Major Market Moves - What Now, Trend Followers? ft. Nick Baltas

[00:00:00] 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.

[00:00:22] Niels

Welcome, and welcome back to this week’s edition of the Systematic Investor Series with Nick Baltas and I, Niels Kaastrup-Larsen, where each week we take the pulse of the global market through the lens of a rules-based investor.

Nick, great to be with you this week. It’s been a little hectic this week and I think, actually, our audience will notice that very quickly because we don’t have a really strict agenda. Both of us have been pretty busy and therefore we’re going to be freestyling, but we’re also going to try and bring a lot of the things we talk about as “theoretical topics” into real-time commentary from a week of interesting market action.

But before we get to that, how are you doing?

[00:01:11] Nick

Hello, Niels. I am doing very well. Look, I was anticipating this week of August, just before I take some weeks of holidays, to be much more relaxed and we know what the market was anticipating to deliver to us.

And I guess to your point, we don’t have an agenda but we have so many things for this agenda that the market has delivered to us. So, I’m doing well. I am enjoying part of the Olympics as much as I can see these days.

But beyond that, markets are keeping us really, really, really, really, busy. It’s been a very hectic week. It’s Friday today, so it’s pretty busy.

[00:01:45] Niels

Absolutely, absolutely. Now, before we get into all the technical stuff, as you know, I always love to hear what’s been on your radar, although I might have a guess about what’s been on your radar. But besides that part, maybe there’s some other non-market stuff that you’ve been focusing on, like the Olympics.

Have you had a chance to watch it?

[00:02:05] Nick

Yes, I’m a big fan of sports and Olympics is something I kind of love watching over the years. I mean, I still remember 20 years ago, when they took place in Greece, and I was kind of spending all my days every single day in a different arena. What I actually find more interesting this year, it’s very personal. My daughter is like five and a half now. She can now understand what maybe Olympics are because she likes sports, but she’s never seen that, obviously. So, I’m spending a good amount of time with her watching different sports.

And I guess on the other side with the newborn, we are also getting busier and busier and busier. So, looking forward to some break, but outside of work and outside of the markets, I think Olympics and family are basically consuming most of my time.

[00:02:47] Niels

Yeah, and I couldn’t agree more.

[00:02:48] Nick

Including taking hours for my sleep, by the way.

[00:02:50] Niels

Yes, there’s a few hours left for that as well. Yeah, absolutely.

And speaking of the Olympics, I spoke with Cem last Friday. He was heading out to the Olympics, and he very kindly sent me a picture from, he was so lucky, he had tickets for the men’s badminton final where, of course, a Danish guy took the gold. So, him sending me a picture from inside where all the action was pretty special. So, I hope they had a great time in Paris.

[00:03:23] Nick

I listened to your conversation about fencing being a volatility trading and marathon being the trend following.

[00:03:31] Niels

What did you think of that?

[00:03:31] Nick

It was quite good. I’m not sure you have any questions for me.

[00:03:37] Niels

No, I think we’ve done all of that. But I must commend ChatGPT for coming up with these analogies, for sure. And if people have no idea what we’re talking about, they should go and listen to the last couple of weeks conversations.

Now, as I said, we have no real agenda. We have a lot of points we want to talk about. So, it may be a little bit confusing. Maybe, actually, to start out with, just to set the scene as we normally do, let me just bring people up to date as to where we are performance-wise. Let’s just do that.

My own trend barometer, probably to some people’s surprise, is fairly high, 52. But people have to remember that it uses shorter term time frames and therefore probably not surprising that it’s looking at this environment as being okay or actually good, given what’s going on in the markets right now.

But the usual indices, as of Wednesday evening, because the numbers for Thursday have not arrived yet, BTOP50 is down 3.6% for the month, still up 1.55% for the year. SocGen CT Index is down 4.45% for the month, up 17 basis points for the year. SocGen Trend down 5.98% for the month and now down 28 basis points for the year. And the SocGen Short-Term Traders Index down 72 basis points and flat for the year.

Meanwhile, in traditional land, MSCI World, as of last night, down 4%, roughly, in August, up 8.18% for the year. The S&P Global Developed Sovereign Bond Index up 48 basis points for the month, up 1% for the year. And the S&P 500 is down 3.66% so far in August, up 12.4%.

Okay, with that out of the way, that gives a little bit of context to what we are going to talk about. It’s almost like, you know, where do we start? But maybe we start with, of course, reminding ourselves that this kind of all started out with a bit of a weak unemployment number, but not surprising because everybody had talked about the US was, you know, in a weak spot, so not surprising. And then a relatively small hike in interest rates in Japan, which again had been telegraphed, and so I don’t think you could say that it was surprising either. And then, of course, all “broke out”.

Nikkei down big time, biggest daily loss since 1987. Then the BOJ comes out and probably says publicly what we all know, and that is, well, the central bank is not going to change rates if it upsets the market. So, this is, if anything else, kind of a BOJ put, and I’m sure there would be another central bank put somewhere out there.

So again, this is not new. Maybe it’s becoming vocalized by central banks to a point where we haven’t seen before. But of course, the market took that as very good news, and then we see a big rebound, and so on and so forth.

It doesn’t change the fact that markets on a month-to-date basis, yeah, they’re down, some of them. Certainly, the yen is up a few percent, but nothing dramatic.

But of course, performance, when you hear the numbers I just quoted, down 6% for the SG Trend Index. That’s definitely a meaningful month for an index like that, but it’s not catastrophic. We’ll talk later about some of the headlines that you kindly sent over to me, some articles and so on and so forth.

