Profiting from Different Investment Regimes? ft. Rob Carver

[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.

[00:00:23] Niels

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

Rob, great to be back with you this week. I hope you’re doing well. I was about to ask you what’s happening or how are things in the UK, but I actually got home from the UK yesterday so I kind of have a feel for what’s going on. But tell me what you’re seeing in the UK.

[00:00:48] Rob

Niels, you mean you’ve entered the UK and you didn’t come and have lunch with me yet again.

[00:00:53] Niels

I know.

[00:00:57] Rob

This is the second time in the year I think you’ve done this.

[00:00:59] Niels

It could have been. But anyways, how are things in the UK?

[00:01:01] Rob

Yeah, they seem to be okay. Yesterday, we had the budget which everyone makes a massive fuss about. I don’t know if this is a uniquely British thing, but this budget thing where every year the chancellor announces all of his tax and spending stuff. It’s a big public event and the media are obsessed with it for about a week before and a week afterwards.

I don’t think any other country has quite the same level of analysis around one single day on this stuff. Obviously, in the US, they have the budget ceiling and there’s ongoing arguments about that. But this one day a year session thing I think strikes me as very uniquely British. It’s quite boring actually.

[00:01:40] Niels

You say it’s boring, but I remember a couple of years ago where there was something to do with a budget and some tax cuts that certainly put a bit of a spinner in the wheel in the UK.

[00:01:50] Rob

If we get to be pedantic, that was a fiscal event and no budget. Boy, was it an event. So yes, you’re right, they’re normally boring. I guess they should be boring. If they’re not boring, then that’s problematic because the last thing you want to do is surprise people.

There was a surprise on the upside because politicians like to as they say pull the rabbit out of the hat. A very modest bedraggled rabbit was pulled out of the hat yesterday. Some people are saying now that chance of an election in May rather than waiting several months to the last possible minute is more likely. So yeah, I guess we’ll see.

[00:02:26] Niels

Anything else more close to our industry trading but not necessarily one of the topics we’re going to talk about that popped up on your radar?

[00:02:38] Rob

Niels, do you really want me to talk about crypto? I can tell. I could just tell. People are honestly listening to the audio and can’t see this, but Niels’ laser eyes are coming through his glasses a bit. I think he’s very much I think wanting me to talk about the fact that obviously Bitcoin, the thing I hate recently reached new heights. That could be because of the… I guess a lot of people are saying it’s inflows due to the fact you can now buy the ETFs, physical ETFs or CTFs. The previous ETFs were purely futures-based. I had a rant on my previous appearance on the show about them so I’m not going to repeat that right now. Everyone knows my feelings.

But yeah, I guess it looks like this net excess demand from people, I’m guessing probably more institutional people who don’t want the hassle of actually working out how you actually custody this stuff safely, so they want to hand it off to an ETF. Of course, ultimately, the ETF has the custody as well. These things are mostly being custodied I believe through Binance. I think they’re the main firm that’s doing this stuff, a firm that’s obviously had run into the SEC, so how safe that custody option is, I don’t know.

I do feel buying a physical ETF and saying someone else could worry about the custody, I don’t have to isn’t necessarily cast-iron protection I would say, but we’ll see how it plays out. I look upon the crypto space in a vaguely disinterested way. It’s like a nephew of mine that I find extremely annoying but occasionally pops up and does something stupid, but it’s not really my problem. I do trade the bitcoin futures but it’s one of 200 futures I trade so it’s not really hit all there for me to be honest.

[00:04:41] Niels

To tell you the truth, that was actually not what I expected in terms of answer when I asked you about what’s been on your radar. I know we talked about it before we pressed record because last time was actually around the day when the ETFs were launched. We did talk about it last time, but okay, I will take it of course. I was expecting you saying, oh, well, it’s just been a super Tuesday. Now we know it’s going to be the same horse race in the US.

[00:05:08] Rob

Is that news? It’s not necessarily news though.

[00:05:10] Niels

It’s not really news?

[00:05:11] Rob

Good news is unexpected, and I think you were looking, as long as two months ago, you would have got pretty large odds on anyone except Trump winning the Republican nomination. So that’s not news. It’s depressing obviously that the US voters, the best two people they can come up with are a guy that’s nearly 80 and completely insane and a guy that’s over 80 and I’d probably say incompetent, but he’s over 80. Is that really the best they can do? It’s a bit depressing. Makes me proud to be British.

[00:05:51] Niels

You may get some nasty messages in your feed labeling like that. I’m sure there will be supporters on both sides listening to us. Let’s leave the politics there.

Besides all of that, going on this week of course is the fact that we’ve just had the Fed, Powell in front of a congressional hearing as he normally does from time to time. And you’re absolutely right, there was nothing really new about that either in the sense that yes, they are ready to cut rates as and when. The when is unknown at this point, so nothing new there.

Let’s quickly do a trend following update before you talk a little about a trend following from your side. Let me just say that of course we had a very strong environment in January and February, and you could say in some respect March so far. Although we’re only a week in, it’s a little bit different because things are relatively calm, pretty flat. What we saw in January and February, from where I’m sitting at least, was a pretty broad based rally if I can call it that, meaning many sectors participated in the performance.

Of course, there were a few standout markets, but in a fully diversified portfolio, they’re obviously not meant to completely dominate. But there were a couple of really good ones. So far this month in March, there are a few of the markets that are still doing well, Mexican Peso, cocoa, equities, and grains. But the fixed income, metals and energies have left the race right now and losing a bit of ground as far as I can tell. So overall, the month is pretty flat.

In fact, the BTOP 50 Index is only up 9 basis points for the month so far, 5.5% for the year. The SocGen CTA Index as of Wednesday completely flat for the month but up almost 6% for the year. Trend Index almost flat for the month but up 7.8% for the year. And the Short-Term traders index is down 22 basis points, up 34 basis points so far this year. Yesterday, I think, was an update by the way for CTAs.

MSCI World up 60 basis points and up almost 6% for the month. World Government Bonds is up this month actually, 70 basis points there abouts. And the S&P Total Return index is up just a little, 19 basis points for the month, up 7.31% for the year. One thing that from time to time comes up, and I don’t know if we talk too much about it actually but when it comes to trend is the fact that we are seeing a lot of dispersion and have been for quite a while.

We’ll certainly say that so far this year, between managers I should say, certainly this year, there is a lot of dispersion in returns from what I’ve seen so far. What I can’t help noticing is that to some extent, if I look at the rankings, managers who tend to trade fewer markets and are using dynamic position sizing seem to be leading the pack.

