252 Systematic Investor Series ft. Rob Carver – July 15th, 2023

[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 markets through the lens of a rules-based investor. Rob, it is great to be back with you this week. As far as I know, you’re a big cycling fan. So I have to ask you what you think of the outlier of having a Danish rider in the lead of the Tour de France at this point, as we can’t really claim to have a lot of mountains for him to train in, in Denmark.

[00:00:54] Rob

Yeah. Well, to be fair, British cyclists do pretty well and England is actually a relatively flat country, but still very impressive. Denmark is not a very big country. In terms of percentage of population, you guys are doing pretty well.

[00:01:13] Niels

Yeah, we probably have a big percentage of cyclists actually as a country so maybe. But as you and I were talking about just before I hit record, I have on one screen the race going on, so I can inform you live if something really important happens because it’s not all about trend following.

[00:01:32] Rob

I’m actually wondering, Niels, whether this is a new format we could explore. We do a kind of joint cycling and systematic trading podcast, obviously just during the big Grand Tour events, so the Giro, the Tour de France and the Vuelta. But we’ll see if there’s demand for it.

[00:01:49] Niels

As long as it’s downhill, I think I would prefer that. Okay, good stuff. Anyways, more serious stuff of course. We have a really good lineup of topics which will surprise people today. I think they’re going to be completely surprised when they hear what we’re going to talk about today.

[00:02:07] Rob

I think people are going to think I’ve lost my mind, frankly. Anyway, we’ll see.

[00:02:10] Niels

Yeah, some people might have thought that already, but there we are. But I’m always curious to know what’s been on your radar since we last spoke, if there’s anything that you’ve noticed from our industry, anything exciting going on from a research point of view.

[00:02:28] Rob

Yeah. It’s been an interesting year half. We’re sort of through the half year now, aren’t we? So it almost feels like an opportunity to take stock and look at what’s happened generally. So actually, I’ve got this report I run which shows the performance of different markets and says which are the most surprising. I measure that by using the return normalized for the standard deviation. So actually, interestingly, top of the charts are Ethereum and Bitcoin out of the markets that I monitor, which may link into some topics later, you never know.

Making up the losers, we’ve got a bunch of natural gas markets and we’ve also got VIX, the US volatility market. So I’m not sure whether that tells a coherent story, but it’s certainly an interesting one. You don’t tend to see these kinds of patterns unless you do this volatility adjustment, because otherwise, markets that are very volatile will always be at the top or the bottom of any kind of ranking and it’s not as meaningful. In terms of the other thing I like to look at is my current risk weighting to give me an idea of how my portfolio is positioned.

To be completely honest with you, one of the advantages for me at least systematic trading, typically with my own money, is I don’t really tend to spend a lot of time actually looking at positions and trades on a daily basis. So I actually find that this podcast is forcing me to occasionally say actually, what is my positioning? Because people often ask me, “What do you think about the US stock market?” I’ll go, “I don’t know. I need to check. Just bear with me while I have a look at my system and see what it says.”

So interestingly, at the moment, my biggest long positions are in foreign equities, although we’re not actually in the US. I’m long the CAC quarante, which is the French market. I’m long Taiwan, I’m long Nikkei in Japan, and I’ve also got some positions on in the bond markets. I’m short bonds, mainly the US 3-year bond I’m short actually, which is interesting. That’s not a contract that we often think about. It’s normally the 2, 5, 10 so we forget about the 3, but there is a 3-year bond contract.

One of the things I really strongly believe in is diversifying across lots and lots of markets. I’m actually adding some more markets to my system at the moment, so I’m pushing up towards probably around the region of, or I hope to get to about 250 markets in my system, of which about 180 will actually be liquid enough to trade. So that’s a little project that I’m working on at the moment.

[00:05:08] Niels

Of those 180 that are liquid enough, how many would you normally have exposure to in any given time?

[00:05:14] Rob

So I’d have positions on them perhaps between 10 and 15, or maybe 20 markets at the maximum. That’s a function of the fact I haven’t got very much capital.

[00:05:24] Niels

Yeah. The optimization process that gives you.

[00:05:27] Rob

Yeah. But it’s just important to say it’s not that I think that the 10 to 15 markets I currently have positions are special in some way. It’s not that there’s some kind of fitting process going on that picks out those markets. It’s purely that those markets best represent the portfolio I would hold if I had a much larger amount of capital. If I had a much larger portfolio, I’d probably still be long equities, but I’d be long lots of other equities. It just happens the three equity markets I’ve listed either because they’re highly correlated with the rest of the market, or perhaps because their own signals are particularly strong, represent my best bet in the equity sector. So that’s quite interesting.

I’m also, let me just check, I’ve got a very small long position in Bitcoin, which again may relate to something that we talk about a bit later. So long equities, short bonds. Again, that feels like a consistent story actually, very much a kind of risk-on growth-on set of positioning to have on. I would imagine that most CTAs are probably not dissimilar in terms of that overall positioning. So it does feel like we’re geared for possibly more inflation because short bonds, equities will hopefully be protective for inflation. Inflation is not going anywhere anytime soon. It doesn’t look like.

The portfolio also thinks about things that were likely to start for interest rates to start, not start falling anytime soon but it’s also saying there’s a significant tightening expected. Also, it’s saying there could be more tightening to come. We’re not quite at the peak of that cycle. That’s what positioning is telling us. One of the advantages of trading systematically is you don’t have to consider the kinds of questions that, I don’t know, a sell side bank analyst will have to answer all the time. What do you think the Fed is doing? What’s positioning, blah, blah, blah?

The portfolio just tells you, based on the signals in the portfolio, you should be short bonds. From that, you can then backwardly interpolate what that says about forecasting what the Feds going to do. In the long run, that seems to be pretty accurate, at least in terms of predicting prices. But yeah, there’s no need to make a direct reading of the tea leaves of the Fed minutes and all that kind of stuff that discretionary traders have today.

[00:07:50] Niels

No, absolutely. Yeah, it has been an interesting first half plus an extra couple of weeks. Certainly a few markets that not usually deliver strong returns have certainly delivered so far this year. Actually, although bonds have retraced upwards a little bit this week or so, it is interesting that they were down flirting with making new lows again after the SVB debacle and completely retraced all of that excitement. So yeah, 2023 could turn out to be an interesting year, another interesting year I should say. Certainly from my point of view, just July so far, just very briefly, it is a little bit soft half halfway into July.