But how should we attack this? There are kind of a few different big topics. We have volatility as a topic, right? We have liquidity maybe as a topic. We have risk management as a topic. We have how different models react to the price action we’ve seen based on how they’re built and so on and so forth.

So where would you like to start? How do you think we structure this conversation so it makes most sense for our audience?

[00:08:17] Nick

I think, how do we start? That’s a good question, right? So, you know, we gathered a lot of data, and we can discuss upon that data.

I think, let me just share my personal perspective, purely living through that week. Like, it’s Friday today. Last Friday was maybe the first bigger move in the equity markets that we saw, and obviously in Japan.

I think living through a week that almost today, you kind of feel it was a bit more of a non-event, even if it was an event, to your point. There was a lot of emotion going into that week. There was a lot of, if you like, concern as to where we’re going to end up. And then you realize that after a week, maybe to your point, S&P was down 6%, and now it’s up by another 2.5% or so. Let’s see what today is going to bring. So, we’re kind of 3% down in a week, which is relatively in line with historical observations.

Or I guess you mentioned the fact that Nikkei had a similar kind of move, still another maybe 10% to go back to flat, down 20%, but then up 12%, if I split that week into two parts. So, I think after the fact, it is crucial to understand whether emotional reaction is necessary for some of the systematic strategies we’re running. And I think number two, we can have a bit more in-depth conversation as to how different systems behaved.

Maybe we should, to your point, try to associate better investor objectives with design choices. Because living through that, on Monday, I was kind of having conversations with clients that they would be more like, we should have a faster system. And then two days later, they’re like, actually, it’s okay with a slower or slower one, because a V-shape kind of dynamic in the market can whipsaw your return patterns.

And I think in that context, we should not forget the fact that entering that regime (and I have some stats from our strategies that came out today, some interesting charts that I can discuss), 2024, as an equity market, up until the end of July, is almost at the top decile of any calendar year since 1928. So, it has been a huge equity rally.

And now obviously, I guess the drawdown we have seen is a tad faster than what we have seen maybe historically. And that’s not just the last couple of days, it’s more like coming from July onwards. So, it’s almost like a mirroring effect.

So, I think after the fact, there’s a lot to be said about a program that might be losing money, to your point. We see the CTA trend index going down 6%. But we should not disregard the fact that, up until maybe a couple of days ago or like a week ago, this was one of the outperformers.

So, the question is how much of that we can give up for us not to suffer on the downside. So, we always know the story exposed. And I think living through it and reassessing objectives is very, very important and managing expectations.

So anyway.

[00:11:23] Niels

So, you know, great points. Before we dive into some of the more sort of the technical side of that, you mentioned [you were] busy talking to clients, right? And so maybe to broaden it out and to get even more context, I’d love to hear, besides the fact that on Monday people wanted faster systems, two days later, they were okay with not so fast systems, besides that, what has come out of the conversations you’re having in terms of, are there any red threats where you could say, well, this is what they’re really concerned about?

And of course, maybe there’s something that’s new that usually, because of course, one of the headlines that I’ve seen (and this is not really from the big press, but from one of our friends in the industry), and it’s a good headline, because it says it all. And it’s something like, ‘Crisis Alpha in Short Demand’, or ‘In Short Supply’, I should say.

But I’d love to hear if, from a shop where you sit, what were the kind of the topics? What were the concerns? And then we can build that into answering some of the more technical side.

[00:12:42] Nick

Yes. So, let me then abstract from just the trend following space and make it a bit more generic. I think any market rotation is a lesson and an opportunity.

The minute you realize losses, it’s the precise minute that another one would look into an opportunity to enter a trade. So, maybe let’s put some context into it. We saw VIX from 16% end of July, spiking to 65% intradayly on Monday, which is a level we haven’t seen since COVID. And it’s now down to 24%, which is maybe a level that we saw earlier in the year, possibly last November.

[00:13:20] Niels

Actually, we saw it on Friday.

[00:13:22] Nick

Exactly.

[00:13:22] Niels

It’s exactly the same level as a week ago.

[00:13:25] Nick

Exactly. You know, the S&P crisis, a year and a half ago, was like around 30. Let me put some more context.

Following the BOJ rate hike, obviously we have been discussing as to whether the Fed is going to start cutting rates. What that led to is this relative attractiveness of dollar and JPY that made the famous FX carry start unwinding. So, some carry losses were incurred by those rotations.

Other things we saw was in a dispersion space. So maybe like a week or two or three ago, we’re talking about the least correlated environment in the US market, in equities, which was the consequence of a high level of concentration purely because most of the return was coming from the biggest cap, AI, hype, tech, and so on and so forth. Obviously, those correlations spiked and then dispersion trades lost money.

Some volatility selling trades did lose money. The skew in S&P got repriced. And from the low levels that we have seen, historic low levels, it’s now at the most attractive levels it has seen in recent years.

So, what we saw was partly some elements of systematic investing getting hit. But at the same time, with some of those implied levels (whether that is correlation or maybe skew or maybe volatility), there was opportunity to get back in. So, when you look into investors, there was a mirroring effect between those that were becoming more risk averse, maybe cutting some of the losses, and those that would find the opportunity to jump into it.