Of course, this is a little bit against some of the discussions we’ve had over the years where some of our guests and friends would argue that you need to have consistent position sizing, you need to trade hundreds of markets in order to catch the big outliers.

All I will say, and it was really not to bring up a discussion about it, all I will say is just that I think what we’re seeing right now kind of confirms the view that I’ve held for many years on the podcast, namely that I don’t think you can say that one is better than the other so to speak. They will be better at different times, but I don’t think you can say that consistent position sizing is better in the long run. I don’t see any evidence of that.

But anyways, it wasn’t to raise a discussion. From your trend following lens, Rob, anything exciting that you see happening?

[00:10:02] Rob

Yeah, just quickly. I certainly don’t want to rake up the dynamic position sizing argument again. But if you look at managers who are trading fewer markets, you’re going to see more dispersion with those managers. Whereas managers who have got lots of markets are going to have a more even set of returns.

If you were to look at the rankings of managers at any given time, you would expect, on average, that the managers at the very top and the very bottom of that ranking will be those who are trading the fewest numbers of markets just because they’re more exposed to random luck. So that’s just a small point. So yeah, no dispersion here. I’m actually pretty much bang in line with the indices. I’m up for the year 5.9%.

If I look at where that came from, so yeah, it’s across a mixture of sectors, but I particularly did well in FX and in equities, and then everything else is kind of a wash overall, some up, some down. In terms of individual markets, interestingly, my best two markets were the Nikkei and the yen-dollar. Obviously, those are related.

I’m also embarrassed to say that I made some money in Bitcoin in Swissy dollar FX and bubbles, and in yen and euro, so very much a Japanese theme there I suppose. My worse markets, not really a pan, but for what it’s worth, my worst market was nat gas. I’m doing okay and I’m closest to my high watermark which actually I set in summer last year. I’ve crept up to that. I crept up to it. Sorry, summer of 2022. I apologize.

I did creep up to it late last year as well but didn’t quite make that high watermark. Hopefully, that high watermark will arrive soon because that’s always a nice feeling. Yeah, just looking more generally, as you said, I think cocoa is the standout market of the year so far. It’s up a lot not just on an outright basis, but also I look at markets on a risk adjusted basis and on that basis, it’s up about six, seven standard deviations for the year, which is honestly a pretty big move.

Crypto is obviously out there as we’ve discussed about Bitcoin and Ethereum appearing in that top movers index. And down at the bottom is wheat. On a vol-adjusted basis, wheat has had a pretty weak year so far to be honest. On a forward-looking basis and looking at the current positions that I’m holding at the moment, there’s a very clear story actually which is that I’ve got quite a big fat short in bonds and a slightly not quite as big but a long in equities, and then in the middle there’s not very much going on to be honest in the other sectors.

That’s very much an obvious pro-growth type trade if you like with an expectation potentially that interest rates are going to rise to tame inflation and equities are going to go up. My biggest long is still Nikkei, so still the Japanese market, and then interestingly, the yen-euro, that’s still a big long so I’d like to say it’s quite a lot. I’m ashamed to say I’m still long Bitcoin as well.

And then on the short side, I’m short US 2s, US 3s, bubbles, so interestingly more the short end of the curve. I’m not sure there’s any meaningful there, it’s just interesting to look at. But my biggest downright short actually is in Swissy dollar. I’m certainly hoping that if I make a business trip to Switzerland anytime soon that it’ll become a lot cheaper.

[00:13:40] Niels

Time will tell. Good stuff. Let’s move on to the questions that we’re going to tackle that came in from our great community. The first one is from Matthijs. He writes, “As always, I love your show and look forward to the new episodes dropping every week. I have a couple of questions for Rob regarding choice of instruments to trade. I know that Rob excludes instruments with very low volatility, say less than 5% annualized, from his trading. There are various reasons to do this. One reason is that you require a lot of leverage to trade low vol instruments, which is dangerous if the volatility of those instruments suddenly spikes. My questions are, one, assuming that the trading costs are not a problem for some low vol instruments, then how can we decide whether or how to still safely include the instrument in our portfolio? For example, take the short-term interest rates future such as Euribor, should we really exclude that from our portfolio? Or should we just size it small to manage the potential jump risk just like we would for VIX futures for example?”

Let’s deal with that first.

[00:14:49] Rob

Yeah. I do exclude low vol instruments from my portfolio. Not actually excluding low vol, I’ve got no problem with low vol per se, but I have got a problem with this high leverage. Of course, if you adjust position sizes for volatility as we do either at the point of the open trade for the classical trend followers if you like or dynamically throughout the life of the position as I do.

The smaller the volatility on an instrument, all other things being equal, the bigger your position needs to be relative to your capital, therefore the more leverage you’re going to have. What that will obviously expose you to is a high degree of a jump risk. If you think about an instrument like the S&P that normally has annualized risk of say 15% a year, in an extreme situation, you might see risk of 50%-60% a year. That’d be like the absolute limit. They go by about four times if you like.

But because you’re position sized for 15%, it’s acceptable risk as long as you’ve got also diversification. The real worry is if you’ve got an instrument with a very low amount of risk and you say this thing has got a risk of 1% or 0.5% a year, something like that. But it’s completely unbelievably inconceivable that the risk could go much higher.

You could have a very sharp move before you’ve had a chance to get out of your position. The classic example which Matthijs actually cites, so the second part of his question would be something like Swissy/euro. You’ll probably remember that the Swiss central bank effectively set a floor on the Swissy/Euro rate of I think it was 1.2 and the rate bounced up against that if you like.

If you actually measured the volatility of Swissy euro, it was very, very low indeed. It was little much less than 1% a year, 0.5% a year, something like that. Then one day, the Swiss central bank woke up and I guess they had a bad evening, a bad night, maybe they’d been out for a few too many drinks in the many, many nightclubs around the Swiss central bank. And they decide let’s get rid of this floor thing, this is ridiculous.

They just said we’re not doing this anymore and then presumably went to get some strong coffee. While they were doing that, the market completely panicked and the Swissy/euro dropped by 20% in a matter of seconds, which was obviously many, many, many multiples of what you’d expect given what the standard deviation would be the day before. That’s the risk you have if you’ve got very high leverage if you just naively keep leveraging up and up as the volatility falls.

Now, the title also said costs are a problem for low vol instruments, and they are. So all of the things being equal, you should look at costs on a risk-adjusted basis. If you need to take huge leverage to trade something, that means that your percentage costs are going to be much bigger relative to your capital. Your real cost a year of holding these positions is going to be much higher. So that’s another reason for not doing it. But if we focus purely on the jump risk size of things, there are other ways we could do it.

