Matt: Welcome to the channel, Abhay. Thank you for making the time. I know you’re all the way out in Australia, and for those who do not know who he is, he is Head of Product for Nerp. Today, he’s going to be going over a few interesting things. So, welcome to the channel!
Abhay: Thanks for having me, Matt.
Matt: Now, I want to talk about… I see you have an algorithm open right now. Can we talk a little bit about this one? And then I want to go into the news that is circulating about a new potential product that you guys are rolling out. I’d love to dive into what that looks like, the returns, and how promising it looks on your side of things.
Abhay: Yeah, for sure.
Matt: Okay, cool. So what’s going on with the Dalio? What do we have open here?
Abhay: Yeah, so this is a new algorithm that we have coming. We’re very close to completion, really in the final stages right now, and we’ve added a few really cool features in this algorithm, which we’ve been working on for quite some time. We’re quite excited about it.
Matt: What are some of the new features?
Abhay: Well, firstly, it’s very, very optimized. So, we’ve gone back and looked at nine years of data and performance for each of the pairs that we’re trading. We’ve really fine-tuned the settings to make sure that they respond the best to that pair. We’ve also made sure that we don’t have a lot of correlation among the pairs that we’re trading, so we’ve optimized the trading basket as well.
Matt: So when you say “correlation,” you mean that it’s not going to take opposing positions on pairs that are heavily correlated because that wouldn’t make sense?
Abhay: Well, in a way, yes. A simple way to think about it is that if you have 10 trading pairs, and eight of them are correlated, that’s a lot of risk concentrated in the basket.
Matt: Okay, yeah. Is there any factor in there that weighs something that wouldn’t make sense? For instance, you can’t long gold and long the dollar. It just doesn’t work out. Is there anything that filters that by any chance?
Abhay: Yeah, and we also look at currencies that typically have economic policy relationships. For instance, the Canadian Dollar (CAD) and AUD tend to follow USD. So, if there’s big USD news, especially related to interest rates, that can introduce a further risk in those pairs.
Matt: Awesome. And, um, so first of all, the Dalio is not a new product. It’s been around for a little bit?
Abhay: Correct. This is a new rollout of that older product.
Matt: Got it.
Abhay: Correct, yeah. We had “Dalio Low Drawdown” and “Dalio High Profit,” which are two configurations. The high profit one was a bit aggressive, so we never ended up launching it. But we wanted to make sure we could still get really solid monthly performance while keeping the drawdown within our objectives. So, this is a blend of the two—highly optimized. The other big feature that we’ve introduced is two forms of hedging in the algorithm.
Matt: Two forms?
Abhay: Yeah, can you explain that?
Abhay: Yeah, the first is just being able to go long and short concurrently on the same pair at the same time, which we had in other algorithms as well. The other one is a more advanced form of hedging, which only kicks in during very strong trends. Because we’re scaling into positions in strong trends, this new form of hedging really protects the account against extreme drawdown and provides a much bigger cushion for the algorithm to continue scaling and make sure the original sequence closes in profit as well, despite being in a very strong trend.
Matt: Interesting. And when will this be accessible to the users or anybody looking to get it?
Abhay: The algorithm is in the final stages of testing right now. We are planning to release it to a limited set of our Mastermind High Tier clients this month. After that first initial lot of testing, I would say by Q1, we should be comfortable releasing it. So, Q1 2025, we could expect the Dalio to be fully accessible to not only existing members but also new potential clients.
Matt: Nice! Well, okay, so the Dalio is a little bit of old news. This gets into the exciting part—the new products! Well, I said “plural,” so how many new products are there coming out?
Abhay: So, we have one new product with three new algorithms that are coming out.
Matt: Okay, so it’s like just a suite, and then it has alternative ones that people could license? What are those?
Abhay: Yeah, so we have Odyssey, Buterin, and Argos as the three algorithms in that offering. It’s a completely separate product, just because the way it’s structured, the offering, and how users interact with the product is quite different from what they’ve been used to with ATA. Hence, why we packaged it under a different product.
Matt: Oh, interesting. And what is the biggest thing that differentiates it from our other products?
Abhay: A few things. Firstly, the product itself has been built more based on expertise for institutions. So, the team that has really worked on developing the product. It’s a blend of our team and the strategic partnership that we formed out in the marketplace with a company that had been servicing a lot of institutions. They come from a very heavy AI focus, so the algorithms are extremely, extremely well—almost all of it is AI-focused and leverages a lot of AI. It was a really natural fit between the two teams because we come more from a technical, market analysis understanding of the charts and price action, whereas our partners come from more of an AI background. So, it was a really good blend of the two, and our objective is to have a long-term strategic partnership to deliver industry-leading products in the marketplace.
Matt: Based on what you’ve seen, how long have these products actually been running?