So, I would say the net effect is almost like a circle that closed as we stand today, the week after what happened last week. Specifically, around trend, which is closer to us and maybe some people in the audience, I guess my observation (which is I think an interesting observation) is that everyone that had looked into it, at least from the people I speak to, were pretty much with the underperformance. This is not the environment that you would expect a trend follower to operate – a massive reversion following a spike.

This was not the environment during March 2023. This was not the environment in the VIX false spike in 2018. This is not the environment. We know what we’re subscribing to. If we allow our exposures to be unconstrained on the equity side, yes, a reversion is going to hit us. Yes, maybe the reversion was a bit sharper than expectation, but we haven’t seen anyone taking a negative stance as to what trend following is here to bring.

Yes, of course, we’ve had discussions around, okay, where is the risk of the portfolio moving now, or how much more exposed you are in Japanese equities versus the JPY versus the treasury markets? But by no measure did we see anyone becoming concerned about trend following as a systematic theme. So, I think that’s the overview.

There is no alarm that I would flag as being, I guess, rung. Surely there was more volatility in terms of… not volatility, variability of trading activity, let me put it this way. But I would say net-net, it’s been, again, a week of learning lessons and capturing opportunities.

That’s how I would summarize it.

[00:17:00] Niels

Yeah. And I would maybe go further and say that the week, once we know what the tally is, so to speak, that performance, that loss is probably not going to be any different from some of the weekly gains we’ve seen in Q1, for example. And I think people need to put it also in the context of the fact that it was a big carry event that took place.

I mean, it wasn’t just a small thing. I mean, it was a big carry trade event that got shaken. It’s probably not reversed or anything like that, but certainly got shaken.

And so, I would actually credit the way the CTA and trend following community has handled it, even if we don’t know all the numbers yet. But just from what I can see, it’s really not going to be a big event.

And actually, it’s interesting. I had a sweet comment, an email from a guy that knows as well. And he said on Monday, it could be interesting if you could talk with Nick about what’s going on during a day like this in the office and the trading and all of that. And I said to him, listen, it’s going to be such a short conversation because Monday was a day like any other day. There was nothing unusual from a trend following perspective. We weren’t any more busy or anything like that. We just follow the rules and follow the systems.

And I think that’s the reassuring part of what we do. Other strategies, I’m sure, would have had frantic activity. I think trend followers, it was really just another day in the office.

And of course, we’ve had days that are much worse. Think Thanksgiving 2021. That was a day that will go down in history, I think, as a one-day event. But yet, the strategies came roaring back, actually, almost beginning on that day.

So, okay, we talked a little bit about sort of volatility. Yes, it spiked, but it’s come back at a little bit higher level than where it was. But of course, it has been remarkably low, given the uncertainty.

Do you want to talk a little bit about some of the things that would have impacted the outcome of how you dealt with these types of things? And maybe from a big picture point of view, maybe we can zoom out a little bit and not just talk about a few trading days, because a lot of trend following systems won’t really react to a few trading days. There’s more to it in terms of a lookback period.

So, do you have a sense of how exposure has shifted kind of a little bit big picture, say, in the last six to eight weeks? I mean, obviously, at the end, the Nikkei has been in focus.

But do you have a sense of how the trend following strategies that you oversee have started to change maybe over the summer?

[00:20:04] Nick

Yeah, sure, sure, sure, sure. So, obviously, the transition is a function of the speeds and the amount of smoothing. But I guess, in broad terms, what we’ve seen, let’s say, over the last couple of weeks, or maybe a month or so, maybe from July onwards, the net exposure in rates was relatively balanced at the time.

I think now the vast majority of markets across regions and tenors are long; so we’re long futures, effectively betting on the reduction in the yield in some respect. And this is a regime we haven’t seen maybe for quite some time now, except for some faster signals switching. Maybe you remember back in December last year, November/December, when we had this kind of quick reversion that ended up kind of falling back into the short territory for bonds. But now they are all long, I would say.

So that was a significant transition from June through July now to beginning of August. On the FX side, not much, I would say, maybe beyond the fact that obviously, JPY from being the weakening and most, I guess, accredited return component to trend followers, now, is long.

Maybe the Swiss franc is now on the long side, but that’s more like a consequence of other moves. And obviously, some of the DM currencies; you’d expect them to transition over the summer on the short side, like Mexican peso is a good example. But I think net-net on the FX side, there was nothing more aggressive than those moves, I would say.

I think the bond market has been a bit more aggressive. Commodities, the commodities are an interesting one, because the beginning of the year was more of a rally. And then July was a massive reversion. I think BCOM was down by 4.5%, GCI was down by 4%. The variety, the vast majority of markets turned, specifically in the energy complex.

But some of the long positions were in gold or cocoa or coffee, and, by the way, cocoa has been relatively muted return-wise over the last month or two or three. It’s not in the news as much as it was back then. But let’s say we’re still long some of the precious metals and coffee, cocoa.

And then kind of concluding on the equity side, this is the more eventful one. This is the more eventful purely because we were long all markets pretty much throughout the year. Now, depending on the speed, a fraction, or all, or most of markets, globally, would have turned.

I think summarizing it all, equities is a big transition. I mean, let’s summarize it in another way. What has been driving returns year to date?

It’s the equity rally, the weakening of JPY, and maybe some rising commodity prices, more into kind of regimes of like one regime and another regime. But these were the three kinds of components we have seen and a very uninteresting bond market (which I think for managers that have dynamic risk exposures, maybe bonds started representing lower and lower and lower risk over the year, just because there was no trendiness whatsoever). I think now most of those moves have been reversed.