I think performance is a little bit soft in the industry, most likely from what I just mentioned before. Fixed income markets are doing a little bit better in price this week. European stocks may be struggling a little bit this week and the yen suddenly having a little bit more strength, so to speak. So I think that’s probably been where the challenge has come from. Then there’s been a little bit of tailwind, I believe, coming from markets like the Mexican peso, the euro, and maybe US and European nat gas contracts have probably also done okay for trend followers, and still sugar. It’s still pretty sweet, so to speak.

[00:09:13] Rob

Oh, dear.

[00:09:14] Niels

I couldn’t help it. Trend barometer, 39, still completely stuck in neutral. If you look at year-to-date returns, you’d probably understand why it’s stuck in neutral because the BTOP 50 is down 39 basis points year to date, down 4 basis points so far in July. SocGen CTA Index down 1% this month, down about 1% for the year. Trend Index down about 1.17% for the month, down 1.04% for the year, and the Short-Term Traders Index down 36 basis points, down 3.67% for the year. Whilst on the other hand, equities doing really well. MSCI World up 1.79% in July, up 16% now for the year. S&P 500 up about 1.33% and up 17.5% for the year, and the World Government Bond Index is flat so far this month. Anything else you want to add before we jump into questions?

[00:10:10] Rob

Yeah, I actually forgot to talk about my own performance. For what it’s worth, just to throw that in there, year to date on down 3.6%. So a little bit more than the industry, but I’ve got a slightly higher vol target and I’m not 100% focused on trend following. I’m not sure that’s meaningful. In terms of the markets that have done badly, my worst market is US 20-year, so again, fixed income not been great for me this year. My best performing market has been the VIX, but a very mixed picture actually in terms of P&L, obvious patterns there to be honest.

[00:10:49] Niels

Okay, let’s start with a couple of questions before we get into the surprise topic of today. Scott writes in, “I’m an avid listener and greatly appreciate all you do. I have a question for Rob. Rob, in your system, as I understand it, please correct me if I’m wrong, you firstly weight each instrument-signal combination on a correlation of returns and then apply a truncated weighting based on historical Sharpe ratios over the whole of your data set. So for example, if you have three signals, two of which are 100% correlated and one which is 0% correlated, I would apply weightings of 25%, 25%, and 50% respectively, and then I would up or down-weight then again based on historical Sharpe ratio on a scale from 0.5 minimum to 1.5 maximum, not down-weighting to zero or upweighting to infinity. This is an elegant way of sidestepping many problems with portfolio optimization, and in my simulations, this is highly effective in reducing drawdowns. This is trend following, but with quite acceptable drawdowns and reasonable Sharpe ratios, which makes me nervous about trading “loose pants” systems in the past. I am concerned that I might be assuming that I can predict which instruments will trend in the future, which I don’t think I can, and worry that I will down-weight something which then starts to trend. Is this a valid concern? If it is, would a rolling window of Sharpe ratio with enough data to be statistically significant be better? Thanks again, Rob.” This is from Scott.

[00:12:31] Rob

Yeah, so just to give a little bit of color to those who may not be following Scott’s question too closely, this is a question of portfolio optimization across different… Well, in this case, the assets are different trend systems so you can do this optimization across different kinds of systems. You could say I’m going to decide how much risk or cash to allocate to my fast-moving averages, my slow-moving averages, maybe to carry as well if you’ve got it in there. You can also do this across instruments. So you could say how much am I going to put into my trading strategy for S&P, how much into corn, how much into sugar, and so on and so forth.

But in terms of the actual mathematics of the problem, it doesn’t actually matter whether you’re allocating to systems, to instruments, or indeed doing some kind of portfolio optimization across any kind of assets. You could do this with the underlying assets themselves. It doesn’t really matter. When you do any kind of portfolio optimization, there’s a few things you want to take into account. One is how risky the assets are. Riskier assets get a lower weight.

Now, the nice advantage of working with a trading system is you would normally set your system up so that all the systems have got the same expected risk; same expected standard deviation. So that problem comes or goes away. But that leaves us with just two of the factors to worry about. The next one is correlation, which Scott talks about. Again, if you’ve got something that’s really diversifying, you might find if you’re say trading, I don’t know, an S&P system, a bund system and a bobl system, well, the latter two are both German bonds, so they’re likely to be very correlated. So you would expect to want to put more of your money into the S&P and less into the two German bond systems.

As Scott said, this is kind of a heuristic rule of thumb in this situation which would be to say, I’m going to put half my capital in the S&P and then split the rest of the other half of my capital up into two German bonds because you can think about saying, that way, then I’d get equal risk to bonds and equal risk to stocks. You’d also have equal risk to the US and equal risk to Germany in this example. It’s not quite as simple as that. There’s a couple of other wrinkles to that, but that’s the basic idea.

The third thing that you need to consider when doing any kind of optimization is performance expected returns, and this is where Scott is running up against a feeling of is this the right thing to do? Now, what he’s actually describing is a methodology that I talked about in my blog but I don’t actually use myself. Basically, it’s a way to essentially adjust these simple portfolio weights according to the backtested Sharpe ratio of the system.

For example, let’s suppose the bund system is just amazing and it completely wipes the floor with both the bobl and the S&P system. Well, instead of giving that 25% of the risk, you might want to increase that, as Scott says, by a factor of 1.5 so you bump it up to about 37%. Obviously, the other two instruments would then reduce their risk. Now, actually, he says it’s elegant.

Well, it’s elegant because unlike a standard portfolio optimization where you take this Sharpe ratio estimate and put into the optimization, most likely, what that would probably do is say you should put everything into the bund, because the bund is amazing, it’s just fantastic. But that’s obviously very, very risky and is assuming that you’ve basically said you can perfectly predict what the performance of your systems will be in the future, which you just can’t do.

So the question is essentially how far you move along a continuum where at the start the continuum, you say I can’t predict the performance of future trend following systems. I can’t predict whether the bund is going to trend or the S&P is going to trend, I just can’t predict it. What’s happened in the past is of no relevance to me. Or you kind of go the naive, full-blown optimization methodology, which would then say, well, yes, put everything in the bund, it’s amazing. It’s great, you don’t need anything else, it’s all you need. You can close down the rest of your strategy, just go with the bund.

