The Derivative - Why Whales Tails Whip Up Market Tremors with Hari Krishnan

Episode Date: September 30, 2021

How do large delta hedging flows of market makers tie in with Central Bank quantitative easing? How do ETF rebalancing’s and structured note issuer movements shift markets? We sit down with options guru Hari Krishnan talk through his new book: Market Tremors - Quantifying Structural Risks in Modern Financial Markets.  Hari's first book, The Second Leg Down, talked about what investors can do when they're in trouble and how to structure actual options trades to keep bleed under control and maximize protection. This time, he's zooming out and asking how to know when there might be trouble lurking. How we can quantify and identify events like Volmageddon (Feb 2018), the Swiss Franc depeg (Jan 2015), Game Stop run up (2021), and more. In a world where we often get bogged down considering how specific items like this gamma, or that Fed decision, or how large ETFs flows will impact markets, Hari masterfully weaves all those 'agents' together to consider how we apply a real world risk to these agents shifting the distributions we rely on to size investment positions. Enjoy the episode. Chapters: 00:00-03:13=Intro 03:14-06:34 = Volatility - The Last Chance Saloon 06:35-15:20 = There’s London Whales Everywhere 15:21-31:55 = Merging Normally Distributed & Networked Chaos 31:56-42:01 = Volmageddon: A Case Study 42:02-54:19 = Big Gamma: Market Making Options / Do Less 54:20-01:09:43 = The Danger of 1 / Volatility Position Sizing 01:09:44-01:17:37 = Takeaways & The Twin Heralds of Risk Follow Hari on Twitter @HariPKrishnan2 and check out his book here. From the episode: Sequencing, Skew, and (option) Strikes with Hari Krishnan The NON-Wisdom of Crowds with Nigol Koulajian of Quest Partners Straddles, SVXY, and (Gamma) Scalping with Logica’s Mike Green The Tail Has Wagged the Dog  The Swiss (Franc) Isn't All that Neutral Don't forget to subscribe to The Derivative, and follow us on Twitter at @rcmAlts and our host Jeff at @AttainCap2, or LinkedIn , and Facebook, and sign-up for our blog digest. Listeners are reminded that managed futures, commodity trading, and other alternative investments are complex and carry a risk of substantial losses. As such, they are not suitable for all investors. RCM Alternatives is DBA Reliance Capital Markets II, LLC. For more information, visit www.rcmalternatives.com/disclaimer

Thanks for listening to The Derivative. This podcast is provided for informational purposes only and should not be relied upon as legal, business, investment, or tax advice. All opinions expressed by podcast participants are solely their own opinions and do not necessarily reflect the opinions of RCM Alternatives, their affiliates, or companies featured. Due to industry regulations, participants on this podcast are instructed not to make specific trade recommendations nor reference past or potential profits, and listeners are reminded that managed futures, commodity trading, and other alternative investments are complex and carry a risk

of substantial losses. As such, they are not suitable for all investors. Welcome to The Derivative by RCM Alternatives, where we dive into what makes alternative investments go, analyze the strategies of unique hedge fund managers, and chat with interesting guests from across the investment world. We had the meme stock explosion with GameStop and so on. We had the Volmageddon in 2018. We had a flash crash in 2010, all sorts of things we've had over the years. And many people would say they're black swans, I'd say they aren't.

They aren't black swans because basically what's happened is there's been some analog of the London whale who is in the market making money, highly levered, with very strict risk controls, who never enters the market unless something big enough happens, in which case that agent, that player, that whale, is forced into the market either to rebalance or to liquidate. And when that happens, the old distribution changes. So the whole thesis of the book is that even if you think that the distribution of historical returns is okay, if a few players get big enough, then they have to react to a random shock of normal size, of moderate size. They'll be forced into the market. They'll have to sell typically. And when they sell, the impact of that selling

will cause returns to be much bigger on the downside, especially, than they otherwise would have been. So it's basically taking a view on how the network behaves versus the distribution. Hi, everyone. We're recording on Tuesday, September 28th here with markets down about 2% on the day. So what a better time to hear the soothing calm voice of one of the best in the vol space here with us today calm voice of one of the best in the vol space here with us today to dig into some of the research and reasoning behind his new book, Market Tremors, Quantifying Structural Risks in Modern Financial Markets.

We have none other than Hari Krishnan. And the book includes a very nice acknowledgement for yours truly. So thanks for putting me in print there, Hari. I really appreciate it. Well-deserved. Very well-deserved. That's my first known acknowledgement. Maybe there's one I don't know about out there, but it's my first known one. So I appreciate it.

And how are you? Where are you? At home? I'm sitting at home, yeah, in some random room. I'm in Duxbury, Massachusetts. So don't track me down when I'm here. Don, yeah, in some random room. I'm in Duxbury, Massachusetts. So don't track me down, but I'm here. Don't track you down. And we've had you on the pod before talking through your background and how you got into the world of options and the hedge funds.

So mostly going to skip over that. But if you could just give us a quick lowdown on kind of what you do in the day out in the options world for clients. Well, nowadays I focus on long volatility strategies. So I do tail risk hedging, adaptive hedging for people who are bearish, and just basically stuff that profits from disorder, downside moves and risky assets, and general uncertainty in the market. So what I do is kind of like an add-on to what people already own. And what does it mean by adaptive hedging?

If they just kind of tactical hedging? Adaptive hedging to me is finding the best hedge given the market, the state of the market now. So if the price of insurance is high, I don't want to buy the most expensive stuff. I want to find other ways to hedge that don't involve buying the most expensive stuff. So I'm trying to figure out ways to protect without overpaying for protection.

And does that bleed into everything? You might go long gold or something or just across the vol surface? Currently for the separately managed accounts I run, it's purely equity based. But I do have some advisory business too, which runs the gamut. So yeah, it could include gold. It could include credit, anything. Which has been a tough month for that.

We're just looking through some data of, right, all the proxies have sold off alongside everything this month. Bonds down, gold down, S&Ps down. Nowhere to hide except natural gas. Well, you know our speech, Jeff. It's volatility is the final frontier. It's the last chance to lose. So yeah.

