Why the Best Trend Models Might Be the Simplest Ones ft. Tom Wrobel & Andrew Beer

[00:00:01] 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.

[00:00:01] Niels

Welcome and welcome back to this week’s edition of the Systematic Investor series with Andrew Beer and Tom Wrobel as well as myself, Niels Kaastrup-Larsen, where each week we take the pulse of the global markets through the lens of a rules-based investor.

Andrew and Tom, it is great to see both of you. Great to have you back this week. How are you doing? Where are you today, actually? Because you’re always traveling, some of you at least.

[00:00:01] Andrew

Well, I’m dialing in from Greece. I had sort of a spontaneous trip to Greece to spend time with my business partner. I’m back in Athens now, where I have better Wi Fi and I head back to the States tomorrow. But it is truly a delight to be here and nice to see you both again.

[00:01:02] Tom

Yeah, thanks Niels. I’m still in London. It seems to be this summer has yet to slow down, although that seems to happen every year. We still see a lot of groups traveling and a lot of people interested in catching up and talking about different strategies.

[00:01:17] Niels

Interesting. Actually, I always felt that the summer period was a little bit slow in terms of travel, but interesting that you see managers come through London. Do you have any summer plans after that or going somewhere nice, Tom?

[00:01:31] Tom

Yeah, just holidaying in sunny UK, in Cornwall.

[00:01:35] Niels

Fantastic, oh, very nice actually. Good stuff.

Well, we got a great lineup today which actually includes some topics we rarely talk about. I think this is going to be fascinating and interesting, as well as a paper that you published, Tom, so that’s going to be super.

Before we do that, as usual, I’m always curious to know kind of what’s come across your desk, what you’re finding interesting at the moment, not necessarily related to what we’re going to be talking about today. So, since Andrew and I were here on our own last time, I’ll give you the chance, Tom, to open the floor with your ideas.

[00:02:15] Tom

Yeah, I think what’s been interesting is we’ve seen the continue of, but maybe the evolution of the multi strat talent war. Multi strats still definitely continue to see talent as their key alpha and edge. But not only are we seeing the hiring and the continued kind of search for talent to bring PMs in house, we’ve seen the expansion of a theme that started a few years ago where they’re getting more and more flexible to how they are structuring the deals with star portfolio managers. So, the continued and increased use of external SMAs, if they can’t bring someone in house.

And then, the largest launches, really, are almost spin outs. In single managers the largest launches are almost spin outs, or multi strat seeds, where they are providing capital, early stage, to talented individuals. And it’s really changed the evolution of how you think about an emerging manager. Because an emerging manager is now often a manager who’s anchored with a very large ticket from a multi strat, and then with some of the interesting complexities that brings – maybe to do with most favored nations or a period of exclusivity.

[00:03:35] Niels

Yeah, that is interesting. It’s certainly different from when I began in this industry, that’s for sure.

What about you, Andrew? As you’ve been traveling to Greece, what have you been focusing on?

[00:03:48] Andrew

Oh, well, I’ve been trying not to focus on too much. But just a funny anecdote. I was doing a webinar yesterday and our mutual friend popped up, out of the blue, to asked, once again, how you can justify using the term fee reduction as the purest form of alpha. So, he was hiding in the tall grass, but I could see that it was him.

So, I would just say I am ready for round two to kind of pick up that discussion again whenever he’s ready.

[00:04:20] Niels

Fantastic, okay, well, putting down the gauntlet, or whatever it’s called, that’s for sure. That’s good, excellent.

All right.

Well, that’s a funny story, actually.

What’s been on my radar is not really that… well, two things, two small things. And I don’t want to dwell too much on it. It’s just something that when I see these headlines, I just find it extraordinary.

Last week I mentioned something whereby, you know, as we in the west, we’re doing “less and less business together” because we are seeing a lot of protectionism, and tariffs, and all of that stuff. China seems to be going completely in the other direction. They’re doing more investments this year, I think, abroad than ever before.

And then comes this headline which is really, to me, I had no idea. But apparently they’re building the biggest dam, water dam at the moment. And once it’s completed, it’s going to supply the equivalent to Britain’s entire consumption of energy in 2024 from one building work, so to speak. I mean, that’s extraordinary when I hear these things. It’s, obviously, probably good for the green transition, maybe not so good for the environment. Who knows? We’ll see.

And then the other thing that has caught a little bit of my attention, in the sense that I hear people talk about it, are these bitcoin yield companies that are setting up where they kind of try to replicate what strategy or micro strategy that has been done, but where, essentially I think, with some of them, they’re kind of asking you to almost pay twice the price for bitcoin because they just say it’s leverage. But why would you do that when you could just buy it yourself?

Anyways, I find it interesting that there’s so much demand for this. And we’ll see how it all plays out at some point.

From a trend following quick update, before we get into our bigger topics, my own trend barometer finished at 52 yesterday, which is strong. So, that very well ties in with what I think is going to be a decent month, if we can keep it up for the last week or so. And also, I think yesterday was a pretty solid day for CTAs. So, the numbers from the day before yesterday are as follows.

BTOP 50 up 1.08%, now only down 2.21% for the year. SocGen CTA Index up about 1%, down only 6.64% for the year. SocGen Trend up 1.2% for July, down 8.92% so far this year. And the Short-Term Traders Index also doing well, up 1.13%, and down 4.23% so far this year. We may not still be able to keep up with the replication strategies that Andrew runs, but we’re certainly giving it our best shot.

MSCI World up 2.3% as of last night, up 12.2% for the year. The S&P US Aggregate Bond Index down 22 basis points, and up 3.5% roughly for the year. And the S&P 500 up another 2.5% so far this month, up almost 9% now this year.

Now from my perspective, where I sit, I think there’s a couple of themes. At the moment we have metals doing pretty well as far as I can tell, and we have equities coming back doing really well. That’s what I see.

But I would love to open our discussion today, maybe with you Tom, giving me your perspective on the current environment. Then we can launch into your kind of bigger year-to-date type analysis and then open it up to all the other topics that you wrote about in your recent paper. And maybe we can also tie it into some other papers that have come out recently.

[00:08:09] Tom

Yeah, absolutely. I think, in our meetings with Global Macro, so more on the discretionary side, we definitely continue to hear very positive news from macro managers that they think this is one of the most opportunity rich environments for macro trading because of the variety of themes that are available. We’ve seen inflation remaining relatively stubborn. We’ve seen global interest rates from central banks being relatively high. I know, not high in absolute terms historically, but high in recent terms. And we continue to see large geopolitical dislocations. And that’s going to play through for macro, but it’s also CTA which is, arguably, part of that macro. If you think about the potential for different trends to emerge across different markets, this is obviously going to be a very interesting period that we’re going to be coming into in the second half of this year.

