We Study Billionaires - The Investor’s Podcast Network - TIP407: The Evolution of Quant Investing w/ Jonathan Briggs, Ph.D.

Starting point is 00:00:00 You're listening to TIP. On today's episode, we are exploring the evolution of quant investing with Dr. Jonathan Briggs. Jonathan holds a PhD in mechanical engineering and applied mechanics from UPIN and has had a long career of researching and developing investing techniques for pension funds such as CPP investment board. He is now the CIO of his own fund, Delphia, where they are implementing techniques developed from his research. In this episode, we discuss the basics of quant investing, the commoditization of the approach,

Starting point is 00:00:30 and what has led to a quote-unquote quant winter, how size and scale applies to the strategies, how machine learning is playing a bigger role in the quant approach, the economic framework Jonathan has devised after decades of research and much, much more. This was a very interesting look into the moneyball approach of investing. Does it invalidate fundamentals? Let's find out in this discussion with the very thoughtful Dr. Jonathan Briggs. You are listening to The Investors Podcast, where we study the financial markets and read the books that influence self-made billionaires the most. We keep you informed and prepared for the unexpected. Welcome to The Investors Podcast. I'm your host, Trey Lockerby, and today I am so excited to have on the show Dr. Jonathan Briggs. Welcome to the show.

Starting point is 00:01:25 Thank you very much. Doctor. Not a real doctor. If you got a PhD, you're a doctor to me, you know? Fair enough. Speaking of which, you've had an incredible career and maybe talk to us about that PhD and then how you found your way into finance and how you've started to develop the approach you have today. So let's start all the way back and go from there. Go to the origin story. So that's a, the PhD is I was interested in applied math, engineering. So that was my first love, which was obviously why you went to a graduate program to be in weekend with. And in particular, I was focusing on something called control theory, which is a set of theories about how you manage to make systems do what you want them to do. So how do you

Starting point is 00:02:04 make an aircraft behave the way you wanted to do. How do you make SpaceX, you know, land vertically on a small ship? So that's all control theory or Tesla how it's self-drives. So that was my passion. There was certainly, obviously, life never goes in one particular direction. They never ended up in the spot you thought you would. And that was true with me. And so for personal reasons, we had to move close to my wife's family after I graduated. And that turned out to be San Francisco, California. And while I was there, I just basically had to look for a job. So any plans of going on into academia put by the wayside? And I had to find someplace to use my talents and get paid effectively. So when I showed up in San Francisco, there were only certain numbers of places

Starting point is 00:02:46 that were hiring. Remember, it was pretty close to the dot-com bubble. So tech really wasn't the place to go. And it turned out that finance was hiring. In particular, Charles Schwab was looking to hire folks with the background similar to mine to help them build. Essentially, we turned out to be a robo advisor. And so folks like myself can certainly help with some of the applied math that goes into building such things. Funny stories, I don't think it was a success early days, but, you know, it was a bit before its time, but has, I think, had some success going forward.

Starting point is 00:03:15 But the person who hired me was looking specifically for somebody with a PhD and with an applied math background. That was my first entry point into actual finance where I learned about stocks and bonds, which I had no exposure to up until that point. Literally, I'd never invested a penny in my life. And there was one person in particular at Schwab, I named Jim Peterson, who he's still there. He's actually the CIO of Charles Schwab, investment advisory, a great human being. He took me under his wing and started to teach me about the Journal of Finance and quantitative

Starting point is 00:03:43 investing, so it included things like Fama French and momentum and things like that. And so that was my first introduction. I spent about four years at Schwab, as I said, to help him to build this robo advisory. I got exposed to their Schwab Equity ratings. Some of you may have seen that. It was pretty clear to me that it was time to move on. And I would say the giant in the industry at that time was out of San Francisco. It was a place called Barclays Global Investors.

Starting point is 00:04:06 They had pioneered the quantitative investing process. Literally written the book on it. There's a book by Goral and Khan where Barkley's Global Investors can talk about at some point, perhaps. And primarily they built an incredible business off of Quantity Equity world, which was they had basically gathered a ton of assets. would continue to gather assets over the next six years while I was there, and really one accolades for their performance over that period of time using a quantitative approach. So to me, I think was formative. That was like going and getting my next degree, so to speak, in quantitative

Starting point is 00:04:39 asset management. And I worked with some of the brightest minds at the time in that space, Richard Gronk, Ron Khan, Mark Britton Jones, Morrie Wakeett, Ken Cronner. These are also the names that became pinnacle in their career in quantitative investing until BlackRock acquired them. in 2010. And my understanding was BlackRock really was interested in the I shares business, which DGI had built along the side of its active investing strategies. And it just turned out, I think, to not be a place for me. It was a different culture, different utility function for the organization at that point. And so I wasn't the only one. There was a natural sort of diaspora, I think, from DGI out of San Francisco at that point. So a recruiter

Starting point is 00:05:19 called me out of the middle of nowhere. It's an interesting human being. He basically called me up and he said, hey, I know what we made last year. I know your wife's name, your daughter's name. I know everything about you. There's this place called CPIB that's recruiting a lot of talented folks. And I hung up on him literally. I thought this guy is a soccer. How can you possibly know this?

Starting point is 00:05:36 Very odd approach. Very odd approach. But what was hilarious is that he said, okay, trust me, go walk across the aisle to the head of the global macro team and go ask him about me, ask him, do you know Bob Shane? I was like, so when I talked to him, this is Ken Croner at the time. He said, oh, yeah, he's a great guy. He helped BGI build from the ground up.

Starting point is 00:05:55 So he ended up turning up to be legit, thank God. And yeah, so the CPPIB was essentially, which is the Canadian Pension Plan Investment Board, was recruiting a bunch of talented folks across the world and bringing them on board in Canada to help build their investments internally. And such began the journey. I started on the global macro team, which is a style of investing that essentially is betting economy versus economy through the various instruments. Bridgewater Associates is sort of the pinnacle of global macro investing.

Starting point is 00:06:22 Most folks have probably heard of them. So I spent a few years working on the macro team. There were some changes going on on the equity side. And at some point, they reached out and asked me and said, hey, why don't you come be the head of the Kwan Equity Research team? That was the beginning of my equity experience as a researcher in terms of actually doing what I would call forecasting of expected returns. That was a journey that was about eight years long.

Starting point is 00:06:46 Talk to us about that pension experience because, as we were kind of discussing beforehand, they did a good job forecasting this gap, this demographic gap with the baby boomer generation and managing that quite well. So I'm wondering if that is an isolated event, if the U.S. pensions have followed suit or if this is an anomaly as far as their strategy and execution? I'm not a pension consultant, so I don't really know that the health of the average pension system in the world, but there are certain countries which have had the foresight to look ahead and say, you know, look, we have a demographic problem. We have a younger generation

Starting point is 00:07:22 which isn't as large. We have a lot of excess requirements going forward for health and benefits. And so they set up a process to essentially do a wealth transfer, intergenerational wealth transfer to take care of, you know, the other generation, their retirement. And CPPD was certainly one of those incredibly insightful and long thinking, long range thinking in that regard. and had of collected assets over the last, I don't know, 15, 20 years to address that. And they're incredibly solid, you know, actuarially, I think they're well provisioned for the next 75 years. And they take those assets that they gather and they reinvest them on behalf of the people of Canada to further the returns as much as possible. And, you know, I think some of the like Norway, I think has done something very similar to some Middle Eastern countries.