The quanty way is to say well, I believe that the historical returns for this instrument are an accurate reflection of what I could see going forward, and therefore, I could do something complicated like measure higher moments like kurtosis and skew and from that get a better idea of what the real tails are in this distribution. You might want to do that in a symmetric way.

You might want to say, going back to 2015, if I’d had a position on the wrong way around, the other way around with Swissy/euro, actually, I’m not so bothered about jump risk because it’s not going to affect me. The jump risk will always be in my favor if you like. Whereas if you have the position on so you’re expecting the peg to hold, then obviously you’d be terrified about the jump risk.

Then there’s a less quanty way of doing that which is you call it a more traditional risk manager approach where you stick your finger in the air and say, well, I reckon worst case scenario for this particular market could be a 20% drop or a 25% drop. Developed equity market, you could say ’87 could happen again, you could see a 30% drop. G 10 FX, you could say you could see a Swissy event again 20%. Bitcoin, you could say 80%. It’s not inconceivable.

Now, the problem with both of these approaches is that they assume that the worst thing has already happened. The worst thing you’re going to see in the future will be about as bad as the worst thing you’ve already seen either in a very sophisticated complicated mathematical way or in a finger in the air way. And you can imagine somebody going into the Swissy/euro situation saying G10 FX currencies don’t drop by more than 5% a day, that’s never happened.

So I’m going to use 5% as my downside limit risk. Of course, that would have been a very expensive mistake to have made. My personal belief is effectively by limiting the leverage, in a very crude way, and saying I don’t want leverage of more than x, what I’m effectively saying is that I believe that a large drop, like a 20%, 25%, 30% drop is very unlikely but could happen in any one of the markets that I trade, and therefore, I want to protect myself against being exposed to that.

It’s a very, very conservative view. That’s purely because I don’t believe you can ever really predict just how bad the future could be based on the past. You can get a good idea. Obviously, all of the things being equal, over the length of enough time, Swissy/euro should be more stable and have fewer jumps than say Bitcoin. But that doesn’t mean to say that there could be an armageddon event in the future that I haven’t foreseen. I just want to protect myself from that happening.

[00:20:49] Niels

I think another way of course of doing it, but it has the same effect I imagine, is just to impose a minimum volatility level, that you just simply say that these markets, like we saw with actually short-term interest rates for quite a while, actually that if you ignored setting a floor, you could have had some silly positions on also in those markets because they were so controlled and stable for quite a while. I think most managers probably operate with a minimum level of volatility for each market like a long-term level.

[00:21:31] Rob

Just to say, Niels, that’s actually what I’m doing. It’s just that the actual calculation of that number, I use 5%. But I haven’t just pulled 5% out of the air, 5% is basically based on me saying what my absolute leverage is and then from that, I can back out what minimum volatility is. So the constraint is actually minimum volatility. It’s just that the point where I start thinking from is, what’s the worst drawdown I can suffer therefore maximum leverage should I take? Therefore, what’s my minimum volatility?

Yeah, it affects a lot of markets. It affects not only the short-term interest rate futures but also the short end of the bond curve. For example, Italian 3 years, German 5 years, they’re all in there and then there’s quite a few FX markets as well that are in that group. Then you’ve got JGBs which again I think the questioner writes about. Everything on the US yield curve in front of five years. It’s quite a lot of markets.

Five percent you could argue is a little bit too conservative for most managers. But because I’m doing this dynamic optimization, my view is I don’t have to trade, say, US 5 years because I can trade 10 years, which is very highly correlated, and just effectively get my US rate exposure through that. But yeah, it doesn’t rule out everything. As I’ve already said, I do have positions in US 3 years and bubbles at the moment because there’s also a degree of a thresholding going on here.

What that means is if you think about something like bubbles, they probably bubble- (sorry, excuse the pun) between a volatility of about say 4% and 6% a year. I don’t know want to be continually pulling them in and out in my system. In practice, what I’ll do is I’ll allow a market that drops temporarily below 5%, I’ll allow it stay in my portfolio. But if it goes below, I can’t remember the exact number, but say it’s 3.5%, if it goes by 3.5%, at that point, I will stop trading. I will close the position probably when I next roll it.

So I think you’re right, most managers use it. I’m a bit more conservative than most but that’s partly because I feel comfortable that I can still get the exposure through other markets implicitly.

[00:23:50] Niels

Hopefully, that was some good solid answers for Matthijs that he was looking for. Then we have a question from Ben.

[00:23:53] Rob

Sorry, Niels. I think he’s got a Part 2 which I’d like to answer.

[00:24:03] Niels

Okay, you answer that. I thought actually you had already touched on the Swissy and the JGB, but go ahead. The question is, “How do we distinguish between benign natural low vol instruments and disasters waiting to happen due to artificial dampened volatility?” So, feel free.

[00:24:20] Rob

Obviously, you can look at the markets and say I think that this market is being artificially controlled by the Swiss central bank, by the Japanese central bank, and therefore I’m not going to trade it. I have some sympathy with that view. Way back in 2011 when the Swissy/euro floor was first introduced, something like that, and I was still working for a CTA at that point.

I do remember saying in a very loud voice, guys, we should not trade the Swissy/euro market anymore. It’s just being controlled by the government. There’s no fair price. The problem with that is there’s kind of a gray area. You could argue, for example, that no one should be trading bonds during the period of QE because the government’s controlling the market.

To me, that’s going a little bit too far. I guess if you’re a tinfoil hat wearing conspiracy theorist then all markets are being controlled by a cabal of people somewhere, and choose your favorite group of people that you don’t like who are controlling them. That approach, I can see some validity in it, but for me, it’s very hard to say. An example of a currency that’s completely pegged, like Hong Kong US dollar, yeah, okay, fine, that’s definitely not something that’s worth trading because the volatility is extremely low because there’s a peg.

The best-case scenario is, well, there isn’t really a best-case scenario in that situation. Why bother trading it? I personally take the view that 99% of the time, it’s okay to trade a market. If the volatility is low, I don’t really care why the volatility is low. My rule is I’m just going to stay away from it because something that looks like as the question says benign low volatility could easily become non-benign volatility very quickly and that’s true regardless of whether the market is being controlled by the government or not controlled with the government. Bad things can still happen.

[00:26:17] Niels

Absolutely. All right. Cool. Now we get to the question from Ben. Ben writes, “I feel as if there is a consensus that longer term trend following products produce better results, though there seems to be a fair amount of debate regarding timeframes such as 20-day breakout versus 50, versus 100 day, and whatever. But I would love to hear the discussion about how timeframes affect position sizing and the ability to get reasonable exposure to a market.