[00:09:09] Niels

Yeah, absolutely. You wrote a paper, as I mentioned earlier, which came out about a good month ago or so. I’d like for you maybe to set the scene, take us through the main findings, what you found interesting, what you found surprising perhaps and then Andrew and I, I’m sure we’ll have some thoughts to share, observations.

[00:09:33] Tom

Yeah, well the paper was really born from a research topic that we started many years ago where we built our own hypothetical paper traded CTA, I suppose, replicator. We looked at if we were trying to build a more transparent sort of way of looking and attributing performance across a trend following strategy, how would you go about it?

And so, we built our own CTA from the bottom up. And part of that process is the selection of markets, but the other big part is the model that you choose. And so, we had done some analysis on the type of model and we kind of narrowed down onto a moving average crossover model as delivering a robust consistent correlation.

And we were kind of really optimizing for correlation, not for performance because what we cared about was being as correlated as possible to our own SG Trend index. And the model set that we chose was a 21/20, so a 20 day short for fast moving average, and 120 days slow moving average which seemed to give a very robust correlation to how a CTA trades.

But then, in the more recent years, we discovered yes, we were realizing a long-term correlation but there were occasionally dislocations which were not worrying, but just areas to look at and think about. And we looked at it and tried to think, okay, how can we turn around what was a negative period of short-term correlation, where we seem to be seeing why the CTAs are maybe realizing positive performance and our trend indicator is losing performance. And we looked at markets, we looked at allocations, and really the key thing we saw was that our model appeared to be a little bit too fast.

And we think of the model that we use, in this trend indicator, as being more of a medium-term model. And what we found is that if we were to move to a longer-term moving average crossover, so the slow moving average is slower, we then immediately realized a much higher correlation during these periods of sharp dislocation. This wasn’t a cause for concern. It was just something that we were aware of, and we started monitoring.

And I think Katie Kaminsky was writing a performance review of last year and wrote about what she called the speed factor. So, looking at is a slower model better or worse than a faster model?

And what we were sort of observing was that maybe the speed factor had shifted to be better performance from slower models. And we did that analysis about, okay, if we correlate our trend indicator over a variety of different time frames to the trend index, what do we see? And that became a nice heat map.

And we see that actually, yes, maybe 15 years ago CTAs did look like they were more medium-term, but there’s been a pronounced shift to them drifting or moving to them employing what appears to be longer-term models. And that made sense because the performance in the longer-term time horizons was better. And we kind of did a comparison and we can see, yes, actually better performance was in the longer-term periods.

But what we continue to see is actually there are periods, occasionally, when shorter term seems to be better. And shorter term seems to have a little bit of correlation to the trend index. And this kind of plays true with what we hear from CTAs when we meet with them in that in their toolbox isn’t just one model, isn’t just one thing. They have a whole host of tools at their availability, and time frame, and the way they deploy models, is one of them, and the way they adaptively allocate risk across those models.

And so, what we’ve come to the conclusion is that trend followers have adapted and evolved their model sets, but they retain the ability to deploy risk in these shorter-term models, maybe during a period of dislocation or market stress. And that gives them more adaptability to a fast-changing market regime.

And when we wrote the paper in Q1, we were actually seeing the perfect storm as a result of the tariff environment. Our flagship CTA index and the specific trend index was realizing a very difficult period of performance and were down I think maybe double digits at that point. But our Short-Term Traders Index, and when we looked at the individual constituents in the Short Traders Index, those shorter term CTAs, over half of them were in positive territory and actually delivering really strong performance.

Which kind of really reinforced, in our minds, the sort of adaptability of the shorter term CTAs. And that a CTA isn’t just one model, it isn’t just a fixed kind of way of looking at risk and a fixed way of looking at the world. There’s a lot going on that’s maybe proprietary to each CTA with how they allocate risk.

[00:14:51] Niels

In the paper I noticed that you created a dispersion chart, and I believe that’s between sort of the managers in the trend index.

[00:15:02] Tom

Was that towards the end where we were looking to show..?

[00:15:05] Niels

It’s actually exhibit one. But my point was really that what surprised me a little bit, I don’t know if it surprised you, is that it looks like, for the majority, that the dispersion hasn’t really, you know, it’s not that huge, even though there will be one or two outliers, I’m sure. But it’s, I mean, okay, I say not that huge, I mean, it’s still +/-10% maybe. That’s obviously, you know, pretty big, or a 10% difference between the big ones that are in the index. But it just showed, it looked to me like, in the early days, that the dispersion was actually bigger than it is in the last five, ten years or so.

[00:15:52] Tom

I think what we’re trying to show with that chart is that, although there is a consistency of what people think of as trend following, there is always performance dispersion.

And what we’re trying to show is that everyone’s use of the same techniques can result in a very different outcome in how a CTA delivers performance. The way you think about correlations and risk can really have a very important impact on the returns that you deliver.

And therefore, your choice of model trading speeds, or how you allocate risk across those trading speeds, your choice of markets, the way you might run stops, or manage risk, or cut risk, is very, very important. And it’s something that’s been really obvious this year, when CTA’s, in the first couple of quarters, realized a sharp drawdown, because of a number of markets reversing and going against their existing trends. The CTAs would have been cutting risk very, very quickly. And so, the shape of a drawdown is very important.

And I think Man group have written some really interesting research on V shaped recoveries, which seem to have become more prevalent. But are they really here to stay and are they really going to always be V shaped? So, the shape of the drawdown is really, really important for how CTAs manage risk.

[00:17:25] Niels

Yeah, I’d love to hear your thoughts on all of this, Andrew, but I’m going to start by asking you a question that I was thinking of when I heard Tom talk about these things. When we’ve discussed this previously, Andrew, you often mention that the wave replication works. Is that it’s actually kind of a slow version of trend because of trading once a week, and so on, and so forth. And I’m wondering, and I agree with Tom when he says that I think managers have become slower. So, I’m wondering whether you think, if you had started doing this 20 years ago instead of maybe 10 years ago, do you think your replication would have been faster by design, if you know what I mean? Meaning, do you think that the change we’ve seen in the underlying managers is also being reflected in how you settled on doing the replication? And, from my understanding, is this is what you’ve stayed with, meaning you don’t adapt your “speed or your design”.

[00:18:33] Andrew

Sure, so, there’s a distinction between what I think we’re picking up on in the underlying CTAs versus the horizon that we look at. So, we rebalance our portfolios once a week. It’s a very short-term bet.

It’s basically a bet that we’re going to be accurate enough on Monday that, with our 10 or however many core positions we’re using, by the following Monday will have given you an accurate representation of the pre fee, pre implementation costs of the space. When people ask me what we pick up on, and they ask are you picking up on a lot of short-term trend? We’ve been saying, which I think mirrors what Tom is saying, is that the principal driver has been longer-term trend.