The halfway house is to say, well, let’s move some way towards that by putting a bit more into bund but everything. So it’s kind of a heuristic way of doing that. The question becomes, how far along that continuum should you move? It’s not actually a very simple question to answer because it comes down to essentially how well you think you can predict the future performance of a given system, and tied into that is how statistically significant results of that system were in the past. So you’re kind of having to make two sets of assumptions.

One is that the results you got from the past are statistically significant, and secondly, that the past can predict the future. I think that’s a joint decision. Now, for me personally, when I look at trading strategy returns, generally speaking, I don’t actually do this. I don’t actually in reality put more into the bund, having this simple example. The reason for that is exactly as Scott says, I just don’t know what’s going to be trending tomorrow. It might be that tomorrow, the bobl does better than the bund.

If I look at the past, yes, there is evidence that the bund is better, but it’s pretty weak evidence. I never really see strong evidence that one system is better than the other, at least when looking across different instruments. Now, that isn’t necessarily true if you’re doing the optimization for different kinds of trading. For example, if you’re doing something like optimizing over different speeds of trading, you definitely for example want to down-weight a fast trading system if you’re trading something that’s very expensive. You also tend to see that, for example, fast trend following even ignoring costs completely just doesn’t seem to do quite as well as medium and slow trend following.

So that suggests that there is enough evidence there to perhaps down-weight the faster trend following in favor of the medium and slow, and of course, if you’re running a trend following fund, you might want to do that anyway because you don’t want to be fitting into the fast trend following bucket. You want to be staying in the medium to slow standard CTA. So I would say Scott’s right to be concerned. That’s why in practice, although I presented this as a technique on my blog that you could use if you did want to do this kind of portfolio optimization, in practice from my own portfolio, I don’t actually do it. So this is very much an example of do what I do, don’t do what I say.

[00:18:51] Niels

A question that came in from Andrew. He starts by saying, “You probably have enough questions for Rob.” All I can say is we can never have enough questions for Rob of course. “However, if not, here’s a few questions. Thanks for your work. I rarely skip any of the podcasts. I purchased all of Rob’s books and I appreciate how specific you are in your illustrations. Thank you for being accessible to average retail investors like myself. Here are a few questions. First, I’m trend following the trend following ETFs using exponential moving averages such as 16 over 64, 32 over 128, and 5 over 20. Is trend following the trend following ETFs a good idea or a bad idea?” That’s interesting.

[00:19:39] Rob

Just to be clear, what we’re talking about here is an ETF that gives you access to a trend following type strategy. I’m assuming that’s what you’re talking about. Of course, such a fund has been launched or is about to launch in the US.

[00:19:52] Niels

Well, it’s live in the US a couple of days ago.

[00:19:57] Rob

I love the ticker of this fund. It’s TFPN, which I believe stands for trend following plus nothing. Listeners can’t see this, but Niels can, that I’ve put as my screen name on the podcasting platform that we use to record the episode. I put Rob TFPE Carver where of course, the E stands for everything because I am a bit more pragmatic, for example, than some people we might say. It’s always good to see another trend following ETF on the market because they’re a great way of accessing the strategy for people who don’t have both the capital and also the expertise to do it yourself. So good luck to that fund definitely. I probably can’t say anything else because of marketing compliance tight restrictions.

[00:20:49] Niels

We’ll let Jerry do the talking.

[00:20:50] Rob

We’ll let Jerry do the talking of course. He’s a better salesman than either of us. Anyway, so this is a fund that’s designed to capture a trend following strategy. The idea is to trend follow those. So it’s almost like a second order trend following or a meta trend following. For me, I could actually think about this in a slightly different way. Because I actually looked at a related version of this question a few years ago myself, and what I said was something else, which is say, well, let’s suppose I’m invested in a trend following strategy, does it make sense to do something which some people call trend following the account curve?

In other words, if this thing starts to do badly, should I reduce my allocation to this strategy, and then if it does well, should I increase that allocation? If you think about it, that’s also related to the first question that Scott was asking because this is a bit like saying, yeah, I want to put more money into things that have done well. But this time, we’re looking on quite probably relatively short time periods. So we might be looking at the last few months or maybe a year of performance and using that as a decision whether it to up-weight or down-weight.

In fact, if I look at the numbers in Andrew’s question, the exponential moving averages, we’re going to be looking at a few weeks or months performance here with those kinds of numbers to decide whether to scale in and out of the trend following strategy. So what this essentially comes down to is, to get technical for a second, the autocorrelation of returns. Autocorrelation returns just means how likely is it that a negative return is going to be followed by a positive return or followed by another negative return.

This is obviously the phenomenon by which trend following makes money because obviously, if markets are positively autocorrelated over certain time periods, then if the market has gone up last month, well, then if it’s positively correlated, it means it’s going to go up this month as well, then you should go long that market and that’s honestly just a very simple trend following strategy. So many markets do show positive autocorrelation patterns at these two time periods. The question then becomes whether a trend following strategy or equivalently for this question, a trend following ETF would also show a similar pattern. Would it show this positive autocorrelation?

At the risk of being accused of trying to plug my latest book, I do actually look at this in there. I look at the autocorrelation over various periods. What I find is that the evidence is mixed. There’s not really much there. One problem is that people have looked at this in the past, but they’ve tended to look at trend following hedge funds. The problem with trend following hedge funds is the performance fees actually distort the autocorrelation patterns, because obviously, if the fund is doing really well, then performance fees will be coming out of it and reducing its returns.

If it starts doing badly, then there won’t be the performance fees applied and that will make the negative returns look less bad compared to the positive ones. What that means basically is it creates a kind of artificial negative autocorrelation. If you’re investing in the underlying strategy or in an ETF that has a management fee and no performance fee, that effect isn’t there. So you need to abstract away from that as well. Winton, for example, have done some work on this where they’ve looked at the hedge funds which had this problem. Anyway, so I found that there weren’t particularly strong results.

Where I found the strongest result and it still wasn’t that strong were looking at the one-year performance. I should also say as well, I wasn’t doing quite the same thing here. What I was doing here was looking at the switching between a trend following and a carry strategy. Implicitly, what I just talked about is shifting between a trend following ETF and cash is not quite the same thing. But I did a similar analysis just on trend following and got fairly similar results. But I did find that the window where it worked best was on a one-year horizon.

