← Top Traders Unplugged
Loading summary
Transcript84 lines
- [00:01]
A
Welcome to top traders Unplugged in markets. Success doesn't come from predicting what happens next. It comes from being prepared for what you can't predict. In each episode we go deep with some of the world's most thoughtful minds in investing, economics and beyond to understand how they think, how they prepare and how they decide and the experiences that shaped how they see the world. No noise, no no shortcuts. Just real conversations to help you think better and invest with confidence.
- [00:36]
B
Welcome or welcome back to this week's edition of the Systematic Investor series with yor Git and I, Nils Caster Blast. Where each week we take the pulse of the global market through the lens of a rules based investor. Y It is wonderful to have you back this week. I hope you're doing well despite the hmm, challenging result in the football last night.
- [00:56]
C
I've got to say as an English supporter, we're kind of used to that behavior. It's very much trend following.
- [01:03]
B
Yeah, that is true. Unfortunately it's one of those trends you don't really want to follow nor set.
- [01:08]
C
No, absolutely. You become very emotionally involved and invested throughout the tournament only for counter trend to disappoint you right at the end.
- [01:17]
B
Yeah, yeah. Well, at least we now have a final with a European team and a South American team. So we get a little bit of diversification in terms of style perhaps.
- [01:25]
C
Absolutely. Diversification is very important.
- [01:28]
B
Good stuff. All right. We've got a pretty full plate today. Great lineup of topics. Thank you so much for bringing them along today. We'll be diving into all of that in a few minutes but as usual I'm always curious to find out what, what's been on your radar the last few weeks or few days for that matter.
- [01:48]
C
Thank you. So the thing that has been on my radar are actually new products coming to market. So we've seen the C CME has announced that they are going to launch 50 odd leading US stocks as single stock futures on the CME which I think is really, really exciting. And at the same time Kalshi is announcing forward curves for compute power for the cost of compute power. So that's like a different, a different type of commodity which we've never seen before which is like using various chips of Nvidia.
- [02:25]
B
Yeah.
- [02:25]
C
And that's really exciting to me. But I think it brings together one aspect that people kind of normally as a mathematician you kind of ignore it. But when you're a practitioner, one of the things that is really important is actually thinking about like the, the tax implications or like various, like Practical implications of trading is. So I've spent, I've spent the last week figuring out what would be the tax implication of trading single stock futures because, you know, the, the treatment of the dividend, the treatment of the price rise, that's actually quite important. It's, it's very clear for, for like index futures, everybody knows what, what it is. For single stocks, less so. And for Kalshi, it's even more complicated. So the, the, the tax status, you know, whatever it is, it's, it's still unclear. And I think if we want to see growth and innovation, you know, it's the practical things which often are stopping you from getting involved in the market which, where you kind of want to.
- [03:25]
B
Yeah, I mean, single stock futures, there is definitely a little bit of a debate among CTA practitioners for sure. And I don't know that that many of our colleagues have embraced them. I do know that some have embraced them and, and feel very strongly about the advantage of trading single stock futures rather than just trading the indices. But at least it gives more choice, I guess. But I will be interested in seeing what the adoption rate, at least among what I consider kind of the CTA world. I'm sure it'll be adopted by the wider hedge fund audience. But yeah, very exciting. And there seems to be a lot of innovation going on in terms of futures and markets almost to a point where you think, well, do we really need another one of these? But maybe we do.
- [04:19]
C
So it's not necessarily just for CTAs, but just the ability to short stock. Right. It's not something that it's very easy for. It's very easy for a retail investor. Right. But if you have a future contract, then it suddenly becomes possible because you don't have to worry about shorting and borrowing and that, that, that, that aspect of it. So I think, I think, actually I'm kind of, I think that will be very attractive even not for, for CTAs, but I think it will be attractive to CTAs as well. But also for the, the audience in general, the ability to short specific stocks will be useful.
- [04:57]
B
Yeah, I mean it's usually a good idea to give people more choice. Although I will say, and that's not the topic for our discussion today, so I'm not, I'm not inviting to, for a big debate here, but I will say with some of the new stuff that's coming out and where you give, especially you have the focus on retail and I'm just thinking is this really a good idea to give these quote Unquote financial instruments to people who do not do this as their profession. I'm not so sure. And it's not because, oh, let's just keep it for the professionals and so that we can make more money. That's really not what I mean. But being short a stock can be know, lethal because it's unlimited upside. Sorry, unlimited downside if you're short and, and, and, and there's a takeover or whatever that might be. So all I'm just saying is yeah,
- [05:49]
C
great, it's, it's, it's no 100%. I mean, I think, I think it's, it's a really interesting discussion, by the way, about the retail investors and I think we'll come to it when we, when we look at certain observations that CFM has made. But, but from our perspective, we look for diversity and we look for, you know, new risk factors and you know, compute powers in AI is, is something which is very much of interest and that's why I think both CME and CALI are looking at it and single stocks is, is, is, is again potentially a way of, to diversify your portfolio and, and, and it's interesting to us as a result of that.
- [06:30]
B
Yeah. I'll try and tie in some of the things you've mentioned in my observations, although it may fail spectacularly. But when it comes to kind
- [06:41]
A
and
- [06:41]
B
all of that on my radar this week was a very energetic and, and, and powerful experience which was, and, and there might be a lot of people who won't share my taste in music, but I did go this week to watch the, the tour of Bruno Mars live in Milan. And all I can say is that if people have not experienced him live, it was quite, quite an experience. And the energy level for two hours straight was amazing. Even, even through the fact that it was 35 degrees Celsius while he was doing all of that. But it was, it was an amazing evening and of course special to share it with your, with your family for sure. The next thing I wanted to mention, not entirely sure where I saw this could have been on Twitter, but there was a. I think it's, it's a company called Sins Capital Macro Terminal. It was a, it was a. What do you call it? Like, not a meme, but a. Anyway, one of these illustrations they had where they had put in rank different hedge fund strategies ranked by blowup forced liquid liquidation risk.
- [07:57]
C
Oh yeah, I, I saw the, saw that.
