Today I speak with David Sun, a retail trader who started his own hedge fund.
Coming from a non-traditional background, David takes a non-traditional approach in his investment mandates. Focused on selling options to capture the volatility risk premium, David believes that markets are ultimately efficient and therefore foregoes using any sort of active signal. Instead, he focuses on explicitly controlling his win size relative to his loss size, and then choosing a strategy with a win rate that bumps him into positive expectancy. By then maximizing the number of “at bats,” he lets the Central Limit Theorem take care of the rest. It’s an approach he calls “expectancy hacking.”
We discuss this approach in both theory and practice, addressing issues such as trading costs and slippage drag, as well as both sequence and event risk. David’s approach is certainly non-traditional, but highlights some unique concepts of how traders may be able to architect a payoff profile around a risk premium.
Please enjoy my episode with David Sun.
Corey Hoffstein 00:00
All right 321 Let’s dance. Hello and welcome everyone. I’m Corey Hoffstein. And this is flirting with models, the podcast that pulls back the curtain to discover the human factor behind the quantitative strategy.
Corey Hoffstein Is the co founder and chief investment officer of new found research due to industry regulations, he will not discuss any of new found researches funds on this podcast all opinions expressed by podcast participants are solely their own opinion and do not reflect the opinion of newfound research. This podcast is for informational purposes only and should not be relied upon as a basis for investment decisions. Clients of newfound research may maintain positions and securities discussed in this podcast for more information is it think newfound.com.
Corey Hoffstein 00:50
If you enjoy this podcast, we’d greatly appreciate it. If you could leave us a rating or review on your favorite podcast platform and check out our sponsor this season. It’s well it’s me. People ask me all the time Cory, what do you actually do? Well, back in 2008, I co founded newfound research. We’re a quantitative investment and research firm dedicated to helping investors proactively navigate the risks of investing through more holistic diversification. Whether through the funds we manage the Exchange Traded products we power, or the total portfolio solutions we construct like the structural Alpha model portfolio series, we offer a variety of solutions to financial advisors and institutions. Check us out at www dot Tink newfound.com. And now on with the show. Today I speak with David son, a retail trader who started his own hedge fund. Coming from a non traditional background, David takes a non traditional approach in his investment mandates focused on selling options to capture the volatility risk premium. David believes that markets are ultimately efficient and therefore forgoes using any sort of active signal. Instead, he focuses on explicitly controlling his wind size relative to his loss size, and then choosing a strategy with a win rate that bumps him into positive expectancy. By then maximizing the number of at bats. He lets the central limit theorem take care of the rest. It’s an approach he calls expectancy hacking. We discussed this approach and both theory and practice addressing issues such as trading costs and slippage drag, as well as both sequence and event risk. David’s approach is certainly non traditional, but highlights some unique concepts of how traders may be able to architect a payoff profile around a risk premium. Please enjoy my conversation with Davidson. Davidson, welcome to the program. This is gonna be a fun little episode. I think most of my listeners are used to flirting with models being a podcast where we try to get folks who are experienced industry professionals. But last season I had Darren Johnson on who is a retail trader, very professional still but not doing investing for other people trading his own money. You are Darren Johnson of this season, I think, coming at the world of investing from a very non traditional approach, forging your own path, really excited to dive into some of the ways in which you’ve done this sort of in an unorthodox manner. So excited to get into it. So David, welcome. Why don’t we start off with your background.
David Sun 03:39
Cory, first of all, been a longtime fan of the show up and falling out episodes really enjoyed it been highly influential in my own trading. So just want to say it’s a honor and privilege to be a guest. So thank you for having me. As far as my background, like you mentioned, I don’t have any formal finance pedigree or background, formal education at all. In fact, my backgrounds in electrical engineering. So how I got into Options was when I was at grad school, this was around 2008 2009 course that’s a period of time when the markets been on everyone’s mind. So I somehow just decided, hey, I want to get into stock market. And back then my investment thesis was watch Mad Money. See what Jim Cramer mentions, buy those stocks, so totally uneducated, that a buddy at grad school who was in options trading. So when he found out that I was trying to get into market, he taught me basically show me about options. And interesting thing is he taught me options from the short premium side selling options. I know most people who get into it more learn more options and from buying options, like as this lottery ticket kind of thing. So I learned it from the short side. And again very unsophisticated back then I didn’t know delta from the data. My plan was sell monthly options on five different stocks again, probably once that Jim Cramer mentioned and it was like okay, If I can get X amount per month, I just got to sell these and I’ll make X percent a year. The ironic thing is that a work for couple of years, I’ve doubled my account over two years or something. So it was working great. And then at some point it stopped working and the stocks that shows all tanked and then gave back a bunch. So I got a little bit disillusioned and got out in market. Coincidentally, because I took my money out to buy a house, my first house. But when I started my first job, this was my engineering job, my buddy there, he knew that I was still an option, because I talked about it a little bit. And he was looking for ways to grow his money, because he had some fun with some