The last chance to lose. The last chance to learn? I said it in the wrong accent, but yes. Okay, and so on to the book, Market Tremors. I was joking with you on Twitter that you needed to incorporate the Tremors movie into the cover, which had a little earthquake scission. I hear you on that one. There are all these

copyright infringement things under the table. So you have to pick an image that's compliance. So you get these things and they look very anodyne most of them look pretty boring and so um the editor and i got together and one of them and she said well why don't you make the lettering a bit jaggedy she did i thought okay that's kind of tremor like so let's go with that yeah the um how does how is that whole process before we dig into the meat how you've written the book your other book uh second leg down i can't remember did you have one prior to that no no nothing at least i've had two bouts of insanity in this career and those two so second leg down was written in 2017 it came out and this one came out recently um so how's that whole process working with the

editor getting it done it's a labor of love i'm sure yes i mean for me the big thing was to build a brand so i try and be a brand i try and do it as well as i can keep the quality high but effectively you you can't be known in this industry for 20 different things because you want to focus on the stuff where you can add the highest value. And for me, it was macro slash volatility. And I knew the macro was a tough space given all of the central bank interventions and interest rates and currencies and so on. So volatility outright seemed like a good place to play. And yet lots of people didn't know about it. I would say even four or five years ago, people didn't consider it an asset class per se. They just thought it was kind of something that occurred on the side of

buying options. And I felt that was an important area to focus on and the two areas for me were what do you do when you're in trouble and how do you um identify dangerous situations so the first book was dealing with bad situations the second book is um how do you smell a rat even if volatility is low so that's what the book is about yeah you stole my uh you stole my thunder there i was gonna say um without saying anything having to do with math or distributions of why you wrote it and what the overall theme is so that's that's that sorry about that yeah no worries um and i'll dive in my next thing so i really liked one of my favorite parts is you really dig into the fact that uh market returns aren't normally distributed which has been covered

far and wide by nassim talab and elsewhere um and even though you point out they appear to be a great deal of the time appear to be uh normally distributed but then you'd say they're not just that they aren't, but actually dig into really what's going on is that the distribution shifts something or more appropriately, maybe someone is kind of shifting that distribution and creating the fat or tails or whatnot. So a lot of times to me, things I read, it's kind of just throws it out there, like the markets aren't normally distributed. You need some other thing, but you, to me, it was different in that you're looking at it more as a shifting distribution instead of just black or white binary of it's not normal.

It's this other thing. Yeah. There are lots of things that have happened post 2009, post GFC. We had the meme stock explosion with gamestop and so on we had um the volmageddon in 2018 um we had a flash crash in 2010 all sorts of things we've had over the years and um many people would say they're black swans i'd say they aren't they aren't black swans because basically what's happened is there's been some analog of the London whale who is in the market making money, highly levered, with very strict

risk controls, who never enters the market unless something big enough happens, in which case that agent, that player, that whale is forced into the market either to rebalance or to liquidate. And when that happens, the old distribution changes. So the whole thesis of the book is that even if you think that the distribution of historical returns is okay, if a few players get big enough

and they have to react to a random shock of normal size, of moderate size, they'll be forced into the market. They'll have to sell typically. And when they sell, the impact of that selling will cause returns to be much bigger on the downside, especially than they otherwise would have been. So it's basically taking a view on how the network behaves versus the distribution. And that sounds fancy. I don't want it to be too fancy. Let me tell you what is fancy. It's saying, oh, let's build a model of the entire financial network. Let's look at every transaction that goes through. Let's look at every buy and every sell, every settled trade,

and so on, and build a model of the economy. That's tough to do. And we talk about that in the book. On the other hand, the distribution, you know, the classical approach, which is everything is normal, or at least everything that happened in the past is a good proxy for what will happen in the future. That's also inadequate. So we tried to fuse those two together in a way that people could actually calculate things, calculate real risk numbers in the presence of whales. Maybe we should rename the book, There's London Whales Everywhere. Right? Yeah. If I had a better marketing sense, I would go. And you say in there that random returns, and I like this part too,

it's not just that the distribution shifts, it's like the rules of the game change. So talk a little bit more about that, of like, how do you view that as, it doesn't really matter if there's, right? Because I could say, oh, there's a spike. Just now we have a little wider tail over there. But you're saying the rules of the game can totally change and not just chip the distribution, but kind of what the distribution is looking at. Yeah, I mean, the simplest case is, let's say that there were a mega player, a whale, who had a huge position and a stop loss 5% below the current price. If the market gets anywhere down, then other people are going to know. There are going to be other agents who I won't name, but in what's called the high-frequency space and the market-making space,

who are incentivized to push prices down through that stop-loss level, causing the sale, which will precipitate a price move even much further down. So, you know, the distribution adapts to, well, big agents adapt to what's going on in the markets. And if they're small random shocks, they can turn into big ones purely based on leverage and positioning. There's a bit of a roundabout reply, but basically the distribution is affected by a feedback loop between biggest agents who either have to rebalance or have to liquidate, given a sufficiently large move and the original distribution and i i was reading it and coming back to uh ben hunt which had a great post and he's been on the podcast and he says the market's a bonfire not a clockwork machine right so if it was a clockwork machine and we could model the whole thing and we'd know okay when this ticks this way this is everything that happened the gears move but it's a bonfire it's hard to model a bonfire right um yeah i mean george soros had this idea of reflexivity which a lot of people know about

sure it's familiar to many people on the show who watch the show but you know a lot of us believed it and we knew it we knew that positioning was a huge factor in markets and cascading sales was a huge factor but being able to model it and come up with some real numbers was was the goal of the book um did Soros write that before or after he busted the Bank of England right because that was the mother of all that was probably you should have put it in the book that was the the first uh kind of asymmetric huge huge whale huge player having to do something. Yeah, I was probably too talented to do that.

That would have been a great case study. Hard to model, though. Okay, so let's get into how you started to think about this, how you merged those two worlds, the normally distributed and the modeling a bonfire, super hard to model. Yeah, that's a great question. I mean, what's been going on over the past decade? Who has become bigger? Central banks, definitely.

Passive investment vehicles like ETFs, definitely. Massively. Massively. Automated market makers who intercept 90% or more of all trades that go through the marketplace, hugely. So you've got dealers or options market makers, you've got central banks, you've got ETFs, ETNs, you've got structured products and systematic strategies. Those are the whales in the marketplace. They drive a huge percentage of all the volumes that go through. And even if I stick a trade into the market and it gets snapped up very quickly, I almost can feel the machine on the

other side of the trade. It picks it up more quickly than it ever did, assuming that the market is stable, but just bails out or it doesn't offer any liquidity when things are getting ugly. So if I want to do a trade now, I can get it off super quickly in most market conditions. But then the quotes disappear if things get ugly. What that means to me, even not knowing anything, even if I didn't know anything about what's going on, is that the market making community is no longer a community that's dedicated to providing quotes with size behind them, but rather a community that is incentivized to do so if there's edge in providing those quotes. So there's a highly competitive marketplace with quotes provided with narrow spreads,

with ease of execution until things get ugly, in which case all of that disappears. So we have a more, we basically get better fills most of the time in the median case and no fills if things are really bad, depending on our position. Did you see, I think it was floating around Twitter yesterday, there was like a market maker PSA video of like, we help pensioners and retirees and provide stability to the economy. Yeah, no, I hear them. And they're not entirely wrong. I have a good buddy.