Starting point is 00:08:09 It's a remarkable decision to make by any society to invest that kind of energy to do that. So as I understand it, CPPIB has something like $600 billion they're managing, and it's understandable at that size that you are essentially, I mean, forced into finding systematic ways to invest, right? Because it's just the sheer size of it. So you can understand this effort to develop systematic investing and that approach. So I know we're going to get into that. I'd like to start with some pretty basic questions around systematic investing or quant investing

Starting point is 00:08:42 for shorthand. Talk to us about what defines, say, a quality factor versus a value factor versus something like a momentum factor. And let's go from there. Yeah. In fact, when people talk about quantitative investing, they basically lean into this concept of value, quality, momentum, sort of the pillars. In fact, you know, AQR, which is now one of the preeminent quantized investing shops. Their founders published a bunch of articles related in the journal of finance, etc. on these types of behavioral ideas, but they go back quite not behavioral, but these types of exposure ideas. They go back quite far in time. So Fama French were the original, you know, creators of the ideas of these broad cross-sectional exposures that explain terms. And what I mean by

Starting point is 00:09:21 that is what they do is they take a group of companies, which maybe it's the S&P 500 or the Russell 3,000. And what they'll do is they'll essentially create a characteristic, which could be a ratio or some quantity that basically describes something about a company. They then take these and they rank them across the entire set of companies that they're interested in looking at. So again, the S&P 500, they're Russell 3,000. And that these didn't describe a portfolio. It's a process of transforming these quantities into a portfolio, which is very straightforward. And then they backtest it or they run it live. Now, it turns out that these three categories, They're very, you know, they become, they're not a single factor, but they represent a bucket of factors.

Starting point is 00:10:04 Value is this idea that like a free cash flow at a price. So it's this idea that there's a certain amount of free cash flow company has. If the price is very high, then that's less attractive than a company that has high free cash flow, the price is very low. So fairly intuitive. And you can rank all the companies of interest in that context. So that's a value factor. This idea that something over price is a value factor.

Starting point is 00:10:25 quality is essentially a measure of profitability, for instance, so gross profits to total assets. Again, it's a ratio to kind of normalize to the size of a company. But again, you can rank the more gross profits compared to the total assets you have, obviously the better at the company. You think it will be. And so you can rank on these things. Momentum is actually very different. It's basically saying if I look at a particular company and I see the trend and its returns

Starting point is 00:10:49 over the last year for six months or however long you want to look for that trend, it again indicates to you that trend should continue. That's the underlying thesis. So then you can rank companies based on these trends. Now, the idea is then you have these three factors, these three buckets that all sort of look alike in terms of within a bucket. And you can create portfolios of each of these and then add them all together. So you're essentially averaging across all of these what we call characteristics,

Starting point is 00:11:14 so valid equality momentum are characteristics that you can average across. Now, you can be bespoke and you know, out of weight you put to each one of these. And that, of course, is part of us the secret sauce for some quantitative investors. But in the end, you end up with an aggregate portfolio, so everything gets squished together and you have an aggregate view. And this then gives you the ones with the highest ranking is the one you want to go long, the ones with the lowest ranking are the ones you want to go short. And so this is a quantitative, it's a very typical or traditional quantitative investment

Starting point is 00:11:41 strategy. You can use it inequities. You can use it in Macquarie and use it in credit. And talk to us a little bit about this idea of smart beta. Is this encapsulated in what you just described? So this is a whole level. So what happened there was, this is a fascinating story. It's the evolution of quantum some ways, maybe not for the better.

Starting point is 00:11:59 But so BGI and AQR in places like this, we're building these factor portfolios and adding and using them for investing. But just in the way I described them to you, I think it should be clear that they're not so sophisticated that they can't actually, can't remember them and walk out the door and go to some other place and reconstruct them. So there's this diffusion process where quantitative ideas became fairly well known in the investment industry. In fact, people published articles about them, obviously they matured with finance, they make books about them. So what happened was that providers of a product sort of took away

Starting point is 00:12:33 from the hedge fund world, the quantitative hedge fund world, and said, well, look, I can just repackage these very same things and sell them back to whoever wants them, whether it's retail, whether it's institutions, at a very, very cheap price. Because again, the process of building this is very, very systematic, you can have a computer do most of the work. They rebranded this smart beta. So they took something which was considered proprietary and exceptional in alpha, which is this idea of being able to beat expected returns with it, comes back and says, okay, well, now this is something very commoditized.

Starting point is 00:13:03 I can sell it to everyone. This became this smart beta revolution where everyone is buying exposures to things, whether it's price earnings and free actual price, well, its momentum, whether it's quality factors. The smart beta just became a way to justify that this thing exists in the context of a particular type of regression framework and in the Qantas like to talk about things in regression framework. So it was just a natural way to start marketing the product. I think it's had some fairly dramatic effects on people's view of quantitative investing because it looks very commoditized at this point. In fact, in the process of sort of talking about what I do to other

Starting point is 00:13:40 institutional investors, one particular very smart investor came back and said, here's how I do smart beta life cycle. You know, it's like I could read it too. It's absolutely hilarious. Okay, so there's 15 steps. I know it sounds like a lot, but it'll go quick. One thing I want to point out is quants basically talk about back test, which is this idea that I can simulate an investment strategy and show you the returns that would have had. And you should feel very comfortable then that this thing will perform like that, the future, which of course, everyone should know that that's probably not the way it's going to play out. But instead, the back testing is part of the whole process of quantitative. So step one, launch product

Starting point is 00:14:16 with amazing back test. Okay, it looks great. Two, experience underperformance. Okay, terrible. Three, show clients that similar underperformance has shown up in the back test. So it's just statistical noise. Four, experience more underperformance. Five, show clients at the level of underperformance has never appeared in the back test, so it's sure to revert. Six, experience more underperformance. 7. Publish a paper. This is critical because only smart people can publish papers and smart people are good at investing.

Starting point is 00:14:45 8. Experience more underperformance. 9. Tell clients that the strategy requires patients and that the back test results are more indicative of the future than the live results actually are. 10. Experience more underperformance. 11.

Starting point is 00:14:58 Claim that smart beta providers as a whole have overpromised that you are absolved in any such wrongdoing because you are criticizing everybody else yourself. 12. Experience more underperformance. 13. tell clients that you are a good diversifier since you are underperforming while others are outperforming. 14, experience some outperformance, but not nearly enough to make up for inception to date

Starting point is 00:15:18 underperformance, declare victory. That's sort of the view now of the allocators to quantitative investing. Now, is that one of the reasons we're seeing what I've heard you call a quant winter that started as of, you know, 2018 or so? Yes, this is exactly right. This is just the result of the investors who put money into quant winter. quantitative strategies, particularly institutional investors, are quite frustrated. So in 2017, 2018, 2019, 2020, this idea of value, quality momentum, just didn't perform well. And it was broad

Starting point is 00:15:50 based across most of the quantitative investing in the universe. And talk to us a little bit more about the commoditization, because what is needed for that to revert? Some new innovation, I suppose, that would need to come and disrupt the market, correct? Yeah, exactly. So actually, this was the choice that faced me. on the research side of things. I said, I looked at this and I said, look, Goldman Sachs has shown up at my door, offered to sell one this thing. And this was prior to 2017.

