long term.

So, let's look at this. Let's look at a fund, for instance, that focuses on producing the smoothest linear ascending equity curve versus a fund that employs convexity. So, a fund that is trying to achieve the smoothest possible equity curve is really only limiting adverse volatility. They're not exploiting the upside volatility.

Now, we can achieve a higher compounded return if we exploit upside volatility. So, what I mean there is that we don't treat all volatility as the same. There is adverse volatility and there is beneficial volatility. So, think of the brakes on our accelerator going around a fast racetrack.

If we have a wiggly racetrack with many curves, if we didn't have brakes on our Formula one car, we would have to adopt an average speed around that racetrack that ensured that we weren't going too fast when the corners came, but we weren't exploiting the straights in that racetrack when we hit those straights. We'd be trying to achieve… We'd be curve fitting or optimizing for an average speed over that racetrack without brakes. We won't win the race that way.

The way we win the race is by having brakes so that when we get to the curves, we can lightly tap on those brakes, significantly decelerate with convexity, and when we hit the straight, we can amplify that straight by hitting the accelerator, and whoosh, off we go. Now, when we look at the overall lap time of someone with convexity and good brakes, they go much faster than someone's trying to optimize for average return over that racetrack, or average race times.

And the question also is, on that squiggly racetrack, we don't know what the curves are going to be in the future. We're coming to an uncharted landscape. We must have convexity in our models. We must have brakes in our models. We can't assume that the racetrack is the same as all the racetracks we've raced on in the past.

We're on a new racetrack here. So, we can't assume that the curves are all going to be the same. We can't assume that the historical track record of how we've have navigated the average speed of every lap we've taken on all of these different racetracks is going to be the same as this racetrack. And convexity is incredibly important when dealing with uncertainty.

So, the problem is that the major risk events, or the major events in any financial market that lead to good times for the investor or bad times for the investor, they are the most material moves in that market. They're not the normal everyday market moves. They're the things that really matter.

e Great Financial Crisis, the:

But the real thing is that uncertainty drives the most important material events we see in the markets. So therefore, you need a model that incorporates convexity into your process to be able to dynamically adjust and take advantage of those material events.

Otherwise, you're forced to use history as a basis to devise what you think your expected returns are going to be. And you're always going to be blindsided by these events that have never occurred in the historical record, like the COVID pandemic, or any of these major material events. You're going to be blindsided, and they will be the major adverse events on your portfolio.

So, it's so important, convexity, but the way that… It's employing what we call a barbell approach to risk management. In other words, the left side of the barbell is keeping our losses as small as possible, as linear as possible, small bites, never letting those losses turn into significant material losses. And the right side of the barbell is amplifying opportunity.

So, for those people that don't employ convexity, always seeking the average speed, no brakes on their cars or whatever, they will always be not responding to the opportunities in the market. They're going to be cautious, nervous nellies because they don't want to go too fast, because they don't know what corn is going come up.

But if you've got good brakes on your cars, you can take advantage of these opportunities with force. You can accelerate into these opportunities, amplify your returns, taking advantage of these opportunities when the market gives these opportunities. You're not letting it occur by your system, you're responding to what the market's giving you.

And with convexity embedded into your process, you're dynamically responding to the conditions of the market at any point in time. It's a very important approach.

So, with this barbell approach, we're forcing this asymmetry, this asymmetrical barbell - small losses, exponential accelerated gains on the right side. So, we're forcing asymmetry into our portfolio returns. And convexity isn't about constant or average speed.

It's about managing accelerations and decelerations, knowing when to take risks and when to pull back. This is why convexity lies at the heart of tail protection strategies employed by people like Dave Dredge, Wayne Himelsein, Nassim Taleb, etc. They are experts in tail protection, how we add that to a traditional portfolio to protect adversity of uncertainty. But as we'll discuss with our trend following topic, we embrace that wholeheartedly with our entire process.

But the goal of convexity is optimal compounding. So, the aim of convexity is to achieve the optimal path of future returns by managing volatility dynamically.