Back in the day, I used to type my trades into Bloomberg chat. So he would pick them up. And he was sitting in London and he would execute them. And he would pick them up and he was sitting in London and he would execute them and he would work hard. He'd get me a tick or half a tick on average for 90 percent of all trades. But the remaining 10 percent would go wildly against him. And he'd be scrambling for his life and I wouldn't be filled and I'd be sitting there pulling my hair out for the ends of it. And that's kind of the way that you get treated nowadays.

You get, in the median case, you get tons of liquidity. I mean, maybe not in huge size, but you get good fills on decent size. But in the cases where you really need liquidity and the market's against you, the incentive-based structure doesn't really go in your favor. But how is, right, it was never a government utility. Like it's always been an incentive-based structure, right? How is it different than 10, 20, 30 years ago? Well, I'm not the expert on this, but in my opinion, nowadays anyone can be a market maker if they trade sufficient volume.

So market making is by definition based on how much you do instead of whether you have bought the seat on the exchange and you wear the jacket and this and that. Now anyone can do it if they have sufficient technology and they do enough size. So it's much more based on how much edge is perceived to be in supply and liquidity instead of being the sort of the central, the sort of the centerpiece for transacting. Right.

And the old days was like, your job is basically to provide this market and you can earn a little spread on doing that versus the philosophy seems like it's changed to my job is just to come in and make as much money as possible providing a spread right semantics but very important in this case well the other thing is that the spreads were much wider back in the day yeah even going from 16th to Nichols was a big change in terms of the amount of profit that designated market makers could extract from trading. And now it's so narrow in so many markets that it's almost as though they're running more hedge fund type ideas than just providing liquidity and collecting

bigger spreads. So the tiger spread is better on average, but perhaps worse at the extreme. And you have the other problem that it's been a winner take all market, right? So the biggest players do what I think you said before, 90% of all the market making, which if some crisis, some liquidity crunch there, they don't get an influx of capital, their risk controls kick in, right? They're just not going to play the game anymore. Yeah, you know, that consolidation is really important because you might know more about this than I do, but imagine that one of these guys went out.

I will not name them they're not many that could be catastrophic to market function at this point yeah well it's been in the news right of the everyone saying can get gripping lied in front of congress with all the gamestop stuff and he did they did stop providing liquidity or you know a market in that stock he said didn't. So we'll leave that for the legal scholars to sort out. But there's definitely something there, right? That's a prime example of like, okay, what was going on with the... And that's the case study in the book of Game Supper. It sure is, but I won't go any further on that point. You got to buy the book.

Well, I was going to say, I'm not naming names. That's the thing. Okay. I'll name them. to say, not naming names. That's the thing. Okay. I'll name them. I don't have a problem. And so I want to push back on this. First, I want to say mean field theory. So that's the basis for all this, of how you brought this? Yeah, it's a jazzy phrase.

I mean, basically, mean field theory says this. Imagine that you were in a room and you wanted to. This is a simple case that's in the book. You wanted to measure the temperature at some location in the room. The temperature really is a function of how much the molecules in the room are buzzing. So the more kinetic energy they have, the higher the temperature will be. But it's a pointless or a hopeless exercise to try and take every single molecule in your room and measure those collisions and those movements. That's just a hopelessly high dimensional problem that's completely intractable. So the right thing to do is to just think of temperature as a macro level quantity.

And if you have a heater on one side of the room, if it were winter, which it isn't, but that heater would generate some excitation of molecules and that heat would sort of diffuse through the room. And there's an equation for that. And so basically you replace every single particle interaction, pairs of particle interactions, with just a distribution. The distribution says, on average, at a given point in time, at a given point in space, this is the temperature. Boom. And that's the way classical finance works. You can start with Markowitz and before. You get distributions of price returns. So what you do is you take time, slice it up into pieces. You assume every piece was generated by the same machine, whether it's God or just the market machine.

It generates all these returns. You collect them, and then you build a histogram. And that is the distribution of returns you can expect in the future. It's a good model to an extent but it's been modified in various ways. People have said well normal distributions aren't good so they're fat or tails. Fine. You can parameterize in different ways, you can look at correlations in different ways. But that doesn't really address the problem in modern markets, which is this. In modern markets, what we're seeing is periods of very low realized volatility that persist for ages. And then suddenly there are all these volatility spikes, spikes out of nowhere from a low level. So one case study that we did, it's very simple, is that there have been more 10-point or greater spikes in the VIX over a five-day period when the VIX started out low, say below 20, than ever before.

There were no such events from 2000 to 2009. So the VIX did spike 10 points or more in, say, 2008, but it did it from a handle of 30, not from a handle of 15. So you get all of these spikes, these bouts of volatility from nowhere. And these are all positioning or leverage-based risks that just don't show up in the price action. And you now have a world where everyone wants to use models and I hate to point a finger at CTAs because I'm a CTA fan but there was some kind of change in the CTA industry maybe 10, 15 plus years ago where people stopped saying well if we have a position that's becoming more volatile but it's on side we'll cut it so that we have a constant volatility budget to say, we'll just let it run, which

is the way it used to be. It isn't today. You then get scaling according to one divided by volatility or one divided by variance, whatever the case may be. So CTAs, trend followers, are scaling their positions inversely to volatility to realize volatility. And then you have all these other players who are big in the marketplace.

You have volatility control funds. So these funds say, oh, I'm going to buy the S&P and I will hold cash. And I will mix my S&P cash allocation so that I can hit some volatility target, say 10%. So if S&P vol goes up, they have to sell their position simply to hit that target. And then you have other funds like risk parity funds, and you have various strategies that target volatility instead of trying to extract alpha

from volatility. And given all of this emphasis on targeting risk, instead of managing risk, we have a situation where many players use leverage and have the same sorts of positions. And if everyone used the same risk model, the markets would be doomed. One day out of the blue, an event would happen that would force everyone

to get out at the same time because their vol limits would be hit. And the market, whatever market this might be, let's say the S&P 500, would go to zero. And this notion that everyone should be doing the same thing on the risk side is really one of the core problems in this business, where if big enough players are using the same model or using a model that's very rigid and everyone knows what that model is that's a recipe for um tail risk but what is might be perceived as tail risk but it's just liquidation in response to a risk limit being hit. And yeah. That's a good mental model.

Like if everyone has a 3% stop to exit their whole portfolio, yeah, the market goes to zero the minute you hit 3%, right? Everyone's selling, yeah. Everyone's selling it. A good heart's law is something I don't know very well,

but it's relevant here, which is that when a metric becomes a target, there's a problem. It fails to be a metric anymore. It actually presages crisis. Right. And to me, and we've talked about this on the pod with a few different guests, it's a much different risk environment, right? It's much stricter.