Now, going back to 2015, though, we did exactly the same analysis that Tom has done. We built all of our own bottom-up trend following models and at different window lengths. And this was when QIS (this is early 2015), so, this is when QIS is… Everyone has their own QIS models.

And we did a lot of work on the QIS models among different banks. And what we found was that the reliability of those models was, if you wanted to establish a strategy in 2015 and have it still working in 2025 (sorry, I’m losing my years, so yeah, yeah, 2015. And you wanted to have it work in, in, in 2025), the question is, would the industry change?

And so, the problem with the bottom-up indices is that they’re done at a particular moment in time and the industry does evolve and change. And what you get is a steady and widening dispersion over time. The correlations tend to drop because people…

Look, I’ve been in this industry for a very, very long time at this point, there are smart people who will evolve, and adapt, and change, and try to make their portfolios, their strategies better. So, the reason we went with the top-down replication… So how would you test that? So, 2015, you say, what model would we have chosen today? Would it be 120 over 20? Would it have been in longer-term, shorter-term, would have been a combination of them? And, by the way, that’s what JP Morgan was basically doing on the quant side at the time.

But then you say, well, what if we’d done the same analysis in 2012? What about 2009? What about 2006? Would we have reached the same conclusion? And the conclusion was, as Tom says, that it changes. So, we could look great for a year or two.

And there are examples of this. Now, Lucas has a great managed futures index which worked really, really well in a period of time. But if you go back longer periods of time, where you don’t have equity exposure, the diversions between the industry gets quite wide.

So, the whole point of doing the top-down replication is we don’t want to make a long-term bet on whether people are going to be more on short-term, long-term, medium-term. Rather, we want something that is sufficiently adaptable. And what you find with a well-designed top-down replication models is your correlations don’t drop, they stay in the 80 to 90 or low 90s because it is that constantly adaptive process.

So, replication always has that weird paradox in that I think what we have been picking up is longer-term trend. If people go really short-term and as we talked about, you know, if everyone becomes Crable overnight, then we’ll see. But I don’t think that’s going to happen.

[00:21:52] Tom

Andrew, I think you’re absolutely right on the over optimization of how you would be choosing a trend model. And I think it’s something that we have experienced with our trend indicator and I agree there has to be some input that allows for the adaptability of tweaking your model parameters.

[00:22:10] Niels

Now I think you and I, Tom, exchanged some thoughts about when you have periods of time, you mentioned that this year there was, in my opinion, a fairly short period of time where short-term managers or models (let’s call them models) instead did much better than trend. But the numbers are the numbers. And, in the long term, short-term doesn’t work as well as long-term.

But then there’s this midterm which a lot of people I think have thought, okay, that’s probably a happy medium to be in. But I think your data suggests that, actually, maybe that’s not the case.

[00:22:49] Tom

Yeah, it’s something we’ve realized because what our trend indicator was trying to do, it would appear (now we’re going to realize this after the event), was replicate CTAs that are by nature employing multiple models. But it seems that the predominant driver is these long-term trend models, where they are then blending in shorter-term models to maybe manage periods of market volatility better but also smooth out the return profile, is the danger you fall into that looks like, on average (the blend of long and short-term) it looks like medium-term. And medium term isn’t maybe what you were trying to achieve.

And if we look at the performance of medium term it can actually be quite painful that you’re not generating the benefits of long-term trend following and you’re also not benefiting from the returns of short-term models during stress periods, and you end up with something that’s neither one or the other. And I think when we look at the return profile and the returns in some of our heat maps, the medium-term can actually be quite a painful place to be sometimes.

[00:24:01] Niels

Any thoughts on this Andrew?

[00:24:03] Andrew

One of the issues we’ve had broadly with the bottom-up designs is you have to specify a lot of parameters to get there. You’ve got to decide how you’re going to size positions, you know, when you’re going to enter, when you’re going to exit them. And one of the issues with the kind of QIS space broadly, which was really designed to try to give straightforward ways of getting exposure to defined trades. And I wrote a paper on this on merger arbitrage. It sounds simple, right?

I don’t need to pay a whole bunch of guys 2 and 20 to basically buy all the stocks that are being acquired and short the acquirers. The problem is, once you get through all of the parameters you specify, do you use only US? Do you use non US? How do you size positions? What if it starts trading above the offer? You know, how long do you wait before the offer is announced? I mean there are a million different things you have to specify. And what you end up with is, instead of single manager risk, you have model risk.

And so, one of the challenges that we found was if we build a model, we know all these parameters that we’re specifying, it may not be the way you, Niels, would build a medium or a longer-term model, it may not be the way Tom does it.

So, drawing statistical inferences from something that we know has a lot of variability, we try to tease out as much information as we can. But we’re careful to try to not draw too substantive conclusions from it because we just don’t know how to look through that noise.

[00:25:29] Tom

And I think that’s what the first chart in the paper, which we can share, was really trying to show was that trend following is maybe seen as this kind of quite homogeneous group and when people talk about correlations, correlations of 0.7, 0.8, 0.9 even, are thrown around. Even amongst that quite controlled group of this SG CTA index, which is only 20 managers, and they’re all large established groups, so, they must be doing something quite similar, there’s a huge amount of performance dispersion in almost every single year.

And I completely agree with Andrew, it’s that model risk of the assumptions and the parameters that you choose. And it’s why a lot of the investor groups that we talk with will maybe start with a single CTA allocation and then move quite quickly to diversifying that with a number of CTAs or a number of different quant strategies which they see as starting to diversify the single manager risk and then starting to maybe dip their toe into other strategies that aren’t trend following.

[00:26:30] Andrew

Because what happens is, if you’re an allocator, you really value predictability. You’re building a whole portfolio and you’ve defined this category called trend or manage futures or whatever it is, and you’ve probably gone to take SocGen data, which is the industry standard, and use that, plugged it into all sorts of capital market assumptions. So, that whole analysis doesn’t work if you then pick one guy who’s 20 points off everybody else over the next two years.

And so, again, that’s model risk, it’s single manager risk. It’s not an accident that there’s a 30 point difference between the best funds and the worst funds this year, over the first six months of the year. Because that’s just what happens if you decide you’re not going to be in equities. That was great in 2022. If you decide you’re not going to be in currencies or something, that’s probably been great for you this year.

So, all of those little specifications, they add up cumulatively driving this very, very broad dispersion which then, from an allocator’s perspective, either drives you, as Tom says, to have multiple managers. The goal of which is to have a small subset of the 20 in the main index, or the 10 in the trend index, that gets you closer to it. And also, there’s also a behavioral aspect of it. If one of your guys is not doing well, but another guy’s doing well, you spend a lot of time shining a light on the guy who’s doing well, or lady who’s doing well.