manager and he wasn’t really happy with performance. So did a lot of research options. Tasty trade came up. And as you guys your audience Berman, I know tasty trade is this online, financial, more retail oriented, educational, financial network that has a whole bunch of content, and really can be a huge source of knowledge. Although sometimes I’ll do much like drinking from a firehose, almost, but found that, and this was around 2017. And it really accelerated my learning curve. So reignited my passion for options, got back and went headfirst, tried all kinds of stuff, learn everything, tried different strategies, and finally started making progress and getting better. Now, fast forward to mid 2018. This was actually only about a year later, I must have done a lot that year, I was getting more confident. And I don’t know what got the idea in my head. But I was funding like, what if I’m successful doing this for myself? Can I do it for others. And again, I don’t know how that idea of having a hedge fund came up. But I was like, let’s launch a hedge fund. At that point, I had no idea what hedge fund was, I was doing research on how to start one, I started my path to become an RA, which obviously is not what I wanted to do. I started studying for my series six, and got the book and everything about the schedule of the test. And I stumbled upon a Facebook group for RAS. And I just randomly posted, I was like, Hey, I’m not sure I’m in the right place. But I’m actually trying to start a hedge fund, instead of joining or starting RIA firm. And coincidentally, somebody there in the group. He had been a hedge fund manager. And he has switched to the RNA side. But he kind of knew the industry and connected me with an accountant, a CPA, who ultimately I started working with. And he was like, Okay, this is what you got to do. And that got me to the right service providers and everything. So basically launched my first hedge fund in late 2018. So ran that for Well, I’m still running that, and then actually started my second hedge fund in early 2021. And then fast forward to now. And here we are, we got connected sometime last year. And yeah, after hearing Barron’s episode, I suppose it that kind of intrigued me because your audience mostly has that institutional professional tilt. But having somebody that was not the same kind of background, I was like, hey, might be interesting for me to check it out to see if I can share some ideas, especially since my trajectory is also quite different.
Corey Hoffstein 08:07
I always love the non traditional path into finance, because I think so much of what we believe philosophically comes from our backgrounds. So people, perhaps who have unique backgrounds are able to see markets in a different way and have different ideas. And we’re certainly going to explore that in this episode. Starting off though at a high level, can you describe the two mandates that you’re managing and the core theses behind them?
David Sun 08:30
One of my funds basically takes a return stacking approach, which obviously that’s a term we’ve done a lot of research on and ideas that you too have them too. And the idea is just to have a core portfolio, and this is a pure beta component. And that core can be whatever you want. I have a certain blend of index funds that I use. And let’s call that the benchmark. And on top of that, I’m using various income strategies as an overlay. To be sort of the alpha component. I think we spoke before you call it a portable alpha, which is kind of interesting term. So taking this overlay, and just stacking on top of your core portfolio. The mandate is simply to outperform the core. So I’ve called my fun, enhance index fund, or enhance beta, if you will, almost a well in fact, it is a buy and hold combined with an active overlay. And that’s the first approach. And then the second one, which was the one we launched later, we stripped out one of our options strategies, which was a zero DTE strategy and the unique factor about zero DTE which stands for zero days to expiration, that’s intraday options with no overnight no risk because everything closes out by enter day. So this second one is pure 100% Zero DTE approach. So the idea for that is just to generate a completely non uncorrelated return with very tight risk management, very low drawdowns and just a pure alpha or absolute return approach,
Corey Hoffstein 09:57
aiming to more fully round out the Picture Can you describe from an implementation perspective how that zero DTE strategy works in practice? What are you trading? What are you looking for intraday? When are you trading intraday? How are you thinking about exiting positions, that sort of
David Sun 10:16
stuff was there DTE were trading 100% SPX index options because those are very liquid. It’s a large notional product. So you actually get a reasonable amount of premium. Being that there’s very little time decay left. So we’re entering, put some calls. So entries Delta neutral, we don’t have a bias, and we put a stop loss on to manage the risk. And we’re entering multiple times throughout the day just to increase the number of occurrences and what zero DTE it used to be for a few years ago, from 2017 and onward, Asterix had three times a week, so many occurrences a year now, recently, CBOE, they released the Tuesday expirations. And then the Thursday’s ones coming out next week, and Becker just got listed today. So there’s a large number of opportunities that we can have, and what multiple entries a day we can have, basically a larger opportunity set. And just to clarify, this is selling options again. So the goal is to harvest that time decay. What makes it unique, again, is that positions are closed out by the end of the day. And the risk management we apply with just using a stop loss. But that keeps the tail risk in check. We do have one other long option strategy as well. And that’s more of a tail risk or hedge because when you’re talking about intraday, you can get pretty large moves. And these intraday black swans that happen more often than you think, because it’s essentially a compressed timeframe. So uncertain days, where you get this big rip, either up or down, we can flip to the long side and capture some of that convexity as well. But the main distinction, again, is the fact that these are zero dated options, meaning they expire that day, everything’s gonna go to cash or basically flat by the end of the day.