In the old days, you could have these blowups happen inside a bank and whomever didn't know guess, like it's, it's a much different risk environment, right? It's much stricter. Like in the old days, you could have these blowups happen inside a bank and whomever didn't know that this was going on. Nowadays, it's one strike and you're out and the risk department's calling you at 345, right? Like they're on your, your stuff saying, Hey, you got to reduce exposure by the close. So in many ways, it's kind of like maybe it's worked. Maybe it is working, like the market keeps going up. But in other ways, yeah, it's much stricter. And when there's going to be a severe downturn, all those risk controls are going to kick in. But to me, it's not the same model, like they're all using xyzriskplatform.com, right? But you're saying

that it's mentally all the same model of we need to protect against you know extreme risk and here's how we're going to do it we're going to put far on it we're going to put uh stops whatever the case might be yeah exactly i mean i i used to know a guy and he um this is probably 15 years ago and he um he knew how to keep prime brokers at bay when he got a margin call yeah so he would get a margin call every at bay when he got a margin call. So he would get a margin call every so often because he was very thinly funded. And he would say, he'd go to his prime broker and say, well, are you sure I have a margin call? Show me.

And so the guy would send back the risk numbers and it might be some span risk system that would spit out a number and say, OK, but who calculated this number? The guy would say, oh, it was my quant. How junior is the quant? And then he would figure out who was the senior guy and the senior guy would invariably be out of town, Paris or wherever. And so he could get five days of margin relief through this sort of process. Prime brokerage is no longer that kind of business. If you get a margin call,

you have to thought you have the cash pretty quickly nowadays. And it's much more rigidly defined. And having a good handle on where those limits are going to be set is very important, even if you're not hitting those limits, even if you're well below them, because you know those are the pressure points in the market. And that kind of brings us back to another point in the book, which is a discussion about technical analysis, you know, things like the Iron Cross, the Hindenburg Cross, and so on. Things like if the 50-day moving average for the S&P drops below the 200-day moving average, why should anyone care? Well, people should care because if other people are doing that,

those are pressure points in the market where selling may occur. So if you can trade around those sorts of positions, you may have an edge. And that's a major theme, which is understanding how other people are positioned and making sure you're not having to liquidate at exactly the same time, or even worse, a bit earlier than perhaps a bit later as well. There was an old trend following program you could buy just off the shelf called Aberration, I think. And it anytime it had a signal in platinum or palladium, maybe, and we would trade this for clients way back when.

So anytime it had a signal in palladium, it'd be up huge overnight because some desk had the system as well. You could buy it off the shelf. Right. So it knew in the morning there was going to be, you know, and maybe it was only 500 lots or something from across the world who had purchased this system. But in a thin market like Palladium, that was enough. So you'd get these five, six percent overnight spikes, you know, when it was supposed to

be the order comes in on the next morning. Exactly. Yeah. you know when it was supposed to be the order comes in on the next morning um exactly yeah one of the case studies in the book is uh volmageddon as they call it feb 18 yeah um we've done a bunch of research on that uh you did a bunch of research in the book. So let's dig in, if we can, a little bit. That's kind of a perfect case study of was fairly easy to identify these players looking backwards, even more so. These agents, whales, as you call them. So, yeah, dig in what you found on on Valmageddon and and why it's important, how it fits in. OK, well, I mean, the chapter in the in in in the book is pretty terse but let me say give me give you the highlights anyway um uh around 2008 or 2009 a lot of people wanted volatility protection and so they wanted to be long the vix they couldn't get long the vix because it's hard to replicate It's a complex formula based on the price of a variant swap that drives the VIX. So there were VIX futures by that

time, by 2009. So there were various contracts that were traded. And the VXX, which was the flagship VIX exchange traded note, or ETN, was born in 2009 in the wake of the GFC. And basically what it would do is it would buy the front month VIX futures and the second month and then reweight the combination. So it always had 30 days to maturity in the blend. That was great because it gave you at least indirect exposure to the VIX. The problem with the VIX, though, is that on average, the futures curve is very steep, meaning that if markets are calm, investors want to buy protection further out. So the front month futures tend to trade at a huge discount to the subsequent months of the back month and so on. So if you try and buy and roll that contract, you're always getting

dinged because you're buying high and selling low. So there's a net carry, which is negative, which is about historically about four% or 5% per month. 5%, let's say. 5% a month is a massive hurdle to overcome. And so the VXX was decaying like crazy as soon as market stabilized. Now, over time, people thought, well, why don't we turn that on its head and trade the inverse fix, which is basically shorting the front month or the front two months and then buying them back and then basically profiting from the role where you benefit from the front month

decaying very rapidly to the spot fix. And you don't lose as much by buying the back month, which is decaying slowly to the front of futures whatever that's a technical thing yeah and um that became super popular and there were people who quit their jobs at um you know well yeah target target i could show up at target so i'm not gonna that was that was the famous guy who was like in the paper like this guy made 82%. Yeah, I remember Jerry Hayworth told me about this. And so I looked it up and it was true that many years ago. And at some point we got to late 2017, which was a very quiet year.

And the inverse VIX ETNs were growing and growing and growing. There was the SVXY and the XIV. And they grew so much that if the flow desks that supported them, remember, these are notes, not funds. So they're hedged instead of replicated. If they had been hedged properly, these inverse VIX ETNs would have accounted for 30% of the open interest across all VIX futures contracts. So they had become the behemoths in the room. Now, the trouble with these ETNs and ETFs that use leverage or inverse products is that they have to rebalance according to a schedule.

And they're highly incentivized to rebalance at the close. Why? Because they want to track anything. They don't want to be just trading in the market and then be subject to a big move near the close that increases their tracking error. The goal of these products is low tracking error and low fees. And they used to use this, well, they probably still do, they used to use traded settlements as a secondary order book to transact so they could try and get close to NAV. Everyone knew on the day that the VIX, the volmageddon occurred, which I think was February the 5th or

6th in 2018, that they would have to trade in mega size and that the futures market could not support those trades. And even if they had tried to hedge in the S&P futures indirectly, that would have had mega impact or massive impact. So one could almost predict how much they'd have to trade to replicate. And using various arguments that are presented in the book that are pretty technical they're somewhat speculative but i think they're pretty good um one could estimate what the follow-through would be so when the vix had closed on the previous friday at 15 spot 6 or something and was already at around 2022 whatever you could guess

quite easily or you could estimate that the VIX futures had to pop to 30 as a function of the impact of forced rebalancing from the mega agent or whale, which in this case was the ETN. And these sorts of cases are great because I've never made myself many friends in the ETF space, but I'm not a massively anti-ETF or ETP person. It's more that I thought they were a really good laboratory or lab for testing the theories about positioning risk. Because these are agents, especially the levered products, that tell you exactly what they're going to do. It's in the prospectus generally, so they have to do what they what they're going to do. It's in the prospectus generally. So they have to do what they say they're going to do.