In other words, if you look at performance over the last one year, generally speaking, it makes sense to hold that particular strategy for another year going forward, but it is quite a weak effect. Remember, it’s a year, whereas Andrew is talking about much quicker periods. So I would say probably not. I would personally not do the strategy is my opinion. Niels, I don’t know if you feel differently.

You could take one view, which is that well, we can trend follow anything and it doesn’t really matter what it is. It’s just that I think in this case, there’s probably good reasons why this wouldn’t work as well as trend following something else, but maybe you think differently.

[00:25:27] Niels

No, no, I agree. I think that most people who have been in this industry for a while would conclude that you shouldn’t trend follow a trend follower. Now, I’ve never done any studies on this but I would just say anecdotally, the best time to buy trend follower is when they’re in one of their normal drawdowns. I don’t have any evidence for that other than trend followers seem to come back from their drawdowns and have done so every single time. So that seems to me like a logical entry point rather than buying the latest new high, which of course is where most people will invest because that’s where they get sucked in from the success of trend following.

This is also why some investors if their time horizon is not long enough will end up feeling somewhat disappointed, because they didn’t make a new high necessarily in the next three or six months or year for that matter. So yeah, I agree with you.

[00:26:26] Rob

Yeah, just to say as well, the effect of performance fees is to make that strategy you described even stronger, because obviously if you’re investing in a, let’s say there’s a fund that’s in a 20% drawdown, the next 20% you own is free of performance fees. So you’re kind of getting an extra 20% for free. Whereas if you invested at a high watermark, then obviously you start paying performance fees straightaway. Yeah, I’d say, the other thing to say is that trend following ETFs are a great way to access the asset class relatively easily.

But if you do want to trend follow ETFs, then just go out and get some cheap long-only S&P ETF that you can get for 5 basis points a year because these trend following ETFs, obviously because they’re actively managed, they do have a higher cost. You’re looking at probably 100, 125 basis points, which for a passive ETF would be very expensive. So it’s quite an expensive way potentially to access the trend following class rather than just sitting there and holding on to it. Alternatively, if you do want to get actively involved trading, yeah, do that but do it with much cheaper ETFs and do it with the underlying instruments rather than with the trend following ETFs.

[00:27:43] Niels

Yeah. I just want to clarify one thing just so there’s no misunderstanding. Just because a manager is in a drawdown, it doesn’t mean that the fund you buy necessarily gives you a free ride. It depends on the way they calculate performance fees, because some funds do it per share class and if you get your own share class if you invest, you invest at a new all-time high. But if it’s pooled where it’s just based on the combined NAV of the fund, then you can definitely, as you suggest, get a free ride essentially until the next all-time high.

All right. Andrew continues, “Secondly, in your books, you recommend when using an exponential moving average, use two factors of 4 such as 16/64, 38/28, etc. Yet, many of the studies I see illustrating profitable moving averages use 10 across 100, 10 across 150, or 20 over 120. What is the advantage of factors of 4?”

[00:28:48] Rob

To be honest, a lot of it is down to neatness. So for reasons that are a bit too technical for this podcast, we’re talking here about pairs of exponential-weighted moving averages and an exponential-weighted moving average crossover trading system here, okay? It would be similar with simple moving averages as well. It wouldn’t make much difference. The idea is that your signal is the fast minus the slow, so it would be if the fast was 16, for me, I’d then multiply 16 by 4 and get 64 as my slope. If 64 was my signal, it would be 60-day moving average minus 64-day moving average.

Now, one of the advantages of 4 is that for reasons that are too technical to go into, basically, if I double the lengths of my moving averages, in other words, if I go from say 8/32 to 16/64, if I do that, then the correlation of adjacent moving average crossovers, the correlation of their returns basically has a consistent pattern. In other words, I basically achieve a maximally diversified portfolio. So that means if you were using, say if you were using 10/100, then the other moving averages you should be using are 5/50, 20/200, and so on.

If you were to go tighter than that, then you’d get correlations of like 0.99 between adjacent moving averages so you’d be achieving very little diversification. If you were to go broader than that, then your correlations would be, for me at least, too far apart. So there’s something very neat about doubling things that have a multiple of 4 because you end up with 2/8, because obviously, 4 is 2 squared, you end up with 2/8, 4/16, and so on and so forth. So it’s quite neat. That’s the reason why I went for 4, because to be honest, it doesn’t make a massive difference exactly what you go for as long as it’s around 3, 4, 5.

Actually, another reason I went for 4, and I think I’m going to probably disclose this now because it’s been many years since I worked there and I’m sure they don’t do that anymore. But when I was working at AHL, their magic number was 3. So they would use something times 3, something else times 3. When I came to implement my own system because I was writing about it in a book, I thought, well, I better not use 3 because I might get sued, which seems kind of silly now. Perhaps I was a bit more paranoid than I am now. There’s this 3, it’s got to be a big trade secret. You can imagine, don’t tell anyone about the 3.

[00:31:27] Niels

It’s probably just because there were three founders so they just picked the number three, right? There’s no magic to it.

[00:31:32] Rob

There’s no magic to it at all, definitely. So I thought, well, I’ll use 4, and the advantage of 4 is I know it’s going to be roughly about right because it’s pretty close to 3. I also know that it has this nice property that when you double the horizons, you get the nice stacking of the numbers. But actually, yeah, if you look at this empirically, and you can do it empirically using two different methods. You can either use real data or you can use artificial data, artificial data which you created to have basically trend plus noise in it.

Those kinds of numbers tend to give you the best compromise between a trend filter that is adaptive, in other words, when the trend changes, it changes direction, but also not too noisy. So if you think about the pair of numbers, the fast and the slow, the bigger the ratio between them, so 10 and 100 has got a ratio of 10, that’s going to be very noisy because there’s not really much, that’s going to be very reactive. The fast is going to react quite quickly to changes, but because there’s a huge gap between the numbers, it’ll be a long time before the crossover happens. So it’s expensive because it’s got a small end, but it’s also not very active so it’s a really poor compromise.

Alternatively, if you went too far in either direction and you had a crossover like, let’s be silly, like 99/100. Silly, really, but there we go. Then that would be something that would be very, very reactive, but it would just be noise. Because those two crossovers are highly correlated, they’ll be flipping from day to day. It’s going to be quite slow. It’s going to be actually quite expensive to trade. So it’s about essentially about compromises in those. It’s actually possible to mathematically show, given a certain return distribution, or lots of assumptions, you can actually derive mathematically what the correct ratio would be.