Equities are more kind of mute or maybe turning direction depending on the speed. JPY is now a long position, I would imagine, for most of the other programs throughout the month of August. The bond market started becoming more directional. I’m not going to call it interesting before we see the results, but certainly more directional on the long side. And then on the commodity side, we have seen some of the markets on the energy complex kind of flipping. So, I think partly the “perfect storm” is the one that commands those directional movements that we’ve seen.

That’s the overall summary, I would say on the side.

[00:24:13] Niels

I agree, to a large extent. I think the way I would summarize it is that the big change happened, as you rightly said, with fixed income as a sector turning from short to long, probably a couple of months ago, starting to see that and obviously increasing in intensity as we came to the end of July. I completely agree.

Of course, long equities would have been reduced. I don’t think it’s changed direction, frankly, because the move is relatively short in time.

[00:24:48] Nick

Only for short windows.

[00:24:50] Niels

Exactly. And actually, I would even go as far as saying that’s probably also what I feel about even the yen, that the window for most trend followers is too short. But there will be some systems clearly that will view the yen differently than before.

So, to me, it looks like, yeah, there’s some deleveraging for sure. There’s some signal strength that’s getting weakened. But there’s also a lot of continuation. And maybe we shouldn’t completely ignore the fact that as July happened, there were a few other sectors that helped mitigate some of this because the downtrends in some of the commodities, like grains, actually, got a little bit stronger. And that would have been beneficial.

And so, I guess it goes to show that there is a lot more that meets the eye in a fully diversified trend following portfolio, which is clearly not being captured by the headlines by some of these financial media outlets. That’s for sure.

I think some of the return dispersion we’re going to be seeing, between managers, certainly also relates to the number of markets you trade. I mean, those who have very few markets, it’s going to be a little bit hit and miss whether they were in the Nikkei or not, or how much yen they had, or whatever.

So, it will be interesting to see. And it is a very good real-time period for people who are either invested in trend following to hopefully get some comfort from the risk management that’s taking place and the controls that we have. But also for people looking at the strategy for the first time and just noticing that it’s pretty well behaved.

[00:26:41] Nick

I’ll give you two comments or at least two data points that, at least to me, were quite interesting to kind of live through. The first one, you just mentioned it: a more alternative commodity complex has actually had positive and good performance over August. Almost every day, those non-traditional, non-benchmark commodities have actually been performing. So, they’re adding performance to more conventional markets. That’s number one. So, the power of diversification in a universe that is relatively insulated by those macro moves is now here to, if you like, justify their importance, number one.

Number two, throughout the year, and if you like, getting into the end of July, this an interesting bond market that we haven’t seen actually for years. Historically, that was the big driver of performance. And over the last year or two, it has not been, I guess, after 2022.

The point I wanted to make is because I personally like dynamic vol scaling and dynamic risk allocation between markets and asset class. I think that’s something I’ve discussed here quite a few times. It was maybe (and that is now a hypothesis because I haven’t looked into the long-term data), certainly over the year, certainly over the recent years, the risk allocation in rates was basically at the minimum we have seen for some time.

To put some numbers in context, looking into, let’s say, four asset classes (the major ones, equities, rates, FX, and commodities), from being equally risk exposed to those four, or starting dynamically allocating around this 25%, looking into rates around 10% or so of risk of the portfolio, it’s a mere consequence of them not trending. And at the same time, you see a lot of trendiness in commodities as well as in equities.

And I think that little exposure in rates can also have an interesting data point that we can discuss later on as to whether we would want to have a balanced allocation with asset classes or not. Because that is also a design decision, which I think is very important when you start doing performance analysis. But I think to me, seeing in a trend-following program that 10% of risk is only being in rates was something that I hadn’t actually seen.

So, it’s worth flagging.

[00:29:04] Niels

Yeah. Okay. So, I’m just going to go in different directions in terms of what comes to mind.

You mentioned dynamic risk allocation. So that obviously takes me straight back to the whole discussion about dynamic position sizing, static position sizing. During the last week, 10 days, clearly there’s been a good period to study the difference between these two.

I don’t know if you have some observations. Obviously, we don’t know performance breakdown. It is harder and harder to track, on a daily basis, the non-dynamic position sizing because there may be not so many people left doing it.

But I wonder how, now of course the managers themselves, they will have chosen one or the other and have a firm belief in that. And I think that’s fine, of course it’s fine. But investors might actually have a different perception of the risk when, if you compare a manager who actually is actively reducing position sizes because the markets are going crazy, or someone who just stays with it until the stop is hit.

Do you have any thoughts on this since it’s rare that we get a period where it’s a good time to compare? Because on a day-by-day basis it probably makes very little difference.

[00:30:37] Nick

So, there are two points related to dynamic sizing. Point number one is volatility scaling each and every exposure, which by design will increase or decrease the exposure in a particular asset. And I think for that, except for a few cases that I guess both of us are aware of, more likely than not I would expect managers and trend following strategies to have.

I think the bigger difference, at least that I personally see, which is the second point (and that’s the one I have more comments on, at least for the recent period), is whether, even after doing inverse volatility scaling or some sort of rescaling in your positions, you still decide to allocate half of your risk in equities, half in bonds, half in, sorry, a quarter in equities, a quarter in bonds, a quarter in FX, and so on and so forth. Because that does not preclude the inverse of vol scaling or not at the individual asset level. We can still have that, but still decide, at the top level, that we combine four asset classes together with 25% of risk.