Abhay: Yeah, so one of them, which is Odyssey, is FIFO-compliant. So that’s the other thing…
Matt: Can you… well, let’s pause there for a second. For those watching, what does FIFO-compliant even mean?
Abhay: It means you can run it on U.S. brokers because the majority of U.S. brokers require FIFO-based accounting, whereas outside of the U.S., it’s LIFO—Last In, First Out—based accounting.
Matt: LIFO, yeah.
Abhay: So, the algorithm has to behave slightly differently to account for that.
Matt: And what is the upside to making it FIFO-compliant?
Abhay: A few upsides. One is you get to run it onshore with local U.S. brokers, which typically have a lot more regulation, or at least regulatory oversight. So, they tend to have a lower risk profile when it comes to the safety of your funds. And then secondly, these algorithms also don’t need as much leverage. So that’s the other thing—because in the U.S., typically it’s 50:1 leverage, so you can’t really get above that unless you’re a very, very big institution, not at the retail level.
Matt: Interesting. Okay, wow. So now that we understand what FIFO and LIFO are and what all this means, it’s basically lowering the third-party counter risk of who’s your custodial service.
Abhay: Yeah, exactly.
Matt: Okay. Now, back to the product. It’s heavy in AI. Does it only use AI, or is there any other type of information that you could give on how the product operates?
Abhay: Yeah, so it is mostly, like I said, an AI-based approach, where data is fed into AI models, which continually update and adjust based on how the market is reacting.
Matt: And how has that played out? You mentioned Odyssey earlier, and you were talking briefly about how long it’s been running. Can you explain what that track record looks like over what period of time? And also, lastly, I know it’s a jam-packed question— is it running with live capital in there right now?
Abhay: Yes, so the track record we have is since 2022, and it’s been averaging about—if we just talk numbers, of course, that’s historical performance, and it doesn’t guarantee future performance—but so far, about 6% per month. On Odyssey, it’s less than a 5% drawdown, which is quite exceptional. So, you have a higher monthly average and a lower drawdown. Typically, what we see across the board for algorithm companies is higher drawdowns and lower monthly averages. That alone is a really big, promising thing.
Matt: Yeah, that’s huge.
Abhay: And it’s also positive pips because a lot of algorithms that we see out in the marketplace don’t capture a lot of pips in terms of percentage point movement of the price, which means they typically have to scale in a lot, which creates a lot of drawdown. Whereas here, we have thousands and thousands of positive pips captured, which our clients will be able to see in the track record. And that means we don’t actually need to scale in. The algorithm is very active, and it doesn’t need to scale in as much. There’s also very robust risk management, so the max drawdown is 3.5% per pair and 30% on the whole account.
Matt: When you say max drawdown, what happens when it hits that? Potentially even if it does or doesn’t—what happens when it hits that 3.5%?
Abhay: We just close the trades on that particular pair. The person will close it or the software will close itself out.
Matt: The software will close it?
Abhay: Yeah, so the person doesn’t have to be there, having their hands on the computer and exiting stuff at 3.5%.
Matt: Yeah, exactly.
Abhay: Okay. And also, the way the product is—these products are structured, or these three algorithms—they’re actually a collection of many, many subsystems. So, it’s a combination of many sub-strategies forming a master strategy. And that also lowers the risk because, like I said, if one particular pair has a huge news event, the algorithm will cap the drawdown to 3.5%. However, if a highly, highly unlikely – I mean, we’ve never really seen this in the last 10 years, or even longer, where five or six or seven different pairs have drawdowns concurrently. But even if that happens, the max drawdown is still capped at 30%.
Matt: Interesting. So now we’re solely speaking about Odyssey, correct?
Abhay: Yeah.
Matt: Can you give me a little bit of info on the other two algorithms that you didn’t really mention yet?
Abhay: Yeah, so the risk management is the same on the other two. So, 3.5% max drawdown, it’ll exit trades. Same thing.
Matt: And what’s the track record on those for performance?
Abhay: So, we have three: one I mentioned is Odyssey. The other is Buterin, which is a crypto algorithm. That one also has a slightly higher drawdown just because crypto has a more volatile nature, but it’s still under 10%. Monthly performance is about the same—6%.
Matt: Okay, nice. So now, let’s dive deep into the other two products. Can we talk about the names, the strategy, and also their track records? So we talked about Odyssey, and now we’ll talk about Buterin.
Abhay: Buterin is a crypto algorithm which primarily trades Bitcoin and ETH. It also has a track record since 2022—actually, the start of 2022. Monthly performance is about the same as Odyssey: 6% per month, and the max drawdown is under 10%. It is slightly higher than what we talked about with Odyssey, at under 5%, just because of the nature of crypto itself.
Matt: Now, what is the similarity between, I know we have Nakamoto, which we haven’t really fully rolled out… am I incorrect by saying that?