So, a convex portfolio achieves better stopping power, which is our risk mitigation, and faster acceleration, capitalizing on outsized opportunities. And this approach contrasts sharply with conventional strategies that optimize for average speed, often ignoring risk until it's too late.

So, these linear equity curves we often see put out by portfolio managers, they're actually the great illusion. So many fund managers, they attempt to present these smooth linear equity curves, but this is a fictional aberration in a nonlinear world.

So how do they create these linear lines in a nonlinear financial market? Well, you'll find that the linear lines are only a temporary line. You'll find that over the long term, those linear lines become concave lines.

You'll find that those linear lines like long-term capital management, for a period of 2, 3 years, while conditions were favorable, they were getting the small win, the small win, the small win, the small win, the small win. They were getting this beautiful straight line, the same as pattern recognition produces beautiful straight lines while it's optimized for that pattern, the same as mean reversion gets beautiful straight lines while the markets are responding to mean reversion in that predictable manner.

Under predictability, we get the smooth line. But as soon as we get a regime shift out of that, we find that the risks become totally unacceptable and totally extreme, which creates concavity to their actual portfolio results.

Because what we find if, for instance, we are applying a trading strategy to a predictable environment, it's always the case. We get a level of return and we think how clever we are in getting this lever of return. And we think we understand what the future's going to bring. We say, how can we lose?

So, what do we do? We lever up. We say, I'm not just going to get this return, I'm now going to magnify it 2, 3, 4 times, times. So, they'll lever up to get this higher lifting power of their predictable return. And what happens? The calamity occurs when suddenly the regime shifts with high leverage and boom, out they go out the back door.

So predictive models, and this is the case with predictive models such as a 60/40 portfolio. The 60/40 portfolio is built on a premise of correlation. It's built on the premise that over the past 20 years or so, we have had stocks that are uncorrelated with bonds. That was the whole basis of the 60/40 portfolio. So, it's based on a backward looking historical basis that these things, this condition of this correlation will persist into the future.

However, we found, now, that has changed to a positively correlated relationship and bonds no longer offer the protection. That portfolio had no brakes. That portfolio was one that relied on an historical representation of what should occur in the future using data.

It's the same as when we do a backtest. We're saying, oh, it survived this backtest admirably, it must go well into the future. But that's not the way to look at risk. Risk is about being blindsided by the things you don't know because they are the most material events in these financial markets.

What blindsides the traditional investment community is things that they never expected or never saw in their detailed historical data sets that they've used to train their models, or quant the hell out of their models to develop the optimal strategy. They're suddenly presented with a new environment they've never faced before and they collapse because they've embedded volatility or they've embedded leverage into their system. They've relied on a correlation level.

So, convexity is not about what history has told us. What it's about is defending your portfolio from the unforeseen by putting breaks into your portfolio and allowing for unlimited upside potential when markets are favorable to you. So, it's about embedding that process into portfolio management. It's not something that Sharpe tells you. It's not something that all of these metrics tell you.

It's something that you've got to enforce into your portfolio. And typically, someone like Dave Dredge or whatever, they'll employ options to provide that degree of protection for a traditional portfolio. So, they'll assess the risk or the unforeseen risk of a traditional portfolio, such as a long only portfolio in stocks. And they'll say, well, there'll be times in the future where we could have a collapse like we've had in the past, or unforeseen events we would never anticipate could strike that portfolio, which has clearly got a long bias and is weak as far as its short side protection.

So, they've incorporated tail protection strategy that takes up for the unforeseen opportunities, or the risks that present them that are possible, as opposed to the risks that have been seen in the historical backtest. And they'll embed this convexity into that traditional portfolio and that will provide a degree of protection.

In our trend following world, our whole process is about convexity. But you're about to say something Niels?

Niels:

Well, I wanted to interject a little bit and just ask you a couple of things here. First of all, convexity may be slightly different from person to person when you ask.

I mean for example, even within our industry people might claim that oh, but trend followers provide less positive skew, less convexity today than they used to. Short term managers will highlight, oh, we are better at providing this tail protection funds that you just mentioned. They will also have their kind of say, but it all depends on the time frame you look at.