So, when I was talking about dynamic sizing, it was more the latter in the sense that I manage my positions one by one, but in aggregation I want to force diversification between the four asset classes without expressing the fact that the collection of equities, year to date, has been trending twice as much as anything in rates or FX or commodities.

So, the observation I have here is that by design, if you end up having equities contributing 60% of the loss, by design a more balanced allocation between the four asset classes would have fared better. But here’s the question. Let’s take ourselves, and I think this is why those rotations are great lessons for investors.

If we take ourselves in the shoes of an investor like two months ago, seeing a portfolio that allocates 25% of risk in rates and rates doing nothing and maybe not even trending at all, can easily raise eyebrows in the sense like, no, why are you exposed to a market that is not trending while equities are rallying so much and you could have performed much better had you been exposed to those. So, I think, to me, it is another design decision that has to be very well covered and disclaimed.

Yes, you want diversification. Guess what? When a particular market is only driving returns, you are going to be less exposed to those returns precisely because you’re balanced.

But you’re going to mitigate losses if you have a single market driving underperformance. Maybe a similar thing happened back in March 2023, I haven’t actually looked at it, whereby bonds, in the exit from 2022, were a bigger risk in the portfolio had you had dynamic asset allocation scaling.

So, what is my observation against the conclusion of that point? There is no right or wrong. I think, to an extent, it’s a trade-off. Whether we force diversification and we acknowledge the opportunity cost of a single asset class performing and not generating 100% of that trend while obviously getting a more defensive profile on the downside versus having more conviction around signals and trying to find defensive components elsewhere; maybe with the speed, maybe with the stop-loss, but with something else.

I think that’s one of the data points that, most likely than not, has caused dispersion between managers over the last week. There are more, but I think that’s definitely the one.

[00:34:09] Niels

Okay, so let’s stick with another design choice, which is something that definitely I think investors will be thinking about and maybe not fully comprehending the impact of and that is speed. Because clearly you would think, on Monday this week, short-term systems would have done better, you know, not massively, but would have done better. They should have made money on Monday, probably, and trend followers lost money on Monday, we know that. And I don’t know about Friday, depending on how quickly your systems react.

But, of course, when you look at the index levels, the Trend index and the Short-Term Traders index, the Trend index, not just this year, but pretty much most years, way outperformed the Short-Term Traders index. But this month, as I mentioned in the beginning, the Short-Term Traders index is flat, trend followers are down six.

But they’re just bringing them back into line because all year the trend index has been outperforming (the Short-Term Traders index). So how do you think about the choices of speed? I mean, of course, I have probably mentioned countless times that I don’t believe that short-term trend following works as well as medium to long-term trend following. I’ve never seen any data on this and even those managers that we know of, who’ve been featured on the podcast, who are short-term managers (and a couple of them have done well over the long term), they’re not short-term trend followers. This is very important for me to stress. There’s a difference between being a short-term manager and being a short-term trend follower.

A short-term manager will not get out because the trend changes. They’ll get out by using a time-based stop or some kind of vol-based situation, whatever. So, there are different strategies.

I wouldn’t just say it’s just the time frame that’s different. It’s also the methodology that’s different. But I know we talk about them as short-term versus medium to long-term managers.

So, what are your thoughts on speed after a week like this?

[00:36:35] Nick

A very good question and I think that’s probably one of the most discussed questions this week. To put things in context, the SocGen CTA Trend index, as in the classic benchmark for medium-term trend followers, follows (and that’s more like stylistic or I guess more statistically assessed) a signal which is around a 30-month half-life on an exponentially weighted moving average basis.

We have done that ourselves to assess it, but interestingly Quantica, I think we’ve mentioned it here, last year I think in September, in their quarterly research piece, they did showcase that for the last 10-15 years or so a signal around this 66-day half-life pretty much allows you to get a correlation with the index of around maybe 90 percent.

So just to put things in context, let’s call a medium-term trend follower somebody that has around a three-month half-life on exponentially weighted moving average. That is not a three-month point-to-point window just because people would associate one with the other. It is not.

We use just long-term data, but we simply decay the amount of information through time so that after three months we have lost half of the importance of the data point as of yesterday. So that is just to put things in context.

Now, how did I experience the week? To your point, before the fact, you do not know whether Monday was a false positive or a true positive. In other words, whether Monday was the start of a massive drawdown or Monday was simply an exaggeration that by the end of the week became a V-shaped recovery. So, case number one, if that was a true positive that the market did turn and it’s now getting into its deepest drawdown by the end of the week, which of course is not, but supposing that this was the case, surely a faster manager would have outperformed substantially.

Having said that, the V or the half V-shaped recovery that we have seen so far made them kind of whipsaw around quite a bit in the following sense. Maybe Monday was better, but from Monday onwards not as well because they turned the signal, maybe reduced the cross exposure so they did not recover. Whereas, I guess, medium to long-term signals were the ones that lost more on Monday, have started recovering, but not as much as to cover, for now, the relative loss that came on Monday versus the short-term signals.

And that’s, I guess, to your point, short-term traders versus medium-term traders having a gap, but I think this gap realized as a widening gap in favor of the short-term traders by Monday into a closing gap up until the end of the week, coming purely from medium-term trend followers capturing some of the recovery that short-term do not. Maybe too much, I guess, too much technicality in my response.

[00:39:39] Niels

No, no, but I would add one thing to it.

[00:39:42] Nick

I have a few more things, but go on.