- [07:59]
B
But it was kind of interesting. So, so the worst one, the nuclear one was short volume tail risk premia Short gamma, that's kind of the one they classify as having the highest blow up likelihood. Then we have a lot of other things. But then of course my eye went straight to looking for CTA managed futures. CTA managed futures came in as number 15 out of 17 with a green color lower it says, it said yeah so really on the lower side of blow up risk and I. Which I completely agree with.
- [08:40]
C
Yeah, completely understandable. Yeah yeah.
- [08:43]
B
So volume targeted diversified macro was the second lowest and then long only hedge fund like equity, whatever that means. I wonder if they meant private equity.
- [08:56]
C
No, I think, I think that was, I think private equity. I can't remember exactly but certainly from the way people operate long only equity less likely to blow cta. Anybody who manages their risks of CTA and volatility driven macros is just. We are more defensive, we are likely to have upside rather than downside. So the, the blowup is, is, is better. I mean it doesn't mean that when you have a position and, and the world blow ups you, you, you're immediately protected. Right. We like second respons but certainly being short Vol is kind of dangerous especially when the world blows up.
- [09:34]
B
Yeah, I mean long only hedge fund. I mean for those who are old enough to remember, I mean if you're a long only NASDAQ in 2001 ish, you would have been down 85% at some point. So to say that nothing can happen by being long only even in an index, it might be stretching a little bit. Speaking of risk, couldn't help notice that IBM lost 69 billion of market value in one day. That's about 25% in one day. Looking at the price of that stock, it traded on 2nd of June only about 6 weeks ago at 332. That's an all time high as far as I'm aware. And earlier this week we were down to 213. That's 1/3 of the the value in only about six weeks. Which again goes to my point. Maybe you don't need to give people too many instruments where they can add additional leverage or risk. When you see what happens just with individual stocks outright in terms of how much they can really move in a short space of time. I mean 25% in a day.
- [10:45]
C
Yeah, that's a big move. I think there are lots of questions about AI at the moment and about the capex actually the capital expenditure and the reason for that is really we are spending quite a lot of money on chips and Nvidia is doing very well out of it. But These chips don't last forever. You know, I mean in the, in the, in the boom and Bust, you mentioned 2001, there was a lot of boom. But you know, the cables that were laid in that time on that money, on that capex have served us very well for the next 20 years, 25 years. But all this money which is being thrown at AI is thrown at chips that will, you know, last maybe for two or three years and then they become old and obsolete. So there's a lot of capex which if it does go wrong and AI doesn't materialize itself, basically there's going to be zero recovery. You're not getting anything out of it.
- [11:40]
B
On that note, I actually saw an interview that was posted on X with Jensen Huang from Nvidia and I will say, not that I have watched much of his stuff, obviously I'm not familiar with him, but he is very impressive actually. Very well argument, you know, he. Argument very well for, you know, why he didn't think necessarily they meant that everybody's going to lose their job. But he did say you might lose your job to someone using AI if you don't educate yourself and get to, you know, know about it. And he might, he might be right on that. But anyways, we've got an interesting conversation he did at, I think it was at one of these venture capital companies could have been Andreessen Horowitz in, in Silicon Valley where he was invited to do like a fireside talk.
- [12:28]
C
100%.
- [12:29]
B
Yeah.
- [12:30]
C
Because you know, people are not necessarily hiring juniors, but they hire juniors because they know how to use it.
- [12:36]
B
Yes.
- [12:37]
C
And you know, us, all us oldies, we need to adapt or you know, to survive.
- [12:42]
B
Yes, no, I'm, I completely agree, completely agree. We also need to adapt in the world of trend following. And July is one another month of a little bit of adapt. As far as I can tell, you know, it's somewhat stable, but I did see some managers with a little bit more downside pressure. Pressure than, than elsewhere. And of course it's, it's, it might feel a little bit more stable than June but you know, there's still a lot of noise coming from the geopolitical scene and so we do see a little bit of a negative bias overall. I mean, oil risk, of course, that's back now with the ceasefire no longer being in place and all the headlines and. Yeah, I mean, could be a will might be interesting. Let me put it like this to see whether our space got a little bit divided in terms of whether we got caught or who got caught short in energy when they sold off and then we got this sudden move back up again. So I think time frame or speed as we like to refer to may have had a little bit to say this month more than usual. But also of course we've got tech stocks selling off in the U.S. japanese, some of the Japanese markets are selling off this month as well. And we also see a little bit of directional change in things like Canadian dollar as far as I could tell. So yeah, what's your observation so far?
- [14:12]
C
No, I think July has been counter trend in general both you know, in fixed income certainly and then FX as you as you mentioned there's some idiosyncratic risk in Japanese bonds. There's a question of repatriation of FX back to Japan. But also I think in commodity you're making exactly the right observation and it was really much on the cusp. You know some people would have been, would have been short oil already and would have been caught out and some people would have been still, if you're a very slow trader, would have been still potentially long and then benefited from this. So I think that we're going to get a little bit of dispersion in, in commodities this, this, this month.