I think that's important to understand because otherwise people just might blend the term convexity into one-and-the-same. And I think that might be a little bit dangerous because we might say as longer term managers, well, hang on, there's too much bleed, it's too expensive to provide convexity on a daily basis because it doesn't matter in the long run, and so on, and so forth. I think Quantica wrote a paper on this a while back.

I just wanted to hear your thoughts and maybe remind people a little bit about what we mean by convexity in terms of time and/or duration just to make it clear.

Rich:

So, typically with our medium to long term outlook with our trend following models, we do have a degree of exposure and risk in relation to short-term adverse movements because our models can't respond quickly enough to those short-term movements. And this is where someone with a tail protection strategy might be using options and get immediate protection from those events.

However, there are costs to those options. So, there's the cost of the option expiring worthless. There's the cost of always having to maintain that insurance in that portfolio over the course of time. There is no cost for our long/short models in our trend following world, but we provide convexity at no cost.

So, we find that, for instance, trend following isn't a pure insurance protection program. If it was, then they'd be embedding trend following into every traditional portfolio because there's no cost to it and they'd love that.

Options are probably the purest way to get protection for a traditional portfolio for adverse risk events because you can align the option exposure to what you're trying to protect, whether it's correlation risk, whether it's market risk or whatever, in that portfolio. You can design your option to give you the required protection to offset the risk of that portfolio.

However, trend following offers better, more diverse opportunities in its ability to exploit opportunities. And the reason I say that is that you get many more markets we can trade in a trend following way than we can apply with options which are much more limited in their diversification in market exposure.

So, in our world of trend following, where I can invest in these unusual commodities or whatever, or get wide coverage, I can exploit these positive attributes of outliers, which is not what tail risk managers are typically looking at. All they're looking at is defending the portfolio.

What we're looking at is how we exploit convexity to not only give us a degree of downside protection, recognizing it's not 100% perfect, because of course we get whipsaws. We get these periods of time, if we don't have short term models where we're exposed to these short whipsaws or whatever, but in enduring material events that have consequence, things that last for a long period of time, we do have a degree of convexity in our models and we exploit those opportunities when our trends turn round.

Niels:

Yeah, and the other thing I just want to highlight, because it does also stem from the conversation that I had with Cem and Dave, a couple of weeks ago. I think it is important when you talk about the race car and the racetrack analogy, that Dave has also written about, and talking about how the curves of the track actually, for us, is unknown because we don't know what the future holds. It bridged nicely into this idea of, yeah, a lot of people have opinions, and they will take action in terms of mitigating risks that they have thought about, they worry about, and so on and so forth. But the question that we need to be talking about as well is, you know, what are you doing about all the things you're not worried about?

And I think that's really important. And, again, in a global portfolio, in a multi asset portfolio, having things like trend following in the portfolio will not necessarily completely alleviate the issue, but it will certainly reduce some of the pain from things happening that people hadn't even imagined could happen. And it goes back to this thing I've been saying for years. We live now in a world where we have to imagine the unimaginable.

I mean, just think what's happened in the last 11 days, not as a political statement, but just the things that are happening. It's inconceivable, you know, two years ago that we would be talking about some of these issues. Right?

So, these are things that are really important and I'm sure we're going to talk more about it, but I do want to pivot because we're already 42 minutes into our conversation today.

Rich:

It segues beautifully into the next thing.

Niels:

It segues beautifully into the next thing. And that is kind of some of the pitfalls we have in the world we live in today, which is obviously predominantly focused on Sharpe. So, let's talk about that for a little bit.

Rich:

So, in modern finance, we find today the pursuit of risk adjusted returns. This has led to a widespread reliance on measures like the Sharpe ratio, standard deviation, and value at risk.

This framework, what David Dredge refers to as Sharpe world, has shaped how investors think about risk. Yet it is fundamentally flawed.

ct we've had so many of them.:

We've got a legacy over hundreds of years. The fact that traditional investment has not been able to account for these anomalies, these blindsides, show that the traditional way of measuring risk is flawed. And this is what blindsides investors.