A strategy might have a better backtest on a longer timeframe, but the stop size would make some markets impossible to trade for smaller traders. Smaller positions allow room for more diversification in the portfolio, but there are so many markets a small trader can access. If I can’t trade markets like sugar or OJ on a 20-day breakout, is it reasonable to do so or is that ‘too fast’? Thanks for the podcast. I learn a lot. From Ben.”

[00:27:17] Rob

Well, Ben, I’m going to teach you something else now. Because your question actually, for me at least, comes from a place of confusion and I think a misunderstanding of how risks and stop losses work together. But don’t worry, this is very common. Actually, I’ve written a series of blog posts about it because it’s something that a lot of people, to me, it’s an error that people make, a sense of confusion.

Basically, your position size should be irrespective of how fast you’re trading or where your stops are set. Position size should be according to the amount of risk you want to take in a particular market. And then your stops level should be set according to the speed that you’re trading. So yes, if you’re trading a very fast market, then your stops will be set tighter.

If you’re trading more slowly, your stops will be set more broadly. The size of that stop is going to be completely independent of what your position size should be. Position size should be based on the amount of risk you want to take in a market on a day-to-day basis, which is going to be based on your risk target, the amount of capital you’ve got, the risk of the instrument and the tick size of the instrument, assuming we’re talking about futures.

With that in mind, the answer to your question is, should we trade slow or fast? We don’t need to worry about your capital because it’s actually irrelevant to be honest. Capital might affect the instruments that you can trade. Absolutely, because some instruments have bigger tick sizes than others. The other thing is of course about low volatility instruments, that you need bigger positions as well for low volatility instruments.

As a smaller trader, you’re a little bit stuck with trading a subset of instruments that have a small tick size and have relatively high volatility. But that apart, your capital shouldn’t determine how fast you trade. I would say generally speaking, you can spend a lot of time trying to optimize the correct holding period. I’ve just gone through one of my periodic bouts of doing that actually, playing around with optimization and the results are on my blog if people want to look.

And every time I play around with optimization, I’ve come to the conclusion that optimization is pretty much almost a waste of time, so there we go. You should probably trade as many timeframes as possible. I would also say, generally, that especially if you’re just starting out, learn towards trading more slowly rather than more quickly because that will keep your costs down and that will mean you’ve got less more room for maneuver in terms of not just bleeding your account dry just because you happen to be trading too quickly.

I hope that makes sense. Apologies if I’m being a bit harsh. I don’t know where this myth that small traders have to trade fast because they can’t afford the big stops of slower trading systems comes from. Suspiciously perhaps from the brokers because they want you to trade more and maybe that’s where that comes from. I don’t know. There’s my conspiracy theory for you, Niels.

[00:30:22] Niels

The other thing just regarding, which I think actually is a good point, maybe we haven’t discussed it in detail and we’re not going to go into it in great detail today, but you start out by saying, Ben, the thing about longer-term timeframes versus shorter-term timeframes. And I’m certainly one of the people who have argued that I think trend following is definitely better in the long term using longer-term timeframes.

What I base that on is obviously the research I have access to and what I see. But another way of looking at it, and one has to be careful, we quote every week, we quote the SocGen Short-Term Traders Index as one of the indices, and people who’ve been following it for any length of time will notice that it really hasn’t produced much of a return for a long period of time. Now, I will say though, I don’t really think you can say that the managers in there are ”short-term” trend followers, they’re short-term traders.

So we just need to be careful about how we label that. But if you use a trend following methodology, certainly the research that I look at would suggest that you have to be longer-term to get the benefit of this approach anyways.

[00:31:44] Rob

Just to confirm that Niels as well, my research shows the efficacy of trend following, most markets start to fall off when you get to holding periods of a few weeks or less. And if you go really short term, you actually start to see mean reversion. You’re right. My suspicion is that a lot of the CTAs who are trading really quickly have got much more mean reversion in their system than they have trend following.

[00:32:09] Niels

Cool. Then we have a question from, well, I would say it in Danish Emile. I’m not sure where he comes from. But anyways, it’s spelled like the Danish Emile or Swedish Emile. Anyways, it’s a question that we actually forgot to, or maybe we didn’t have time, or maybe I forgot it last time we spoke because it’s been with me for a while. So I apologize about that.

But it’s a completely different type of question but actually it’s an interesting one, and he specifically wanted you to answer it Rob because of your situation. He writes, “As a full time systematic trader, how do you manage your expenses with up and down months/years? On the discretionary side, it’s easier to pick and choose your entries in a month.

Also, if you find your losses clustered, I suppose this means that certain market conditions are not favored. Do you put anything in place to try and identify the issue?”

[00:33:08] Rob

I’m going to answer these questions in reverse order, Emile. I’m going to start with the last one. By definition, a systematic trader should not be meddling with their system. They should be running it consistently. So I just said, for example, that I’ve had a flat P&L curve for last two years. I reached a new high watermark in summer of 2022, and since then it’s gone down, it’s gone up, it’s gone down, it’s come back up again and now it looks like it’s near the high watermark again.

And that’s just due to the fact purely that given the expected Sharpe ratio of my trading system, unless your Sharpe ratio is very, very high, indeed, you’re going to have periods of months, maybe even years when your trading system is not producing any profits. Specifically for trend following, although my system is not purely trend following, with trend following specifically, because it’s a positive skew system, that means that your return pattern is going to look something like 6, 7, 8, 9 months of losses and then 3, 4 months of gains.

And those 3, 4 months obviously, on average, in the past, will take you well above that previous loss level, but it does mean that you’re going to be suffering from long periods when nothing seems to be working. I can go back to an earlier period when again there was probably one, two, maybe even three years of fairly flat performance, and then 2022 happened and obviously that was a big gain.

This kind of, I guess you could call it, down the stairs up the escalator pattern of curves means that that’s just the way it’s going to be. Although there’s nothing to stop you obviously researching and trying to improve your system, the last thing you want to be doing is saying I’ve lost money over the last six months therefore I need to change everything. That’s a huge mistake.

[00:35:00] Niels

Can I interrupt you here, Rob, because actually, there’s a great example that I had mentioned last week. But I think it’s a great example, and that is that we’ve been talking about cocoa for the last say few months as wow, what a trend. It’s been driving performance of many funds. But as far as I can tell, and I’ve also heard it from other people in our industry, cocoa didn’t make money for trend followers for about 15 years or more. So I think that just goes to the point you mentioned there, Rob.

[00:35:32] Rob

Yeah, I have my suspicion about cocoa because my youngest daughter is addicted to this stuff, and I think she’s probably single-handedly responsible for driving up the price. Anyway, so that’s all said on meddling.