- [14:57]
B
Yeah, yeah, yeah. So it continues to be an interesting period as it always is. My trend barometer kind of reflects that. It's, it ended yesterday at 45. That's a completely neutral reading and I would say it's fairly much in line with what we see in terms of the performance numbers. So these numbers are as of Tuesday evening as they only update later on today for Wednesday. But beta 50 down 3 basis points are pretty flat, still up 8 and a half for the year. SoC gen CD index down 38 basis points in July, still up about 9% for the year. SoC gen trend down 33 basis points up about 8.76% for the year. And the Short Term Traders Index interestingly enough as we will gonna be talking about short term a little later and that's hurting a little bit more. Down 1.12% in July, still up 4% for the year. But that is a lower volume index. So losing more than the trend index means that it's probably under a little bit more pressure this month. MSCI World in the traditional world is up 1.1% so far in July as of last night up 11.15. We have the S&P US aggregate bond index that's down 43 basis points but still up 41 basis points for the year. And then the S&P 500 up 1% for the month of July, up 11.32%, so doing pretty well overall. All right, so Marcus wrote in a couple of weeks ago, but we did not get to it until today. And Marcus writes Thank you for the fantastic podcast podcast. I've been listening for about three years and I have learned a tremendous amount, so we appreciate that feedback. Of course. My question specifically concerns systematic trend following in non financial futures, where different contract months can represent meaningfully different economic exposures due to seasonality, storage, carry and delivery dynamics. Most trend following implementations appear to generate signals from a continuous front month future series and then preserve exposure mechanically through contract rolls. However, I've never heard the more fundamental modeling question discussed. What return process are we actually trying to forecast? Are we trying to model the underlying spot process, the return process of the currently tradable futures contract, or the return process of a continuously rolled investment strategy? In markets such as net gas and agricultural commodities, neighboring expiries can have materially different behavior and even different risk characteristics. If that's true, what is the theoretical or empirical justification for treating the continuous one month series as the main object for signal generation? Has anyone tested generating signals solely on the history of the currently selected tradable contract and revalidating positions when changing contracts, rather than inherently change inheriting exposure mechanically through the role? I'm not asking about proprietary implementation details. I'm generally interested in the modeling philosophy behind systematic trend following and whether this question has been explored in the literature or by practitioners. Many thanks for everything you do, and I hope you find the question interesting. Actually I do, Markus, so and I have the perfect guest to answer it for you, I hope. And therefore Yoav, over to you.
- [18:33]
C
I've got to say, that was such a detailed and actually knowledgeable question. The guy like Marcus literally answered the question himself.
- [18:40]
B
Okay?
- [18:41]
C
But he's not going to be surprised to know that yes, we have looked at all of these, okay? And within our company we have a commodity fund that actually does trade the full set of sort of the full curve, so to speak. And I think the question actually makes a lot of sense because unlike normal assets which stay, which persist, S and P doesn't disappear month to month in commodities. You consume the commodity and then like it's a new commodity a month later or a few months later, which like the only thing which is connecting potentially different months is storage, right? So if you have inventories in wheat, then you can connect a little bit the behavior of this month to next month. But certainly there is a lot More variety in terms of what are the risk factors that affect different months, okay. And when we look at the front month inventories, for example, is really important aspect of what's happening. And the other thing which is funnily enough is important is actually the composition of the people playing in the market. So you would find that in the front contract of oil you will see a lot of CTAs, a lot of market makers, a lot of retailers playing a little bit of a game in oil futures, oil options actually. And then as you move out to later contracts, you will find that the people who participate are the people who care about genuine production and genuine consumption. So if you go out, you know, three years out into the oil curve, it is mainly dominated by physical players who really want to hedge their production. So, and they want to sell it or they want to buy because they are consumers of oil, okay? So the price dynamics at that point is about long term supply and demand, cost of production, that sort of thing. You know, if you look at cotton, the back of like a year, or wheat or any of the agriculturals, then a year out, we are talking about a new crop, okay? So the, and that is actually very different from, so that the front of the market is really dominated by demand because we've already produced, so to speak. We kind of know, we know very early on throughout the year like what is going to be the crop, the size of the crop. What can surprise us is the demand shocks as you go to further out on the curve. Suddenly you're moving to the new crop. You really don't know what is going to be the supply. So you are actually, you're actually more and more worried about whether there's going to be a plague, what is going to be the weather, it's going to be different risk factors, okay? So I'm talking a lot about various risk factors. It's really exciting. But you have to ask yourself the question about like what is how much diversifying risk I have When CTA start looking at what to trade, the first question is to say, you know, let's capture 80% of the risk with just trading the front contract. And it's very useful. It's very useful because the front contract is super liquid and trading costs are lowest, okay? And lots of people are trading it. And that means that there is actually good, good volumes to trade. It's very easy to roll the position. So it's very convenient to trade the front market, the front contract. And you see, that's why most of the liquidity is There and then the question is, what is relevant to this market, to this contract? I'm now trading the June contract. What is more relevant for me in terms of generating the signal? Is it the history of the June contract or is it the history of the front contracts? Okay, and these two answers are both, both these answers are valid actually. And to some extent the industry has made a choice that we are looking at essentially, most of the industry is looking at essentially the history of the first contract. And there's a good reason for that because as you approach the front, it is all about short term supply and demand. It is about inventories. So the front contract, in some sense the history of the May contract, last month, when it was the front contract, is actually more relevant to the June contract. And we see that, for example, as you approach expiry, the volatility of the June contract that maybe have been less volatile a month ago suddenly goes to match the current front contract. Sometimes it does make sense to separate them. So you're talking about natural gas. That's a really great example where summer and winter contracts are very different and the volatility is different. So when you roll the position from one to another, you're actually rolling potentially from one contract, which is more volatile, to another, which is less volatile, or vice versa. And at that point actually you have a decision to make which do you keep the volatility of the individual contract or do you keep the roll contract? And it's not, it's not that we don't do it. A lot of the CTA is just do the, the front contract. And that's actually perfectly valid sometimes in very specific contracts, in very specific markets, when there is obviously a material difference. And I think natural gas is an example. You know, for example, we would calculate volatility on a specific contract or we would make the adjustment to when you roll the position in order to manage the risk slightly better. But that's not, that's not a, you know, that's, there's no real truth here. All of this. If you look at the. How much risk. What's the, how much risk is explained by what we do already with just with the front contract. We explained in, in financial universe, like 95, 90 to 99% of the risk in commodities less so. So sometimes if you want to capture a different risk factor, you really have to trade further out of the curve. I think the one thing which is not trend following is, is other strategies like carry or trying to calculate the roll yield in commodities. There you really have to be aware of the seasonality. Like if you're trying to calculate a carry in natural gas between the June and the December contract. Good luck to you. It's like the two contracts are just completely, you know, there isn't any connection between them in, in many ways. So understanding roll yield and convenience yield and stuff like that in commodities, you really have to pay a lot more attention to the specific of each individual contract.