So everyone knows they're coming. Yeah. And these are sort of the extreme cases of the positioning risk problems that we try to deal with. And I'll throw out, we had Mike green on the pod before who was on stage with Chris Cole, another Volpro at a conference at EQ Derivatives or some conference, Volconference, basically calling this out and arguing with the ETF creator. I can't remember his name. But there was this public discussion of how this was a possibility and how it would spike.

So it wasn't just like if you were paying enough attention to the prospectus, like people were actually out there talking about it. And then Mike Green told us on the pod, he structured a trade for Peter Thiel. They bought, I don't know, fives or tens of millions of deep out of the money puts in that one product. And then once it pierced its low, it went out of business and he made a lot of money. I didn't ask if that was in the infamous now Peter Thiel, $5 billion IRA. We'll have to ask that. Well, Mike Green is a friend of mine, so I'm not going there. But yeah, I mean, that was a fairly predictable case,

given what we know today. People didn't know as much back then, I mean, that was a fairly predictable case, given what we know today. People didn't know as much back then. I mean, people were less aware in 2017, let's say, about the potential risks of exchange traded products. And they didn't even know about lever products. And one of the trading books that I ran from 2012 to 16 was a book that effectively shorted badly, badly designed ETFs. So the simplest case is imagine that you've, you found a levered ETF. The levered ETF has to rebalance every day to maintain constant leverage.

So if the reference index before leverage is mean reversing, they're buying and selling at the wrong time every single day. So that thing is going to underperform a constant borrow two times leverage investment in the index. So stuff like that was easy to do. And, you know, the way that ETFs and ETMs work, and I don't criticize the business model, the business model is, let's throw anything against the wall and see if it sticks. So it's not a, let's do what we believe in. It's, let's see what people will buy. Let's paper it up carefully. Let's design it as best we can, not being money managers ourselves.

And run with the stuff that wins. So they're playing a big options game as well, but it doesn't necessarily benefit the average guy or girl. I think the Schwabs of the world have gotten a little more careful of allowing the customers to trade the levered products and basically pointing out of like, Hey, these are for trading. It's not for investing, right? If I'm bullish natural gas, I shouldn't buy the natural gas three X bull ETF. Cause it's going to have a rebalancing issue. Um, so I think the industry has gotten better at that. Um, but I'll definitely agree that it's like Oreos these days, right? If I'm in the grocery store and they have like pistachio Oreos and orange Halloween Oreos and Star Wars Oreos, right?

It's like any type of Oreo you can get. It's just shelf space. And if something takes off, they're going to, right? They'll mass produce those like crazy. So it's the investment banking model. Yeah. We're going to put it, right? It's almost like a venture capital deal

right of like hey we're gonna launch all these companies one's gonna be a huge winner eventually um and so be it if we had 10 losers it doesn't really matter and they don't have to be losers performance wise but just losers it didn't get enough investor interest right yeah exactly exactly next you go into the uh i'll i kind of i'll bundle it the gex squeeze metrics jim carson lily all the uh gamma people out there right so lily frankison and jim yeah yeah so that's become the like hottest kind of topic and from for me in terms of like agents and big players and market makers right so yeah did they inform your thinking or you were already working on this how did how did that work well i worked for market making firm for a couple of years back around 2000 and i knew that i knew knew Blair Hall. Blair Hall is a friendly acquaintance of mine.

So I knew that one of the great innovations in that space was just taking on positions and then managing to macro hedge them on the backside. You know, that was what that business was. Instead of saying, I need to be flat at the end of the day, It was more, I need to be hedged at the end of the day. And so it was kind of a second generation approach to making markets. And so I was interested in that. And then this notion that institutions like to buy puts and sell calls was well known to me. I knew these collar structures were big, and they had been big for a long time. And it's understandable why institutions want to do that.

They're worried about losing their jobs and buy puts. And they need income in a zero interest rate world, so they sell calls to monetize the premium in that. So assuming that the market makers had the opposite position, they were then the dominant agents. Now, these are not dominant agents in the sense that they have billions of dollars of capital. They can be fairly thinly capitalized, but they intercept the majority of trades that go through the market, increasingly so. position where they were short puts and long calls, i.e. had the reverse position of the large institutions that were looking to collar their positions, their equity index positions, lots of interesting things could be thought of. And I love the stuff that the squeeze metrics guys do and Jam and Lily and so on, But I use it more as an indicator of potential pressure points in the market than as an indicator of direction. And even if you look at the squeeze metrics research, it doesn't really give you direction. It more says that in areas where

there's high open interest inputs, if the market goes down there, expect a lot of oscillation. It might go down a lot, go up a lot. It's going to wiggle around hugely in that zone. And so what the book tried to say was a lot of things that were considered to be a fundamental origin, like in the initial phases of the COVID crisis in February and March 2020, you saw this move in, say, the S&P. I'm a little overly focused on the S&P in the book, but where you see this jaggedy down move in the index, a lot of people said, oh, there were varying opinions about how serious it would be, what the response would be, but you could explain it legitimately just based

on hedging from the options market maker community. So whenever the options market makers are short puts in size, and there's a small random shock to the downside, they have to sell. They have to hedge by selling. And if there's another random shock to the upside, they have to buy. So they're aggressively selling and buying in response to random noise that's going through the market. And I think that's a pretty compelling case for positioning risk, overwhelming fundamental information in terms of market movements. And that's a big theme nowadays. The whole Mike Green passive argument is an argument against fundamentals dominating pricing in this modern market. It's more an argument that if flows go into passive, if there's a relationship between performance and flows, then passive buys. There's no latitude there. And so you get these self-reinforcing feedback moves,

you get this distortion of the historical distribution that's based on positioning or based on this market structure. It's not based on a view. So little is based on a view now that the very notion of fundamental value or equilibrium, which I find a pretty bizarre notion to begin with, is out the window. Right.