This is what leads to major material loss events, things that are never expected or never seen. If you expect something, you prepare for it. If you never expect something, you get blindsided by it.

So, the industry, which I'm having a crack at here, they've got an inaccurate understanding of risk. And one of the greatest misconceptions is the belief that risk equates to volatility.

So, this assumption underpins so much of modern finance, modern portfolio theory, the efficient market hypothesis, standard risk management practices. But history has shown us that the most damaging risk events are not the ones we expect, but the ones we don't see coming.

So, if risk could be forecasted using historical distributions, then risk management would be easy. But every major financial crisis from Black Monday to the Global Financial Crisis, occurred because the models failed to anticipate the unexpected.

So, as Dredge described Nassim Taleb's quote, he puts it, “understanding is a poor substitute for convexity”. So, what he means here is risk isn't about predicting the next crisis, it's about building a portfolio that can survive and exploit uncertainty. It's not about prediction, forecasting, it's not about that.

So, let's get into this flawed assumption of Sharpe because it's rife throughout finance. So, traditional financial models, they rest on several flawed assumptions. One is that risk can be quantified using historical data. Two, markets behave in a linear fashion following normal distributions. Three, correlations persist into the future. Four, expected future returns can be estimated with confidence. And five, all volatility is bad and should be minimized. That's the way traditional finance looks at risk.

So, these assumptions have repeatedly failed. The reality is that markets are nonlinear, they're prone to fat tail distributions, and are constantly shifting in ways these models (traditional models) cannot predict. So, as Dave Dredge puts it, risk is a possibility distribution, not a probability distribution. Such a great one-liner, that one.

So, what this means is that traditional models, they typically try to assign probabilities to past market behavior. So, assuming that similar conditions will yield similar results in the future. But risk isn't about the past, it's about what could happen in the future. So many investors mistakenly believe that long periods of low volatility indicate a safer market. But history shows that stability breeds instability.

Variance at risk, for example, assumes that the longer a market experiences lower volatility, the lower the risk of an extreme event. However, in reality, low volatility environments encourage excessive leverage, and they may make markets more fragile and vulnerable to shocks.

Now this isn't just a sign to the financial markets, Niels. Let's look at avalanches. Avalanches occur after long periods of snow, progressively building without disruption. It looks safe. Over a long period of time, if you analyze the fact you've had all of this snow slowly accumulating, you don't see the risk of the avalanche suddenly, catastrophically infecting the whole system.

Floods, major floods are devastating, typically, when they follow long droughts that harden the ground. Financial crises happen when extended periods of complacency and excessive risk taking takes place. So, we can see that if we use historical measures, data measures, to try and quantify risk, we're going to be hopefully wrong.

And a good example is the forest fire started with a lightning strike on a tree. So, we could look at our data, historical data, and see that the prevalence of a lightning strike on a tree has X percent. But that's not the real risk. The real risk is the fire having contagion risk. Once it hits that tree,

suddenly all of the undergrowth around that tree and over vast areas suddenly erupts into flames because of the contagion risk.

That is not measured by the traditional VAR model. That is something that we know through logic, is that over long periods of dry periods, debris is going to build up and it's going to provide contagion risk so that when one spark comes into that forest, we're go to get a significant event, which is going to be a calamity.

So, the assessment of VAR is going to totally underestimate the calamity. It's only going to assume that there's a lightning bolt hitting a tree and what is the magnitude of that event? What it's not going to assess is the widespread chaos that was created because markets are connected. Markets are complex adaptive systems. All of the things these traditional models fail to assess in their historical, backward looking, quantitative ways of assessing risk.

So, the problem with variance at risk, Sharpe based metrics is that they estimate the probability of a tremor, but not the probability that the tremor will trigger a catastrophe. They measure the likelihood of fire, but not the potential contagion effect of the fire.