Abhay: Correct, yeah, we haven’t rolled it out because we’re still making enhancements to it.
Matt: So what’s the difference between Nakamoto and Buterin, besides the names?
Abhay: With Nakamoto, we are actually revamping it quite a lot. We’re trying right now to make it a single-entry system, which looks for very specific conditions and primarily trades a wider suite of altcoins. That’s what we’re building right now. But because we changed the vision for Nakamoto’s strategy to really make it a standout algorithm, it’s going to take a little bit more time.
Matt: So, Buterin only trades Bitcoin and ETH, nothing else, and Nakamoto will find other altcoins?
Abhay: Yeah.
Matt: Interesting. And in comparison to the Odyssey and Buterin, Buterin holds more risk from what you’ve seen in the data, compared to Odyssey?
Abhay: Yeah, I think the risk management is effectively the same. The drawdown is slightly higher—still under 10%, which is pretty impressive—but it’s more volatile.
Matt: Yeah, okay, just because crypto moves are… yeah.
Abhay: Exactly.
Matt: And now, the last product—what’s that looking like?
Abhay: That’s Argos, which is very similar in terms of performance. So, it’s about 6% per month with a max drawdown of 7.5%.
Matt: Wait, okay, so hold on. Odyssey: 6%, 4% drawdown. Argos: 6%, 7% drawdown. Why would I not just use Odyssey?
Abhay: Odyssey actually has a minimum account balance of $50,000, whereas Argos is $20,000 or $30,000, which we’re still finalizing.
Matt: Why is that?
Abhay: It’s just because of the risk profile of the algorithm and how it behaves. With these ones, the more balance you have, typically the lower your risk profile, because it can handle drawdown situations and margin better in the account. So, with Odyssey, it’s a higher requirement for a minimum account balance.
Matt: Interesting. Okay. But strategy-wise, are they still the same, or no?
Abhay: They have differences, that’s why they’re two separate algorithms, but they’re built on the same underlying principles.
Matt: Okay. So, when it comes to setting up these softwares, especially the ones that are coming out now, what is the barrier to entry for an individual? Am I going to have to be tech-savvy? Do I have to do anything in particular to have these operating?
Abhay: So, the nature of setting up and watching these algorithms is different, and how, like I was saying, the client interacts with the product is the reason why we packaged it into a separate product instead of just adding these three algorithms into ATA. The reason for that is, with the feedback that we received from clients, we noticed that due to broker differences, spread slippage, etc., a lot of clients weren’t actually getting the same performance that we were getting on the same algorithm on our accounts, on the company accounts. And because of that, we changed just the way they execute in this product, where the client can effectively set up on our platform and their trades will mirror the trades from one of my master accounts. That way, we make sure they’re taking the same trades at the same time.
Matt: And also, it’s a much simpler way for them to set up—they don’t have to install any license, any algorithm files. They’ll be able to see the trades as they are taking place, and they can close out the trades if they want. They’ll also be able to select their risk profile as well, but beyond that, they don’t really have to do too much.
Abhay: Exactly. So as a user, this eliminates—for instance, I’ll speak from experience here—sometimes my algorithm will take different positions, or at least at different times, compared to my friend who’s running the same one. So, having it all linked to a master account, does this eliminate that?
Matt: Yep.
Abhay: So, we’ll all basically be getting in at the same price since we’re just using one brokerage?
Abhay: Yes. So, we are looking to have a few options of brokerages, but within those, we look for brokers which have very minimal slippage. Because naturally, if one broker can’t execute the trade at a particular price, it will be slightly higher or lower because of slippage. But if we select particular brokers to work with, and ensure that they have very low slippage, then we should not have clients experiencing that.
Matt: And the other big difference is, on an ongoing basis, there are no dynamic changes that they have to do to the account. Like I said, the only two things would be they can select the risk relative to the master account, and they’ll just see the trades. But they don’t have to go and update the pairs when we release a new version, or really watch the margin on each pair, or do anything like that. So, really, as a user, this makes it far easier to manage and run the software because now, I’m not really having to go in there and maybe pause auto trading or do any of these things that I previously would have to do?
Abhay: Yeah, it’ll be much more hands-free relative to ATA.
Matt: Nice, interesting. That’s a big thing, and that definitely eliminates a lot of the risk and possible headaches when it comes to licensing, right? Because sometimes we can have issues with licensing, and different people experiencing things with their licenses, and a lot of those errors come up when you’re issuing thousands of different licenses. So, having it all linked to one kind of eliminates all of that.
Abhay: Yep.
Matt: Sounds like we just need those.
Abhay: Yeah, we’re working on it!
Matt: Nice. Okay, well I’m excited, and thank you for making the time today. You guys will definitely be updated when this comes out because I will probably be funding an account in which you can document the whole thing with me.
Abhay: Thanks for having me, Matt!