Corey Hoffstein 11:59
This is where I think things get really interesting because you describe this process to me as having absolutely no signal to it. You’re not making entries and exits based on some sort of conditional market signal, you view it as a pure probabilistic play. I was hoping you could explain that to me.
David Sun 12:21
One thing I want to point out this no signal approach, this is by choice. And when you say signals, what does that mean? People talk about some kind of Moving Average crossover, or looking at RSI or something that gives you a feeling that you can predict direction. Now for one, we aren’t biased, like I said, we entered Delta neutral at entry. Now, it doesn’t stay Delta neutral depending on which way the market moves. But the idea is we’re gonna enter Delta neutral, no matter what the markets doing, either. And it looks like it’s trending or doesn’t look like it’s trending, depending on volatility is we’re just going to enter the same number of entries every day. And it’s basically like, if you have a loaded die, you know that the odds are, you’re going to get one through four, a certain percentage of the time, let’s say one through four is a win, and five and six is a loss, you have a higher chance of the ones before. So we just want to roll the die as many times as possible. And again, with the higher frequency of expiration cycles now and the frequency at which free entry, we can get a few 1000 occurrences on the same strategy in a given year. Because normally, with longer data strategies, depending on the risk management and how long you’re in trade, you’re in a position to enter 45 days, exit 21 days later, you’re gonna get so many occurrences in a year. But with something like the zero dated option, we get a huge number of occurrences. And really let the central limit theorem and the law of large numbers that can actually really play out.
Corey Hoffstein 13:49
You describe this in your own podcast that you host, as well as on a pre call we had together you describe this idea as quote, expectancy hacking. That was a unique way of phrasing what you’re trying to accomplish with your approach. Can you expand on that idea what is expectancy hacking,
David Sun 14:10
when we talk about expectancy, or EV, it’s basically a function of three things, your win rate, your win size, and your last size. And when size the last size is basically risk reward ratio. The thing is, I found, especially for newer traders, maybe even more experienced ones, a lot of people focus on the wind break, they want high winrate or they want a high hit rate. And there’s various reasons for that, but everyone wants to win. No one likes the feeling of losing, and you just feel good about closing trades and putting wins on the board. But ultimately, if your losses are too big, it doesn’t matter how high your win rate is, you’re gonna have negative expectancy. So if you apply rigid loss mitigation, and let’s just use a simple example. If you have a trade where you’re risking $2 To make $1 lose two on a loss make one on a win, I’m not counting for slippage or fees are just perfect to the one, you’re going to have about a 66.66%. Break even win rate. So if every three trades, you essentially went to and you lose one, you’re going to net zero. If your win rate is above that, you’ll be positive, even if you’re below that you’ll be negative. So the idea is using risk management, we’re essentially fixing the win loss ratio, you can never determine all three variables. But rather than focusing on trying to somehow fix the win rate, we fix the risk reward ratio, and then design or enter trades that have a probability of winning higher than that break even win rate. That’s what I call expectancy hacking. Again, you can’t control all three variables. But if you can control two of the three, then your net profit net expectancy is only determined by that one. Options are kind of a probabilistic instrument anyways, generally speaking, lower delta equates to higher win rate. So if I have a stop loss at x level, and I’m entering trades at 30, delta 20, Delta 10 Delta, now you got to test this out. But basically, the lower the Delta, the higher the win rate, generally speaking, and at some point, if you go low enough, that win rate is going to peak above what you need for breakeven. And at that point, you’ve essentially, last time we spoke, you use the architecture or architected this payoff profile. And that’s what I mean by expectancy hacking, you’re really focusing on two aspects of the expectancy equation are easier to fix, and can control and then letting the cards fall where they will, on that third variable, which is the win rate.
Corey Hoffstein 16:48
So focusing on that win rate for a second, another key feature of your approach, you’ve called, quote, premium capture rates, and you’ve actually described as being the single most important factor to your approach. So again, I was hoping you could expand on what is premium capture rate? And why do you consider it to be so important,
David Sun 17:10
we’ve focused primarily on premium selling. So when you’re calm about purely premium selling strategies, or selling options, premium capture rate is synonymous with the expectancy. So the term should be fairly self explanatory. It’s the amount of premium you capture or net, after accounting for all losses for fees, commissions, slippage, whatever it is. So it’s basically if I sell $1,000 in premium, and at the end of the day, after all the trades, I’ve netted $250, that’s a premium capture rate or PCR of 25%. And the reason that’s important is because if you’re running various strategies and try to compare metrics, or let you compare strategies, on an apples to apples basis, now there are some limitations, obviously, because you have to keep in mind different contexts. If you’re comparing a super low delta strategy to a super high delta strategy. Basically, you can’t always expect the premium captured be the same. And what I mean is, if you’re doing a very high winrate strategy, you’re probably getting not a lot of premium on an option, because you’re going forward in the wings, now you might have a higher PCR, but that p&l In absolute terms may not be greater than if you went and sold a at the money strategy, a straddle, and you were able to capture 10%, because a high PCR of a smaller pool of premium may not be more and that flute terms than a low PCR have a huge pore premium. So you just have to be aware of the context when comparing. So it may not make sense to compare the PCR have a really high delta to a really low delta strategy. But as long as you’re on that same regime and let you have a normalized factor, and it gives you something to shoot for, if I know that based on a testing, I want to capture, let’s just do 25% As an example, that’s like my golden standard. And as I’m live training or for testing, and let you gauge Am I above or below my typical expectancy on a strategy.