Starting point is 00:16:17 So in expectation, something this commoditized and other providers, too, whether it's Vanguard, for instance. So should I expect outperformance based on these things? And I made the decision that, no, I don't think that's likely to continue the performance we've had in the past. We need to do something different. And then to do something different, you obviously have to invest time, energy, and resources, which is never an easy decision and an ongoing concern, particularly of an asset

Starting point is 00:16:41 manager. So you dug into some research. I'm curious what you found when you were assessing people like Buffett, Clarmann, Soros, et cetera. You've written about this underlying theme that you found that came from studying those three in particular. So I'm curious, what was the thread that kind of tied those three together? The idea was, you have successful investors.

Starting point is 00:17:04 These were folks who had a long track record, had shown. resiliency across many market conditions or market cycles. So the question really was, from the quantitative perspective, what we've been told always or intuited ourselves or argued amongst ourselves was that folks like these shouldn't be successful by definition. Why? Because they don't have a massively diversified portfolios. They don't have three, four, five thousand stocks of their portfolios. They're actually very focused on understanding companies at a very detailed little bit of company. They're not thinking about these things as, you know, if I back tested this idea, It would it give me a great result?

Starting point is 00:17:39 They're really saying, hey, this is a company. This has a purpose that exists to do something and is it a good thing. And that can be applied to macro as well. It doesn't have to be just company, but applied to credit, can be applied to currencies. Whatever it is, you can really sort of dig in and try to understand what it is the real economy is doing and then trying to understand what investors think about it. And so that was as it became more and more evident to me that what we had was commoditized on the quant side, it became more and more of it as I said that in the fundamental space, there is a lot

Starting point is 00:18:06 a rich understanding of what markets should be doing or could be doing or probably or might be doing. And as a quant, we've kind of ignored that channel, right? You may have a superficial interest in it. So for me, it was, well, if we're going to look someplace new, there's this whole body of work, which we've disregarded because it isn't statistically interesting. What is it? We have some shining examples of folks who have outperform the markets for a long time. We should learn something. And that was, you know, it's a bit of humble pie, right? It's basically being able to say, hey, you know, I've adjusted, whether it's a defined math background of a PhD, plus all the statistical analysis we did. Mark is a little step back and maybe somebody else

Starting point is 00:18:46 could teach you something new. And that was, that was really provocative for us and for myself. What in your experience did you see as sort of the shortcomings of that quantitative approach? Is it that obviously it's needed for the scale we talked about in the pension arena, but is it applicable to use that same approach when you have a much smaller account? Does it break down anywhere along the way? The concept of quantitative investment, which is this idea that your returns or your risk-adjusted returns are a product of two things. One is the skill you have, which is intuitive to everybody, multiplied times the breadth or the number of attempts you have to express that skill. So think about it as gambling. If you can count cards, go to Vegas.

Starting point is 00:19:30 You get thrown out of the casino probably, but let's assume you don't get thrown. without you can count cards and you sit at a table and you play blackjack, single, get a single deck blackjack. The advantage conferred upon you by counting cards is small. It's not huge. And so to really sort of reap the rewards of that ability to count, you have to sit at the table and play many, many, many, many hands. And so that's a perfect example of the combination of skill,

Starting point is 00:19:50 which you have to beat the house plus the number of times you can make those bets. It's kind of like gravity. Maybe you don't believe in it, but it's true. It's exists. It's statistically closest thing of proof you can do. So particularly in investing. So that's a long way to say that those, that property of investing skill times the number of bets you make is true for anyone.

Starting point is 00:20:12 Obviously, the trade out between skill, the number of bets you make can be made, which is why we see fund-in investors taking very few bets, but they potentially have very high skill. Whereas potentially quantitative investors have the lower skill, but have many more bets. So you can kind of make out for those two things. If you think about it, like somebody trying to do this on their own, that number it becomes really problematic. So the number of bets you have to make becomes almost unmanageable if you're managing a portfolio with 10,000 securities. And it's almost impossible for an individual to do or too well. Let's take a quick break and hear from today's sponsors. All right. I want you guys

Starting point is 00:20:48 to imagine spending three days in Oslo at the height of the summer. You've got long days of daylight, incredible food, floating saunas on the Oslo Fjord. And every conversation you have is with people who are actually shaping the future. That's what the Oslo Freedom Forum is. From June 1st through the 3rd, 2026, the Oslo Freedom Forum is entering its 18th year bringing together activists, technologists, journalists, investors, and builders from all over the world, many of them operating on the front lines of history. This is where you hear firsthand stories from people using Bitcoin to survive currency collapse,

Starting point is 00:21:24 using AI to expose human rights abuses, and building technology under censorship and authoritarian pressures. These aren't abstract ideas. These are tools real people are using right now. You'll be in the room with about 2,000 extraordinary individuals, dissidents, founders, philanthropists, policymakers, the kind of people you don't just listen to but end up having dinner with. Over three days, you'll experience powerful mainstage talks, hands-on workshops on freedom tech, and financial sovereignty, immersive art installations, and conversations that continue long after the sessions end. And it's all happening. in Oslo in June. If this sounds like your kind of room, well, you're in luck because you can attend in person. Standard and patron passes are available at Osloof Freedom Forum.com with patron passes offering deep access, private events, and small group time with the speakers. The Oslo Freedom Forum isn't just a conference. It's a place where ideas meet reality and where the future is being built by people living it. If you run a business, you've probably had the same thought lately. How do we make AI useful in the real world? Because the upside is huge, but guessing your way into it

Starting point is 00:22:35 is a risky move. With NetSuite by Oracle, you can put AI to work today. NetSuite is the number one AI cloud ERP, trusted by over 43,000 businesses. It pulls your financials, inventory, commerce, HR, and CRM into one unified system. And that connected data is what makes your AI smarter. It can automate routine work, surface actionable insights, and help you cut costs while making fast AI-powered decisions with confidence. And now with the Netsuite AI connector, you can use the AI of your choice to connect directly to your real business data. This isn't some add-on, it's AI built into the system that runs your business.

Starting point is 00:23:16 And whether your company does millions or even hundreds of millions, Netsuite helps you stay ahead. If your revenues are at least in the seven figures, get their free business guide, demystifying AI at Nessuite.com slash... The guide is free to you at netsuite.com slash study. NetSuite.com slash study. When I started my own side business, it suddenly felt like I had to become 10 different people overnight wearing many different hats. Starting something from scratch can feel exciting, but also incredibly overwhelming and lonely.

Starting point is 00:23:50 That's why having the right tools matters. For millions of businesses, that tool is Shopify. Shopify is the commerce platform behind millions of businesses. around the world and 10% of all e-commerce in the U.S. from brands just getting started to household names. It gives you everything you need in one place, from inventory to payments to analytics. So you're not juggling a bunch of different platforms. You can build a beautiful online store with hundreds of ready-to-use templates, and Shopify is packed with helpful AI tools that write product descriptions and even enhance your product photography. Plus, if you ever get stuck,

Starting point is 00:24:26 they've got award-winning 24-7 customer support. Start your business today with the industry's best business partner, Shopify, and start hearing sign up for your $1 per month trial today at Shopify.com slash WSB. Go to Shopify.com slash WSB. That's Shopify.com slash WSB. All right. Back to the show. Now, my understanding is, or at least originally, was around this idea of quant solving for

Starting point is 00:25:00 the psychology aspect of investing, right? Kind of taking the human out of the equation because as humans and our biases, we're prone to make so many mistakes and really shoot ourselves in the foot, so to speak, right, when it comes to investing. So how does the quant approach solve that exactly if you're trading so often, as you kind of mentioned? Obviously, there's trigger points that are being, you're being alerted on to get out of a position, at certain points. So talk to us a little bit about what that looks like.