So, you either go that, or you go a multi manager route with somebody like Abby or, or Efficient, and then we’ve tried to argue that we’re a third option to try to get broad based exposure to this space.

[00:28:10] Niels

And this is what you probably have to just then explain for the audience here, Andrew, because they might say, well hang on, Andrew is also making choices about which markets to trade. He’s also selecting not to have too much equities, not to have too much, you know, commodities or whatever.

But you seem to be kind of avoiding, if I can put it that way. You know, it seems like replication strategies are maybe not prone to the same challenges in making those selections in terms of dispersion.

[00:28:46] Andrew

Well, by targeting a large pool of funds, whether it’s the SocGen CTA index, whether it’s a pool of mutual funds, a pool of UCITS funds, etc., it is a meaningful drop in single manager risk. That is meaningfully dropping your dispersion relative to if you just try to pick one of the 20 and put all of your eggs in that basket, You can’t get it down to zero, but you’re going to get it down a lot.

But, in terms of the way that we ended up building it, it comes down to sort of a philosophical issue of what’s the signal in managed futures, and what’s the signal and how do you get efficient exposure to that signal? And so, what we found is replication is by no means perfect. It was just the most efficient and effective way we found of saying that you’ve got a lot of talented people out there who are building and trying to find ways to extract more alpha and more returns from the market. Can we find a way to basically tap into that signal in a way that we can not only deliver efficiently but also be able to put it into vehicles that will tend to be much more accessible to the investors who don’t have the ability to invest in these flagship hedge funds.

[00:30:04] Tom

And so, have you done some analysis of that reduction of single manager risk via your replication process? I’d be really interested.

[00:30:14] Andrew

Yeah.

So, if you look at our long-term correlations, when you do it, look, our correlations get into the 90% range, which is better than the mid-80s or low-80s that you get with a lot of different managers but it’s not perfect by any means. It tends to look more like the correlation structure of a multi manager portfolio.

So, Abby will tend to have a lower tracking error, to the broad industry, than any single manager. And we’ll be closer to Abby, but with the difference that (and again, this is also where it gets a little bit confusing) we’re trying to aim for what we view to be the pre-fee, pre-trading, pre-implementation cost returns, which will actually increase our tracking error relative to the reported indices or reported mutual funds which themselves are fully loaded for those costs. So, it’s a little bit nuanced.

[00:31:09] Tom

And if you (staying on this topic quickly), if you’re doing some sort of, I don’t know, principal component analysis, or factor loading about how you should be allocating your 15 asset portfolio to replicate the similar exposures, do you monitor how well you’re doing in replication? Is there ever a scenario where you don’t find you can allocate to the assets that you need and that your universe of assets isn’t then going to generate the similar return exposure?

[00:31:40] Andrew

So, we’ve never changed our models. So, in November of this year…

[00:31:44] Tom

But it’s not necessarily changing the models, it’s more, do you have the assets available that can generate the similar exposure?

[00:31:53] Andrew

So, my contention is that the only thing that matters in this space are major themes at an industry – at an industry level. And I think we’ll talk about this paper from Aspect. But you know, when people talk about what works and what doesn’t work, they don’t talk about heating oil. They don’t talk about esoteric small markets. Do you get inflation, right or not? And that’s going to play through the major market.

So, the conclusion that we drew, back in 2015, is that the alpha generation in the space is detecting changes in information early. So, you can put on contrarian trades that typical allocators are not capitalizing on. And the most efficient way to implement and execute those trades is through the biggest, largest and most liquid instruments.

[00:32:40] Niels

It’s very interesting, in a sense, because you’re absolutely right. I think from an index point of view, obviously big themes mean a lot. But then you look at individual managers, and you’ll find that “some of the best managers” in recent years, it’s been down to one or two markets that they’ve got right. So, it’s this really interesting sort of contrast between, on an industry level, yeah, it’s probably the big themes, so to speak. But actually, if you wanted to do your homework and try and find really interesting managers that could give you that outperformance, you probably would want managers who trade these smaller markets, and where we have seen some huge numbers being put on the board.

[00:33:36] Tom

There’s a perfect example at the moment, and it’s in sort of many headlines, of Mulvaney’s performance. Last year they are unashamedly very dedicated to pure trend following and being overweight in certain markets like commodities. And they generated huge outsized performance on the upside. I know that some of those themes would have come from cocoa, from coffee, where there were very strong market trends and that couldn’t be captured by someone who was looking at major markets. But then you seem to have the follow-up in 2025 when they go through a much tougher period of performance for those markets.

[00:34:17] Andrew

Remember guys, we’re trying to be the beta, right? I totally believe that there are people who are going to do things that are idiosyncratic and special. But our whole view on this space, and when we came in it as an allocator… In the same way that somebody says, look, if somebody says, I’m going to pick four people out of the SocGen CTA index to give me a rounded out portfolio. They’re not going to pick four people that are identical. You’re going to pick four people who think, you think give a balance – maybe one who’s longer-term, one who’s shorter-term. And the combination of them is supposed to get you back to something.

Ideally, it’s the index returns plus 100 basis points because you’re good at picking those four. But what they’re trying to get back to is the beta. They’re trying to get back to the overall beta of the space. And so, we as allocators said, we love the beta, we want the beta, how do we get that as efficiently as we can?

[00:35:07] Tom

But that’s strange because we don’t have any inquiries for someone asking for an investable structure product of the SG CTA index. I wish we did, but we don’t seem to have those inquiries.

[00:35:21] Andrew

Look, I think, I mean, but this goes back to the structure of the allocator. If I’m an allocator sitting at a pension plan, I’m picking four managers. It’s a lot more fun. I put out a search, you know, the centimillionaires and billionaires are going to fly in to see me. I’m going to, you know, I’m going to hear about all their statistical innovations. It’s my job, I like it and I’m going to keep doing it. That’s, that’s the way a lot of the industry is structured. And that’s okay. There’s nothing wrong with that.

But if you’re an allocator at an REA who, you know is trying to figure out a way, how you can bring some of the benefits of this to your portfolio and don’t have the ability to invest in four or five underlying funds and don’t have the time and energy to due diligence, buying something that gives you broad based industry exposure as efficiently as you can is really valuable. And that’s always been our focus. It’s how we got into the space being on that side of the table.

[00:36:13] Niels

So, I actually have a question I’ve been pondering for a little while, Andrew, because you talk about the beta. And I’m curious how you define kind of the… I mean this year you’ve done great outperforming the index, but still, where do you draw the line between kind of tracking error? I mean how much is okay and where does it become kind of, it’s not actually doing exactly what it says on the tin because it’s not replicating close enough even though it might be outperforming. But of course ,it could also be underperforming. But how do people know if you’re doing a good job?