This is why risk cannot be managed exclusively through predictive models. It must be built into portfolio design. So how do we escape this Sharpe world thinking? How do we embrace convexity?

So, the solution to these flawed risk models is convexity. A portfolio approach that builds in protection against the unknown while maximizing upside potential. Here we go back to the brakes and the accelerator again.

So, the role of convexity in portfolio design. We want to mitigate adverse volatility. We want to have effective braking power. We want to exploit, actively exploit beneficial volatility, strong acceleration during favorable regimes. We want to position a portfolio on the right side of the return distribution to optimize compounding.

In other words, we're shifting the distribution to the right. So, we've got lots of small losses but occasional massive wins, which is shifting the distribution to positive skew. This is optimizing compounded wealth over a long time. The thing that really is problematic for compound of wealth over a long time is this concave risk - risk seeking behavior.

Things such as mean reverting models, negative skewed models, they are taking small wins, small wins, small wins, small wins, and suddenly they get hit with a massive risk event which significantly cripples them. And then that creates significant compounding drag and totally dilutes the compounding effect.

So, in a nonlinear world, survival isn't about eliminating volatility, it's about leaning into it - leaning into it. It's about learning how to control and exploit it. So, this principle extends far beyond finance. So, a surfer for example, they don't try to flatten the waves, they learn to ride them. They adjust their stance to avoid being wiped out while maximizing momentum on the crest. They are applying a convex model to their surfing.

A pilot navigating turbulent weather doesn't aim to remove turbulence, but adapts the aircraft's response to maintain control and stability. A skilled chef that doesn't eliminate heat fluctuations. What they do is they leverage temperature changes to cook the perfect dish. They know when to apply high heat for searing and when to let ingredients simmer. They are applying convexity in their process.

This is in any nonlinear system that exhibits these nonlinear patterns, we need to apply convexity to addressing them. We don't apply a square peg in a round hole, a straight line into a wiggly market. That is a disaster waiting to happen.

Niels:

All right, I have a couple of questions because I know we have a final topic we also want to get to. I have a couple of things I know we talk about and people would have heard me say the same that, you know, volatility is not risk. So, I do want to ask you, just for a very brief comment, what do you consider as risk?

Rich:

Risk, to me, is permanent loss of capital.

Niels:

Okay, I knew you were going to say that. I mean and I didn't know it, but I had a feeling you would say that. And so, I heard Cliff Asness talk about this this week.

I was out walking in Miami, and I was listening to a podcast with him and they were talking about that and he said, well, yes, but what's the real chance of a permanent loss of all capital? Very, very low. I mean, how many strategies in our careers have we really seen blow up completely?

Rich:

In the trend following world hardly any. Why? Because they apply positive convexity. In mean reverting world, LTCM, all of this world it's huge. And this is, Niels, the reason we can't find historical records of validated track records of fund managers who exploit anything aside from trend following over 20, 30 years. They don't exist, Niels. And this is because these risk events cripple them.

Niels:

Yeah, that is a fair point. All I'm just saying is that I would love, if there was something in between. Do you know what I mean? Because it doesn't happen that often, even in the other world, that people completely lose all their money. It can happen. And on the other hand, volatility can feel like risk even though I agree that it is not the same. So anyways, I wanted to give you that piece of information. I thought that was some interesting thought.

He also discussed this idea between the depth of drawdowns versus the length of drawdowns and how actually the length of a drawdown could be much more damaging to our psychology than actually a quick, larger order. Anyway, I don't want to go there today. We wouldn't have time, but something to talk about.

Oh, yeah. I heard another podcast recently, that was a Grant Williams podcast, where they talked about that probably the possibilities as well as the probabilities of something crazy happening is probably higher today than ever before. I mean, the range of possibilities is actually much bigger today, but so is the probability of something happening. I thought that was interesting. I completely agree with that.

And then the final thing you talk about is VAR, and now it gets into math and all that. That's not my strong side. And you certainly shouldn't do it, “live” on a podcast. But I did want to ask you, just very briefly, with VAR, you can calculate the uncorrelated VAR and the correlated, meaning the one that takes correlations into account. And I imagine that most people in our business would use the one that takes correlations into account.