Starting point is 00:25:29 There's a bunch of interesting points you write. The one is how do quant trade? And then the other thing is the biases when you are trading, you know, how do you, the second judging, the triple judging, the quadruid, you know, looking back, maximizing your regret. So yes, indeed, the systematic process, which is tied to this idea of a back test, is to try and remove the biases. So you create a thesis and that thesis you believe in and make a peer, evidence around it being properly vetted and properly tested. So like a scientific method almost. Then when you let it go, the impetus to interfere with that is very low.

Starting point is 00:26:07 So there's a big barrier to jumping in to try to overcome that. And so that prevents you from introducing your own biases. So certainly in your research, you could have introduced your own bias, which happens a lot, by the way, even in quantitative space. But getting the process moving, does the thesis is to try and remove that. that effect. And also the idea that you can test things in time and geography. So you test an idea in Japan, test in the US, Canada, Australia, emerging markets, and you test it through multiple cycles and macro cycles. So that helps you build up a body of evidence that whatever you're doing

Starting point is 00:26:39 probably doesn't have as much bias as you might have if you're just to do it off the side of your desk off the top of your head. Now, that's the idea that, you know, it can introduce some firewalls to the bias problem. But quantum doesn't necessarily have to trade quickly. So So when I said that we made lots of bets, there are two ways to make lots of bets. One is to make lots of bets across many stocks and one is to make lots of bets through time. So high frequency traders might trade three stocks, but do it every millisecond. So that's a lot of bets. As long as their bets are independent of each other, then that's a great way to bring into that idea of breadth.

Starting point is 00:27:15 But there's another set of quads like myself and partners in aQR who are not really looking for speed. They're looking for understanding a thesis that that says I can make not many bets in time, not too many bets in time, but a lot of bets across names. When I say cross-section, it's the company dimension or the time dimensions. The cross-section is the company dimension. Time dimension is a whole different game. And it turns out that you want to use as much of both as you can to increase the statistical significance. But I want to caution people not to think of this quant fast trading all the time. That's certainly like the Flash Boys and things like that

Starting point is 00:27:52 became very popular, and are easy to understand speed as an advantage. But I would flip you over towards the fundamental space, which says speed isn't the only advantage, but being able to predict is another advantage. And fundamental investors do that all day, every day. And so perhaps a mix of those two ideas, which AQR and BGI sort of extended into, and I think that we extended further in that direction, is can you make things that are more forward-looking, more prognostication to help you with, and then use that in less speed, but more future looking. The primary reason you want to do that is because high frequency trading has an impact on markets that erodes whatever returns you because it's just your paying costs as you trade more and

Starting point is 00:28:35 more and more and markets are smart. As you trade fast and you trade more and you trade with more dollars, more and more people are attracted to that behavior in front running. So it becomes a sort of endless, sort of like a fruitless exercise to try to trade too fast because then you erode all the returns that you may have had by understanding something better. So if you really want scale, if you really want to trade a lot of assets, a lot of dollars, and this is why institutional investors are attracted to these ideas, is that you want to trade slower and you want to hold for longer, and then that reduces your, your, your, your trictions or your transaction costs. So quats have traditionally never really gone out as far into the future as fundamental

Starting point is 00:29:15 investors. Fundament investors were really king of like Buffett. I mean, I don't know maybe he's looking out 10 years, maybe seven years, five years. But they're really pushing the boundaries of forecasting it into the future. And quants have not done that to date, you know, these ideas of quality value momentum. You're looking at about a three-month horizon, maybe a little bit longer with some value factors, maybe six or seven, maybe eight months out. So you've really left, they've really left behind this whole idea of long horizon forecasting. And I would argue that actually quite deeply that quants are not forecasting in the strong sense that a fundamental investor is doing, if that makes sense.

Starting point is 00:29:50 Yeah, it does. And what's kind of coming to my mind is for, especially for retail investors, understanding these three-factor approaches and how they integrate together potentially. So if you wanted to build a portfolio with them, I'm wondering if any of the factors tend to overlap or if they're kind of all in their own island. So, for example, value and momentum almost seem to be opposites, right? Because you think of earlier companies like Facebook, Amazon, when they weren't generating any earnings, you could easily look at that on a value factor basis and say, well, this is a

Starting point is 00:30:22 terrible investment, right? There's no earnings. Whereas, you know, if you were a momentum trader, you would have done quite well. So, and value has been underperforming for a very long time. So obviously, if you're putting a packaging, a portfolio together, probably want some exposure to each one. I'm not looking for financial advice, but I'm curious, is that a common approach even on the institutional side where they're trying to put.

Starting point is 00:30:42 a certain allocation towards these different buckets. So this is true even in an R framework, which you get deviates from quality by momentum, but to your point, that's exactly right. So this is diversification. Diversification is the number of stocks, but it's also the number of ideas. So momentum as an idea or value as an idea or quality as an idea. You want them ideally, what you'd like them to have is different return signatures through time.

Starting point is 00:31:07 And so what that allows you to do this, build a portfolio, which is robust across different regimes potentially. And so that's exactly what you're looking for. So diversification across factors is part of the lexicon for quantum investing. All right. So you've done all this research for many years now and you've developed this economic framework that I'd love for you to kind of lay out for our audience and talk to us about how the elements of quality, value, and momentum kind of evolve from here. So as I said, the ideas behind quality and value momentum had my expectations on them had paled a bit. And I've also mentioned And the fundamental investing world often seemed like a very fruitful direction to go look and learn from.

Starting point is 00:31:46 And so that's exactly what myself and my team we did. We basically took a pause at the time. And we sat back and we said, let's try to reexamine some first principles, what it is that we actually believe that markets are doing. And let's see if that ends up at quality value momentum again, which is fine, in which case we've come for a circle. Or perhaps we end up in a different place. And luckily, I had a thought partner in this, his incredible human media. Frank Arachi, he's now the senior managing director, global equities, fundamental equities for CPPID. I've known him for 12 years almost now. And he and I and my team at the time,

Starting point is 00:32:21 we sat down and we started to think about, okay, first principle, what do we believe in? Well, first of all, what we said was, you know, markets are forward-looking. So price is a function of things expected in the future. It's not a function of things looking backwards as a thing looking forward. Of course, that expectation of some futures is predicated on all the information available to now. So that includes sunspots, that includes how many people drive in the morning, it also includes everything. So it's a very general statement, but it's quite powerful. So it allows you to sort of ask other questions. Like, okay, well, if prices are very forward-looking and prices change through time, then can I say something about the change in price? So it turns

Starting point is 00:32:59 like you can. You can do some very simple manipulation of that idea and you combine some things together. And this gets a little bit on the mathy side. But what pops out is that, well, okay, It changes in price, which are really returns, are described by some sort of attention based on surprise. So what I mean by surprise is, markets expected something in the future, that got something else, and then there's a correction. So this idea that markets are forward-looking and they react to revealed information and the size of that reaction is based on the attention being paid to it.