Corey Hoffstein 19:13
I want to talk a minute away from the theoretical and more towards the practical because when you trade in options, you’re trading in a instrument that has much wider spreads than say something like cash equities. So we have this theoretical concept of the win rate, the wind size, the last size you’re trying to fix to and then choose a strategy such that the third bumps you into positive expectancy, but the actual implementation, realities are going to move that wind size and loss size, potentially off of target. So how do you account for things like spreads and slippage in your premium calculations and how you try to minimize them from an implementation risk perspective?
David Sun 19:56
From the calculation standpoint, you just bake it into your or math, I use the example of a perfect two to one risk reward is going to have a certain breakeven win rate. With slippage, your loss might become more like 2.2, or 2.3. All that’s going to do is shift what’s required to break even, it’s basically going to pull all of the EVS down. So it’s not that you can really do anything about the math, it is what it is, obviously, you have to hit a higher win rate, so to speak, to get the same premium capture after you’ve accounted for the slippage. But from an implementation standpoint, that’s one reason why we basically only trade SPX index offers because liquidity you want to focus on a liquid instrument to minimize the slippage. On the other hand, we don’t use market orders anymore, because depending on market conditions, it can be wide, you have really bad slippage. So you can use algorithmic orders that can prove the execution you can use various limit orders, even Interactive Brokers, I don’t know if that’s more concerned, retail or professional, but they have technology and execution algorithms that anyone can use that are available that can help with execution. And that’s one way to minimize the slippage. Just be smarter about getting a good fill. Basically, you just got to bake it into the map. I’ve heard about stop orders and options. And there’s always the warning of okay, never use stop, never do this options are illiquid this one work? Well, it won’t work perfectly, but it’s just about understanding, okay, how do they work, and then taking what you know, and just applying that to the system and just adapting. That’s really our approach.
Corey Hoffstein 21:33
Maybe you can dive into that component a bit more, because it seems like a huge part of your processes, fixing that last size, which depends on stops and stops can have potential problems. I’m thinking in my head potentially, like situations where you’re selling deeper out of the money option. And as you’re losing, it should be increasing in Delta. That may be a scenario where it’s adverse for you to potentially close the position or perhaps it actually goes in your favor. I’m curious as to your experience with those stop losses. Do you risk gaps? Do you risk having to cross the spread more aggressively to close out the position which increases the loss? sighs how does that actually play out in practice?
David Sun 22:20
Let’s talk about zero DTE for DTS intraday strategies. So there is no gap, because we’re out of the market once the markets closed. So the flip side is there’s no gap and zero dt, but there’s gamma. So things move a lot faster. And to account for that really is just if you have a fairly tight stop, you can get out or hopefully try to get out before the gamma really picks up where the volatility or the Vega really hit you. In fact, if you ever look at the option chain, and these are defined by the CBOE rulebook, but the spreads are actually wider, when the option price itself is wider, and there’s certain limitations, they have to quote the bid ask spread. So if you’re getting into a trade at $1, and you’re getting out of it at two to $3, are you still in that regime, so the spread is consistent, and it’s not too bad. But if you let a trade balloon to like $10, then you’re gonna have five or 6x and the spreads even wider, and obviously that’s not good. But mainly with zero DTE gap risk, presumably is not present. And again, if you apply the execution algos that help with execution, and give that better price, then that can keep you safe. Now for the nonzero dt. This is why I’ve recently moved out and my longer data strategy is more into the 90 tt regime. Because when you’re far out there, that gamma curve is a lot flatter. So like you said, your deltas increasing as it moves against you, but it’s moving a lot slower. Not only that, at further durations, the true impact of Vega is a little bit more muted as well. So it’s not going to hit you as hard as you may think just based on what your brokerage platform shows in terms of the nominal Vega. And even the live trading I’ve done. I’ve been trading through COVID and everything and the backtest. I’ve done if I’m trying to limit my losses that to x, for example. Now under normal circumstances that might be 2.0 2.03 with the slippage in COVID. With the gaps even, I think the worst we saw was 3.3 3.4. So like an extra multiple, but it’s manageable. And again, that’s also something you just have to work into. Because if you have a perfect two point, whatever your numbers are going to show your average loss is going to be whatever that number is, but on a longer dated study, including those gaps that occasionally happen. That’s going to bump it up a little bit. Again, you have so many occurrences that more or less, they’re not going to skew things too much.