I’ve not done it myself, but I’ve seen other people do it for fun at parties, because that’s the kind of parties I go to. I’m joking. It wasn’t at a party, tt was at a conference. So I would say to people don’t get too tied up about 4. Anywhere between probably between 2 and 6 will work equally well. Whatever you do, don’t go and optimize this number either, because then you’re fitting in and other parameters. So that’s why I always use 4. That’s why I don’t optimize and fit and say I’m going to use 3, 7, and I’m going to use 12, 50, which is very close to 12, 40. Don’t optimize it. By picking one number, I’m effectively reducing the degrees of freedom on my trading system, making it less likely I’m over optimizing.

[00:34:06] Niels

Yeah, and it’s kind of robustness, just picking numbers that are, as you said, not completely fine-tuned. All right. Well, third question from Andrew is, “Using a free back testing system online that only goes back to December 31, 2002, testing against five major ETFs,” and he lists five major ETFs, “I cannot find a single exponential moving average combination outperforming buy and hold over that timeframe. Am I missing something? Is the backtesting timeframe too short? Should the EMAs be outperforming buy and hold or simply reducing drawdowns?”

[00:34:46] Rob

Yeah. Also, just to say, the five ETFs he lists are one is gold, one is Inflation Protected Treasury Securities, one is the NASDAQ. So they’re kind of big markets basically. There’s nothing weird about them. I think we should say that upfront. I always kind of get really nervous when someone starts a sentence with “using a free backtesting system online” because the story normally ends with them saying, well, I’ve found this strategy that makes 10,000% a year, I’m going to be rich, what do you think? So it’s nice almost to hear a story that starts that way but doesn’t end up with that punch line anyway. But I’ve got a lot of questions really.

So one question would be, he says outperformance. Well, how is he measuring outperformance? Is he using just outright returns? Is he using risk-adjusted returns? The other question that isn’t obvious to me from that question from the way he’s described it is, is he running a long-short system or a long-only system, for example? So if he was running a long-only system and if he was running it just measuring returns and performance, I would be not surprised at all if it underperformed the long-only system, because basically, you’re going to have a lot less risk because there’s going to be a lot of times when you are completely out of the market.

Inevitably, there’s going to be times when you’re missing out on big bull runs or something like QQQ or the NASDAQ. So your risk will be low, your drawdowns are going to be low. Your absolute return, yeah, probably will be lower, to be honest. The only time you would definitely outperform a long-only system would be if you had a long-only trading system with a moving average that basically just kept you flat the whole time. In which case, your outperformance would be great because you’d be flat the whole time and the long-only system would just be consistently losing money. But none of these markets are kind of like that. So that’s one question I have to say.

Having said that, I’ve got a couple of hundred markets in my dataset, I could easily find 20 or 30 in which trend following for any moving average hasn’t worked in the last 21 years, very easy. Effectively, what I’m doing there is a kind of reverse data mining. I’m cherry picking to basically move bad rather than good, which is what you normally do. So it would be very easy for me to write. I’ve seen people write blog posts like this saying, oh, trend following is awful. Look, I pulled out these five markets over the last 20 years and it’s terrible, they do really badly. Trend following is awful.

But of course, someone else could equally pick five, find something else, another trading strategy which in reality is hopeless, but I bet you could still find five markets over the last 25 years or 20 years that have done really well in that particular strategy. Again, put them on your blog, say yes, strategy X is amazing. But what we’re doing here is working with a limited data set, both in terms of time period in the last 20 years, but also in terms of market. So I would be surprised if no trend following system works on any of them.

It may be because it’s a free backtesting system that’s not really working very well. It may be due to the way performance has been measured. It may be because it’s long-only. I’ve got so many questions here. But even if all of those questions have the right answer, it could still just be bad luck at the end of the day.

[00:38:08] Niels

Yeah, I think a lot of people who come to trend following, Rob, they will have read a book and probably their first inclination, I think will be ah, this should work on equities. I also think that a lot of people actually promote trend following specifically on equities saying, yeah, this is really great. But I’m not so sure necessarily speaking of that timeframe, the last 20 years or so, certainly the last 15 years or so. I don’t think necessarily, at least if you trade indices, and some people will say, “Well, that’s exactly why you shouldn’t trade the indices.”

Well, again, that’s a matter of opinion. But I do think a lot of people may not feel that that has been a great sector to do trend following because we’ve had these big reversals in the last 12, 15 years. As you say, it tends to take you out and then it takes time before you get back into the long side of the trade. So comparing to buy and hold, I wouldn’t be surprised. That doesn’t sound surprising to me. Now, I know he mentions a few other markets, etc. But just from memory, equities hadn’t been necessarily a great sector for trend following.

[00:39:21] Rob

Yeah, he’s got five ETFs. Three of them are equity indices. So you’ve got Dow Jones one, a global equities one, you’ve got QQQ, which is the NASDAQ, and then there’s this TIPS is Treasury Inflation Linked Securities, and GLD which is which is gold. So that sample was just five markets. Second of all, as you say, it’s heavily biased towards equity indices, which to be fair, it’s not just the last 20 years, equity indices have been, in my book, I look at this equity indices have been the worst sector for trend following out of all of them. But does that mean you shouldn’t trade them at all? That comes back to the question we were asking before. Maybe down-weight them a bit as we discussed earlier.

[00:40:07] Niels

Yeah, this is interesting, because of course, some people love equities and have it as a big portion of their portfolio. This is something I feel that people miss when they get introduced to trend following a lot, certainly if they want to do a DIY, and that is really the importance of true diversification. Because that is really the secret sauce that we talk about, but we don’t really always make it clear that don’t start doing trend following unless you can get very, very decent diversification in your portfolio.

All right. Well, you mentioned markets that’s been advertising 10,000% returns and are sort of little bit out there. Of course, what we’re going to be talking about now is why on earth you’ve gotten yourself, as a crypto skeptic, a self-proclaimed crypto skeptic, you’ve got a gig now, I assume a paid gig to be somewhat of an advisor to a firm, a crypto hedge fund as I understand it.

[00:41:19] Rob

This is my guilty secret.

[00:41:20] Niels

It is. That is really a guilty pleasure, isn’t it?