Starting point is 00:33:36 Again, these are very general concepts, but that's a really key part. So markets are always looking forward, looking to expect something. And then they have to react and that something shows up and it's different than what they expect it. The next piece is, okay, well, how are we going to discover what those things are the markets are expecting? We want something that makes markets move a lot. Big returns are available to you. We want it to move frequently, which is this idea of breadth. So we want many bets at the game.

Starting point is 00:34:02 And we certainly don't want the game to go away, whatever it is it we're looking at. We don't want it to end in three years or five years. We want this thing. whatever this surprise thing is, we want it to go on forever. And the final piece was, well, capital markets aren't there. Casino, you know, capital markets are there to allocate capital. This is the Benjamin Graham, you know, the voting machine, weighing machine, right? So this short horizon, sure, maybe there's a voting component, but in the long horizon,

Starting point is 00:34:26 capital markets are there to allocate capital efficiently. That was the foundation. So it was this idea of surprise, which is based on expectations. So what do markets expect? Can you understand that? This rearrangement once markets are surprised. This idea of surprise has to be worth the pursuit. And then the idea that, well, okay, let's look at fundamentals and let's look at it

Starting point is 00:34:46 through the lens of an empirical investor. So let's think about this. It's designed a set of experiments to test all of these. And that's really what we did. We basically spent about a year and a half thinking about these ideas and testing them. The test is interesting. So bear with you for a second. So if you look at how fundamental investors invests.

Starting point is 00:35:05 So you can go online, you can say, you know, look at KPIs. For instance, that's kind of like the language of a fundamental investor, KPI. So KPI is telling fundamental investors a lot about the state of a company, you know, what it's likely to do in the future. You can buy KPI libraries, by the way, there are thousands of thousands. And so assume you take all of those KPIs and you say, well, I'm going to be the perfect fundamental investor. I'm going to assume that I can predict the KPI's perfectly at any horizon, one year,

Starting point is 00:35:32 or two years, three years. Then I'm going to take those predictions and I'm going to put them in a back test framework. I'm going to basically be like a quant. I'm going to create a quant portfolio out of these ideas. And I'm going to run a back test and see what happens. See if any of these ideas create really lucrative portfolios. Interestingly enough, most of those ideas don't produce lucrative portfolios.

Starting point is 00:35:51 Before you go too far down that rabbit hole and say, well, you know, that's absurd. But as a quant, we're like, oh, of course, we knew this. You know, fundamental investors don't know anything. So, of course, this is obvious. And I'll come back to that because that's actually quite arrogant, and we had to eat pro on that. But what we did find was that various measures of cash flow across different horizons were incredibly powerful, meaning if you could predict these cash flow-y-type variables in a portfolio context, you almost achieve perfect foresight on returns themselves, which is a remarkable concept.

Starting point is 00:36:21 It means that, oh, my goodness, maybe fundamentals, cash-fluy fundamentals, and returns are tightly coupled together, are tied together. I don't want to say causally linked because that sets up a whole argument about specific mathematical definitions of causality. But let's just say for the sake of argument, there's a causal linkage between these cash flow growth, growth of cash flow through time and returns. There's a horizon component to this. In a short horizon, not so much, right? That tie to returns in a one-month horizon is not so strong. That tie to returns at a three-year horizon is almost 90%. So you're really starting to see this. interesting interplay between, again, the Benjamin Graham, the voting versus weighing, that in the long

Starting point is 00:37:04 horizon, these growths of cash flows really describe a causal relationship that markets are weighing on constantly. In the short horizon, there's a bunch of other stuff going on that don't necessarily at least in the variables we looked at, explaining terms. So this is remarkable. So the question to ask then is set aside quality value momentum for a moment, which have none of that forecasting power at all, like not even close to that. These other things, Wow, I mean, we couldn't have accidentally found these because they're so close to returns themselves. They have to be statistically significant.

Starting point is 00:37:36 There's no human bias that could have introduced that. And by the way, it works, again, globally across all time. So then, oh, well, these seem like really rich things to understand. So kudos to fundamental investors to really think of discounted cash flow models. I think that's the right thing to do. But again, all these KPIs are useless. So the question was, as a quant, can we predict these things? And this fits all the difference.

Starting point is 00:37:59 So now we lean into the predicting realm of things and how do we do prediction, which opened up a whole new world for us. So then, right, well, what do you use to predict? You have tools, right? So regression is linear regression is one tool. But everyone listening to your podcasts heard of the AI, the revolution. So linear regressions are linear, they're very simple relationships that they can capture. Machine learning and AI capture very, very complicated relationships. The problem is with machine learning and AI is that it requires a lot of data to help sort of organize themselves to pick out complicated relationships.

Starting point is 00:38:36 The worst thing you can do is to give a bunch of very little data to a machine learning algorithm and tell it to figure stuff up because it'll figure out relationships that aren't really true. We call this overfitting. The problem we found was in trying to forecast returns with machine learning, which has been done by many, many organizations. I think some had some success. I think very few have had success. is that the complexity of returns themselves don't lend themselves to scarce amounts of data. So you really are operating at a dangerous end of the deep end of the dangerous part of the swim, I don't know if that's a real expression, probably mixing my metaphors there.

Starting point is 00:39:10 It's a dangerous part of the ocean because you have this thing which is very dynamic. Returns are very dynamic and you have this very limited pool of observations to train your machine, your nonlinear machine learning algorithm on to try to predict returns. And so this creates a problem, right? So if you start to use this technology, it's kind of like giving somebody who's untrained, a weapon to somebody who's untrained, you know, it's quite dangerous. And so what's happened is this, there's been a bunch of folks who, a bunch of managers who tried to use it with disastrous results.

Starting point is 00:39:39 So in sample, it looks like, you know, when you try it, you test, it looks great, you try it live, it looks terrible. I think one way that that has been short-circuited is that in the really high frequency space, again, you go back to lots of observations. Hillary millisecond is a thing. And I think you've seen some success. You've seen, because, again, because the numbers of observations are really, really high, it gives the machine learning something to play with.

Starting point is 00:40:00 On the other hand, in the long horizon, when you're talking about three months, six months, three years, there's just not a lot of independent observations. And so the machine learning really struggles to create good forecasts that you would expect to work out of samples, so when you go live. But what we observed was quality, the properties of cash flows is not the same as returns. While those things, remember we've said are going to be causally linked together, they're a very different beast. They're not being impacted by the speed of human thought.

Starting point is 00:40:28 They're being impacted by the speed of human interactions, which is buying and selling and producing goods and services, which is just not as fraught with chaos. And so about structuring a problem where we're trying to predict a fundamental outcome, you basically change the dynamics of the thing you're trying to predict. And I think fundamental investors really knew that. They basically said, look, price is really, really hard. But if we assume this co-integrating relationship between price and fundamentals, let's focus on the fundamentals. And again, we'll think about it in the longer horizon.

Starting point is 00:40:56 So absolutely brilliant. If we set about to replicate that behavior, let's do this forecasting exercise. And this thing turns out to be well-behaved. So fundamentals turns out to be well-behaved relative to returns. So we started the march down the machine learning path. We started linear stuff. You know, we slowly increased the amount of degrees of freedom that this thing was allowed to play with. And then we've said, okay, we're going to let the dogs out.