On the second part of the question, on the discretionary side, it’s easy to pick and choose your entries in a month. I may be making the wrong inference here, but it seems that the inference is what he’s saying. If you’re a discretionary trader, you can just choose when you print money. That’s the inference to me. A discretionary stock trader, all you do is say I want to make say $2,000 this month, I’ll just buy this, pick these entries that are going to make me $2,000. Easily done.

[00:36:11] Niels

NVIDIA comes to mind.

[00:36:13] Rob

Indeed. I’ve sort of toyed and flirted with being a discretionary stock trader over the years, and I’ve never been very good at it. I’ve certainly never done it professionally. So I’m probably speaking from a place of ignorance. But I’m pretty sure that it’s a lot harder than that. So just generally speaking, you’re going to have a smoother pattern of return, so more kind of consistent returns, if your system has, broadly speaking, a high Sharpe ratio ignoring the asymmetry around trend following versus one trend following. We’ll just move away from that for the moment.

So if you want a consistent pattern of returns, you need to have a high Sharpe ratio. And that means you need to do one or two things, trade systematically, or be one of a very, very small number of people, very gifted people, who can do better trading systematically than they can in a discretionary sense. And even those very gifted people will never be able to make money consistently month after month after month.

The only people who can really do that are high-frequency traders. I guess you could call them the ultimate systematic traders because the sort of trading they do is impossible to do for human beings. So, yeah, I don’t think there’s any fundamental difference in terms of how you should think about managing your money or generating returns depending on whether you’re a discretionary trader or a systematic trader. It’s just going to be how profitable you are.

For me personally, I’m much more profitable as a systematic trader than I would ever be a discretionary one. But anyway, so that’s the second question. So the third question is a more interesting one, which is as a full-time systematic trader, how do you manage your expenses with up and down months and years? So let’s ignore the systematic bit.

Let’s just say you’re a trader, and you’ve got expenses. In my case, three children, one at university, a mortgage, two cats, a wife, and as I’ve already said, a massive monthly cocoa bill to meet. And you’ve got a trading system which as I said is unlikely to be producing regular paychecks. My trading system does not send me a check for a few thousand dollars every month, which is what would be happening if I was still employed.

So there’s a number of different ways to deal with it. The main point to make is that if you’re really relying just on your trading income, then you need to be trading very conservatively. You need to be with quite a low level of risk because you essentially want to be able to say, well, let’s say if I need know I must make 20% of my capital every single year, then that’s very problematic.

Because although you might make 20% on average, in the years when you’re down and you’re not making 20%, you’ve still got to take that 20% out of your account. Your account is going to get lower in value. That’s going to make it subsequently much harder to make 20% the following year. If on the other hand you only need to make 5% out of your account, well, why bother because that’s pretty much where the risk-free rate is, right?

But the hope is that if you’re a good trader, you’ll be withdrawing that 5% and making more on top of that, and that can sustain your account and accumulate. And it gives you a buffer against the fact that if you have bad periods, years when you lose money or a couple of years in a row when you lose money, then you can take some of the money out of the account and it still doesn’t affect you.

The other way of doing it is by having a separate cash buffer. So what you say is well, okay, I need to make 20% for my account every year, but I know there’s going to be years when I don’t make that. So what I’m going to do is maybe say be really, not really pessimistic but relatively pessimistic and say, well, I’m going to assume there could be five years in a row and I don’t make money from trading, and so I need to have a separate pool of money to cover those five years.

So that means in practice, just completely arbitrary figures, assuming you’d thought you’d need say $500,000 trading capital to meet $100,000 of living expenses, you need to have another $500,000 of capital somewhere else that you can draw from. Now, from a kind of economic point of view, that’s equivalent to saying you need to make 10% on a million dollars.

So it’s actually economically equivalent to reducing your expected rate of return that you need from 20% to 10%, or 10% to 5% if you’re going to be more conservative. From a behavioral finance point of view, it feels safer to have this five-year cash buffer there and then this trading account that’s doing crazy stuff. Although it might not be as efficient for things like margin usage and interest and all this kind of stuff, for that, you probably will be better having all the money in the one account.

In terms of making you sleep at night and also arguably lower counterparty risk towards your broker, it probably makes sense to have that separate pot of money there. And then what you do is when your trading account makes money, you basically take that money out of the account and you pop it into the savings account, into the surplus account if you like. And so the value of that surplus account is going to be kind of moving up and down.

As money comes in the trading account and then leaves again as you’re spending your money, the value of the trading account is going to be hopefully going up, spitting out some money when it reaches a high watermark, then going back down again and so on and so forth. That’s basically the approach that I take.

The difference is that instead of having a big pool of cash which is very inefficient, I have a big pool of long early investments, mainly ETFs but also some stocks, those also generate income, but they generate in the form of dividends which obviously is not a guaranteed income stream but is a steadier income stream. Psychologically, that feels better. So that’s the other way of doing it.

But yeah, just generally speaking, you’ll need a lot more money to live off if you’re trading full time than you think you do, which is why for the vast majority of people it’s probably a really bad idea. Even I don’t live entirely off my trading earnings. Obviously, I have the massive royalties from the millions of books I sell every year plus the 10 million pounds a year that you pay me Niels for this podcast. Oh, sorry, I shouldn’t have said anything. I’m going to upset the other guests.

[00:42:45] Niels

The other thing I was going to say just to wrap it up, because we need to move on, and that is just that if you are a systematic long-term trend follower, you can easily have a job next to being a long-term trend follower because it doesn’t take many minutes per day if you’ve structured it correctly to run such a system. All right. Now we’re going to move on to today’s main topic I would say. It is a paper from our friends over at Man Group.

It’s from a series called The Road Ahead series, and it’s written by Henry Neville, who is a portfolio manager at Man Solutions. And it’s called “Regime Based Investing.” Now, I’m going to let you drive this, Rob. But essentially, if I can set the stage, essentially, it’s looking at different investments and how they perform during I think seven different regime definitions. They look at six different assets or strategies, that being equities, bonds, commodities, long-short equity. What else do they have? Equity cross sectional momentum and trend following.

Now I’m going to let you talk through the different regimes and what the key takeaways are Rob because I actually thought was an interesting way of discussing some of the things that we love to talk about.

[00:42:20] Niels

Yeah, absolutely. I’m a bit of a fan of Henry Neville. I’ve never actually met him because we didn’t overlap at Man Group. But, Henry, if you’re listening, I’m now your biggest fan and I’m going to read everything because you write very well. You’ve got a nice writing style. It’s easy to read. And also, your stuff is quite short as well. You say this report, it’s a bit long on page count, but it’s mostly pictures, very few words, which is great. A lot of people who write this stuff, particularly on the sell side, do tend to rather spend a lot of time on workout. You’ve got a point to make.