[00:41:24] Rob

Well, I’m not sure it’s a pleasure or I’d be doing it for free. They are paying me as you say. Yeah, I’m a crypto skeptic and I still am a crypto skeptic. I believe that that crypto is a bad thing, I suppose. I can’t see myself ever, for example, buying crypto on what you might call a cash or a spot basis. I know this is an industry that’s full of crooks and scumbags to be honest. I’ve traded the two CME listed crypto futures for a few years now, Bitcoin and Ethereum. People have said to me, doesn’t that make you a hypocrite? I don’t think so because I trade lots of things. I trade crude oil, for example. I still believe that there’s a climate crisis and we should be using less oil.

I go both long and short crypto as the system wants me to. It’s not that I have a strong opinion either way. I have personal opinion that it’s bad, but I’m perfectly happy to hold the future while it’s going up and make money from it. I don’t see that as a conflict. So you could argue, but what I’m doing now goes beyond that a bit. I’ve known the guys that run this firm for a few years now. I’ll mention them just once. They’re called Stylus Digital. There you go. That’s the promo but I don’t want this to be a promo episode.

Over a few years ago, they came to me a little while ago saying, “Do you do consultancy? We didn’t know you did consultancy.” “Yeah, I do consultancy, but only for people I like, who I know well, and who I trust. So please, don’t start writing and offering me work. It’s not going to happen.” These guys are like, “We’re friends, we know each other well, you trust us, we’re good guys.” I said “Yeah, but you run a crypto hedge fund.”

They said, “Yeah, but we’re a systematic trend followers. We bought all of your books and our system is basically from your book. So we’re about as closely aligned to your views as possible to be.” I said, “Yeah, but you’re a crypto hedge fund.” I said, “Well, okay, I’ll think about it. But one thing I want to be very clear, I’m still going to be a crypto skeptic and you’re still going to see me on Twitter saying all bad things about crypto.” In fact, just this morning, I reposted a thread of somebody criticizing some of the slide decks of a crypto hedge fund back in the day. I haven’t changed my opinion. They’re like, “Yeah, that’s fine. We think our clients will appreciate your independent thoughts.”

So yeah, I’m basically taking the crypto shilling or coin or whatever it is. I’m being paid in fiat currency I must say. I’m not stupid. But yeah, I was just saying to you before we started the podcast that the crypto is something I would like to go away because I think it’s a bad thing environmentally. A lot of people lost a lot of money, a lot of people have been ripped off and it’s encouraging gambling, but it doesn’t look like it is going away anytime soon. It’s kind of intellectually interesting as an asset class and there’s some interesting things about it.

I’m of the opinion that if someone’s going to really trade this, they should be trading it properly and hopefully my advice will help with that. I think also that my next book, dare I say it, it’s going to be at least probably a year and a half before I start writing it because I’m too busy at the moment, but I think my next book will include some stuff about trading crypto purely because I’d like to say to the average on the street if you really must do this crazy thing, here’s how to do it in the safest way possible in a systematic way. So this work is potentially helping me with that as well. So what can I say?

[00:45:14] Niels

Well, you can say a little bit more about trading crypto actually, because one of the topics that you also mentioned, and actually, that’s going to tie into another topic that we’re going to come to later in our conversation today and that is about trading things safely, let’s put it that way. Obviously, I’ve heard a podcast where you and the people you work with were quizzed by someone that clearly came from the crypto space, and you’re right, you certainly have some different opinions, and so on and so forth. But then you raised this question about can it be traded safely.

I imagine when you think about that, you think about yeah, it can be but you have to go to the exchange. That limits the choices down to two futures contracts as far as I can tell. But if I misunderstood that, how do you think about trading crypto safely even if you follow trend following rules?

[00:46:15] Rob

I guess what you’re trying to do is reduce counterparty risk. So there’s a whole separate question about position sizing and leverage, so for example, crypto returns a very fat tailed, and that means you need to think more carefully about vol targeting. That’s a separate issue purely in terms of counterparty risk. Obviously, what you don’t want to do is give all your money to the next Sam Bankman-Fried. You don’t want to do that. You are looking for an exchange that- obviously, a lot of people thought that his exchange was perfectly safe, FTX was a premiere safe exchange.

I’m not going to endorse any particular exchange because I don’t use them myself. But obviously, you’re looking for something that’s definitely domiciled in an onshore jurisdiction, so in the US or in the UK or in Europe, which limits it to a very small number, and ideally, something that has some kind of regulatory approval. Obviously, there’s an exchange in the US which is listed in the US so it’s kind of got approval from that perspective, but simultaneously, the SEC is trying to shut it down.

So the fact is and the honest truth is there is no exchange that would be safe enough for me personally to put my money into and that’s why I only trade with CME with derivative contracts. So it’s a question of degree. So on the one extreme, you’re going to have some very dodgy offshore based fly by night crypto exchange that probably won’t be here next week. On the other extreme, you’ve got the CME. So it just depends on where your personal threshold for counterparty risk sits on that. Now, one thing that’s kind of interesting is that crypto hedge funds that are sensible do both. You could just trade on the CME, but then you’re limiting yourself to two contracts.

As you said just a few minutes ago, diversification is the most important thing in trend following. So just buying two out of however many coins are out there that are liquid and you can trade, there’s probably now 30, 40, 50 coins that are what I would describe as institutionally liquid. So they’re liquid enough that a reasonable sized hedge fund could trade. Going from a universe of 40 or 50 down to 2 is obviously going to reduce your diversification benefits.

Having said that, it’s not going to reduce your diversification benefit that much because all of those 40, 50 coins are actually quite correlated, because they’re all in the same asset class and the asset class is a very tightly knit asset class. So it’s much more like something like US bonds as an asset class where all the instruments have got pretty high correlations of about 90% because they’re all very similar. It’s much less like say, I don’t know, the agricultural complex where you’ve got soybean here and you’ve got lumber here and you’ve got wheat here, and canola and sugar in that the correlations don’t tend to be as high and there we can get a lot of diversification benefit. So the losses from restricting yourself to just those two markets are not as big as they would be normally.

So normally, I would say as a trend follower, if you’ve got the money, definitely trade more than two assets per instrument per asset class. I wouldn’t trade just, I don’t know, wheat and sugar out of the commodities asset class. I want to trade some other things as well. With crypto, there’s less of a benefit in doing that. There’s a loss from not diversifying, but there’s also a gain because you’re making your exposure to counterparty risk lower.