- [25:05]
B
Super great stuff you have and great question, Markus. As I'm sure people can tell. I really did have the perfect guest to answer this question. So thanks for that. You have. All right. You already alluded to it. We are going to discuss a blog post, link it to a paper from our friends over at CFM in Paris written by two different group of people. The first one, the blog post, one of the names that is one of the authors is actually Philip Seeger who's been a guest on the podcast. And the other, the actual paper which we touched on last week with Alan. One of the authors there is Jean Philippe Bouchot who has also been a guest on the, on the show. So, so it'll be good to dive in and see how these two things link together to. To some extent. So again I'm going to head over to you, Yoav, to, to gently take us through these somewhat. You know, it's not the easiest stuff to, to get your head around. So. So let's break it down.
- [26:18]
C
I'll do my best. I'll do my best. But I really love, I really love this, this blog and it's not trying to make any huge statement. It's really very descriptive. It talks about the evolution that we have seen in equity markets over the last 20 years or so.
- [26:35]
B
Yeah. I should just mention. Sorry to interrupt. I should mention that the blog post that people are looking for it is called 7 degrees of market Structure. So just to make it clear.
- [26:44]
C
So it's a wonderful piece of descriptive information and it makes seven observations about things that have changed in the industry. Industry. First of all, we've seen a lot of democratization of the market and that's part of a lot of money being injected. For example, during COVID into the, into the retailers and the retail doesn't know like it's got lots of money and, and it is not. It's not busy working at work. So what is it going to do? Retail investors started flooding, started flooding the equity market and that's, that's really quite exciting. And the other aspect that they talk about is the rival of the rise of systematic investment because of course you can. Now if you have a broker account which has an API, you can literally create your own systematic trading and trade from home. And that's, that's absolutely wonderful. Okay. But the object observation that I really found interesting were they made three different observations which I think are actually related to each other. And I want to speak a little bit about that. That so, so the one thing that they make an observation is about the options explosion. We've seen a huge rise in the volumes of what's called zero day to expiry options in the index. So and like normally when I was a little lad and I first entered the financial market, you would buy, you would like S and P options were like a monthly affair, okay, A month, a monthly expiry affair and then they became weekly I think in 2012. Now they're like you can get basically an expiry every day. And what you've seen is a huge increase in the volume of trading and funnily enough driven by retail. We were talking about short volume. A lot of the retail is about selling at the money straddle, so both a put and a call at the money and basically harvesting what's known as the variance risk premium. And we've seen another change in the market that we've seen over the last couple of years which is that the market seems to look like they are becoming more idiosyncratic. And the way that we measure this we look at what's called the dispersion index, which is. Well, actually looks at the implied correlation that we can get from how much volatility do we have in each individual stock stock in the index. We can add all the variance of that and then we can subtract the variance of the VIX the the index and that number is becoming very big. So there's a lot more variance in the idiosyncratic side than there is on the VIX side on the index side. And that's, that's been a, it's been actually a global phenomenon not just in, not just in the S P but in pretty much most equity markets. If you look at the S and
- [29:33]
B
P
- [29:35]
C
sort of sheet where they track it, there is a, they've got a dashboard tracking a correlation across multiple equity indices. It's an correlation is at an all time low, I. E. Idiosyncratic risk is at an all time high. And the final observation that they show is actually the behavior of the index becoming more mean, reverting and they call it like the buy the dip psychology, historically buying the deep was mistake. You, you, you bought the deep and then you, you, you were like catching the knife, the falling knife. But over the last, over the last 10 years, buying the deep actually works much better. And if you look forward a month later, you would have made back the money. And we've seen that, we've seen that in the Trump Liberation Day. We've seen that in, in the equity market recently with the H. So seems to work, seems to work. And that's, that's really nice, really nice observations and I kind of want to link them together a little bit because that kind of offers the mechanism, mechanism to understand what's going on. So with a little bit about the options market, you know, normally when you, a retailer or when you or I trade against and, and buy the buy or sell an option in the index futures index index futures options, then we are trading against the market maker. Somebody like Morgan Stanley or JP Morgan or Goldman Sachs. They market make in the options market and then what they do is they go out and hedge their position. So it's not like a naked position for them is they, they basically look at the delta of the options that they are holding and they go and how to edge it. And in the past when you used to trade monthly options, okay, then you know, today I would buy, I would trade a stock like if I'm a retailer, I'm a fund manager and I trade at 100. Tomorrow I might be trading at 105 because the market has moved. And a day later I'm trading at 120. And the, each time that I'm trading an option, the market maker would take the other side. Then we'll put on a hedge and that hedge will change slowly because there's not a lot of gamma. Every time you need to change your, your hedging is when the delta of that option changes. So that's, we measure it by looking at gamma, the second derivative of the price. So by the time the gamma has got a really interesting behavior because it's very spread out when you are far from expiry. As you reach expiry, Gamma gets concentrated at the money. Okay? So when you are near the strike, when the price is near the strike, then hedging becomes very expensive. As the price moves around, suddenly your delta changes a lot because the gamma is high and you need to hedge more often. And that thing creates trading pressure. With the zero day to expiry. Retail has been selling options at the money near expiry, okay, Literally at expiry. And that means that creates a very Very strong positive gamma for the market makers who are holding the position. Okay, so that, what does that mean? Right. Unlike in the past where all the strikes were spread out and very few of the strikes were at the money near expiry, nowadays there is a concentration of Ghana at the money. And that means that what happens if the price starts going up? The market maker is like long a lot of delta and they hedge it with being shorting the future. They enter the market to short the future. So what does that do to the price? That reduces the price so that pulls back the price back to the strike. Similarly, if the price starts dropping, the the the market makers will have a negative delta which they need to hedge out and they start buying the future essentially to compensate. Okay. So to me that's, that's kind of anti trend, okay. In a way that you know, when, when, when we, we could. We think of CTAs as somebody who buys straddles, okay. And what we've seen a flood of retail people who are selling straddles and money. Now you know, as you pointed out, selling Vol is very likely to blow up. But you know that's retail for you. Okay? They are just happy to collect the coupons. But the effect, the net effect of that is that you having a counter trend, a huge counter trend player in the market. And we understand the mechanism for that. Every day, every day a new counter trend because it's a new zero day to expiry option coming to the market and is playing anti trend game in the equity market. So the effect that we saw, we see out of that is the auto correlation of the internal like intraday returns on one day returns in equities have turned negative over time. Okay. And that really, that really reduces the overall volatility of the index. Both the implied volatility and the realized volatility. Because every time the price goes up, the retail is basically causing the market makers to sell back and to reduce the volatility of the index. Okay? So we've seen the index volatility to be very contained. Single stock options, Single stock options or single stocks, they are a lot less affected by this process. Okay. So when you measure the implied dispersion, which is basically the single stock variance less the index variance, what you see is that is actually at an all time high. So what they are describing is essentially a reflection of that process. The options explosion causing mean reversion, causing the buy to dip psychology and also causing the amount of risk available in the index becoming lower versus the amount of index of Risk available in the single stock. Right. So that's actually relating to our discussion earlier. Why might one look at single stock futures rather than index futures? And the answer is that there's now more risk available there's in the, in the single stock versus the, the index than it ever has been because the index has been depressed in terms of the volatility. So that's a really, really interesting paper and it's a, I think it's, it's, it's a really, really interesting observation and I think it is related a little bit to the, as a possible explanation to what another CFM paper, another great paper is trending still your friend that you discussed with Alan.