Starting point is 00:41:19 We're going to try machine learning. But again, not completely crazy because we're saying we're going to focus it on particular variables related to real economic fundamentals. And that started to work. It's like, oh, this is really interesting. But again, we never achieved anything close to perfect foresight on these variables, on these cash flow type variables. So we're really struggling.

Starting point is 00:41:38 How do we get our accuracy up? So, okay, we did something interesting here. The returns that we can generate the portfolios like this when we back to us and don't look commercially viable. Like no one would pay for these things. Now what? Okay. A couple of things.

Starting point is 00:41:50 One is, first of all, we probably should put really interesting things into the machine learning process, something we know has value in forecasting the future. And here's where I come back to my eating pro moment. Oh, KPI might work. Bingo. So now you start putting in KPI into a forecasting problem for fundamentals. And oh, my goodness, all of this fundamental knowledge in the world was not for nothing. I mean, this is a huge compendium of human intellect.

Starting point is 00:42:20 People have spent time denoising business models. They look at business models and they say, let's throw away the insignificant things and look at the significant things. And so fundamental investors have really spent a lifetime understanding how certain variables interact with each other within a certain business model. Once you bring those into the forecasting exercise, you really start to see a lift in the forecasting accuracy. Now, we went back to the fundamental investors that we'd spoken to about this.

Starting point is 00:42:44 Of course, they're like, oh, yeah, you idiot. I mean, we never said these things had value in forecast returns. We said KPIs had a lot to say about cash flows. And cash flows have a lot to say about returns. And so as a quad, basically, where was I? I'd basically gone down the yellow brick road and ended up being a complete believer in fundamental investing. And actually, it was this idea of surprise. Remember I talked about surprise.

Starting point is 00:43:06 Well, if you give machine learning a forecasting problem, it's going to spend a lot of its time and energy and budget. I mean, you give it a certain number of variables, a certain number of observations to work. with to help calibrate it, it's going to end up giving you a lot of things that are already well known, which is unfortunate, right? It may do a great job, but it's telling you stuff you already knew. So, okay, well, that's not good. Let's add some structure. Surprise is the structure. So you have to understand what markets expect, subtract that from what the outcome will be, which is surprise, and then force the machine learning to focus on the stuff that's not known, which goes back to this idea that surprise is what defines returns. And so, but we did it.

Starting point is 00:43:46 this way it was through a fundamental channel. Let's take a quick break and hear from today's sponsors. No, it's not your imagination. Risk and regulation are ramping up and customers now expect proof of security just to do business. That's why VANTA is a game changer. VANTA automates your compliance process and brings compliance, risk, and customer trust together on one AI-powered platform. So whether you're prepping for a SOC 2 or running an enterprise GRC program, VANTA keeps you secure and keeps your deals moving. Instead of chasing spreadsheets and screenshots, VANTA gives you continuous automation across more than 35 security and privacy frameworks. Companies like Ramp and Riter spend 82% less time on audits with Vantta. That's not just faster compliance, it's more

Starting point is 00:44:35 time for growth. If I were running a startup or scaling a team today, this is exactly the type of platform I'd won in place. Get started at Vanta.com slash billionaires. That's vanta.com slash billionaires. Ever wanted to explore the world of online trading but haven't dared try? The futures market is more active now than ever before, and plus 500 futures is the perfect place to start. Plus 500 gives you access to a wide range of instruments, the S&P 500, NASDAQ, Bitcoin, gas, and much more. Explore equity indices, energy, metals, 4X, crypto, and beyond. With a simple and intuitive platform, you can trade from anywhere, right from your phone.

Starting point is 00:45:20 Deposit with a minimum of $100 and experience the fast, accessible futures trading you've been waiting for. See a trading opportunity. You'll be able to trade it in just two clicks once your account is open. Not sure if you're ready, not a problem. Plus 500 gives you an unlimited, risk-free demo account with charts and analytic tools for you to practice on. With over 20 years of experience, Plus 500 is your gateway to the markets. Visit Plus500.com to learn more. Trading in futures involves risk of loss and is not suitable for everyone.

Starting point is 00:45:55 Not all applicants will qualify. Plus 500, it's trading with a plus. Billion dollar investors don't typically park their cash in high-yield savings accounts. Instead, they often use one of the premier passive income strategies for institutional investors. Private Credit. Now, the same passive income strategy is available to investors of all sizes thanks to the Fundrise Income Fund, which has more than $600 million invested in a 7.97% distribution rate. With traditional savings yields falling, it's no wonder private credit has grown to be a

Starting point is 00:46:31 trillion dollar asset class in the last few years. Visit fundrise.com slash WSB to invest in the Fundrise income fund in just minutes. The fund's total return in 2025 was 8%, and the average annual total return since inception is 7.8%. Past performance does not guarantee future results, current distribution rate as of 1231, 2025. Carefully consider the investment material before investing, including objectives, risks, charges, and expenses. This and other information can be found in the income funds prospectus at fundrise.com slash income. This is a paid advertisement. All right. Back to the show.

Starting point is 00:47:12 Now, surprises can happen either up or down, right? So what I'm kind of curious about is there's been a lot of surprises, especially in the last couple of years. And, you know, when we had the original COVID market crash, that was obviously surprised. And then there was a surprise to the upside following thereafter. So how would a machine learning approach have performed through some event like that, both up and down? So to your point, expectations are not a single beast, right?

Starting point is 00:47:39 There's multiple cohorts of investors out there investing in a particular style. And so they have their own expectations to avoid being stuck with one set of expectations, which can be right or can be wrong, in which case you've kind of limited to your ability to earn returns if you're always fighting one set of cohorts when another one is actually dominant in price. You have to understand the heterogeneous nature of investors. So Bridgewater talks about this a lot as well, but we've internalized these ideas very deeply ourselves is that even if we take a very simple example.

Starting point is 00:48:09 There's a dominant view in the market, which is expressed by the dollars being invested in a particular view. So let's say, Trey, you say, sales growth is going to be high for this company. And there's a bunch of people like you, and you often in a lot of money into that view. Well, if that were the only view in the market, price would shoot up, it'd go crazy. But we know there's a set of people who disagree with you in the market. And they'll invest a certain amount of money to disagree with you. And they'll impact price as well.