[00:39:44] Niels

I just want to add one thing to this, because I think it is important for people to think about. This is obviously where managers experience and research makes things a little bit more interesting, and that is you can’t assume that lookbacks are symmetrically applied, meaning just because you have a certain entry period, so to speak, meaning, yeah, let’s say it takes people on average three months to get into a full position (let’s just, for argument’s sake, use that number), I mean, it doesn’t mean that it takes them three months to get out, so to speak.

I mean, managers can do things in their design that makes even a medium to long-term manager sometimes react more quickly, is my point. And I think this is where edge and uniqueness, even though people will say, well, in the very long term, a lot of these managers look the same. Yes and no. I don’t think they really do, but they can from time to time.

So, there are a lot of moving parts that go into this that may not kind of meet the eye when you just look at the numbers, how they were made up. Of course, each portfolio will be different from the number of markets they trade and so on and so forth.

So, it’s really hard. I actually think it’s a really hard job for an investor to really truly understand the difference between managers.

[00:41:21] Nick

I mean, over the longer term, I would agree with you. To make this simplistic example, market falling by five and going up by five, if you invert that, you get to the same point at the end. So, over the longer term scale, some of that is fast enough to capture a rotation, but then is differently positioned so that at the end they get to the same point, might make the end result, in terms of like no realized sharp ratio, look very similar.

But I think the dispersion, specifically around market rotations, is very, very important for investors to take a look at. Because the reality of remaining solvent in a drawdown is harder than in an up move. And I think this conditionality is very important because it’s the mere condition for you to remain solvent and see the end result.

And I think sometimes we kind of tend to forget the fact that we have to live through the drawdown before we see the new all-time high. Because yes, the new all-time high can be there, but if you’re not solvent to realize it, then it’s just paper money.

I agree with you. I think the nuances of assessing a strategy should focus on those turning points and market rotations and how they can preserve capital, specifically when other things are suffering. So, I would agree with that point.

[00:42:41] Niels

Yeah, and then on top of that, you could say, well, seemingly similar managers, when you see some of the numbers and you think, well, how on earth did they not lose money during that month? And this is, of course, because there’s also been another trend (pardon the pun) in our industry. And that is that with many managers, and there’s nothing wrong with that, it is not pure trend anymore. So, it looks like trend, but it’s not pure trend. And therefore, you will get other factors having an effect on the performance.

[00:43:14] Nick

Which, by the way, it’s not good or bad, but it’s just another different feature that people can add to their profiles. I think, for myself (we’ve discussed it quite a few times) I’m looking into reversion dynamics, I’m looking to carry dynamics.

So, the reversion, for instance, did help month-to-date because it was very quickly shorting the Japanese market. Sorry, what am I saying? No, the other way around. It very quickly went long the Japanese market, purely because it realized such a big loss. But that’s a reversion component, right? And it has absolute trend. But combining the two, yes, of course, can bring value. But to your point, that is also another consideration for comparison purposes.

I think the last point that I would say on the speed, interestingly, looking into equities or creditor effects, the difference between signal speed months to date wouldn’t have been substantial in the sense that, yes, maybe being faster would have reduced some of the losses and then missing some of the rebound as we’ve discussed. But they’re all kind of moving in the same direction.

I think the two asset classes where you saw some differing behavior, specifically around a very shorter-term speed and longer term speed, were clearly the markets that over the last two or three or four months have not had established trends. And these are commodities and rates.

Commodities went through this kind of cycle of picking up by early Q2 and then kind of going into a normalization and then picking up just before June and then kind of crashing, well, “underperforming substantially” over the last month or so. And obviously rates have been kind of whipsawing around.

So, this is the bigger dispersion by being shorter term or medium-term trend follower over the month of August with shorter term speeds performing significantly better both in absolute as well as in relative terms in rates and commodities.

That’s the other thing I would flag.

[00:45:07] Niels

With your experience, Nick, are there any similarities to what we’ve just gone through, because not all of our listeners will have gone through something like this before, but some will. Obviously, it was only a year and three or four months ago we had Silicon Valley. It was six years ago and change since we had Volmageddon.

By the way, what’s an interesting stat about Volmageddon is that the streak that the S&P had, going into Volmageddon, in terms of not losing more than 2%, and that streak going into this one is almost the same duration. It’s something like 356 days versus 351 days. So, I think most people probably should make a little note in their calendar if the S&P goes for about 340 days without a 2% drawdown they should start getting out of their long equities, anyways. Similarities or are these shocks always different?

[00:46:15] Nick

So, I think a very interesting comparison is last year. So, March 2023, we’ve had however many sigma moves for the two-year point in the Treasury market. It was like what, six, seven, eight? I have forgotten the number. But anyway, it was kind of unprecedented for like three, four, five decades and so on. And this was pretty much one and maybe the main reasons of the substantial reversion we have seen in trend following profiles.

Even more so, whether a program was fast or slow, or equity defensive or not equity defensive, the dispersion was not too wide versus, at least in my perception, in my experience so far (and you can tell me if you agree or not), the dispersion will sink in the beginning of August this year. Because what is happening now is that equities, which is typically the growth risk that becomes the differentiation factor between managers and investors, is the reason for the drawdown. We have investors that would want to see a very fast speed and equity defensive, a reduced equity participation program. And some that look into a more medium-term speed, unconstrained, I want to have some equities and so on.

These two examples would perform significantly differently month to date, but also January to end of July. Those two would be relatively similar back in March last year. Because if the target market is equities, however quick or slow you are to it, however constrained or not you’re to it, you’re still exposed to rates.