- [36:51]
B
Just one question for you before we move on to that just, and this might be a very naive question, but if successful in launching all these futures on single stocks, would you expect that we would get then because it becomes easier to trade against the momentum, so to speak, go short and all that? Would that also, would that reduce the dispersion among single stocks and actually make it look more like the index itself?
- [37:20]
C
No, I think that the mechanism that causes it is more the options. It's more the.
- [37:24]
B
But if you have futures on single stocks, I mean you probably already have the options so.
- [37:30]
C
Well, no, that's the thing. So there are options already, there are already options on single stock but there are a lot less liquid. So even if you look at the futures, right, the CME is not launching futures on all 500 stocks.
- [37:42]
B
Right.
- [37:43]
C
But only on the top 50 which are very, very liquid. So each, each, each dynamic is slightly different. I would, I would expect most, you know, we've seen growth of the option market in equities. Then we've seen a little bit in like in bond futures we've seen that. So we've seen again moving to weekly options at different days, expiry a little bit in oil and stuff like that. So, so basically the, the, the short term implication to, to trend following is actually quite interesting. But it's, I think it's, that's, that's not going to be the driver. I think what's driving the, the single stock futures is more to do with the fact that retails want to short.
- [38:27]
B
Right.
- [38:27]
C
I think everybody is kind of calling the, the top of the, the AI, not the top of the market but the top of the valuation. From a valuation perspective, the S and P is very, is very highly valued at the moment and that gives a very easy mechanism for people to be able to, to short and that's, that's I think where it's coming from. Okay, but I don't think that's, that's, it doesn't affect CTAs that much.
- [38:58]
B
All right, well, you already mentioned it. We're heading back to the paper Alan and I just touched on last week, which is the microstructural account of the demise of short term trend following again from the good people over at cfm. So tell me. Well, maybe I should make one thing clear and maybe, maybe you can clear it up completely. I think last week, and I probably did not spend enough time on the paper, but I think last week I might have left the impression that it was kind of the tick size that had changed somehow and that made it or that impacted the short term trend following, so to speak. But actually it's the volatility normalized tick size and that is very different from just the tick size. So maybe you can just set the record straight and just explain what they mean when they say the normalized tick size. Volatility normalized.
- [39:58]
C
Yeah, but it's exactly what it says on the team. So when, when the, when the, when the CME or any other exchange sets up the, the market, they would set the minimum size that the price can move in. Okay. And it's normally set up in the way that the market trades at the time and relay and it's, it's reasonable for the size of their market. Okay, but, but you know, events happen. Okay. So if I look. So when you set up a certain tick size for an equity market that might have been 30 years ago, the market has risen in value. So unlike bond, bond futures, for example, which I love and trade, the price stays roughly at 100. Okay. Give or take. It might go to 110, 120. But you know, like the TY is trading at the moment at 108. It's, it's the, the market is that now the tick size is, that is then the volatility adjusted tick size is basically the size of the tick divided by the volatility of the underlying price. Okay. Now in equities you set up the, the market 30 years ago, the market has gone up. You know, we've seen the equity markets going up, you know, hundreds of percent. And now when I look at the volatility, suppose the volatility of the index is still 20%, but it's 20% of a much bigger number. So you know, the S and P has changed from about a thousand in 1995 to about seven. Like it's, it's, it's gone up enormously. Right?
- [41:23]
B
It has.
- [41:24]
C
So, so, so the, the, the tick size in the equity space has become much smaller. So if I look at the, the CFM paper, if I've actually done the, the maths. So you go back and there's a very large like the proportion of equity market in the small tick size is much higher than the proportional market in the large tick size. Now they do have a little bit of the. I think at the end of the appendix, appendix F at the end of the paper they look at within each asset class what is the variation. The picture there is actually a lot more nuanced I would say. But I think, I think part of the result is essentially the effect that we are seeing here which is to do with equity markets not performing well. Now the question why equity markets have not performed well? You know you, you can take the. They have, they provide an explanation about the sort of the sparsity of the order book. I think what the, this paper, this, this blog provides an alternative or a complementary explanation which is to do with the nature of the players that have entered the market and they are very much playing the anti trend position. And it's kind of nice because here we understand the mechanism. It's much clearer to see what the mechanism is that is driving which is playing counter trend.
- [42:44]
B
All right, so before we leave the papers and the blog post maybe just talk a little bit about the relevance for trend and maybe specifically for short term trend following of all of this.