And, therefore, this is obviously not a statement of facts, it's just that it has certainly worked well for those of us who use VAR in the right way. Not to say it couldn't mean we can't have big losses, but I think if you use VAR in a right way, and don't get blindsided by the weakness of some of these things, it's probably still a decent measure compared to some of the other ones, which are, I mean, again, Sharpe.

Rich:

We’re talking about lesser evils here.

Niels:

Right, exactly. Okay, fair enough. That's fair enough.

Okay, so good. Let's move on quickly to the last point because that is obviously, we've been leading up to this, and that's kind of the application of these things within trend, but just maybe, we are limited by time today.

Rich:

How long have we got, Niels?

Niels:

Oh, you have 10, 15 minutes, for sure.

Rich:

All right, so what we've been talking about with convexity is not new for trend following. We've known this, and this is why we've developed our models, our processes this way.

What we're looking at with trend following is our ability to cut losses short at all times and let profits run. So, what we're doing is we're applying a barbell approach to minimizing adverse volatility and exploiting opportunity with our process.

And we've got a few other tricks. We have very small bet sizes. What we're doing is we are limiting the risk of any individual trade that we do to being a very small risk, but we allow each trade to explode in positive opportunity where suddenly that could be contributing a significant component of what I call unrealized equity to your portfolio. But we're not compromising the risk of what I call the core capital. So, the core capital is the capital that we start out trading with when we start our models and we enter a trade. We've got what I call core capital comprising the capital worth of all of the closed trades at that point in time.

Then we're only exploiting a very small risk bet of our core capital to our trade. So, our core capital is protected with a very small risk bet. That possible bet can explode an opportunity where we're holding significant unrealized risk. And that unrealized equity could be 5% of the value of the portfolio. But that was the percent of the portfolio at this point in time, not the percentage of the portfolio when that trade commenced.

Niels:

Okay, so I do want to, sorry, I really don't really mean to, but I do want to interject one thing because otherwise I will forget. But I think this is important.

I think what you're describing there, because this has been one of the discussions we've had over the years on the podcast, should you look at closed equity and open, or should you look at the total equity? And I think what you're describing makes sense for those who want to do it.

However, I don't think it makes sense, honestly, when people manage other people's money, because what we've seen last year was that, with some managers, the open trade equity became so large that people who bought into those funds at a certain time had open trade equity of 50%, 60%, 70% and therefore was much more at risk with their capital, with their investment, not the people who had been in for years.

And so, I do want to just mention that because I think this is something that we had not really seen for a while, that you could have that much open equity from one or two really big trades. And so, I think to me at least, and people don't have to agree with me, but I think this is much more a principle that makes sense when you manage your own money and when you have that infinite runway.

Rich:

Yes.

Niels:

But if you buy into a fund that has that much open equity and where the stops are so far away that you can dive into a straight 50% drawdown, it changes the way I think about that concept.

Rich:

If you apply a process to allocate AUM that way. For instance, you could apply a process that applies it based on closed equity. The unrealized equity is not a problem of the new incoming investor. That is something that the existing investors in the fund have.

So, there are different models you could apply for that. But also, the problem is that what we're looking at now is the business of trend following as opposed to trend following.

So, the philosophy of trend following to me is do not do trend following unless you're prepared to do it for the long haul. Do not look at the individual results of individual trades. Look at the last 2,000 trades, look at the last 4,000 trades. All of these things are designed to ensure that there's patience, discipline, long term horizon.

Everything with what I'm talking about is the philosophy of compounded wealth over the long term. To use trend following for other purposes, for instance, as allocations into traditional portfolios, all of these things, put in other concerns that have to be addressed for those people that are doing that, it's not so straightforward for them.

It's also not so straightforward for people trading other people's money with AUM, for instance, for the investor coming in now with a sizable investment and there's this significant unrealized equity and they lose that in the next short term. That is unfair. So, there are all of these additional problems. That's the business of trend following.