I think what I wanted to highlight is that there is similarity in the sense that yes, there is a primary driver, a principal component, a duration component that ends up propagating throughout the portfolio. But that in itself, at least in my own experience, was never truly targeted as a market that we would want to be heavily defensive against – this rate shock we saw last year. I think that was more contagious than the equity down move and maybe volatility move that we saw last week.

That’s why I think we have some winners, some losers, and I think there is wider dispersion. We have some investors that have portfolios with no long exposures in equities. By design, they just don’t want to have it because they want it for defensive purposes. Guess what? The portfolio did not move. Genuinely, it’s maybe close to 0% month to date with all the others averaging out to zero.

So, I think this is a key example whereby purely focusing on equity risk, growth risk, cyclicality risk, and defending that type of risk would have implications with regards to slow speed or fast speed, capping equity exposure or not capping equity exposure, balanced allocation between the asset classes or not. And obviously, the difference between those two scenarios can now obviously raise the question of what are we trying to defend against?

If it’s a 60/40 portfolio, maybe we should be more thoughtful about the diversification we force or how we cap exposures in the two markets we care mostly about, rates and equities. So, I think that’s another point I wanted to flag today that there is no one size fitting all and different investor objectives would require different treatment with pros and cons.

There’s a disclaimer that you’re leaving something on the table. But again, to my earlier point, solvency in a down market has, in itself, a significantly important consideration, right? That’s my point.

[00:50:17] Niels

Yeah. Talking about dispersion, maybe not for you and I to talk about because I actually want to talk to Andrew about this. So, this is a little heads up, Andrew, for when you’re coming next. I’m not entirely sure which week that will be. But there was a big dispersion between replicators and the benchmark leading into July. Replicators doing much better than the benchmark, frankly. So good on them. But a lot of that dispersion has disappeared in the last six weeks. So that’s another interesting angle to all of the things we talk about here week after week.

So, I’d love to talk to Andrew about that, what observations they’ve made. But then the other thing, maybe before we completely leave this thing about, well, was this normal? Was it not normal? Was it just the carry trade being unwound? Or what about the whole idea that this time around we had zero-day options?

Now, I know nothing about this market. You know something. So, I mean, is that a factor as well? Obviously, we know it’s been a while since the VIX really spiked. It did spike on Monday, for sure, up 100% on the day or more, I guess. But were they part of the problem?

[00:51:45] Nick

Not that I’m aware of, but I would disclaim the fact that I haven’t spent too much time looking into that particular point of the market over the week. Surely, selling volatility in those markets did come at a negative return over the last week. But this is no different to selling longer tenors.

By design, you’re exposed to a realized volatility shock.

[00:52:10] Niels

Except maybe just for me to fully understand, the reason why I say that it may be different is that, and this is where you can correct me for sure, but it seems to me a little bit more difficult to hedge an option that expires in a few hours. Do you know what I mean? Compared to hedging something that runs another 30 days.

Or maybe that’s not a big issue. But on a day like that, where things go crazy, hedging a zero-day option, to me, appears to be more challenging.

[00:52:47] Nick

I mean, certainly to the extent that the market is moving quickly intra-daily, you have to catch your delta and your gamma is increasing. And obviously, you have to be active on the market. So certainly, to some extent, this can have implications.

But the question is, what is the direction of the exposure? Whether your short gamma or whether your long gamma can actually exacerbate or actually contain the movement.

I’ll give you this example whereby in Nikkei, for instance, there was substantial intraday momentum that took place on Monday. And had you been exposed to those intraday moves, and following those intraday moves as they opened in the morning, you would have more than just benefited from that move.

So, I think it was a combination of events. And other people, and I think that was another topic that we discussed prior to the call, is to say, well, is it risk parity funds that aggravated the losses? The typical trio of shame that people talk about, you know, the CTAs, and the trend followers, and the risk parity funds, and the vol targeting funds, and all that.

So, I don’t think there was one driver. I don’t have any particular data to discuss specifically whether zero DTs in one way or the other were impactful. But I should say that investors that had long volatility, long gamma exposures did realize some good return purely exactly when they would expect it to realize. You know, the market goes south, and you’d expect your hedges to perform.

Fine. That was the case. And selling volatility on the other side, specifically correlations and dispersions were the ones that were hit the most. But yeah, looking into the reasons behind that, I think it’s a multidimensional problem. I think liquidity is also something we should consider.

[00:54:42] Niels

Yeah. Yeah, it would be interesting to do kind of a post-mortem at some point when we have the data. Yeah, to see how different agents in the market, let’s call them that, reacted and played a role in all of this.

I’d like to maybe end our conversation talking a little bit about something that has been at the heart of all of this, this week. And that’s risk management. I’ve mentioned in the past that the way I think about trend following is that first and foremost, we’re risk managers. That’s the part of the equation we have some input on. We have no say in terms of what performance we can get from the markets. But we have some control over the risk we take.

A lot of managers today, as we talked earlier about it a little bit, have evolved from maybe a classical risk management approach, using stop losses and static position sizing, to a more value-risk-based model. I’d love for you maybe to put a few words on this as a methodology in terms of robustness, pros and cons. And obviously also in light of, well, what if everybody uses value at risk? Aren’t we getting ourselves a little bit into trouble because it may trigger, to a large extent, some reaction pattern among a lot of people more or less at the same time? So, I’d love just for you to think a little bit outside the box about these things.