Corey Hoffstein 24:44
The premiums you would be collecting it say when VIX is 15 versus when VIX is something like 50 are going to be really different. And there might be this path conditional situation where you’re more likely to lose when VIX is 50 Then when VIX is 15, curious how you deal with the sequence risk nature of the strategy.
David Sun 25:06
This is the other really key factor to expect to see hacking, expecting this hack. And we talked about just fixing the risk reward ratio. But if you’re, let’s say, selling out a fixed Delta, when VIX is higher, you’re going to get more premium. And if you look at a lot of these backtest studies, it’s not necessarily that the win rate is going to change. Because if you believe in efficient markets and delta as a proxy for probability, you should get fairly similar win rates across all market regimes at a certain delta. However, if you just happen to lose the larger trade, he collected a boatload of credit and high vix and you lost that one and you only won the small ones, that’s going to have a terrible impact, it’s going to completely skew the probabilities. And like I said, that sequencing risk, now you couldn’t have it the other way around, where you win the big ones, there’s the small wins, but we don’t want to consider that because that’s just luck. This was precisely because during COVID, when we were trading, we were collecting a lot larger credits. And we thought we were safe, because we could stay at the same delta as we’re used to. But we would just take these huge losses, sometimes on a stop. So this idea of credit jargon really was born from the experience of trading through that market. So the idea behind credit targeting is I want to equalize the amount of premium I collect for every tray, people know, you should be consistent with sizing. But traditionally, sizing has been purely in contract sizing, or maybe in buying power, because the capitals equalize. But for me, because the losses are determined by my stop, which means that there are multiple of the credit, which means my credit is really the proxy for my risk. So if I’m trying to collect the $1,000 on a trade, you can collect two contracts at $500, or 10, contracts at $100, you’re still going to get that 1000. Now, for various reasons. If it’s that extreme, again, you don’t want to take too much leverage by going too crazy on a contract says But generally speaking, that idea of fixing the credit, that is what ultimately is going to minimize that sequence risk, because that’s going to fix that risk reward ratio. What’s interesting is what premium selling where people who learn options, because volatility is such a big component, we’ve been taught, high volatility is opportunity. high volatility is when you scale and, and I used to do this, it didn’t work that well, in COVID, again, and learn so many lessons. And not that it can’t work. If you have the bankroll and the buying power to manage losses, and to wait things out, you can take advantage of the high volatility. But because of the mechanics that I use of hard stops, they basically don’t fit with that kind of approach. Because you get stopped out and you’re done. You’re either the trader took a big loss. So by fixing that credit, and let you really smooth out that sequence risk and a mix each size, truly equal size. And one interesting aspect is when you have high credit, if you’re trying to target a fixed amount that actually lets you in a higher IV environment, you can go down and contract size to click the same amount of premium. Or you can go down and delta or both. So the way I look at it is, what happens is, I look at high volatility, not as an opportunity to make more, but as an opportunity to make the same. But hopefully with less risk. Because I can scale down, I can size down I can go further out and money. So that’s the byproduct of this idea of credit targeting as well.
Corey Hoffstein 28:39
So you’re very explicitly targeting this expected return and backing out the trade size necessary, sort of asymptotically approach that expectation over time. But it strikes me that by locking in the expected return in many ways you’re letting the risk vary. And I’m reminded of this idea of risk ignition by Aaron Brown, who’s the ex Chief Research Officer at AQR and the idea of risk ignition is that when you take too little risk, you’re ultimately failing to maximize the opportunity in front of you. And if you take too much risk, you’re gonna blow yourself up. So given that you are sort of locking in the expected return and letting risk vary to a certain degree in these different environments, at least based on how much leverage you’re taking on, how do you know what expected return to target? Is 5% safe is 10%. Safe, why not go for 20?