[00:37:03] Andrew

I mean we do a lot of checks on what we’re seeing in our portfolio. So, for instance, I mean Tom has mentioned these bottom-up models. We have lots of bottom-up models that we run where we can essentially do the same kind of PCA analysis on those models to determine whether we’re picking up on the major themes.

So, we go through periods of divergence and performance, and again it’s… What’s a little bit apples to oranges here is if you’re comparing us to the SocGen CTA index or to the Morningstar category… Morningstar category is a little different because it’s accessible if you want to invest in 25 or 30 individual mutual funds, you can create it that way. I don’t know why anybody would want to.

But you’re comparing something that has a completely different fee structure, a completely different level of implementation cost relative to something that’s reported that is generally reported after lots of additional fees and expenses that you can’t invest in. So, what it was, what we did first and foremost, was an investment product.

Now what we’ve now done with SocGen is actually worked with them to create an index using the same replication model. And so, my argument to somebody is that if you’re a pension plan investing in the space, you’re using the hedge fund data because on your business card is, I am the Head of CTA Hedge Fund Allocations, and I’m looking for hedge fund data, and someone has given me a hedge fund hammer and I’m going to look for hedge fund nails all day long.

If you are a wealth manager who is saying, how can I get efficient exposure to this space? You are much better off using that SocGen replication based index to build your capital markets assumptions, to benchmark individual managers. And if you decide to then invest with us, now you have something that has no tracking error because it is apples to apples just like an SPY has no tracking error to the SPY.

[00:38:58] Tom

But isn’t that just the same as using Niels’s Dunn capital fund returns as the index in your capital allocation models?

[00:39:07] Andrew

I think we have a stronger argument that we are more representative of the broad industry than any single manager fund because of the dispersion issue. I mean you can look at our numbers over time. I mean, there’s plenty of variation on an annual basis, but we’re hovering above the index now, nine years running, and that really has to do…

And again, maintaining a high correlation over that period of time, of course with periods of time when correlations are going to break down, particularly when markets get more volatile. But the argument that I make to people is what I think is compelling about what we do is that it’s self-adaptive.

So, if we’re missing something over, you know, a week or two weeks, or copper is flying and everybody’s making a bit of money on copper, the model itself is self-adaptive. So, it’ll tend to correct itself over the coming weeks. The same thing happens at inflection points.

[00:39:59] Niels

I want to move on, but I have to say one thing about what you just mentioned. Maybe I’m not getting it fully correctly, so pardon me in advance if I get it wrong. It sounded to me like you said, okay, if we want to go out and we want to sell replication as kind of, we can give you the beta, but then at the same time you invent the beta by saying, well, use our replication model to create the beta. Of course, there’s not going to be much tracking error in doing so, but to me that’s a little bit cheating. But it’s a great marketing way of doing it.

And I think to seem to remember that in our conversation with the guy who asked you that question on the webinar recently, I think they had kind of, you alluded to, they had kind of done the same by creating a benchmark and then saying, oh, but we can outperform that or whatever.

My point is, from my perspective, I think if you’re saying you’re going to replicate the CTA index, well, then the tracking error must be calculated from the CTA index and not from your own model that’s just being replicated by another group. That is my personal opinion.

[00:41:04] Andrew

Right, so, you wouldn’t use the index. But you’re also not in the model construction business. It’s a very different business.

And look, and I think if you said, in five years, over the next five years, which is going to have closer performance relative to the broad industry, a DBI replication product or you as a single manager? I don’t think that’s a close race.

[00:41:26] Niels

No, absolutely.

[00:41:28] Andrew

And again, as I always say, look, what we do is we don’t have a 99% correlation. There are a lot of reasons for that. But what we’re trying to do is solve problems for people on the allocator side. And if you’re starting with just saying, what is better?

I mean, is it a better alternative if you’re a wealth manager to use an index of hedge funds you can’t invest in? You know, there are a lot of benchmarks out there that just don’t have a lot of utility from a practical perspective.

So, look, again, we’re just responding to the feedback that we’re getting from people. How do we make their lives better? How do we make it easier for them to invest and grow the pie?

[00:42:14] Niels

Before we move on to the next topic, I just want to mention and give a little bit of a shout out, actually, because you had written a paper, Tom, that was kind of inspired by the SocGen Trend Index having been around for 25 years now. But actually Aspect Capital also has done a paper recently where they provide a lot of details to some of the tricky periods throughout those 25 years. So, if people feel a little bit on the edge at the moment because of the performance so far this year, there’s some good evidence that this is not the first time it’s happened, and it certainly won’t be the last time. But I don’t think we have any time to go in to talk about it.

Now, interestingly enough, just before we move on, my calculations can of course be completely wrong, but maybe both of you will see whether they are right or wrong. I looked at the 10-year rolling returns of the SocGen Trend Index. And maybe just to give people a little bit of context, so, the highest 10-year rolling return of the index since inception is 180%. The lowest is actually a positive 1%. And the average is 60% and the current 10-year returns is 96%. It actually surprised me when I saw those numbers.

I thought actually it seems better than what people write about saying, you know, trend hasn’t done anything for a while. Well, I mean it’s done 96% in the last 10 years. That’s pretty decent. Anyways, let’s move on.

Andrew, you also brought some topics that we would love to get into. So, I’m going to hand it over to you and let’s see where we go.

[00:43:59] Andrew

Yeah, so, one of the curious things is that, when we looked at the space back in 2015 initially, we had very much of a two-pronged approach. So, my partner and our third member of the team who he was working with at the time are extraordinarily quantitative and they were basically doing everything they could to build models and pull everything out.

I do what I do, which is I talk to everyone I can find, and I knock on doors. And there was a curious disconnect in the way people were talking about the futures space or these strategies in that (and again remember this is the time where people were talking about commoditization of the space and QIS products were coming in) what you saw was this bifurcation that some people talked about trading futures contracts as basically almost frictionless. They’re the most liquid contracts in the world. Of course, you can just go in and buy anything you want.

And then there was another group of people who were talking about what are called implicit trading costs. So, implicit trading costs are, you know I want to buy XYZ futures contract and I start to try to buy it at 2:30pm, the mid between the bid and ask, let’s say, is at 10, and by the time I’m finished buying it at 4 it’s gone to 10.2 or something or 10.1.

So, the people who… (and these were both allocators as well as people who were coming from inside of running the trading desks at large firms)… The implicit costs were much, much, much bigger than people expected.

And so, when people talk about what we do, we’re saying we’ve used a round number of maybe 200 basis points of all-in costs across the industry as the implementation costs associated with a typical CTA. And I think that’s actually probably a bit light.

So that whole idea of implementation costs I think is very relevant today because one of the ways that a lot of firms have responded to threats of lower cost products like what we do, like the QIS products, I think, has to introduce more and more complexity into their portfolios.