One of the things I’ve been doing is trying to actually quantify that because one thing you could imagine doing is I’m going to trade on this exchange, but I’m going to buy some kind of insurance like a CDS or something like that against this exchange defaulting. If you could do that, what will be the price of that insurance? If I do that, am I then fully compensated for all the risk I’m taking on? Does it then make sense diversify?

It’s an interesting question because what it allows you to do is almost quantify the benefit you’re getting from trading on this potentially dodgy exchange. You could back out from that implied probability of default and say, what I’m saying here is I’m okay with this risk if I believe that the chances of this particular exchange going bust in the next 10 years is 5% or lower, and there is actually academic work that’s looked empirically at that, so I’m getting quite nerdy now. I apologize for that.

[00:50:57] Niels

You sound like a bit of a crypto expert.

[00:51:01] Rob

Not at all. One thing that is interesting about crypto because it’s a relatively new field, there’s a lot of stuff you can take from traditional finance, as I said, from a kind of academic point of view and apply it to crypto and it’s still quite novel and interesting.

[00:51:16] Niels

We are running a little bit out of time and we don’t necessarily have to go into all of these points that you raised. Again, you can comment on as you wish. You say, yes, there are things you can take from traditional finance. Clearly, you’re taking the fact that they trend and there is momentum to be had in these strategies. Some people come to crypto because they’ve been informed that this is a hedge against inflation or this as a hedge against Fiat or something else.

[00:51:44] Rob

I’m shaking my head now.

[00:51:45] Niels

That doesn’t seem to proved out, or that it’s some kind of risk offset, which didn’t work either last year. Probably we don’t even have enough data points in the short history of crypto to say anything about that. But I agree with you. If you’re just taking it as a speculative instrument, yeah, there’s lots of volatility. Actually, that makes it a “fairly liquid market” for trend followers to trade because the more vol, the less we’re going to trade it, so to speak, in terms of number of contracts. So that’s not a bad thing from a trend following point of view. But I do want to make the transition.

You talked about dodgy exchanges. We’ve seen things that have gone wrong in the Bitcoin space, and then you mentioned to me in a in a separate email, let’s talk about the LME, trading LME, and I’m thinking, well, there’s a link there in terms of dodgy behavior. We all know that there was a nickel issue last year I think it was or the year before. Was there something about LME trading you wanted to point out?

[00:52:53] Rob

Yeah, well, it is interesting actually because if you did listen to this podcast I did with these crypto (I was going to use a rude word there but I won’t), these crypto people, one thing they’re quite fond of is taking things that have gone wrong in traditional finance and saying traditional finance is all bad. If you trade on a two transfer exchange, your trades could be canceled. I’m like, well, I can think of two specific examples, the LME copper, and the flash crash in 2010 when some of the more offside trades were cancelled for the CME. I would say that probably amounts to 0.0000000, as many zeros as you like, 1% of all trades that have been done on traditional changes in last 20 years.

Whereas I think I said for all the LME’S faults, you’re not going to say the chief executive of the LME being up in the US court for massive fraud and stealing billions of dollars anytime soon. Whereas with crypto, that’s an ever-present risk sadly, but anyway. It was part of the fact that, as I’ve said earlier, I’m expanding my list of markets and one of the things I’m doing actually is adding on some LME metals, although I’m having to do it in a slightly not dodgy way exactly, not even a crazy way, but basically, as a individual retail investor, you can’t trade on the LME. It’s a restricted exchange.

Accessing those futures is tricky, but at least one broker, I won’t name them, but at least one broker does offer effectively a synthetic version of those contracts. So that’s been one of the big holes in my portfolio has been; actually, ironically, it’s been UK exchanges I’ve been missing. So I’ve been missing a lot of the UK quality things like Brent, for example, gilts, FTSE 100, all those financial exchanges, but also the LME. So that’s why I bring it up. But yeah, it seemed like a good time to remind ourselves that crypto is awful, but every now and then, something goes wrong in traditional finance as well.

[00:55:01] Niels

Absolutely. Speaking of the LME, wasn’t it very recently also that there was like a physical delivery where they realized that it wasn’t the material, I can’t remember which country it was where it was just a bag of stones.

[00:55:15] Rob

Yeah, this is hilarious. Actually, I know that I shouldn’t be endorsing another podcasts, but there was a very good episode of Odd Lots where they talked about this specific issue and about the kinds of stuff you have to do. We trade this stuff. We trade all these physical commodities, giving no thought whatsoever to concerns like storage or delivery or warehousing or anything like that. It’s almost magic. It’s great.

As long as you roll before expiry and don’t have to worry about delivery, you can just exist somewhere. But yeah, somewhere there. I’m trying to think what positions I’m currently holding. But while I’ve got this physical… let me see. I’ve got some soybeans and live cattle. So someone somewhere is looking after some cows for me. Presumably, there’s not like a massive risk.

[00:56:06] Niels

So when you’re short, it’s the cows looking after the farmer. Is that how it works?

[00:56:11] Rob

I have no idea how it works. This is the beauty of it. Presumably, there’s no danger anytime soon of someone replacing a cow with a bag of stones. But maybe they’ll replace the cow with a less good cow and there’ll be a problem. I don’t know.

[00:56:23] Niels

It could be a bull actually.

[00:56:27] Rob

Well, then I might end up with more cows, which would be great if I was long, but maybe not as good if I was short. But yeah, the whole physical commodity space is kind of fascinating. I do feel like when I’m trading financial markets, like you Niels, I come from fixed income market so I understand the fixed income market. Obviously, everyone understands how things like stocks and FX work. I used to trade options. So things like the VIX, I kind of understand those. But you start asking me questions about live cattle and about wheat and grain and corn and sugar, and I’m like, well, I’m trading.

Actually, I probably now know more about the mechanics of Bitcoin and Ethereum than I do about pretty much the whole of the commodity complex. You’d think metals would be quite straightforward because unlike cows, you don’t have to feed them. But yeah, there’s apparently this ever-present danger that you go in the warehouse to get your nickel, tin or whatever, and you look in the bag, and it’s not there. It’s gone.

[00:57:34] Niels

Yep, such is the world of finance of course, surprising from time to time. Now, Rob, I think people have realized it’s Friday afternoon, both of you are a little bit loose on the tongue here. But before we do leave the audience to have a nice hot weekend, at least if they’re in Europe, I’ll say two things. One is I think everyone should go and read Howard Mark’s latest memo or listen to it on their podcast. It’s called Taking The Temperature. I’ve thoroughly enjoyed it.