David Sun 29:32
Just to clarify, because I might have skimmed over a bit when you’re referring to backing things out. The point is when we’re approaching designing a system for sizing the trade, we take this top down approach. Typically people take a bottom up approach where they’re looking for different opportunities and different positions and different trades, and then they put them on and then they manage them. And at the end of the day, you have all these opportunities, you manage the risk and there’s always this phrase, make as much as you can or take what the market gives you now ultimately all of us take what the market gives us. But when I say a top down approach, using the PCR, and again, let’s use 25%. This gives us a top down approach where there’s a Million Dollar Portfolio and one and make 10%, which is 100,000. If I know I expect to capture 25% or premium, then I should sell for under 1000. Because I’m going to capture 25, which is 100,000. So at 400,000, or premium to sell, then back in that out, I just take the 4000 divide that by the number of trades I’m going to do in a given period of time, let’s say a year, and that gives you that credit target. So the idea of having a tour of the credit target lets you take a very high level, top down approach and take your end goal and back that all the way to the one step that you got to do over and over. Now, that gives you the approach the goal, but what should the goal be? That’s your question. So a lot of people talk about back testing. And back testing is a tool. Sometimes, one of the criticisms of back testing is, for one past performance is an indication of future returns. And of course, just because it worked in the past isn’t gonna work in the future. But back testing gives you context. And when I do back tests, I have a reasonable expectation to hit my target. But what’s the path? What I’m getting at is the drawdown. So the way I build my test is usually I’ll have some dynamic user input. So all the trades are there. But I can flip the target 5% 10%, whatever it is, and that’s going to dynamically scale all the trades. And it’s going to compound based on a net lick of every single day. And it’s going to scale dynamically. And I can look at the equity curve. And I’m looking at the max drawdown max drawdown over the entire period of time max drawdown over any given year. And you can just play with that. And essentially, at that point, it’s up to the individual trader, if your risk tolerance is x. And if you look at Oh, I can target 20% return with X percent drawdown that’s within your tolerance. That’s the first step. I say that’s the first step. Because the thing is with risk tolerance, as I was saying that you usually don’t find your risk tolerance, it will find you because you always think you can tolerate extra Nm. But at some point, it’s always gonna surprise you. But at least having this kind of approach gives you a guidepost and let you know where to sit your initial guardrails. But then once you run it, having that live feedback, but that’s the start. We look at the risk adjusted return and what kind of drawdown do you expect, based on a certain target return that you’re going to set for the whole system?
Corey Hoffstein 32:34
How do you think about fair compensation for risk? Because you might be setting the expected return. But it’s possible that your risk adjusted return is high during periods that you are selling when vol is expensive, and quite low when you’re systematically selling cheap vol. Without any signal. Why is this ultimately better than an approach where we’re just naively long other positive expected return asset classes like stocks or bonds.
David Sun 33:06
So there’s a few angles, which I want to talk about this because he talks about fair compensation for risk and quote, unquote, cheap of all, what is cheap, cheap relative to what, and if you believe in efficient markets, you would imagine that risk is actually priced in. Now, when we go forward, if the implied volatility is higher or lower than the realized volatility, no one can really predict that. So if our V is higher than IV, then the Evolve was cheap, in hindsight, and the opposite is true. First of all, if you believe in the ERP, volatile risk premium, and the edge of selling options, that has to be the fundamental thesis behind why we sell vol in the first place. And you can craft different strategies with short options, all options, but at the end of the day, they’re just different ways to try and harvest out that risk premium. So you just need to use your risk management because there’s a risk in theory versus risk and practice. If I sell $1 and premium on this high leverage private or high notional people talk about max loss, oh, you’re gonna risk 30,000 or 40,000, whatever it is, to make 100 bucks. That’s a hugely negatively skewed risk reward profile. But that’s risk in theory. Number one, is this stock or this index? Gonna go to zero? What is actually your chance of getting the $40,000 max loss is zero. I mean, it should be pretty close to that, especially with an index. Now, if you don’t use risk management, then what is the probability of X amount of loss and that’s just based on the probability distribution, but beyond that, we are using very tight risk management. So this is why it comes back to I say that we’re defining our own risk. The risk in theory is not the same as risk and practice. So I Do believe what the proper risk management, you can achieve stronger risk adjusted return. And again, when you talk about risk adjusted return, like what is the risk, we look at more like the risk and practice the actual risks that we plan to take. In the context of our risk management, we’re able to harvest out that little bit of edge. Nobody knows how much actual edge there is at any one time. And again, that’s all in hindsight. But it’s because you apply to risk management that you can extract that edge and, quote, unquote, safe manner. So whether or not is cheap, or not cheap, that’s not really something we look at, we just look at, okay, if we do this trade, and these probabilities hold up, we can have this expectancy show up.
Corey Hoffstein 35:44
Let’s keep diving into the risk side of the equation. The other thing that immediately jumps to mind for me, particularly in this market environment that we’re in in 2022, is event risk. There are very particular days on the calendar that typically represent higher event risks, things like elections, or FOMC meetings, certain economic data prints, how do you think about zero DT strategies on those days versus say, non event days, where maybe the tail risks might not be as great,
David Sun 36:17
this comes back to at least my belief in the efficient markets and how everything’s priced in on FOMC days, for example, and this is kind of timely, because I actually just went back and updated from our research on our success rate on FOMC days, FOMC. These are basically intraday binary events, because they have the speech at 2pm, or they released the minutes, so there’s always something going on, like once a month, and for one, you’ll see options hold their price until 2pm, the theta is just a lot slower. So on those days, the intraday obvious higher, you’re going to get further out of the money for the same amount of credit. And conversely, you can’t scale it as aggressively because essentially, this time comes to a standstill, everyone’s waiting for that event. So to that extent, I think that’s priced in. So having said that, again, we just take the approach of just play the odds and let the currencies come out. And from a tail risk perspective, we have tight risk management anyways. So that’s basically, from an implementation standpoint, really, we don’t see any major change on those events, because we’re going to have those risk mitigation components in place, regardless. And again, I mentioned this is timely, because this may be some reasons in bias, but we always thought about, okay, if these are particularly volatile days, and always felt like we’re actually not winning, it’s always a loss on FOMC days, like should we go back and just skip those days, or do something different or no days. And I finally pulled out our back test log and our live trade log, and just looked at those event dates, the days and where they do the speeches or the days where they released the minutes, and it looked like the win rate was slightly lower. But it wasn’t enough to compel me to sift them, because again, the way we construct it our risk reward profile. If that win rates above the breakeven, you’re going to make money. This was done kind of quickly. But again, looking at the numbers, it just didn’t really compel me. And this is another instance of you can always think of as a signal, okay, FOMC days event data, I’m going to skip this trade or do this or do this differently. So I know a lot of people, when you don’t look at data, it’s easy to have that bias. And even the last I think two or three FOMC days, we didn’t have a losing day. So like that recency bias kicks in. And that’s when we start going, Oh, should we skipped this? Or is there more risk, but really, we want to be data LED? So we looked at the data and like right now, I’m just not compelled to do anything different necessarily.