So not like those passive flows are coming in and the portfolio manager of SPY is like, it looks a bit toppy to me here. I'm going to put that billion on the side for a month. Yeah, exactly. Or even- You have to put it in there yeah or even kathy wood now or in arc and all those right it's just here's the stocks we're targeting and it's it's going into those stocks and and so back to the game is that as you're writing the book as it's coming through like this game stuff stuff is exploding kind of late in the game for me when when no

pun intended when the game stuff thing happened yeah uh so i kind of just mentioned it the editors were saying you really need to analyze it but i'd throw my hands up by then but i was aware that downside risk could be exacerbated by market making options, market making activity. I didn't predict, to be fair, the upside melt-ups, the melt-ups that occurred in AMC and GameStop and so on. But they're perfectly logical in the context of this model, where retail used to be pretty small. And perhaps it still is pretty small in terms of individual retail investors. But one of the big things about options is you can get a ton of implied leverage.

So if I buy an option at a buck or a cent even, that's 100% out of the money and the market starts rallying up there, suddenly I can make 100 times, 500 times what I paid. So my tiny investment, my $10,000 or $50,000 or $100,000 investment can become a $10 million notional position pretty quickly. At that point, it's significant. That implied leverage that anyone can get by buying low delta options or short time to maturity options is significant in terms of distorting the market structure that we used to have. That's a big thing nowadays and it's hard to trade against. It's made it really hard for short sellers. RAOUL PAL for short sellers right well the the next edition would have a whole chapter right of that the retail's kind of become a new agent right there because they're not it used to be each one was doing their own

thing and i think with social media with these message boards and with kind of right if they're all pushing they figured out if we all push together and use social media to push, we're an agent and we can spike this thing up and get, you know, get them to cover. So it seems to me. We need to write a book about meme stocks, Jeff. I'm not ready yet. I'm done on this one. But it seems to fit perfectly with the theory, right? Of like, yeah, they become in mass. No, not one of them, but in mass, they'd become an agent. Which kind of got me thinking too, of your, your theory of everything's networked, right? Have we become more so with social media and with globalization, right? That that's,

it's even become more interconnected and, and agents can drive, you know, you had the graphs of all the nodes, right? They. Not only have they become more connected, but some of them have gotten bigger, which will filter down and shake the spider web, so to speak. Yeah, which begs the question, what's the endgame of this for me? And what's the endgame for the viewer, the audience of this? The endgame for me is we are actively building models of what mega agents are doing, how they act, how they could impact markets and so on. And we're trying to trade around. So we think that there's a i'll give you the original example i gave in this discussion where there's a whale who has to sell at 95 the market's trading at 100 so i know that if the market gets down to 97 it's going to blow through or 96 it's going to blow through so maybe i buy the 96 foot or maybe I do something else, but I'm trading

around where the whale is going to have to be balanced. That's point number one. Point number two is more for the viewer, which is, and hedge funds have made this mistake too. There are some very talented people in the hedge fund space. I remember Everest Capital tried to fund a lot of their positions by financing in Euros versus Swiss when there was a PEG, and they got blown out doing that. And that was not a function of their investment acumen. It's a function of being overconfident in a low volatility,ing trade, positively yielding trade that blew up against them. And so my number one takeaway for the readership is do not scale positions as one over volatility, especially when volatility is low. You're going to get yourself taken out one day if you do that and you don't have regard for positioning risk. So don't do that. Do less than you think you should in low volatility regimes and maybe be

a little bit more active with all spikes because then the genie has been unleashed and volatility is actually expressing true risk. In this environment, I think the right approach, and whoever watches this is free to correct me, is to do a little bit less, maybe not today, but in general, less than you think you should. Because if you try and gear up to the maximum level possible, or to your target level, you're going to be exposed if something happens out of the blue. And other people will as well. And so there'll be this vicious feedback loop where you're one of the sellers and so some of the easy takeaways that don't scale as one over volatility scale in a way that's a little bit different from what these programmatic strategies do because otherwise you'll be

the vixen in this you won't make enough relative to the risk in the trade to justify what you do. Do something a little bit out of the ordinary from a risk management standpoint. I'm not saying don't do something that's not defensible, but don't follow the crowd in terms of managing positions. So would exhibit 1A in the trial of that theory be bonds, right? Like super low vol and they're going to have a flight to safety and everything everyone loves about bonds. But maybe that's a perfect example of like, there's some hidden warts there we just can't see yet. And they seem to be more likely to pop out sooner than later. Absolutely. That's a great, great case. Yeah. Where you cannot. Yeah, exactly.

Bond vol has been pretty low. You cannot scale positions according to bond vol versus equity vol. Because, you know, what you're seeing nowadays is that equity vol is actually quite elevated relative to volatility in other markets. Fixed income and currency volatility is pretty low. It's artificially low for good reasons, which I won't go into now, but equity volatility is still not that low. So to do relative sizing on that basis, it's dangerous. Yes. Yeah. Even if it's not relative, even if it's just, I'm looking at my treasuries and I think it's totally safe

and totally low vol standing on its own, perhaps consider elsewhere. Come back for a minute just of how you're quantifying this. Are you coming up with a single number? Does it inform each market differently? Is it all connected? Touch on some of that if you could. Okay, well, I'm trying to do something a little bit more concrete than what the Charlie McElludds of the world do. And I think he's very good at Nomura and various other people do, which is to come up with various models of positioning risks at risk and to figure out where,

what structures or what strategies are growing disproportionately and not necessarily following them, but realizing what could happen if they have to rebalance or liquidate and to use that as a basis for hedging. Because, you know, you can have a lot of views and I've gone on the air on various stations, not as good as yours, but I've been on various stations

and I've said, well, how solid is the Fed put? Is it a hard put on credit? Is it a hard put on credit plus equities, which is a bit of a speculative leap? Or what? Or is it no put? Now, if you do believe that there is a Fed put, whatever's been going on with the governors recently, then you really want to protect against an air pocket move down in risky assets, sort of like a 10% down move in two weeks or a week or whatever.

Because they have everything beyond that covered. Because they have everything beyond, well, yes, exactly. Although if it is down 10, are you really going to risk your career or your wealth that it won't go down much further? That's another question. So really hedging against those sorts of position or liquidation risks is a good way to play the air pocket because you understand that on the basis of liquidations or overzealous positioning or margin changes in margin so so you can cover that even if you think the fact that is so. Love it and you mentioned would great minds think alike because I was saying like one of the right Charlie McElligott you mentioned um no more on his yeah always out with the CTA positioning