- [42:58]
C
Oh absolutely. So I think it's important whenever you do a bit of maths is to have a little bit of, of back of the envelope understanding of how much does that impact the trend. So the impact that we're talking about is about reduction in say one day autocorrelation of equities and we can figure out what does a 1% negative autocorrelation mean for trend following. So if you were just trading, you take, if you're a trend follower which takes yesterday's price and uses it for today, then that will be a 16 basis point sharp negative 16 basis point sharp on your, on your, on your performance. Right. 1% of correlation times 16 because it's square root of 12 the number of days in the year you get hurt. 16 basis points sharp. Okay. If you are a fast trend following, even if you like within our industry we feel we are trading like the example they gave was a 520. The amount of weight that you have on that short term effect is something of the order of the magnitude 1 in 5. If you're trading at 520, so maybe 20% of that number. So your single market drop in Sharpe to 1% drop in autocorrelation is going to cost you something like three basis points. Three basis points of Sharpe. Okay. And that's not, not that much actually. Okay. It magnifies itself. If I look at the equity portfolio as a whole, okay. So there is an effect here which is more general. So if I, if I do that times like you, some diversification in the equity market you don't have that much of a diversification. So maybe we multiply by two. So that costs you a 6.6basis points of sharp. 0.06 sharp for short term trading. That's not that significant actually. And that's one of the reasons why I think that trends are driven really by macroeconomic events more than just the order book structure. Of course, the order book structure and the intraday behavior becomes very important as you move faster and faster. But even for fairly fast short term traders, this is not 1% is not that significant. Now we have seen maybe a drop of 3% in auto correlation in the equity market over, over the last 20 years. And that's kind of becoming, becoming painful, right? It's becoming a little bit painful. So we said six, it's about 20 basis point sharp for short term traders. For long term traders, the slower you are, you're basically staying away out of the game sort of. So it's a, it's a lot less. The, the alpha is driven a lot more by long term trend like secular trends that you observe in the market and macro events and so forth. A lot less than sort of the intraday. And that's, I think one of the reasons why the industry has drifted to slightly slower because you're actually harvesting, you're actually harvesting a different source of like you're not playing in the same market as the, as those retail traders.
- [46:07]
B
Yeah, I think another way that I'm thinking about these things is that just from memory, the conversation we had with Toby Crable not too long ago where he talked about that, that he had noticed that for example, markets that generally traded sort of higher at the open in the old days, and I'm not going to quote the correct numbers, but it's going to be fairly close. But I think he said in the old days markets that opened higher, they would generally also move higher and close higher during the day something like 58% of the time. Now in the last few years we're down to 54, 53% of the time. That doesn't sound like a lot. He said it actually makes a big difference for if you're a short term manager, whether you get that follow through.
- [46:50]
C
No, absolutely. If you're playing in that short term space. This is brutal. Right. And it's a headwind and on top of execution costs and it's painful. But as you move to the two month horizon that CTAs are trading now, the amount of, of risk that you have on that one day observation is a lot, is a lot lower.
- [47:15]
B
Yeah. All right, well, let's change, let's change direction a little bit here. The next paper that you brought along is called Thematic Investing as Missing Factors by Wei Li. And again, I think a lot of the papers that we touch on here in the podcast, for those who are not on my Sunday email list, that's probably where I tend to share them. I'm not, not so good and consistent in terms of sharing them in the actual show notes. But anyways, you brought this along and yeah. Very interested to hear how that ties into today's conversation.
- [47:52]
C
Absolutely. So first of all, a shout out to Robert Kozowski and the Imperial College Conference on Quantitative Investing. It's a, it's a wonderful conference that has been running for 18, well, more than 18 years, but this is the 18th conference and it's organized by Nick Baltas as well. So a shout out to Goldman Sachs and for supporting it as well. And it's a, it's a wonderful conference. They choose really good papers. And I was listening to that paper and I thought this is really relevant to the equity discussion because when we talked about the dispersion, we talked about what's, you know, a lot of the risk is no longer in the index. Okay. And, but we, we have to understand that like the question is where's the risk? Where what is driving equity returns, single stock equity returns, if it's not the index? Right. So, okay, so in the good old days the index might have explained 35%. These days it might be explaining 30 or 25% of the index and then of the total variance. And now the question is what's, what happens to the remaining variants? And you know, quants have already been, you know, kind of asking this question for many, many years. And of course industry is another one industry specific. So it might not be, you know, it health industry right now during the World cup. It might be the entertainment business industry in, in, in, in, in U.S. no, no, no, it's, it's, it's, it's a real, it's A real, there's industry explains it a little bit. And then of course, in the 1970s, there was a Rosenberg guy who, I can't remember his first name, and he came up with the BARA factors. That's another amazing, amazing innovation that it was like a risk model. It says, well, I've noticed that, you know, value stocks move together and growth stocks move together. There's a certain factor like leverage or growth or value that also explains the co. Movement of stocks. Stocks, okay. And what I find really interesting about the, the current situation, there's an evolution of thematic investing in the industry. And themes are another way of explaining the volatility. Right. So we said the index may be explaining, say 30%. When we add industry specific volatility and we add bara factor industry volatility, we get up to maybe 50% of the volatility. And I think the paper and the presentation by Whaley in the, in the, in the conference was amazing. Really, really good. And he's saying, well, okay, so we've explained 50 of the volatility. We understand what's happened to the other 50% themes. What are themes? Themes are like risk factors, but the things that are transient and they are driven by news. So World cup is a theme. Okay. Or, you know, straight of homooze, it's a theme. Okay, Covid, that's it. You know, these are like very big themes, but there might be small themes. So, you know, it might be single stock, like semiconductors. Right. Semiconductors are a great example because it's both a theme and also an industry sector. Okay. So, so I think what the paper says, well, okay, how much extra volatility I can assign to semiconductors, which is not explained by just the semiconductor industry moving together. Okay, how much can I harvest that? And they've done a lot of work in trying to sort of basically first of all, mapping news and mapping sort of events that happen. Right. The semiconductors is playing it at the moment because there's a huge demand for obviously for GPUs. And he says, well, there are two observations that he wants to make. The first one is that the total amount of risks that we can explain is another 5%. So we have explained 50%. There's 5%. There's still 45% of risk that we like, which is genuinely idiosyncratic. Right. It's maybe we can't explain at the moment or like there's no. Which is fine. Like we have idiosyncratic behaviors of companies. They are just idiosyncratic by nature. But then, but the other observation, he says that if, if investors are interested in it. Okay, how do you harvest it? Okay. And I think his paper is really interesting is that we've seen a growth in the ETF industry. Right. So if you go to Morningstar and say I want to buy, you know, a military industry something.