Corey Hoffstein 38:45
Another very common risk management technique. Probably the first that many people turn to because it’s the lowest hanging fruit would be diversification. And you have made the explicit choice only to run these approaches on the s&p 500 index using SPX options. Why not expand the basket? Why not trade this on multiple indices or multiple underlying stocks where you could get even more at bats?
David Sun 39:14
When it comes specifically to zero DTE? The main reason is because SMP or SPX has the most number of expirations, especially now with the five days a week coming up. So the most amount of opportunities and liquidity and one thing I touched on earlier with that we’re doing zero DTE, there’s not a lot of time left. So the premium with respect to the notional size of the product actually is quite small. So it is a leverage strategy. Because of that, a smaller product is going to have, relatively speaking with the same on premium, but when you don’t have that kind of skill, basically fixed costs come into play. If it takes SPX of 4000 Price product to get $100 of premium, then there’s going to be the cost cost of buying your wings to spit off the wrist is commissioned as a slippage. All of those are basically fixed costs. So when you have a smaller product, from a proportional standpoint, the probabilities may be the same. But once you add the fixed costs, that’s going to basically depress your expectancy because those fixed costs basically trumps everything else. So what’s the odd specifically, it’s just SPX has the most number of opportunities, and it has the product size that makes the trade actually work at scale. Now from longer kind of strategies, like what the 90 D, I’m not opposed to looking at other products, but for the research I’ve done and everything that’s involved with our strategies up until now, it’s been focused on that product because it’s something we knew. And number one, the liquidity, I’ve done some research into trying to get some uncorrelated underlyings like applying those to TLT or us O or some commodity ETFs. For one, I think the fact that they also don’t have the same number of expiration cycles, like even with spy right now, you’re gonna get a good amount every month and then once that comes closer, does the weeklies and everything. So having less expirations, it doesn’t allow me to spread out that sequence risk because we talked about the credit targeting with entries. But if I’m trying to enter daily, for example, and there’s only one expiration month, and let’s say my preference is enter a 90 DTE, what’s going to happen is if there’s only a monthly expiration, I’m going to enter a 90, but then 8988, all the way down to like 60 before the next one comes out. So you get a lot of positions basically a concentrate on the same expiration cycle. So that introduces a degree of sequencing risk as well. So there’s some constraints just as far as the opportunity set. Liquidity is one issue as to why that’s probably the most liquid ETF there is to trade and liquidity is a huge factor. We talked about slippage and using stops and everything. But again, I’m not opposed to it. It’s just I haven’t had the time to find something that works the same. But I think that principle of diversification it makes sense. And that’s something I’m going to continue to explore. Who knows one day, I may be running this on multiple underlyings.
Corey Hoffstein 42:11
We spent most of this chat so far talking about the zero DTE strategy, you’ve mentioned the longer dated 90 DTE. Obviously, those two approaches are going to have very different risk factors. The most obvious that comes to mind is there zero DTE is going to be heavily tilted towards the gamma versus a lot more Vega in your 90. DTE curious, what other risk factor differences are there between the mandates? And how does that ultimately affect strategy design,
David Sun 42:39
I think that really just comes down to the sizing. So with the fun I mentioned, which is return stacking, the beta of the whole strategy is already going to be basically one plus, because you have the beta of the core plus the beta of the option strategies itself, whereas with the zero dt is going to be more of a pure Alpha approach. So we can lean more heavily into the options strategy itself. Again, looking at the sizing in the context with what is your end goal, and what is construction of the overall portfolio because, again, what the return stacking approach, we’re taking mostly the US index ETFs. But you could build a core portfolio of whatever you want, you could have mix of bonds, Misaka, Mondays, or whatever it is. And that can be your core. But that’s going to come with a different risk profile. So if you chose a core that’s going to be less volatile, you may have the latitude to step it up in the sizing on the option strategy on that side. So really, it’s just the overall risk profile, and the deviation of returns, that is within the risk tolerance. For me sizing itself really is a mechanic a lot of people don’t really think about sizing as a mechanic. But I’ve come to think that that’s what it all comes down to sizing is the mechanic that really lets you steer the path to a degree, we can never control the exact path. But with the sizing with the credit targeting, minimizing sequence risk, that gives us some hand and kind of steering the path that we take to the end goal.