So, we’ve talked a lot about people adding more, and more, and more, and more non-core markets. And it’s almost like a badge of honor how many markets you trade.

So, the question that we’ve always asked (and we’ve been extraordinarily focused on implementation) is, it’s great, but do you really think market 193 trends so much more than market 14 that if you really got through all these implicit costs, and you’re a big player in the market, and that adds 400 basis points of incremental trading (implicit trading costs) in what you’re doing, do you really think it’s trending so much that you’re going to get another 1000 basis points of returns on that instrument, that net you’re still adding 600 basis points to what you’re doing on it?

So, I think that’s a big issue right now where a lot of the more complicated strategies, things with many more esoteric markets, were supposed to do better in a period precisely like this, and they’re not. And so, I think that there is (I’ve called it) a complexity crisis. What is going on?

A lot of really smart people have added these different things to their portfolios. They don’t seem to be working. And I think this is the Voldemort of the space that I think people have not focused enough on the allocator side as well as to whether these things are worth doing once you work through all of the incremental costs associated with it.

[00:47:29] Niels

I find that interesting and I’d love to hear your thoughts, Tom, on that.

[00:47:33] Tom

Well, yeah, I think you’re right, but I believe the way CTAs look at markets is that when they are adding markets, costs and transaction costs and slippage have to be a huge part of the process of adding that market. And I think CTAs have really positioned themselves to be market implementation experts over the last 10, 15 years.

Futures, many years ago, were seen as an exotic market. And slowly what was previously an alt market has become a more traditional market. And that process of erosion and more standardization seems to be continuous.

So, I would counter that by saying if the transaction costs are too high, I don’t believe a CTA would be adding that market to their CTA portfolio. But, if you can generate a positive information ratio, Sharpe ratio, from adding that market, then you’ve got to remember that, in isolation, each market has probably got quite a low… I think this has been spoken about before. A single market with a trend following strategy has an information ratio of maybe 0.5, 0.4, 0.3, maybe even lower.

But if you buy into the kind of classic portfolio theory that diversification is the only free lunch, you would want to add that market because then, in aggregate, over the large number of markets that you can add (and we do see CTAs, as you correctly say, trying to add as many markets as possible)… You are benefiting from the marginal benefit that extra market is giving you.

Now, that’s assuming that the market isn’t highly correlated to all the other markets that you’re already trading. And I think the danger is in flipping the problem on its head. If you aren’t adding that market, it isn’t just a decision that you’re not adding it, it is an assumption that you’re making.

You’re making a reverse assumption that the markets that you have chosen are going to be maybe outperforming or doing better on a cost-adjusted basis than the potential market that you could have added. And that is a dangerous assumption to make. And you’ve got to be very aware and cognizant of that assumption.

Is there an element of luck that major markets have performed better in recent periods? Maybe. But it all goes back, in my opinion, to why are institutional allocators allocating to these types of diversifying strategies? They are trying to diversify their portfolios, of which what typically is equity risk. Equity is the primary risk asset that a lot of groups will hold. And therefore, they want the performance characteristics of these types of strategies to diversify that risk.

I don’t like the term crisis alpha. I know it’s a term that gets used a lot, but I think they aren’t necessarily looking for crisis alpha, but they are looking for the ability to build risk mitigating portfolios. And having those types of markets in the portfolio, you know, the more esoteric markets, maybe markets like commodities (which I know that you exclude), that is potentially generating the interesting performance during these periods when people do want their CTA portfolios to deliver outsized performance gains that can diversify losses in other areas.

[00:51:02] Niels

The way I look at it, and I don’t go into the nitty gritty about, you know, does it cost $2 more to trade cocoa than it costs to trade the s S&P 500 or whatever it might be? I think, in a sense you’re right, Tom, about, I’m sure we as managers, we really look carefully before adding these markets and so on, and so forth.

But I will say, and I’ve been on record many, many years saying this, that the narrative that started like maybe 10 years ago by now, where managers came out saying, oh, it is so much better to do trend following on a portfolio of 300, 400, 500 markets, and then you include all of these markets. I’ve always kept saying, “show it in the data”. Now that doesn’t mean that it’s not going to be different.

If you want to be different, of course, if you were trading Chinese markets, I think in 2021, you probably had more trends and opportunities and you would have done better. I’m not sure whether it was 2021 or 2020, thereabouts. But, if I look at the last 10 years of returns or even rolling 5 years of returns, I don’t see these ultra diversified portfolios doing better than those who focus on say 70 odd, 100 odd markets of the most liquid and well established futures markets. I just don’t see it in the data. And nobody has come forward and sayed, oh yeah, I can show it to you.

So, my question is actually the following. Is it a true belief of these managers, where they can show the evidence? Or maybe they can only show the evidence in backtest because obviously there you might find some argument, but the future data, since they started doing this, has not supported it.

Or is it just a really great way of positioning yourself, from a marketing point of view saying, oh, we’re going to do this differently to all the others. Because that’s the first question we get as a managers, “How are you different from all your competitors?” So, you have to come up with something. And this was definitely something that worked well, in my opinion. I know it’s a little bit harsh, but that’s how I feel.

[00:53:15] Andrew

Now listen, I mean, it’s a great idea. There’s just no evidence behind it. And, again, the idea that you’re getting a meaningful diversification benefit going from 150 to 151 positions is just, like, that’s storytelling. I’m sorry, I mean, the problem is that also people have an incentive, as you say.

This has been an area that has had lower cost, simpler competitive threats. One of the very awkward statistics is that the simple managed futures ETFs in the US, that are derided for their simplicity, are doing meaningfully better than the more complicated mutual funds and doing meaningfully better than hedge funds this year. It could be luck. It could just be one small episode, but it doesn’t appear to be an entirely isolated thing.

Look, all I’m saying is, if I’m an allocator, and somebody comes to me and says I’m moving into all these markets, I want to know what kind of work are you doing on the implicit cost?

I’ll tell you what we did, and we trade 10 of the most liquid markets, is as far back as 2019 we were the first futures-based client of an algorithmic trading firm called BestEx. And so, we analyze every single trade that we do to understand exactly, you know, when did we start… what was the arrival price at the time, how did we do relatively? We are constantly benchmarking ourselves relative to whether our execution is as efficient as possible. And we’re trading easy stuff.

So, the question that I would ask is… And the thing is, allocators have not had a strong incentive to ask these questions because they like the story. And when people like the story, they don’t ask hard questions about it. But now we’re at a point where there are a lot of examples out there that we can point to where it looks like it’s backfired. And that’s usually when you have this, you know, kind of soul searching as to what are the questions we didn’t ask at the time. Was this predictable, in a way?