- [52:29]
B
Sure.
- [52:30]
C
Because it's, there's a war going on, then there is an ETF for that. In fact, you get very specific ETFs for like healthcare and healthcare manufacturing and healthcare heart surgeons and healthcare kidney. Right. You can get ETFs for almost everything. And like there's thousands of ETFs explaining it. The second one is of course the QIS. So the, the, you, you speak to Morgan Stanley, you speak to Goldman Sachs, you speak to, to, to these guys, they will say we can create QIs, basket harvesting, world cup harvesting, Iran war. Okay. And, but what he says is actually, sometimes these do not actually capture the volatility. The paper is talking about capturing additional orthogonal volatility. And he said, he observes that a lot of the ETFs are actually too much exposed to basically to the industry or to the currency or to the country or to the general bara factors that we've seen and are actually not representing. So if an investor wants to have more pure theme based exposure is better off essentially constructing this on his own or her own for that matter. So it's a really nice paper because it's less about prediction, it's really about risk, about trying to identify where is the risk coming from. And I think when we design, I think it goes back to the question we asked in commodities is like, how much extra risk do you have in going to, you know, single stock futures? How much extra risk do you have in going to the third or the fourth contract on the commodity curve? All of these, like, these are, this, this is much more important to have sorted out before you even try to predict things. Things. Okay. So I really, really, really like this paper because it's like, sets up a nice framework to understand, okay, what is thematic investing, which is all the crazy the moment, how much, how much risk is there in that
- [54:37]
B
and, and parts of, of the whole thing about themes. Obviously a lot of people will know the AI theme at the moment, but it's, it's narratives at the end of the day. And funnily enough, there's another paper that you brought along by the title of Narrative Momentum. So I think we probably spend a little bit of time on that before we wrap up completely. But this is a paper by I think Ronnie.
- [55:08]
C
Ronnie Satka. Yeah. So Ronnie Satka, an amazing academic quant, is done a lot of, of work on seasonality and is, and this paper is about a really unique data set that they've done. They've curated news and like millions of articles.
- [55:25]
B
About 13.3 million articles.
- [55:28]
C
Yes. So I mean like it's, it's, it's a Sisyphean work of like between like over 11, 12 years and then, and so, and what it does give them, which is really nice is like a real point in time of the evolution of stories. Okay. Because nowadays if you run a search about what was, what was the news in 2012 like, like, good luck finding out. The Internet does not remember and, but they do. So it's, it's really nice that they have like a really good point in time data set to say, okay, can we understand how themes and narratives evolve and what does that imply to prices? And I'm going to cut a very long story short. What they do is they do, they, they look at the themes which are like, like war, which investors tend to like in react instantly, which is not surprising. Okay. So, but a lot, a lot of stuff is like, it just evolves very, very slowly. So it has a lot of themes that take months to evolve. In fact, it looks at like six months horizons. And what he finds is there is the under reaction. Like it takes time for investors and we are very familiar with that. We as trend followers. Indeed. Exactly. And in many ways like the whole point of that is that in the absence of complete information in the market, the fact that the price has gone up or the fact that other people are talking about a theme actually makes you evaluate your own valuation in the absence of complete information. Right. So one of the reasons, the way that we form our own valuation of what a commodity is worth or what an equity is worth is through essentially seeing what other people are talking about or, or buying. Right. Or the way the price action behaves.
- [57:15]
B
Yeah.
- [57:16]
C
So, so it's a really, it's a perfectly understandable and I would say explainable and I would say legit rationale. Okay. And he spends a lot of time then constructing exactly like the first paper sort of a beta for each of the stocks to the news. And then he shows that, that indeed actually does make money. You can essentially trend follow the news. And as the news collapse, you, you start then, then, then you kind of essentially move away from the, from those stocks and you make money, you make money that way as well. So I was, I Was, I really liked it. He looked at the backtest, he looked at, he looked at sort of the correlation to general momentum, like momentum that we, we kind of know in life. Because of course the first question is like how related it is to price momentum and it shows that it is unique. I think my main, there's like, there's a mathematical issue I have with the paper and there is a, there is a reality check on the paper. I think the, the mathematical check is actually to do with the calculation of the beta of the stock to a narrative. And that's actually something which is difficult for us. Right. So if I'm trading, if I'm trading the first contract or the second contract or the third contract in commodities, we do have a good history to understand what is the correlation between summer natural gas and winter natural gas. Okay. There is, there is a, we can do some maths over a long period of time, but to understand what the beta or the correlation between a certain stock and a narrative, it's a much more difficult exercise. Especially as we've already established that only about 5% of the volatility is driven by the narrative. Okay. So that makes the estimation error on like being able to construct a genuine basket difficult. It's not a, it's a very subtle process and you know, it's mathematically possible to diagonalize matrices. But whether what we get is correct is not. I mean, I think I've spoken to one of the big, one of the big banks and I think they do a QAS product and in that case, actually I actually like the way that they do a human overlay of the basket for the themes that they are doing. So they're looking at the, they're saying, well, this stock, it might not show in the price, but we kind of understand why, you know, Coca Cola will benefit from the World cup. Okay. For example. And I think I like that approach. Of course, completely non quantitative makes it very difficult for us to trade. But from a mathematical point of view, I think that's, that's, I think the weakest part of the, of the paper. But I think I have a much more generic issue which is that the way we consume new news is changed. Right. So the paper is definitely correct for, you know, the universe between 2012-2023, if I remember correctly, that that's the periods that he was covering. But with the rise of AI, with the rise of our ability to harvest text and news and I mean you see that in the market, you see that analysts these days, you know, machine learning will basically churn all of the news on the stocks that they are interested and will present it like, like instantly. It's just insane that, that ability. And it's not just. So this is, and this is the, this is the, the human analyst, right? The CTAs, the, the, the systematic, the systematic funds, they do that, you know, already part. You know, it's, it's, it's, it's a done, it's a done problem now over the last five years has become a very much a solved problem. So I think that the debt under reaction overreaction is potentially going to be out the way quite quickly. So that's, that's my issue with that market.