[00:44:57] Niels

Yeah, and just before you jump in, I’m just going to say that I will link to this paper of course in the show notes of this episode.

[00:45:03] Rob

Yeah, definitely. So this regime analysis is something that people like to do a lot, interestingly. But it’s quite nice to see all the different styles of regimes wrapped into one place. So I guess there’s this, broadly speaking, I guess you could say two main ways of categorizing regimes. One would be like a macro view. So you’d look at something like inflation or growth or the unemployment rates, something like that, and the other thing is a market regime.

So you could look at say bull or bear markets. This paper has got a mixture of both, which is nice. That the set of assets that Henry’s analyzing just to say it’s equities, bonds, commodities on the long-only side, and trend, and then two long-short equity strategies, value and then cross-sectional momentum which obviously is going to be correlated to tread but not quite the same. So it’s quite interesting.

Just one thing to say on the study, I think a lot of people make the mistake of using it thinking they can use it as some kind of way of switching between strategies according to the regime that you’re in, because that’s what you assume is going to happen. But I’ve read this quite carefully and Henry’s not making the claim that you should do that. That’s because his regimes are classified in such a way that you can’t actually predict what regime you’re going to be in say over the next six months.

The regime classification is done in a way that you can’t necessarily tell which regimes you’re in today. It’s almost a purely backward-looking exercise. Now, you could do that. My experiences with that approach is that it tends not to be very successful at predicting returns. For example, very trivially, if you knew in advance that the next six months was going to be an equity bear market, well, go short equities and you’ll make lots of money. It’s much harder to do that than you might think.

It’s a bit easier to say, I think there’s a recession coming, we’re reasonably okay in predicting recessions over the next three months or so, not a bit harder to go further out unless of course we get a massive exogenous shock like a COVID or something or UK government fiscal event for example. But there’s a weaker link between recessions and investment performance then there is, say, between say a market-based regime. So all of those caveats either way, it’s still a very interesting thing to do. It’s very hard to summarize. It’s 11 pages.

[00:47:41] Niels

Let’s talk about the regimes. The first one is inflation.

[00:47:44] Rob

Yeah. So let’s structure it like that. So inflation regimes, basically, we’ve got four different regimes. We’ve got reflation and deflation, and we’ve also got fire and ice.

[00:47:59] Niels

Maybe I should just explain. So, fire is when CPI, year on year, moves above 2% through 5%, and then it peaks. And ice is where CPI moves below 1%, then below minus 1%, ending either in a trough or where it goes back through -4%.

[00:48:17] Rob

Yeah. So, one thing I have to say, although I’m a fan of this paper generally, I’m not really a fan of, again, if this was a quantitative exercise, then categorizing regimes that way would very much be overfitting I would say. I would probably do something simpler like, is inflation low and falling, high and rising, high and falling, or low and rising? I would do it like that because that’s the simplest way of doing it and less chance for overfitting. These regimes roughly approximate to that, but not quite.

[00:48:57] Niels

What do you take away from the inflation regime? Is there anything you take away from their analysis that stands out in your view? Because he talks about the returns, and he talks about the hit rate for each of the asset classes.

[00:49:13] Rob

He does. Now, these aren’t quite the same thing. All other things being equal, you want a high hit rate, and you want higher returns. But as we’ve already discussed, if you take a strategy like trend following, it’s got quite a low hit rate. But on those occasions when you are right, you make more money. So it’s quite nice to see hit rate as well as returns because it gives you a slightly more nuanced flavor of what’s been going on.

Henry seems to favor hit rate because he kind of colors in the hit rates, maybe because it’s more obvious what a bad hit rate is and what a good hit rate is. A good hit rate obviously is going to be something well above 50%. A bad hit rate is something well below 50%. Inflation is interesting because there’s often these stylized facts. So, we know that bonds do badly in inflation periods, right? That’s kind of well-known.

It’s less obvious what equities should do in inflationary periods. Actually, he finds that both equities and bonds do quite badly in the fire regime. So that’s kind of interesting because it’s not necessarily always whether equities should be inflation hedges or not. On the other hand, commodities do extremely well in the fire regime, as does trend incidentally. So both trend and commodities would look like good hedges against the high and accelerating inflation that the fire regime implies. And of course, that’s what we saw in 2022.

[00:50:29] Niels

What stands out to me also is that there’s actually only two asset classes that have a positive return in all four different regimes. One is long-short equity value, but in two of the occasions, in fire and reflation, they make no return, but they make small returns in disinflation and ice. But actually, trend following has good returns in all four.

[00:51:03] Rob

Yeah, spoiler alert people. We probably wouldn’t be talking about this paper if it wasn’t the case. But trend following really wins in this report in pretty much every type of looking at regimes. It’s the only strategy that performs consistently across all the different regimes that are looked at. So you can stop listening now if you like or you can continue while we talk about a few more. The next regime he looks at is real growth. This is divided into recovery, slump, boom and recession.

So this is, for me, a more natural way of doing it. So recovery means growth is low and rising, slump means growth is low and falling, boom means growth is high and rising, and so on and so forth. So you get the idea. And then the recession categorization is the standard NBAR definition of a recession. And of course there’s a bit of hand waving going on here because a lot of these categorizations are done purely on the US economy.

Whereas, of course, certainly, definitely commodities and trend following is going to have exposure to lots of different countries. But the US is the biggest economy and global economic cycles are kind of correlated, so it’s not really a big issue. So let’s take the recession as a starting point. That’s the one everyone is less keen on. So both equities and commodities do badly in recessions, bonds do well. You’d expect that, right?

Because generally speaking, central banks cut rates during recessions and real assets do badly. In recoveries, the picture is pretty much reverse. So bonds actually don’t do so great because obviously central banks are increasing interest rates. I guess that’s what we saw certainly late 2022, early 2023, definitely that kind of regime, whereas equities and bonds do well. And then the other kind of regimes are, there’s not so much of a clear pattern.

But of course, once again, trend following, all the kinds of long-short strategies do okay across all the different regimes, but trend is also a consistent performer across all the different regimes. So another win for trend following. So the yield curve for me is quite interesting because I’ve actually looked at this myself. Going through the early 2000s, everyone was saying, oh, CTAs have done really well. Well, a lot of that return has come from the bond sector. A lot of that’s come because we’ve had a secular downtrend in interest rates, and the yield curve has been upward sloping, so bonds have just been as perfect straight line of yields falling, positive carry.