Actually, if you listen carefully, there are so many lessons even though he has nothing to do with trend following. There’s so many lessons that go straight back to the trend following mindset. So I just enjoyed that.

Now, not that it’s going to take more than a couple of minutes, but there was a new piece of, I don’t know if I would call it research, but there was a paper out from Man Group. They always do some good stuff in terms of reminding people of certain attributes of what we do. The latest one, I liked the title, A Different Point of Skew, which of course is very nicely put. In terms of the essence of this paper, the reminder they want us all to think about Rob, what’s your takeaway before we wrap up?

[00:58:46] Rob

You used the word reminder because of course we know this stuff, and probably a lot of listeners to our podcast to do as well but not necessarily everybody so it’s worth saying. Basically, what they’re pointing out is that skew is different for different assets depending on the time period you measure it. So if you look at say equities, they’re negatively skewed, and actually, they’re negatively skewed at most time horizons. So they look at a week and a month, but the same would be true of days, the same would be true of probably even a year I would say actually.

But trend following is not like that. Trend following actually, the skewness basically improves the less frequently you look at them. So in this particular analysis they do, the weekly skewness actually is as bad as it is for equities. It’s minus 0.7. But when they get to monthly, the equity skew MSCI World is down to minus 1. For trends, it’s up to plus 1 which is obviously the opposite end of the spectrum. The reason for that is basically the following.

If you look at very short time periods for trend following strategies, assuming they’re not trading that quickly as well, so they got an average holding period that stops super quick, then effectively, what you’re seeing when you look at them, when you look at the skew, is you’re going to see the skew of the typical portfolio of assets they’re holding. So that means if they’re typically long negative skewed assets, then their returns will themselves be on a daily basis typically negatively skewed.

Now obviously, generally speaking, a trend following portfolio is going to be long a lot of things that have negative skew like equities. On average over a long period of time, it’s going to be for example short volatility futures, either VIX and the base stocks, which are positively skewed. Short position means negatively skewed so it’s not surprising that on average, you get negative skew. Not every asset is like that but that’s kind of a rule of thumb.

Once you move on to a slower time frequency, of course, what’s began to happen then is that the long-only equity strategy is still holding all of its equities so it skew profile is unaffected. The trend following strategy will start cutting losses in things. It’s this process of cutting losses either with a stop loss or with some kind of trend following filter that produces a positive skew. But that doesn’t manifest itself until you’ve got to a time period that’s commensurate with the time period of the strategy you’re holding. If you go all the way out to a year, you’ll see very strong positive skew in trend following strategies.

Just looking at my own returns, I think my worst ever year, I was down 8%, so you’ll see small losses, and then in a good year, I’ll be up 25%, 30%. So that is quite a strong positive skew when you move out to really long time periods. So it’s a nice reminder of that. They have a nice little case study where they look at the returns around reversals as well, so the last sharp reversal we had with our Silicon Valley Bank thing that happened and that caused a sharp reversal in a lot of assets which tended to be correlated with trend following positions. So there was a bit of a sharp drawdown in CTA performance then.

They show that typically, things tend to improve after these sharp drawdowns. That’s why that kind of sharp drawdown that would contribute to a negative skew in a short time period, when you move to a longer time period, because we recover from those, they disappear and then you end up with a much nicer skewness profile. So that’s a nice little piece, not very long, quite easy to read, written by Graham Robertson, who is a colleague of mine someone I know actually.

They kind of finish saying, can we improve skewness? Well, we can. But we’d probably need to, for example, speed up our strategy because faster trend following for the reasons I’ve described will have better daily skewness. But it’s got increased costs, as we talked about at some point in this podcast, it tends not to be as effective as slower trend following. It’s a different kind of strategy really, so you wouldn’t necessarily want to do that.

[01:03:00] Niels

If you’re a long-term investor, you really don’t need to worry about daily skewness anyway.

[01:03:02] Rob

You really don’t. Absolutely not. It’s the same with drawdowns. My long-only portfolio, I look at it very rarely. I try look at it only once a year, in practice more often than that. It looks so great, because it’s really not very volatile at all. Of course, that’s because I’m only looking at once a year, whereas my trend following strategy because it’s trading leveraged futures, I tend to look at the P&L more frequently, and therefore, it appears more volatile. But this is private equity, right?

[01:03:31] Niels

I was just going to say, Rob, you are coming out as a private equity investor and a crypto analyst.

[01:03:38] Rob

Just to say, there’s an interesting thing in the UK at the moment. The government is trying to push pension schemes to hold more private equity. Indeed, I’ve actually seen a piece of this as well. Yes, this will increase returns by… whoa, hang on a second. If that was an individual investment firm making that kind of claim, then the FCA, the financial regulator would be all over them, but the government marketing BS apparently is fine.

So yeah, I’m fortunately not going to be one of those people whose pension gets forced into this. But actually, I might be because I’ve got a university pension scheme which I don’t have control over. So maybe I’ll write to my MP, my Member of Parliament and see if I can lobby against this crazy, crazy attempt to, I don’t know.

[01:04:29] Niels

Well, now that you’ve moved into crypto, nothing really would surprise me. So even if you move into politics just to get this overturned, Rob, I would not be surprised.

[01:04:38] Rob

Everything you thought you knew about me is gone. How are we doing in the Tour de France, Niels, before we wrap up?

[01:04:44] Niels

Actually, I have to say, I don’t know the exact position right now, but I will be checking it once we’ve recorded.

[01:04:52] Rob

Yeah, that’s good. We don’t want people to listen to this and get spoilers about the Tour de France if they are watching that on catch up as well.

[01:05:01] Niels

Anyways, when they listen to this, it’ll be old news anyways. All right. Well, Rob, you’re off on your extended holiday as usual at this time of year. I hope it’s going to be a great one for you of course, and we can’t wait to have you back in a couple of months of course with lots of new ideas and thoughts and whatever experiences.

Now, if you enjoy these conversations, why don’t you head over to your favorite podcast platform, leave a rating and review on iTunes, Spotify, Amazon, wherever you get your podcasts. We so much appreciate those ratings and reviews.

Next week, I’m joined by Andrew for another fun and insightful conversation about CTA ETFs no doubt and what’s been going on in this space. We already alluded to it, another new exciting entrant this week. So make sure you send us your questions, info@toptradersunplugged.com is where you should send them.

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

[01:06:05] 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.