Corey Hoffstein 44:10
I want to come back full circle to the introduction in your background, you discuss how you’ve had to navigate your way as an outsider into figuring out how to launch your own fund. I was wondering if there are any lessons you could share for other individuals who come from non traditional backgrounds that are interested in launching their own hedge fund.
David Sun 44:30
For me, one thing that was surprising was it’s not necessarily that hard to start a fund the startup costs and getting into service providers. You can do an incubator fund, for example, which is one where you establish an entity for purposes of just having a track record but you don’t have any outside investors. And once you do the full launch, you’re going to have your overhead and you can take us investors actually starting up as a business isn’t necessarily the hardest thing to do, or the most expensive thing. What’s going on. be a lot harder than you think is raising enough capital to sustain your overhead or even make a living, because depending on the fee structure you do, you’re going to have to hit a certain threshold as far as the Aum. So I think it’s having realistic expectations, I probably should have done this more, but maybe asking around in your sphere, depending on who you’re going to try to gather assets from getting pre commitments or something, so you have an expectation of what you’re going to start with. The other thing is, depending on your audience, because for me, our investor base right now is still basically 100%. Retail, we don’t have the scale for really institutional investors. So that audience base, they’re going to perceive what you do, and how you explain your strategies differently. So if you’re primarily presenting to a lay audience, what they prioritize, and what they want to know, and don’t want to know is going to be very different. I would imagine, if you’re trying to pitch to an institutional allocator, they’re going to do a lot of due diligence, they want to turn every stone over and know where the risk is exactly. And everything, every single little detail, but lay audiences, they’re not really gonna be able to understand that anyways. So ironically, what it really comes down to is just, they just want to be confident in you as a person, and that you’re intelligent and understand what you’re doing. It’s almost more like that human side of it almost is more of a factor. It’s know your audience, and know how to play to that in terms of when you’re trying to showcase what you’re doing. Because I spent a lot of time in the beginning, had a lot of I thought was kind of dumbed down PowerPoints and graphs and everything. And a lot of that really just went away. And it was more just talking about high level. What’s my thesis? What’s my outlook? What’s my investment philosophy, rather than the deltas and the datas. And everything? Those two things were, I think, two things that I learned as I went on my path, because I don’t think I really had enough AUM to even cover my expenses for the first two years. Again, maybe I was little naive in the way I started. So suddenly, and I just rip the band aid off and as because I did an incubator. And I was like, You know what, I’d rather have a live track record with a little bit of investors than having a longer track record with no investors. For someone that just felt different from I don’t know if that’s true or not, but it was just different. I have actual outside capital, I’m gonna focus a little bit.
Corey Hoffstein 47:29
Well, David, we come to the end of the episode here. And the question I’m asking everyone at the end of this season is to reflect back upon your career so far. And just think about what was the luckiest break in your career that you’ve had,
David Sun 47:45
I think the luckiest break I had in my career. And if you’re talking about specifically, the trajectory with launching a hedge fund, and the progress of that really was the timing of market conditions. And this was pure luck. If you remember I said I started in 2018, q4, that was not a good quarter to really be doing anything with volatility because we had that really large drawdown. So I went out the gate with a large drawdown, but it recovered in 2019. And 2019, was a fairly, quote unquote, easy year. So I was able to immediately redeem myself, like, okay, these strategies actually do work. So 2019 ended up being a good year in terms of performance. And then 2020 came, that was, of course, the monster wall, Black Swans because I was in the market, so to speak to in 2008, but not really actively engaged, it’s just dipping my toes in. So having that experience of 2020. And having gone through 2018, I had a re tighten up my risk management a little bit because of that experience. But 2020 was more lessons learned more adjustments to the strategy. And then somehow by miracle 2020, ended up being a fairly good year as well. And then 2121 And And now 2022. So the fact that I was able to learn those lessons at the right time, because if I had started right in 2020, or didn’t have that experience of 2018, I could easily have maybe gone on too big or going to respect the risk as much. So I don’t know if that’s the answer people want to hear, because that’s not something you can really teach. It’s just having the dumb luck of that timing. Everything worked out to teach me just enough. So I learned from it but not hard enough lesson where I couldn’t recover from the big drawdown or whatever.
Corey Hoffstein 49:34
Well, that’s precisely the point of the question is, it’s what’s the dumb luck that when we reflect upon ended up being some of the best parts of our career So David, thank you so much for joining me. This has been a fantastic episode really enjoyed getting to hear sort of your very non traditional approach to attacking these markets. And I think there was a lot of great stuff to come out of it. So thank you.
David Sun 49:53
Thanks again for having me.