I sometimes take offense as being deep in the CTA world of like hold on I'm behind the scenes seeing actual positions and that's not totally what's going on or they've exited so sometimes I'm a little pushing back on that um which you touched on in the book a little bit of like we don't necessarily need to know everything about these positions but if we have the general idea um correct right you have any thoughts on that i do i have a friend mark malik at conquest yeah i've known him for years and um i don't want to misrepresent what he says but he used to have this idea that um he could replicate the returns of any cta or any cta index that you gave him by mixing and matching the asset allocation or the weight assigned to shorter longer term trend signals in any given system or in his system

so in other words you can take a return stream and say, if I focus more on short-term signals, let's say, and more on rates or equities, I can more or less match the return stream of a given fund. Now, generating alpha over and above that is irrelevant. It's more a function of knowing what they're doing. And so using a system that you have in-house as an inference tool to guess how people are positioned is pretty powerful. You don't need to trade that system. You just need to use it as a risk indicator. 100% accurate, right? Yeah, it could be 80%, 70%. If you get the gist or the gist of the gist or the gist of the gist of the gist yeah that's okay and mark's old partner nigel who's been on the pod and he went into

right he can replicate almost any hedge fund just with short vix and um s&p exposure right i've just basically just levered one way or the other of those kind of tools. So that's super interesting too. And you see that, okay, it doesn't really matter what names they own. At some edge case it does, because if just one name sells off and they have to exit, but if there's one big move down and all these names are getting liquidated, it's kind of the same thing. Absolutely. I fear that discretionary macro is largely a function of one or two decisions made over the period over a 10-year period so let's say that you were long or short fix but just before a crisis you decided to turn the dial down and you got out of it yeah

that's a hedge fund career yeah and yeah um um and Conquest have kind of gotten onto those concepts. Yeah, definitely. And I wanted to just ask, like I've spoken to many quant managers over the year who kind of say, and maybe they're old fashioned or we can get into it,

like none of this matters. I can see it all in the price stream. That's the ultimate tell of where positioning is and everything of where prices are at the end of the day and i'm or even even at the end of the minute or whatever but right that and you there's these people price tells all i don't need to know all this information what what do you say to that i'm not putting forth their argument all that well but you put it forth pretty well i'd say how did you size the position if you size them as one over ball you're in that game anyway you're in that space anyway i'm telling you don't size it it's one over ball unless you understand what the risk is in that game anyway like your risk and you think the price is the only thing that matters

you still are making an allocation across your signals so So if you have nat gas as one position and euro dollars as another position, obviously you're sizing the euro dollars bigger in terms of notional exposure than you are in the nat gas. How are you doing that? Are you basing it on realized volatility? If you are and there's positioning risk in one or both, you're still making some implicit assumptions that are not encoded in price. They're encoded maybe in the volatility of price or distribution, but not in the price direction itself. Well, yeah, I don't know.

You could argue that it's embedded in the price, right? Like it doesn't, not as many contracts are bought because there's positioning risks. So I dialed down my exposure and then it doesn't drive the price up high enough or something. I don't know. It's a great, great debate. We could have that one. But let's say I ran a trend follower with one signal. I'm going to look at the 10 day, 100 day moving average. And today the 10day moving average is higher than the 100-day, but it's been a rocky road. It's been super choppy. If I don't take volatility into account, I'm going to be potentially under or over-allocating to that position.

So if I don't take positioning risk, it perhaps isn't expressed in terms of the jaggedness of the path into account. I may have a different allocation than you would if you did take it into account. I hope that's clear. It's basically that sizing seems to be based on just some combination of fairly crude indicators. It may take some path dependence into account, but don't take positioning risk into account directly. And by positioning risk, we mean that there's... Everyone else is doing the same trade. Yeah, yeah, yeah. And they're more levered than I am, so they're going to have to bail out.

Sooner. And would you argue some of this not taking positioning risk into account is some of the CTA struggles over the last 10 years? Maybe they, you know, are so systematic focused that they've forgotten that there's other players in the game? Because basically we're saying you can't just play the game. You've got to play the players at the table, right? Yeah.

I mean, any CTA, many CTAs, I don't want to speak for all of them because I'm not universal on this, but they probably made a lot of money being long fixed income. And so they have been playing the positioning risk game to their benefit for many years, which is that central banks have allowed that fixed income game to work well. Yeah. Whether it's with keeping the short rate depressed or QE,

which keeps longer bonds, it gives up a support to longer bonds. So a naive trend follower would not take this into account. But if things should change, and I could go into lots of speculations about that, if the central bank should be less active in this, then there could be some very ugly surprises in store for people who are simply basing their allocation on trend with a scaling rule that's based on one divided by realized volatility.

That's why we focus so much on vol as a really structural hedge. And then what, so if I'm doing that model, I'm saying, cool, natural gas volatility is, well, I'm going to use dollar terms, right? So if I have $10,000 per contract, I have a million dollars, I'm risking 10 bps, so I'm going to do one contract. Somebody check me if my math was wrong on that. But, right, you get the point of like, I have a vol number, I get a signal i'm doing one contract based on that vol number you're saying that's dangerous right because that's just based on some abbreviated look back period of vol but then how do i fix that so do i extend the look back period do i just add some random uh numbers you're saying now you need to quantify what that um you can quantify quantify yeah otherwise do, do 50% of that.

Yeah. There's nothing better to do. And of course, the book doesn't cover every case. Positioning risk is hard to quantify in every case. Just don't do too much if you think you're dealing with a pegged currency or an artificially depressed asset. Don't go hog wild. Do less. Just do less.

And do you have any ones that you can share that are staring everybody in the face or that people are talking about, such as the Swiss peg or the short VIX ETFs or the British pound back in the day? Well, this isn't my idea, but I'd like to present it, which is that Asian currencies have been unusually correlated recently. In other words, the cross

Asian currency volatility, especially excluding the rupee, but other ones, they've been extremely stable. Now that could suggest political influence. I'm not going to go into that or geopolitical forces, but it also could suggest to investors that these are safe currency pairs. I would be more cautious, not because there is an edge in getting the higher yielding currency vis-a-vis the lower yielding one or trading against the megacountry. All I would say is that don't expect that to be persistent in the future without the potential of a left-tail event.

So there are things like that where there's political compression or geopolitical compression of exchange rates or geopolitical coordination of interest rates where one could see real blowouts. Now, I don't want to be the guy, the zillionth guy who comes out and says, oh, yields are too low. They blow up here and everywhere else. That's a hard trade. It's a hard trade for two reasons. One, because there's tremendous political incentive to keep yields low.

Also, because there's roll down. Even if the 10-year has a very low yield, if the five-year has a much lower yield in percentage terms, you're fighting against the tide. Yeah. Still, I would be very cautious about these. And three had been wrong for 20 tide. Yeah. Still, I would be very cautious about these. And three have been wrong for 20 years.

Yeah. I don't blame them. Yeah. I'm not going to fall into that trap. I'm not the pundit who's going to come out and say, oh, it's all going to blow up next month. But I'm just saying there are a lot of structural risks. So don't scale according to Markowitz or risk

budgeting. And it seems to me, my brain just rolls with all these possibilities, right? Of like Michael Burry and the credit default swaps, right? Or the mortgage-backed loans, right? A similar thing. There's these huge agents playing this game if it just trickles down a little bit it's going to kick off this cascade where all those have to default so right there's a million examples like that but a lot of the examples are people played that on offense and you're kind of saying just consider them if you find them why not play them on offense but especially be concerned about right and a lot of people lost a lot of money trying to play that short housing, right.