Niels:

Yeah, and I didn't mean that it was unfair. I just said it opens up a risk that people should be aware of because it's slightly different.

Anyways, I, I was sidetracking you. I apologize for that. But it's just something that came up when you talked the closed equity versus open equity and so just wanted to mention it. Anyways, back to your…

Rich:

So, where I think trend following has an added extra nugget to the traditional tail risk protection strategy is that trend following is really aggressively looking to amplify upside. And it's doing this by virtue of its models that let profits run. It's doing this by virtue of some models that aggressively concentrate into material trends.

So, there are some trend followers out there that, you know, they might pyramid into trends. I apply an ensemble of models into trends. What I'm doing is I'm not just accepting… What I'm aggressively accelerating, I'm aggressively accelerating with my models into opportunities or what I call material outliers. I don't know what, in the future, is going to occur.

But once my signals say we are possibly on a material outlier here, that's where I dip my finger in the water, take a trade, and if it continues to go my way, I, what we call, apply an anti-Martingale degree of progression into it. I'm applying a new strategy, a new strategy, a new strategy. This is different. So, an anti-Martingale is a positively convex approach.

A Martingale, which is a betting strategy applied in casinos where people might lose $10, they'll then double up $20, lose that, double up $40, lose that, double up $80, they might win that, but then they only win $10, they have exposed $80 of risk. That is a concave strategy. That is a Martingale approach.

An anti-Martingale approach is where you get a small loss, a small loss, a small loss, a small loss. Boom. A big win that significantly pays for all of those losses and more. Ed Seykota, one big trend pays for more. That's positive skew. That's anti-Martingale. That's convexity.

So, with trend following, we ante up the exploitation of upside volatility. Because something that Dave Dredge and the others don't really talk about is diversification. And trend followers, we are heavy into diversification.

So, what we are doing is we are spreading the net wide. And we're spreading the net wide for two purposes, to achieve a degree of uncorrelated properties in our portfolio, but most importantly, we're doing it to find those outliers wherever they may roam. Because these things are unpredictable.

You can't say it's going to occur in this market in 10 days time. You don't know where the next outlier's going to be. So, you need a wide net to be able to exploit those opportunities when they occur. So, we are amplifying, many times over, the upside potential of our portfolio strategy.

So, we have talked about tail protection as possibly being with options as an optimal strategy to protect downside for a traditional portfolio, and the slight weakness that exists in trend following for very short-term corrections. We don't provide that degree of convexity.

But where we really throw the edge at the tail protection mob is that we are here not just to provide insurance. We're here to provide amplified compound growth over time by exploiting or harnessing outliers and by harnessing positive volatility. So, we cut losses short, which gives us a shield of preservation, discipline, quick exits. Preserving capital is paramount in our outlier hunting.

Small controlled losses are seen as the cost of doing business, allowing traders to stay in the game and position for the next big opportunity, which you never know in advance. We let profits run - exponential potential. What we're doing is we're striving for a term called Infinite yield. Now what that simply does... Infinite yield. I like that term infinite yield.

Niels:

Well, it's one of those things you could never advertise, right?

Rich:

Yeah, you can't say, I'm going to offer you an infinite return. But the principle there is a return that pays for all of the costs you've had over the course of time. It's something that has significant exponential upside growth opportunities. So convex strategies shine here. It comes to capitalizing on these trends.

Outlier hunters design their approach to ride winning trades for as long as the trend persists, maximizing the potential payout of each successful position. But this requires the art of patience, and it requires the power of asymmetry. So, these single outlier events, they can offset dozens of small losses.

Whether it's a surge in commodities like cocoa or a massive rally in equity markets. The ability to ride trends until they naturally exhaust is what turns a good trader into a portfolio defining one.

The other principle is keep bet sizes small. This is about surviving volatility. So, outlier hunters, they prioritize survival over short term wins.

So, their focus is on survival by being in the game so that when those outliers arise, they're there to participate in it. Because there's nothing worse than being knocked out of the game.