You guys are going be in real trouble when interest rates start going up, you’re going to be in real trouble. Because of course, as everyone knew, in the early 2000s, it was actually illegal for CTAs to go short fixed income, apparently. (I’m being sarcastic.)

So I’ve done work on this myself. The definitions here for the different regimes are again to me quite logical. So you’ve got bear flat and bull steep, and a bull flat and a bear steep, and then a yield curve inversion. If you go back and listen to the episode we did with Cam Harvey a few years ago now, you’ll know that an inverted yield curve is one of the best predictors of recessions. And spookily, there was one in 2019 which predicted the COVID recession even before COVID was a thing. That’s how good the inverted yield curve is at predicting recessions.

[00:54:35] Niels

It hasn’t quite predicted the current regime, right? Because there’s been an inversion since and we’re still waiting for the recession.

[00:54:42] Rob

Well, you say that, but in the UK, we’ve had three quarters now of falling growth.

[00:54:48] Niels

I was talking about the US economy, but there we are.

[00:54:51] Rob

Of course you are. Maybe we should have Cam back on again to defend his record. So yeah, given that it’s correlated with recessions, you’d be perhaps surprised to learn that bonds don’t do very well with an inverted yield curve. But that actually makes sense if you think about it, because of course, an inverted yield curve means you’ve got negative carry.

So although you’d expect to gain as the curve kind of uninverts and the short end comes down, generally speaking, while the yield curve is inverted in that almost pre-recession part of things, you’ve got the headwind of negative carry, so you’re not doing so well. The single biggest numbers here are on the steepeners. So a bull steepener is incredibly good for equities, incredibly good for bonds, pretty much good for everything.

Trend does pretty well in that environment as well as you’d expect. A bear steepener, on the other hand, is very good for commodities. Everything else doesn’t do so well. The bonds returns for the yield curve regimes are almost kind of like a mathematical expectation of what you’d expect, the others are perhaps more interesting. Okay, so equity markets.

So this is the S&P 500 and we’re just doing bull and bear markets. That’s obviously quite an easy regime to work out. Is the market currently going up or going down?

[00:56:12] Niels

I think we should qualify and say that I think all the regimes had to last for a minimum of six months. Is that right? That’s how I remember it, something like that.

[00:56:22] Rob

I can’t remember the footnotes. But for a lot of these definitions like recessions, technically speaking, a recession has to be long. So that’s by construction up to six months.

[00:56:33] Niels

But people should read the footnotes of the paper. They should not just trust our memory.

[00:56:38] Rob

I don’t think that’s true generally, because for example, the October ’87 crash is marked as a bear market that lasted for about two months in the data. So I don’t think it’s true that all the regimes had to last for six months.

[00:56:54] Niels

It may not be related to the equity markets but some of the other inflationary regimes and all that stuff I just seem to remember.

[00:57:00] Rob

Anyway, so let’s look at the equity market regimes. You’ll be shocked to hear this, Niels, but apparently, equities do really well when there’s an equity bull market and really badly when there’s an equity bear market. So that’s surprising for me to be honest. The other strategies I was thinking they’d be more interesting. Bonds, not really. They do actually better in an equity bull market.

That’s maybe just reflecting the fact that the correlation of stocks and bonds for a lot of this period, because this particular dataset goes back to the mid-1920s. So for most of that period, stocks and bonds have been correlated. there’s this weird period in the last 25 years up till about 2022 when they were negatively correlated. That was actually unusual historically.

Commodities do better in an equity bull market. That makes sense to me. In an equity bear market though, you really want to be trading value, cross-sectional momentum and trend. So the kind of alternative assets doing much better in a bear market than a bull market. So that’s nice, because if you’re buying them as an equity hedge, at least historically they’ve given you that performance, that’s good. So there’s been clearly some real work behind each of these tables.

It just occurred to me that for example to calculate an equity long-short value strategy back to the mid-1920s was actually quite a feat of data engineering. So I don’t know whether he did that himself or sourced that from another paper, but kudos to whoever did that because that must have been quite a bit of work.

Okay, we’re getting through them. Monetary policy, so this is what’s happening to the Fed funds rate. Is it tight or is it loosening?

Obviously, that’s going to be a bit related to the yield curve stuff we’ve seen later. So loosening monetary policy, you would expect to be good for equities. It is. It’s also good for bonds. So that’s almost a mathematical fact that if interest rates are going down, you should own fixed income. Most assets do pretty well I guess, although tightening cycle low equities still makes money, although not as much. Bonds lose money. Again, mathematically, that’s what you’d expect.

If interest rates are rising, fixed income should lose money. Trend of course does equally well in both environments, although a little bit better in the tightening environment than the loosening environment. So again, it’s sort of a nice little hedge against what’s happening with the Fed.

[00:59:34] Niels

Actually, one thing that surprised me when I just looked at the table right now, and that’s obviously because we don’t really think about what happened all the way back to 1947 or whatever the number is. I feel that my memory is that we live in an environment that most of the time is loosening, right?

In fact, over the full period, there is actually 56% of the time that is a tightening period, which is quite interesting. It goes to your point from before that the period after GFC and maybe leading into it was very unusual both with correlations in terms of stocks and bonds, but also certainly in terms of central bank policy.

[01:00:17] Rob

Yeah. This particular data goes back to ’47. Obviously, from ’47 to about 1980, we had a secular uptrend in interest rates. So there would have been a higher percentage of time spent tightening in that period. And then since then, we’ve had a secular downtrend in interest rates only interrupted in the last couple of years.

So yeah, we see both of those but there’s obviously a bias towards tightening in the first bit of the period and the bias was loosening in the second bit of the period. And this is where these sorts of exercises become a little bit less. When you’re basing all of your facts on what’s happening in a tiny environment and what happened between 1947 and 1980 pretty much, is that really relevant to what could happen in the future?

As I said, as a backward-looking exercise, there’s nothing wrong with this. As an exercise in trying to infer what you should be doing in the future, you have to be a bit more careful about that because of the bias effect.

The next one is the commodity curve. This is whether the futures curve is in backwardation or contango. Every time I talk about backwardation and contango, I say two things.

Firstly, if I ever go out and buy two dogs, I’m going to call them backwardation and contango because I think that will be a cool name for dogs. Sadly, as I said earlier, we’re actually a cat family, so that’s never going to happen. And also, my family would never permit me to call the animals that name.

Secondly, I can never remember which is which, which is a bit embarrassing as someone who’s written a book called “Advanced Futures Trading Strategies”, you kind of expect me to know which is which.