Cause it persisted for a long time. So would have been wise to be aware of that risk. And we're in a very weird period in the markets where the standard strategy over the years for me was sell the, or sell insurance against the risk of a moderate down move and massively buy insurance against the risk of a moderate down move and massively buy insurance against the risk of a mega move. But people have calmed down to that. So they are bidding up the tails quite a bit.

I think Nassim Daleb and various others have been responsible for that. But conceptually, that's the right way to play it. If things are going to break, they're going to break very badly. So currency pegs are a great case study for this sort of thing. The question is, as a money manager, can you manage that position? So if I sell nearby insurance and massively overbuy faraway insurance, saying that if the currency peg breaks, it's going to break big time, and it breaks somewhat and i'm sitting at my desk on a given day or i have a rule in my system what do i do then because if nothing happens there i could actually lose money on the hedge so betting on the extreme

tail event is requires a lot of skill yeah but it's also just it's back to nigel and we'll put a link to that but right just be positive skew right like that kind of takes away any of these outlier move risks um right if you just have kind of a a bent towards i'm going to capture these asymmetric moves instead of you know sell those asymmetric moves. Exactly. I hear that Nigel has a meditation. Yes. Get into it. With a name like mine, I think I could probably

read mine out for a little bit more, but more power to him. I think you've mentioned before, but like main takeaway you want, you want people to get out of the book. The main takeaway is that the major drivers of market movements nowadays are leverage and positioning risk. Those are the twin heralds of risk. If you don't have a handle on them,

you need to be wary about the way you put the way you put your money into the market, because those things can rear their ugly heads at any time and the fact that there are more episodes of large spikes in risk from nowhere should make people cautious now that doesn't mean they shouldn't be in the market being in the market's a pretty good idea i think my green has said a lot of things about the impact of passive flows on the drift of the S&P. The median returns have gone up. But long plus hedge or long plus a genuine diversifier is a good way to play. You don't need to do it through tail hedges.

You can do it in other ways. But diversification is not working, will not, in my opinion, work as well as it has over the years. And the reason is that we live in a flow-based marketplace. And anyone who doesn't understand that is vulnerable to getting everything go down in size at the same time. Maybe it will bounce back. Maybe the Fed will step in. If you're in that position, if you're in that seat, you might not feel too good about your position at the time. There's the old saying that every endowment has a 50-year horizon. Does the guy sitting in the seat have a 50-year horizon? No.

Well, apparently last year they do because they're all posting 30 40 50 60 percent returns right but that's that's the crazy thing so um the a lot of people are they are we too clever by half i'll say right like is this too smart for its own good of like when all the winners are right question yeah right of like hey wash you put up a 56 saying who cares maybe the flow is why they are all in on equities you know i'm not going to discount how smart they are but maybe they're saying yeah all these macro agents are pushing things to the moon and we're going to be fully invested and take advantage of that well there's a world of average returns and there's a world of compound returns. And maybe the ultimate,

the final frontier is diversification between the world of average returns and the world of compound returns. So I work for a firm that does a lot of machine learning. We always look at average returns in the ML stuff. Why? Because average returns are not susceptible to leverage alterations. If you have a positive average return, there is some level of leverage that will make that a good strategy if it's sufficiently positive. But hedging is the world of compounded returns. negative drift strategy that saves your, you know, saves your skin when things get really ugly and also allows you to compound more aggressively using the hedge as a way to access capital when you need it. And I think, you know, I haven't given this speech before, but it just kind of, I think it is a good one, which is that diversifying between the world of average

returns and the world of compounded returns is the ultimate form of diversification and that's really what one needs to be focused on yeah 56 whatever i won't guess what they did maybe 56 of it is unlisted assets right that they had marked themselves. Who knows? More power to them. But if it is, A, all of that stuff has liquidity risk, and B, they're not focused on compounded returns perhaps as much as they should be. Now, I agree that you do need to follow the flows. I don't resent anyone saying, look, all this money is flowing into passive, all this passive is flowing into large caps. The large caps are accelerating. Fine. There's no reason not to follow that, but that cycle can reverse pretty viciously. And explain real quick what you mean just by average returns versus compound return.

I think what you mean by average returns. Let's say that I had a strategy that made 1% if I was right and lost 0.9% if I was wrong every day. It was a coin flip. That's a winner, right? That's a winning strategy because my edge is 10 basis points and I'm on that trade. Well, it's 0.5 times 10 basis points, but anyway. But if I gear that thing up, if I do it five to one or ten to one i'm going from one percent up on up days to minus 0.9 on down days i'm going from that to 10 up days

and nine percent down days that's a losing strategy because the down days require a bigger return to dig out of that hole. So the world of looking at the median outcomes is the world of average returns over whatever horizon you trade in. It's not the world of compounded returns. You need to find an edge over in the average return world. And that's what works. Again, I'll quote Taleb in Mediocristan. It doesn't cover the tails. The tails have to be covered elsewhere. But that's the only way to do model building.

But it's not the way to think about managing the structural risk in your portfolios. I'm always open to questions. So, you know, if you want to pop something on Twitter or anything else. Yeah, you're sort of new to Twitter. When did you come on board? Last year? Yeah, in 2020. In 2020, right in time.

2021, sorry. Yeah, so it's been less than a year. Yeah, it's been a good follow. So follow them on Twitter. Where do they get the book? How do they get the book? The best deal is on Springer. Springer has a 25% discount for all books until September the 30th.

Time is short. Awesome. Well, go, everyone, read the book. Follow Hari. You won't regret it. Dish in some of the best knowledge out there in the vol space and just one final comment for me is the acknowledgement to you and uh Taylor and Jason was heartfelt because you guys have really been thought leaders in the space

so thank you guys appreciate it yeah um with some other good company to be in there i think so yeah uh well all right great thanks so much we'll see you whenever you get out of your house i don't know uh we got to get out that way and visit each other one of these days always welcome thank you john all right thanks so much the derivative is brought to you by CME Group. CME Group is the world's leading and most diverse futures and options exchange. For more information and educational resources about futures and options, visit cmegroup.com. You've been listening to The Derivative. Links from this episode will be in the episode description of this channel.

Follow us on Twitter at rcmalts and visit our website to read our blog or subscribe to our newsletter at rcmalts.com. If you liked our show, introduce a friend and show them how to subscribe. And be sure to leave comments. We'd love to hear from you.