And so that's why we have these small bets. They're the ultimate form of stop. Stops themselves can be broken. We know that in very fast moving markets they're not always observed. But small bet sizes are the things that really keep our losses small and linear in nature.

Staying in the game is such an incredibly important thing if we want to exploit principles of compounding, because compounding is something that takes time to achieve and you're required to have skin in the game for a very long period of time to exploit the power of the exponential growth that compounding gives. The fourth great principle of trend following, which is not necessarily entitled protection, is diversification as a safety net.

It's not only about a safety net, but it's also about spreading already net wide to capture those outliers, expanding the opportunity set, and also risk reduction through diversity. So effectively, what I'm saying here is it's not something new to trend following, but the term convexity, what Dave Dredge is bringing up, is something that really resonates with us because it's doing all of the things that naturally come from our philosophy and our approach of how we deal with it. So, that's basically a wrap up.

Niels:

Yeah, this was wonderful. And I think maybe you and I can coin a new term that we as trend followers we follow the ‘ni tram’ strategy, which is Martin spelled opposite.

Rich:

I like that one.

Niels:

Anyways, this was really great. I think you touched on a lot of really Important points, and I think that these different narratives, these analogies that, you know, we do live in a very narrative driven world and if we can come up and, as we've often discussed, I mean, maybe we haven't found it yet after all these years, but if we can come up with some explanation that really resonates with people, I think it will help more investors understand.

Rich:

Yeah.

Niels:

Why they need it in their portfolio. That it is not just something that we talk about, but actually that there is real value to be had. Also, as I think we talked about in the very beginning of our conversation today, it feels, at least to me, that the world is becoming much more divergent in so many ways, politics, trade between countries, you know, whatever it may be.

Rich:

No longer is there this coordinated central bank intervention. No longer is that the case. Each economy is doing their different thing. Currencies are going their different ways. Everything's becoming less…

Niels:

And we're not all friends anymore. Right?

I mean, you know, we only have to go back 10 years maybe. For those who are relatively new to finance, they may think, well, you know, what are you talking about? I mean, have we have ever been great trade friends with China and Russia?

Yes, we have actually, because the world thought, that's the way to bring piece to the world was just trade with each other, but that's not necessarily how it panned out.

So anyways, I think, personally, that we are in a period that will perhaps shine even more positively on these types of principles and therefore these types of strategies. And I think it's probably more important today than ever before that people take a really good look at their portfolio in terms of what is the true profile?

Back to this idea that people think they might be diversified, but in reality, they're full of short vol strategies essentially.

And also want to give a shout out to Meketa. I met with them in Miami and they have a wonderful paper about first responders, second responders for portfolios, which ties beautifully into our conversation today.

However, we have run out of time. Rich. I hate to say it, but we have. But I think if people will just take a few minutes to show their appreciation, I saw a few really nice comments and reviews coming in this week.

So, if you appreciate all the time that Rich has put into preparing our conversation today, do go and leave a rating and review on one of your favorite podcast platforms and we will certainly read it, and we certainly appreciate that very much.

ut on his first appearance in:

So, if you have any questions for Nick, as always, you can email them to me at info@toptradersunplugged.com. That's it for this week from Rich and me.

Thanks ever so much for listening. We look forward to being back with you next week. And until next time, take care of yourself and take care of each other.

Ending:

Thanks for listening to the Systematic Investing Investor podcast series. If you enjoy this series, go on over to iTunes and leave an honest rating and review. And be sure to listen to all the other episodes from Top Traders Unplugged.

If you have questions about systematic investing, send us an email with the word question in the subject line to info@toptradersunplugged.com and we'll try to get it on the show.

And remember, all the discussion that we have about investment performance is about the past, and past performance does not guarantee or even infer anything about future performance. Also, understand that there's a significant risk of financial loss with all investment strategies, and you need to request and understand the specific risks from the investment manager about their products before you make investment decisions. Thanks for spending some of your valuable time with us and we'll see you on the next episode of the Systematic Investor.

Chapters

Video

More from YouTube