Starting point is 00:48:35 So you end up with this equilibrium state, which gets created in price, where there's a certain some outgoing one direction and certain going, and there's certainly one that's dominating, but that dominance is not so severe that price becomes crazy. It doesn't go one direction another very, very, very rapidly. The chaos of cohorts. Right, the chaos of go. I like that. So you want to understand because not everybody is going to be right. At some point, there's a truth moment where your bet is marked to market. In a contrarian view is marked to market. And typically, markets figured out shortly before the reveal, so there's a lot of correction that goes on, but there's a marking to market. And in that moment is where you see

Starting point is 00:49:16 returns becoming really, really interesting. So in the case where you were right, let's say you were the dominant view, let's call it consensus for argument's sake, and the contrarian view has to correct itself. So it would have been losing money all along the way because it was, you were dominating price. Price was moving against their view. Contrarians look at you and they say, oh my goodness, you were right. They switch directions and they match your position. You would get an extra bump at the end of the day. Returns would go your way. And so what you would see for Trey, which you would see for your portfolio, if you were betting or your stock position, but you could extrapolate to portfolio these, you would see a slow diffusion of returns

Starting point is 00:49:54 heading away. You'd have a nicer ramp up in returns and then followed by once the information was revealed, maybe some slight bunk still, some post-announcement drift. The person who was betting against you, of course, would lose all along the way. But at the other hand, maybe I was right in trade withdrawal, even though, and you dominated your price returns up to a certain, up until just before announcement, you would say, oh, my goodness, I need to switch my position. You would flip your position to match mine. In that flipping, you would move all your price, all your dollars towards my position

Starting point is 00:50:23 and change the path of return. So for me, as a contrarian investor would have been losing money. And then as you push, you realize that you needed to switch, I would have, my returns would have shot up And then again, a post-ouncement drift where I would have had a decay. Now, what's really cool about this is we've just described to you the return signature to the factors that we talked to you about earlier, which was momentum and value. So as a consensus being right, you would have seen in the impact of price, people following your view and piling in.

Starting point is 00:50:54 At the very end, contrarians would have realized you were right. You would have had an extra bump and then a decay. This is like a very nice momentum-shaped pattern. Conversely, when contrarians are right, you see this negative. return followed by a pop in return space and a decay. That's a value signature. And so we made full circle, we came back and we said, oh, my goodness, this idea of surprise, we've already evidenced in different quantitative investing strategies. One is value and one is momentum. These represent different cohorts in the market betting differently, and yet there's a convergence

Starting point is 00:51:25 at the end. The idea is we can measure both of them and have a better forecast in the future. we can arbitrage both sets of cohorts. And so you end up with a beautifully diversified set of return streams based on who you're arbitraging, who you're trying to bet against. And the probability is in a cross-sectional sense, you're getting it right in multiple dimensions. When I hear you talk about that, what's happening for me is just this resolving to buying and holding.

Starting point is 00:51:52 Because, you know, I think so often we're not really thinking about the cohorts in play. And a lot of people are just thinking, hey, it's, it's, it's, It's right or wrong. It's me and this guy on the other end or gal on the other end. And this is the trade. But we're all sitting around the same poker table, but we're not. One person's playing poker, the other person's playing craps. It's just like there's, this is this wash or this pool of different people with different

Starting point is 00:52:17 time horizons with different strategies that is creating so much noise that just makes me want to sit back and say, look, this is why I don't want to look at my screen for 10 years. That's right. In which case, you know, that sense you're just buying the equity risk premium, just buying an index fund, allows you to coast along with all that noise and ignore it and just get the value of being long in equity, a group of equities, which are likely to go up over time. So that kind of sounds like a robo advisor to some degree. But there are robo advisors, though, they're just looking at your age and your time horizon

Starting point is 00:52:53 and your risk tolerance, et cetera, et cetera. And they just say, here's your package. It's not very active beyond that. I understand it. Is that correct? That's exactly right. So the activities and what's the weight of fixed income versus equities or commodities or whatever yet, that's their active bet. The big addition for institutional investors in particular is that they're looking to increase the beyond the asset allocation. They're looking to increase their return because they have obligations. They have like

Starting point is 00:53:18 Canada Pension Plan Investment Board that has obligations in the future. So being able to increase beyond just passive exposure to or semi-passive exposure to equity or risk premium, to provide some extra juice to help cushion the blow that they're going to have in terms of beating their obligations. In fact, when you go to a pure long, short portfolio, you're not even really, it doesn't really even matter that you're in equities or fixed income commodities, not in the sense that you still need to do security selection. You don't have to understand the underlying system that you're working in, but really you're not trying to harness the risk premium within those asset classes. You're really just adding one risk. It could be anything, right?

Starting point is 00:53:55 You don't really care as long as you're right, as long as you're getting those bets and the numbers of table, the numbers of times the table is large. But what's so enticing about this approach is, you know, the only other reference point I kind of have on it that comes to mind is Jim Simons and his fund. And you hear about these outsized returns that are just like 60% plus a year for 20 years. So it's really enticing and it draws you in. But as I understand that that approach also had limitations or at least they found, some sort of optimized scale that said, you know, look, we're only going to run a fund of this size and we're not letting any more money. So talk to us. Is that a limiting factor? Do we hit a ceiling with the approach that you're describing as well? Yeah. So I think actually any investment

Starting point is 00:54:40 approach will hit the ceiling. So there's only so much you can move markets, move around in markets before you move the market. And then you know that you've lost whatever edge you're going to have. So the bigger you get, the less nimble you are in terms of being able to express your positions without people being able to follow you or front run you. So rent tech, they hit a limit. And so they kept up the amount of capital they could go in. To be fair to Rent tech, you know, I mean, they certainly invested a lot of time, energy, and money into building machinery.

Starting point is 00:55:11 Working with data, it's very expensive. It's only natural that at some point they have to close and to maintain the returns that you're trying to be able to compensate themselves for the amount of effort they put into to that. That's true with any strategy. Fundamental strategies as well, right? They'll reach a capacity, albeit if they're working in large-cap space because of liquidity, they'll reach it later than, say, a quant who's potentially operating at smaller scale or in smaller stocks than they are exclusively, if that makes sense. That does make sense. When you were talking about machine learning, one other question that came to my mind was around the inputs and comparing it or juxtaposing

Starting point is 00:55:47 it with the Robo-Advisor approach. This machine learning approach sounds much more complex and nuanced. And I'm wondering if, is it capturing headlines and news on certain companies? Is it incorporating all kinds of like this plethora of data from all over the world? Yes. So we take in a ton of data. Look, our costs are huge computing data. And I'm sure it is with rent tech and two sick. But it's, it's really, you know, and in some ways, we could use more data. So we're always looking for other data sets because this helps inform our forecasting exercise. And I think you can't play in this without. out a ton of resources.

Starting point is 00:56:22 As an individual, you could never compete at this scale for these things. It's an arms race. Perhaps there's an arms race for distance to the exchange, which is high-frequency players played in for a long time. And that's sort of been tapped out. We don't hear about people building any more microwave towers, or maybe we are, but not as many as the Hummingbird Project, whatever, is not the thing anymore. Prediction is the thing.

Starting point is 00:56:46 That prediction is going to create a new set of constraints in arms race around data in all likelihood and around talented machine learning and around frameworks. As I said, how do you take all these very powerful things and focus them on the way that has a high probability of success? We found one way. Perhaps others will find others, but that process of understanding is very hard. It takes many years of experience to understand how this plays out. And so it's an intimidating and daunting thing to think about it.

Starting point is 00:57:13 We haven't sort of grown into it over the course of 15 years of your life to sort of internalize all of this and understand it in this sort of coherence. Well, what's coming to my mind is sort of like deep minds alpha go example. They're projecting out these probability trees essentially and each thread has a certain weight, I guess. And then as it proves out one position or the other, it's learning that and predicting. And I'm guessing this is not that dissimilar as far as the approach where it's, there's laying out these probability trees, if that's the right vernacular.

Starting point is 00:57:45 Good enough. Okay. Bear with me. You're totally right. So I've spent some time talking with DeepMind here and there and folks who work there and researchers and things. And yes, you're exactly right. I mean, AlphaGo is actually quite beautiful in many, many ways, almost exquisite.