[00:56:36] Nick

And just to clarify, is now value at risk a way for us to assess and monitor the risk of a portfolio? Is it about portfolio and position sizing? Or is this about a stop loss at the asset level and the implications that this can cause by the forced selling?

[00:56:57] Niels

Yeah, I think of this as a portfolio tool. This is not…

[00:57:01] Nick

As a portfolio tool, right?

[00:67:02] Niels

Yeah, as a portfolio tool, yeah.

[00:57:04] Nick

So, look, I mean, if we look at it purely theoretically, value at risk is just a point in a return distribution. And this is only a threshold of a loss that would be substantially harder should the threshold be crossed. So that’s part of the reason why people would typically look into conditional value at risk or what we call expected shortfall, which is the expected loss beyond a particular threshold.

Now, and I think that’s, I mean, classically in risk management, that I think is in lesson 101, that basically says value at risk is a threshold, but your actual loss is more likely to be substantially larger. So, you have to look into the expected loss conditional upon the fact that you’re crossing the threshold. So that’s number one.

Number two, obviously looking into historical realizations, you’re just bound by what you’ve seen. So, you can calculate that by saying, I’m expecting that the distribution of returns is following some distribution that I’m going to fit using my historical data, but then I’m going to, as we tend to call it, have a parametric definition of value at risk, which comes purely from theory of a particular distribution that somehow is matched to the real data.

There’s a way to say, look, instead of doing that, I’m going to run a Monte Carlo simulation, because I’m assuming a model that my return series is following. That’s not too different to what I mentioned earlier on, the parametric one.

At the end of the day, if people follow similar rules, this cannot, in itself, have a, I guess, a contagion effect, except the possibility that losses happen at the same time, and therefore they have to act on a synchronous manner to start unwinding exposures.

But that is, I think, to a certain extent, inconsequential to the measure used. I think the question is more about whether this is the right measure to use and whether people over or under react to it. And I think the way I would look into risk management would be more along the lines of correlation dynamics. I think that’s key. Externalities that are caused by investors holding very similar portfolios for similar objectives that they’re not aware of that.

And to give an example, why is FX carry now so important? Maybe because too many investors have similar FX carry portfolio positions without acknowledging the fact that they’re all going to react to a global growth shock. Because if that is not the case, then why do we care about FX carry? So, I think this is a good example whereby a trade that historically might have been more idiosyncratic starts becoming dominant.

And it’s hard to see with historical data how realized correlations were low and eventually play out higher, purely because you and I, not knowing each other, follow similar rules in a similar portfolio, and both of us react on a similar market event. And I think those externalities that we inflict to each other can have contagion risk for the overall market. I think the market doesn’t just drop because liquidity is low.

The market can drop because liquidity is low, and we all act on the same signals, even if we thought that we’re well diversified to the market based on the backtest. So, to me, that’s more how an efficient risk management system should operate. Try to foresee correlation spikes between assets and asset classes.

I think that is not what value at risk can achieve.

[01:00:41] Niels

As they say, the only thing that goes up in a crisis, Nick, that’s correlations. And there we have, you know, a good example of that. And of course, this is also this discussion about endogenous effects and how, maybe, even though most people knew that there was a big carry trade going on, but maybe it was bigger than people realized.

[01:01:05] Nick

Funding liquidity is one of those catalysts. To the extent you cannot lever up anymore, this can be a signal that the market becomes quite constrained. And you don’t need more than one or two big moves to have a bit of a vicious circle.

But when you have those vicious circles, just to give like the last sentence for today, maybe, when you have those vicious circles, I think it is getting a bit too aggregated, this whole narrative about every time that we see a drawdown, risk parity, vol targeting, and CTE funds, I think it’s a story that has been played so many times. I cannot see how differently a discretionary manager would act if they hold the magnificent seven. I’m sure they’re cutting their losses.

So, I think it’s an easier target, purely because there are some systematic rules that we can easily look at and maybe monitor. But I just think it’s overplayed now.

[01:02:03] Niels

Or maybe private equity was just funding all their acquisitions in the end. Who knows? A wild guess here.

Now, let’s finish off our conversation with a few of the headlines from this week that you kindly sent to me. I don’t know whether it was from Bloomberg or where. It doesn’t really matter.

But of course, the usual headlines, one being, Fast Money Quants Are Hammered After Everything Went Wrong. Another headline, Momentum Traders Appear to Be on The Wrong Side On All Their Crucial Bets. And there’s another one here, Volatility Rocks Quants Threatening New 170 Billion Selling Spree.

So, there’s definitely no shortage of colorful headlines when it comes to describing a week, which I actually think, from a trend following perspective, will be very similar to many other weeks that we have been through.

If you enjoy these conversations and all the hard work that Nick and all the co-hosts do to prepare every week, please leave us a rating and review on your favorite podcast platform. It really does help. And it also gives us a moral boost to see these ratings and reviews coming in.

Next week, I’m pretty sure I’m joined… Well, I was going to say I’m joined by Mark, but that can’t be true. So actually, I can’t remember who I’m being joined by, but that doesn’t mean it’s not going to be a great conversation. It will be. So, if you have any questions, by all means, send them to me and I’ll make sure they get tackled.

Nick, have a wonderful vacation. I really truly hope that you get a chance to relax after a week like this. Nothing more deserved than that.

So, from Nick and me, thanks so much for listening. We look forward to being back with you next week. And until next time, stay safe in the markets and take care of yourself and each other.

[01:04:05] 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.