- [61:18]
B
So. So yeah, so before we move to the last kind of small point where, where we wanted maybe to. To sum it all up and, and talk about kind of different ways you can trend follow equities when you, when, when you just said that before it, it actually reminded me of, of a, of a section of the, of this interview I mentioned in the beginning with Jason Wong where he talked about that with agentic AI as it comes, you know, news is actually not anymore being pitched from some kind of database. It's actually being generated for each of us individually. So, so it's kind of. So what is the news? Well, it really depends on who you ask so to speak. As far as if I understood him correctly, which is. Yeah. Kind of frightening and intriguing.
- [62:10]
C
No, it's a, it's a great, it's a great observation. It's, it's actually something I haven't thought about which is that the. We all live in our own bubble. Like the social network basically presents us with the news that we like to think about. Right?
- [62:23]
B
Yeah.
- [62:23]
C
And that creates our own bias in the trading strategy.
- [62:27]
B
Sure.
- [62:28]
C
Right. So. And you're right in that kind, in that sense the, the truth, the, the narrative truth is almost individual.
- [62:36]
B
Yeah, yeah, yeah. Anyways, just to sum it all up before we, before we wrap up up so to speak. So how do you create a diversified strategy in equity?
- [62:48]
C
So so like we've covered like a huge, a huge range. Obviously us as CTAs, we trade equity, you know, equity index futures.
- [62:56]
B
Sure.
- [62:57]
C
That we understand now covers about 30% of the equity risk that is available to harvest. And we've talked about sort of short term facing challenges but medium term and long term equity trend actually is all nice and well. Okay. And then we think about industry and industry is actually reasonably easy to trade. There's lots of plenty of ETFs which offer it There are a lot of actually futures on, on specific sectors and that, that will give you a nice diversification. You have to be aware that you don't want to double count, right? Because you, right. So like at the moment trading the semiconductors and the sb, you're basically trading the same thing. So construction, you start, have to, you have to start thinking about construction and then we move on to the BARA factors. Right. At that point you really have to think about the construction, you have to think about shorting, you have to think about long like you're constructing it from individual stocks. But it's difficult, right? And it requires you to normalize to get a truly diversified portfolio. And it's. And it doesn't. There's less and less risk and of course narrative is even worse in that respect is because at any time it might be covering 5% but your risk factor changes over time. So you are always not, you're not just catching up to trend, you're also catching up to the risk factor which is trading right now. And that's pretty difficult. But I think what is nice is that you can actually create sort of a cta, a trend program in equities which is, is like almost uncorrelated to the first component. Right. So all of these things, the aim for that is to create and, and you know, I'm not, I don't think I'll be telling huge state secret is that like, you know, some of the companies that have worked in the past have been doing that. So, so, so it's, I'm describing, I'm describing something process that a lot of people have already gone through and, and that diversifies and that's a good thing thing. But equities is a really rich universe that you can, you, you can create a, a diversified trend portfolio in it.
- [65:07]
B
Yeah, well we've certainly made some managers happy to hear that trend following on. On. On single stocks is, is. Is an interesting and, and potentially profitable venture. Whilst most CK probably still stick to the index for, for, for their purposes in terms of putting on exposure anyways, this was super. You're very educational. I really appreciate that. And to everyone listening today, if you want to show some appreciation for the preparation which is not insignificant for these conversations, then do go to your favorite podcast platform and show your appreciation for the co host to come every week and share some of this really insightful information. Next week I'll be joined by another great co host and that is Nick Baltus that we just talked about earlier today. And no doubt that's going to be timely, fun, insightful and if you have a question for Nick, as always you can email them to info top traders unplugged.com and I will do my very best to bring it forward. Now you have the World cup final. Won't have one of our teams. Denmark didn't even make the World Cup. Switzerland is out. So is England. England, yeah, but it might still be a good game. So I think on that note we have something to look forward to other than the release of this episode this weekend.
- [66:33]
C
Looking forward to Spain winning.
- [66:35]
B
Looking forward. You look forward to Spain winning? I look forward to the best team winning, whoever that might be. So anyways, Joav, thank you so much. So from your with me, thanks ever so much for listening. We look forward to being back with you next week and in the meantime, as always, take care of yourself and take care of each other.
- [66:55]
A
Thanks for listening to Top Traders Unplugged. If you feel you learned something of value from today's episode, the best way to stay updated is to go on over to your favorite podcast platform and follow the show so that you'll be sure to get all the new episodes as they're released. We have some amazing, amazing guests lined up for you and to ensure our show continues to grow, please leave us an honest rating and review. It only takes a minute and it's the best way to show us you love the podcast. We'll see you next time on Top Traders Unplugged. This podcast expresses the views of its hosts and the guests appearing on the podcast as of the date of its recording, and such views are subject to change without notice. Top Traders Unplugged do not have any duty or obligation to update the information contained herein. Furthermore, Top Traders Unplugged maybe make no representation to its accuracy and it shall not be assumed that past investment performance is an indication of future results. Moreover, wherever there is a potential for profit, there is also the possibility of loss. This content is made available for educational purposes only and should not be used for any other purpose. The information contained in this podcast does not constitute and should not be construed as investment advice or an offer to sell or a solicitation to buy any securities or related financial instruments in any jurisdiction. Certain information contained herein concerning actual economic trends and performance is based on or derived from information provided by independent third party sources. Top Traders Unplugged may believe that the sources from which such information are obtained are reliable. However, Top Traders Unplugged cannot guarantee the accuracy of such information and has not independently verified the accuracy or completeness of such information or the assumptions on which such information is based. This podcast, including the information contained herein, may not be reproduced, copied, republished or posted in whole or in part in any form, without the prior written consent of top traders.
- [68:28]
C
Unplugged.