Starting point is 00:57:59 Good for them. But they were able to harness the concept that we can't in the world of investing. So they can create data, so they can play many, many games. They create this adversary relationship between different cohorts. They can create cohorts who play one way and another way and another way and they can force them to play each other at super high frequency and learn and learn and learn. And so this is ideal for machine learning or AI. On the investing side, we just have to sit and watch paint dry, right?

Starting point is 00:58:24 You just have to wait a thousand years for more observations for more co-arts, particularly if you're going directly for returns because the dynamics of returns themselves are so problematic. So you have to provide structure for machine learning in that context, which we haven't seen, you know, don't want to take anything from deep mind, but in the context where you're limited observations, it's a very, very hard game. But you're exactly right. we are looking at probabilities.

Starting point is 00:58:48 We're not looking with certainty. And so that's how our portfolios evolve. So we look to the future at various horizons and we say the probabilities are changing at various horizons because you can imagine companies being successful in the short term and not in the long term, vice versa. And so we're playing this probabilistic system and letting our ideas sort of flow into portfolios that try to capture these through time. And it's quite complicated.

Starting point is 00:59:12 Well, to add to the complexity, you also have our own government now. creating more and more liquidity to enter the markets in really unforeseen ways. And so I'm wondering how does the computer accommodate for that sort of outlying factor? So for instance, Robin Hood, that was a, they created a whole different world for retail investors. It's a free trader and lots of access to leverage. But remember, those are cohorts. You know, they're not, so there already were cohorts. There's hedge funds or mutual funds, their ETFs.

Starting point is 00:59:44 There are flows. And now we just have another one. which is very, very strong. So if you try to imagine there's a cohort of instance of one cohort, which is for simplifying it, but one cohort of retail investors, where would they line up on these expectations of future events? And if you're better than they are,

Starting point is 01:00:01 then obviously they're arbitrable. The deeper question is, will retail look more and more like game stop shenanigans or will it look like fundamental investors? So where would they end up and how would they impact prices? Because if they end up in a game stop sort of situation, where fundamentals don't matter, then it's problematic for our strategy. It doesn't mean that it's impossible for us to work when, but if they dominate the market

Starting point is 01:00:24 over all horizons, then that's a problem. What I would hypothesize is what we've seen so far is that's a short horizon thing to get, right? So this idea that voting machine, weighing machine, we're seeing a big impact on the voting side of the equation through retail flow. Will that stay that way? I don't know. It'll be interesting to watch. But again, they're a cohort.

Starting point is 01:00:43 We can look at what they do. Now talk a little bit about your strategy because now you have created your own firms, Delphia, and you've implemented the strategy. Is it meeting your expectations? Yes, it's meeting our expectations because we're seasoned at this. We look at our back tests. We look at our thesis and we say this is what we think we should expect out of sample going forward as we invest. And we've met those.

Starting point is 01:01:06 I've been quite delighted that we haven't run into the things that we know are out there. So even when you look at a back test, there are certain conditions in the world where every strategy is likely to, maybe not at the same time, but every strategy has a weakness. And the way I look at it's like, we've had a nice nine-month run. We know there are bad states in the world out there. None of them have shown up in that nine months ago. We know they're there. It's kind of like some horrible venomous fighter running around. You can step on it any time, but it hasn't happened yet. And so we're quite happy. It's been a joy to watch that. You know, I'm tempted to ask you about as you were kind of entering into the finance space,

Starting point is 01:01:41 what books and resources really inspired you. Most people default to like, you know, intelligent investor security analysis. But with you, I'm like, it's probably money ball or something that's like, you know, but what resources have made like the biggest impact on you that maybe others can at least grab for themselves? By the way, the money ball is hilarious. So we call this process of trying to understand what moot's markets. We call it money ball, obviously.

Starting point is 01:02:03 It has a nice ring to it, right? Yeah, so there are two books which I look at as foundational for me and I have my team. them obviously. So one is active portfolio management by Richard Arnold and Ron Con, which doesn't tell you anything about, you know, forecasting the actual returns, but it tells you about all about portfolio construction. So as soon you have some forecasts, how do you put portfolios together? And this book was, it was seminal in the field and continues to be sort of the Bible, the rock on which fundamental investment, every quantitative investing built upon. The other one surprisingly is John Cochran has a book called Asset Pricing, which talks about

Starting point is 01:02:41 in a very generalized and quite beautiful framework, how anything can be priced, whether it's a pair of tennis shoes or wine or equities or fixed income, insurance. So again, hopefully you can see the connectivity. That view of the world is very aligned with me now and our team and our investment style, but in the context, again, of the active portfolio management book, which is this portfolio construction handbook. Now, is your strategy only available to institutions or what does the outside capital look like? So we have two sets of strategies.

Starting point is 01:03:14 One set is available to institutional investors. And other strategies just become recently available for retail. The reason you have to create separate strategies is the legal requirements for investing with institutions are very different than for retail. So I'd love to be able to say that we can take exactly what we do for institutional investors and give it to retail bit. But the law will allow you to do that. Perhaps there's been some evolution in the last year in SEC regulations.

Starting point is 01:03:40 So perhaps there's some things you can do a closer approximation to what we can do for institutions. I would say the biggest difference is that one uses leverage. He's pure long short. That seems to be suitable for institutional investors. We can handle that kind of approach. And then if you go to the other side, you think about long-only or likely-levered products that are available for retail. But as I said, there's things revolving. So perhaps there's some way to make one available for the other in the U.S.

Starting point is 01:04:05 The retail side is only available in the U.S. Institutional side is available to anyone with all. Very cool. Well, Jonathan, this has been a real pleasure and really enlightening. I would love to do this again sometime soon. But before I let you go, I definitely want to give you an opportunity to hand off to the listeners where they can learn more about you and your research and Delphia and anything else you want to share. We have a website, Delphia.com. We'll be creating an institutional website as well in the near future. There'll be papers and we think our thought processes, primarily on the

Starting point is 01:04:36 institutional side, there's always some interesting tidbits about our thesis on the retail side. I would say, just look forward to, you know, we're just getting started. We're only nine months in. I'll be talking to folks in different channels shortly in the near future. Thank you for having this. This is very cool. You have a great podcast. You know, it's an honor to spend an hour with you. Well, I appreciate it. Let's do it again soon. All right. All right, everyone. If you're loving the show, don't forget to follow us on your favorite podcast app. And remember that we always love to hear your feedback. You can always reach me on Twitter at Trey Lockerbie. And if you're just starting out, go ahead and Google TIP Finance. You can find all the courses and resources we've built

Starting point is 01:05:14 for you there. And with that, we will see you again next time. Thank you for listening to TIP. Make sure to subscribe to Millennial Investing by the Investors Podcast Network and learn how to achieve financial independence. To access our show notes, transcripts or courses, go to The Investorspodcast.com. This show is for entertainment. purposes only before making any decision consult a professional. This show is copyrighted by the Investors Podcast Network. Written permission must be granted before syndication or rebroadcasting.

We Study Billionaires - The Investor’s Podcast Network - TIP407: The Evolution of Quant Investing w/ Jonathan Briggs, Ph.D.

There aren't comments yet for this episode. Click on any sentence in the transcript to leave a comment.