And again, I mean, when you’re talking about relatively… So, you divide implementation costs between the explicit costs, which are ticket charges, and then you can do very sophisticated analysis on what are called implicit costs, which is basically the ripple effects that you create in the market.

And the general rule of thumb is whatever your explicit costs are, it’s probably three to five times that when you’re getting into implicit costs as you become a scale player in this space. We see it in a very, very small degree with the stuff that we’re trading. And we’re trading in 10 of the most liquid futures contracts in the world. So, you know it’s a lot bigger. You know it’s a lot bigger. And the question is, what should you demand as an allocator? How much excess return should you expect? How much diversification value?

And I think what’s happened is that, you know, kind of talking about the nature of the typical allocator in the space, it’s the story they want to hear. And it’s, and it’s a story that managers want to give. But I think it’s time to kind of sharpen the pencils and look at this because… I have a metaphor that I’ll tell you in a second, but sorry…

[00:56:25] Niels

Okay, sorry to interrupt you. I really have a question for you here. So, you talk about implementation cost. So, in a sense what you’re trying to get at, I guess, is implementation cost but also slippage, in a sense.

Okay, so here’s my question. Well, in order to measure slippage, you must have a reference price. You must say, okay, I’m aiming for getting the price at 12 o’clock on the day of trading. So, if you trade every Monday, you want to get the closest possible to the price at 12 o’clock in all the markets you trade – just an example.

Okay, fine, that’s fine. And you could do what you do to optimize that. And I think that’s good to do analysis. But my point is, it’s kind of random because what if you said, I’m just going to get the closest price to the price that is at six o’ clock that day on the market? The price would simply be different. Do you know what I mean? And we don’t know. So, for me, I don’t disagree with the fact that, of course, the managers need to be paying attention to how they execute.

But still, to me, I think it’s all about, okay, well, if you’re a manager and you run backtest, you use, say, the close price, or you use the open price in your data feed for your backtest. And so that’s what you should target in your live trading because, you know, we trade what we test and we test what we trade.

So, for me it’s kind of a little bit random whether we got a good price that day because the price could have been even better had we decided to do the 12 o’clock fill instead of the 6 o’clock fill. And so, the small marginal gains you can get, whether you get a small difference on the actual point of trading, I think most managers today (at least the managers I feel I know well), I think they do an incredible job to get very close to what they feel is the reference price. And I would be surprised if there was a lot of slippage, either positive or negative, on an annual basis from that reference price.

[00:58:33] Andrew

There are a lot of different ways you can measure it, right?

Like a lot of things you have to specify your parameters. So, you can either do it in a way where you’re trying to honestly grade yourself. That’s the indoor voice. And then you can also pick your settlement price. You can do whatever, in a way that games the system in your favor.

A meaningful number… And by the way, Transtrend (shout out to Transtrend), they wrote a great paper on how bad the UCITS rules are, how inaccurate the rules are around implementation costs. Because there are large CTAs out there that in their regulatory filings on their UCITS funds are showing zero implementation costs. We know that number is wrong. Whether it’s 180, or 250, or 79, or whatever, we know that number is wrong.

The question becomes (and I’ll use a metaphor for it), there’s a signal underlying any strategy. Now, the trend following signal is not a natural 2 Sharpe ratio strategy. It’s maybe .6, maybe it’s .5, maybe it’s something.

So, let’s imagine though, that you trained an AI whatever, on 50 small micro-cap US companies. And you’re trying to figure out what these guys view of the economy is. And it’s a large enough sample and you’re analyzing absolutely everything you can possibly analyze. And the metadata comes out that, oh my God, these guys think, actually, their order books are falling apart. They think the economy is much worse than they expected and it’s not reflected in the equity markets. So, you got this big shiny red sell-signal.

What do you do? Do you then walk out and say let’s try to short 50 small micro-cap companies? It could cost you 10 points before you get exposure to it. Or do you go out and short the S&P 500 or the Russell 2000 futures contract? So, in most quant strategies, you do all this research and you know that settlement price is going to be wrong, or the open price is going to be wrong in terms of where you can effectively trade during the course day.

So my point is, look, we’ve been for six or seven years, a huge part of the research that we conduct is to make sure (because we believe that in a space where you have a naturally low Sharpe ratio for the underlying signal) it’s getting that as efficiently as you possibly can, where nickels and dimes matter an enormous amount over time. And I think that’s just an area that because, again, the industry structure and what allocators are looking for, they haven’t asked these questions and I think they should.

[01:01:12] Tom

I think these are valid concerns. I would go back to your question about moving from 150 markets to 151, and I would agree that doesn’t provide huge marginal diversification, but I do believe moving from 15 or 50 to 150 does provide much more marginal diversification.

And if we are talking about the performance of alt market CTAs, I mean we can question it, but you only have to look at some of the largest alternative market groups. I don’t think you can deny the success of Man Evolution, Man Frontier and many other groups like Systematica who have built up fantastic track records and performance in their alternative markets products.

[01:02:02] Andrew

But I think, actually, the data on that is quite interesting in that they worked incredibly well for a period of time and a lot of them are working terribly right now. So, look, clearly if you’re the first guy to do something, and you discover it and you build around it, hats off. Man Evolution, Florin Court, Yyou guys built incredible businesses around it.

You know, look, I founded a commodity firm in the early 2000s. I mean, it was trading things like power markets changed enormously over the next five years. Things that was shooting fish in a barrel in 2004, 2005, became difficult to do in 2006, 2007. So, markets evolve and markets change.

And the question is, is it worth it today, when you do have Florin Court, and Man Evolution, and Systematica, and AQR, and all these others who pushed into more and more esoteric markets, the question that every allocator should be asking is yeah, maybe that had a thousand basis points of excess returns and you’re paying 500 basis points to get exposure to it. So, that’s great, hugely valuable. But 2025, 2026, the answer might be, it’s not true anymore.

Look, I don’t have the perfect answer to this stuff, but these are questions that I have seen, again going back to 2015, I knew people were asking these questions back then. When we talk about the space that has been broadly disappointing, in a lot of different ways, over the past few years, this is a question I just haven’t heard a lot of people asking. Maybe it’s not worth it. Maybe you should look at 150 markets and then synthesize that down to 15 liquid positions and implement it at a much lower cost.

[01:03:49] Tom

But it comes down to the predictability of trends. And I think we’d all agree that it’s impossible to predict periods where trends will occur. But if we are saying that CTAs and trend following are trying to capture extended periods of market directional moves, maybe up or down. The causes of those moves are varied and many. It’s not simply a structural price drift as the market digests information. It can be an inefficiency in, for example, an alternative market which is more thinly traded, which can build a price trend and allow a price movement and direction to persist.