[01:01:56] Niels

Do you want me to help you here? First of all, I will say if you do buy two dogs and you’re not sure which one is which, then buy a white dog and a black dog. That might help in the first place. But I can help you a little bit because as far as I remember, contango is kind of what I think of the norm, meaning that the further out you go, the higher the price, meaning you should be paid more if you think about it in terms of yield.

So the further out you go, you should theoretically get a higher yield for taking more risk. So, in contango, future prices are higher than current prices, and backwardation is the opposite.

[01:02:33] Rob

Right. Okay.

[01:02:34] Niels

You’re not sure but there we are.

[01:02:38] Rob

Yeah. I think about positive and negative carry. To me, I don’t see any need to learn to pieces of jargon that l have no need to know, unless it comes to animal naming. Anyway, having said that, so they look at the commodity space in aggregation. And I don’t know exactly how they’ve measured this, but anyway. So the market spends 62% of the time in contango, and that’s again since the mid-1920s.

Interestingly, there’s been more contango I would say a little bit more on average in the second half of that period of the first half of that period, so the backwardation tends to come from the kind of- particularly the 1940s were a big time for backwardation apparently. So you wouldn’t expect much of an effect in the equity in the bond space. Equities do a little bit better when commodities are in contango.

Bonds do a little bit better as well. I’m not sure how meaningful these figures are to be honest. I suspect it’s what econometricians would call a case of a missing correlated variable. So basically, it’s quite likely that backwardation and contango is driven by things like inflationary environments, monetary policy environments, and so on and so forth. And that’s why the equities and bonds have got a different pattern.

I’m not sure you can say there’s a direct driver between what’s happening in commodity curves and returns of equities and bonds.

[01:04:03] Niels

But what is interesting about the numbers is that equities and bonds are the only two assets that do better in contango, meaning in normal environments. All the alternative strategies and commodities do better according to their numbers when things are in backwardation. So in a sense, you have to assume these numbers and just take them as they show up.

[01:04:28] Rob

We’re talking pretty small sample sizes here and some of the differences are very small. So for example, long-short equity value, the hit rate in backwardation is 63% and the contango it’s 60%. I haven’t done the math, but there’s no way that’s statistically significant. I mean, really the only two asset classes that are interesting for me here are commodities and trend, because obviously, commodities are the things that you know.

So unsurprisingly, the figures are quite interesting for commodities because if you look at the returns, I think in backwardation, commodities made 8% a year, and in contango they made 2% a year. If you look at the hit ratio, it’s the other way around. So, in other words, in contango the hit ratio is 80%, and in backwardation it’s only 50%

[01:05:15] Niels

80%, sorry. 80% in contango.

[01:05:18] Rob

Yeah, exactly. So from a hit ratio point, it’s better. But again, this comes down to this argument about asymmetric returns and skew and so on and so forth. But just focusing on the return numbers, those are what I would expect. So basically, this is all just telling you that carry works in commodities because if you look at the shape of the curve and take a position based on that, that’s effectively a carry strategy. Trend is better in backwardation, quite a bit better on the return, a little bit better on the hit rate. Again, that makes sense to me to be honest.

[01:05:55] Niels

All right. Let’s do the last one.

[01:05:56] Rob

Yeah, so value versus growth. So this is looking at whether the Fama-French value factor is doing well or doing badly. For example, you would expect that going into the 2000 tech bubble, that growth would be great and value would be bad. That’s what you would expect, right? There’s been a long period until about 2021 when value looked pretty bad.

Cliff Asness at AQR was writing articles saying, guys, this is a great time to buy value because it’s just getting cheaper and cheaper. And everyone was saying, is value dead? And then it kind of has come back a bit. Although, just recently, these tech stocks, it’s hard to markup as value. So you’re doing extremely well. And Warren Buffett’s kind of saying there’s not a lot of value out there. So maybe it’s not a good time.

Anyway, so they split regimes into value and growth. The market spends more time in a value state. All that is saying is that value works on average more than growth. Interestingly, equities do better in that value state. At least on the return, the hit rates are the same. I’m not completely sure what the interpretation of that is. Maybe it means that on average, more equities are valued. That’s certainly not true of the current S&P 500. So I’m not sure how to interpret that.

Obviously, again, for most of the asset classes here, the link is going to be tenuous. Trend for example does bring positive returns in both regimes although is better for it, but I couldn’t tell you why. I think that would be a data mining exercise. So what you really want to focus on is long-short equity value. That of course is the one you’d expect to see a very clear pattern.

Naturally, that does very well when value is doing very well and does very badly when growth is doing very well. That’s exactly what you would expect. Long-short cross-sectional momentum and equities, which is sort of to a first approximation going to be value with a minus one sign in front of it. So that would be true if all the counting ratios stayed the same basically. That does very well in the growth regime, not so bad in the value regime. But yeah, be careful with this stuff.

There are lots of reasons that the said regimes aren’t balanced, you get biases through time. But there are some conclusions at the end. One of course as we’ve already said is that trend is great. It doesn’t matter how you split your regimes, it just makes positive returns in both of them. So that’s wonderful.

I think beyond that, I’m not sure that there’s much mileage in for example saying, oh, well, we’re now in a bull steepener, growth, recession and it’s a bull market, therefore, we should buy commodities. I’m not sure whether you can take the inference to that level.

[01:08:58] Niels

No. And I don’t think Henry was expecting anyone to do that.

[01:09:01] Rob

Of course. Just a warning sign because people I think do think that this regime classification is an excuse for changing your portfolio allocations, and it shouldn’t be.

[01:09:10] Niels

No. For me, the big takeaway really is I think it’s good to visualize some of these things, as you point out. And of course, we didn’t write this paper, but we like some of the conclusions, some of the numbers we can take from it. But I think to me it actually just confirms my own intuition, and that is trend is a pretty robust and adaptive strategy.
Therefore, it’s not surprising to me that it does pretty well in most of these different regimes. That’s kind of how I would leave it for today.

Unfortunately, we have to leave it for today. We’ve gone a little bit over our usual time. But I appreciate, Rob, for you coming on and for taking us through all of these different topics and questions.

Next week, I believe I’m joined by Katy. So that’s going to be another fun and insightful conversation no doubt. If you have questions for Katy, send them to info@toptradersunplugged.com and I’ll do my best to pitch her your questions. If you enjoy these conversations and all the work that Rob puts into preparing them, go to iTunes, go to Spotify, go to Amazon, wherever you listen to your podcast, and kindly leave a rating and review so more people can find these conversations.

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

[01:10:40] Ending

Thanks for listening to the Systematic Investor podcast series. If you enjoyed 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.

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.