[01:04:29] Andrew

One of the interesting things when you’re trying to understand a space, talk to really smart people who decided not to do it. That was actually the way I sort of figured out, in 2012, 2013, that a lot of liquid alt products, that were bringing hedge fund strategies to the masses, were not going to do well. Because I went and I found all the hedge funds who said I can’t do it. I’m paid to generate returns and I can’t do it with these kinds of constraints on it.

So, there are a lot of quants out there who will tell you that there is no statistical evidence that alternative markets trend more, but they don’t have skin in the game. And so, one of the challenges you have as an allocator is, how do you get people who have an incentive?

So, look, I don’t know what the answer is on implementation costs, but if I’m running a complicated CTA that is facing lower cost products, and I drill down (as I know these firms can do), and look at the implementation costs and the result is kind of equivocal when I think about the added value. I may still do it because it’s good for my business, even if it’s not a Sharpe ratio maximizing activity.

[01:05:35] Tom

But I believe they’re doing it more because actually they believe the investor is going to receive a better (maybe dangerous to use the word better,) a more diversifying and potentially beneficial return stream.

[01:05:49] Andrew

I hope so. But as Bill Ackman has said again and again, incentives drive everything. And if your investors want you to show up… If you just lost a mandate because somebody walked in with 15 new… they’ve got the Hungarian interest rates or something that you didn’t have, and the person on the other side values that, and you had concluded that you didn’t think it was going to add value net of transaction costs, there’s a decent likelihood next time you go in you’re going to have the Hungarian interest rates.

[01:06:23] Niels

But I wonder whether the same argument could be made in the opposite direction. In a sense by going in and saying, well, I’m not trading all these markets that are difficult to trade, and where you need to spend a lot of time and money on all of that. So, actually. it’s much better for…

I mean, I think the point is, we are always going to position ourselves to fit the product we are bringing. And I think the best way to do that is show me the numbers. I mean, let’s talk about data, we can talk about what we think will happen and why we think it’s better. But let’s look at 10, 15, 20, 30, 40 years of actual returns and then people can decide what actually has stood out as being both better, more robust, more consistent, whatever it is.

And then they’ll probably end up putting together a few managers that are different. And there I agree. People who trade alternative markets, they certainly should be different to those of us who only focus on the liquid markets. They may not perform as well, they may perform better at times, but maybe the key for the investor is actually just that it’s different.

[01:07:35] Andrew

All I’m saying is you should have the data.

[01:07:36] Niels

Yeah, exactly, I agree with that.

[01:07:38] Andrew

You have somebody saying to you, this is why we think these markets are great. By the way, be aware that we’re charging you a 1 1/2% all-in fee or something. But be aware there’s 400 basis points more of costs, but we still think it’s worth it.

Now, going back to Tom’s original point, that’s what the multi strats say, but they do it openly. They’re like, you’re going to pay 1200 basis points a year on it. But guess what? We think we can do 25 and you’ll end up with 13 at the end of the day, that has no correlation to anything under the sun. That is absolutely, truly valuable. But they’re delivering a 2 Sharpe and you’re giving back half the Sharpe to them, basically.

The problem in the managed futures space, I think structurally, is you’re starting with a lower Sharpe and you’ve had a push toward more expensive things, which again, we’ll know this in three years. We can do this again in three years and we’ll have a lot more data. My argument is, I think people may have pushed into complexity, underestimating the implementation slippage, implicit costs, etc., coming along with that because they should be doing better. Five years after this, they should all be knocking the daylights out of something simple like replication. But they don’t. And that implies that whatever they’re doing is not covering the incremental costs, or the fees are too high, or something else.

[01:09:01] Niels

So let me just finish with one final thought, two things actually. Tom, you probably remember this, but I’m pretty sure that there was a large manager, who trades alternative markets, at one of your conferences that (I didn’t participate but this is what I’ve heard anecdotally) actually said, alternative markets don’t trend better.

I don’t know if you were at that presentation…

[01:09:26] Tom

I can’t remember the manager, but I can definitely agree that that could have been said. And I, and I think it’s probably factually correct. I think it goes back to the point of adding incremental markets.

[01:09:37] Niels

Yeah, so, it was just to say that I completely… I think we all agree that it’s very difficult to say that a market trends better. It just depends on the time frame. I mean, if people would have said that about cocoa three years ago, saying that’s the worst market you could trade, and then look what happened. So, I think that’s one thing that I take away from this, today.

The other thing I just want to leave the audience with, it sounds a little bit negative when we say, well, you know, trend following or CTA hasn’t really done much for you the last five years or whatever. No, but I think if we do look at when trend delivers, it is usually when all the other strategies don’t. And that has huge value in a portfolio.

And this is why when people talk about 0.3 Sharpe 0.5 Sharpe, honestly, Sharpe wasn’t invented to look at individual strategies. It’s a portfolio measurement, and we should look at what it does to increase the Sharpe of the portfolio. And I think there multi strat/not multi strat, whatever, I think people will find that trend following competes incredibly well in improving the overall Sharpe of any portfolio of stocks and bonds.

[01:10:48] Andrew

That basically was said, all of that, on the webinar yesterday. So, that’s.

[01:10:54] Niels

Well, there we are.

[01:10:55] Andrew

I mean, one is a bigger proponent of the space, as a portfolio diversifier, than I am. I think it’s more diversification bang for the buck than anything else. My point is how do you then choose…?

And look, we’ve made specific choices. They’re not the only choices you can make. But to me the issue is, if you go through a period like the long winter in the late 2010s, then how much of that was due to the cost structure of the industry as opposed to, if you had a more efficient way of doing it, would you have done better over that period of time? Because we could be facing another long winter, and that’s when these things really matter. No one cares about them in 2022.

[01:11:30] Niels

And maybe these winters are just a function of the fact that there are a few big trends that we can participate in. I have to stop it here because I know Tom has a hard stop, but I will say this was fun and I really look forward to continuing this conversation in a few months.

I know Andrew and I, we have one in between, on our own, but this is great. And I love the fact that we can agree, and we can disagree, and we can do it in a very respectful manner. So, I really appreciate that.

I really appreciate all the work you put into putting together these topics that we wanted to discuss today. And if you, the audience, feel the same, please go and leave a very positive rating and review on the podcast platform of your choice.

Next week I’ll be joined by Mark. That’s going to be super interesting. He always has some interesting ideas and views to share. So, if you have a question for him, send them to info@toptradersunplugged.com and I’ll do my best to bring them up.

From Andrew, Tom and me, thanks ever so much for listening. We look forward to being back with you next week, and in the meantime, take care of yourself and take care of each other.

[01:12:35] Ending

Thanks for listening to the Systematic Investor Podcast series. If you enjoy this series, go on over to iTunes and leave an honest rating and review. And be sure to listen to all the other episodes from Top Traders Unplugged. If you have questions about systematic investing, send us an email with the word question in the subject line to info@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.