Predicting the Market with the Wisdom of Crowds | Vuk Vuković | 630

And welcome, welcome, welcome. This is Peter Winick. I'm the founder and CEO at Thought Leadership Leverage, and you're joining us on the podcast, which is Leveraging Thought Leadership. Today, my guest is Vuk Vuković, and Vuk is the CIO and co-founder of Auriculum Capital, which is a hedge fund that uses wisdom of the crowd and a proprietary network analysis of social media bubbles to predict weekly equity market moves. He's got a PhD in political economics and is just an all-around super smart and interesting guy. So welcome aboard, Vu.

Thanks, Peter. Thanks for having me. Appreciate it.

So tell us a little bit about your journey, 'cause there's an academic piece and then there's the hedge fund piece. So it's, it's really interesting that you've, you sit in a bunch of different areas here, but give us the short version of your journey as a thought leader.

So yeah, so the, the academic piece kind of translated into the hedge fund piece, right? So that was kind of the path, right? I used to work in academia before, so I have my, as you mentioned, my PhD from Oxford and with my two colleagues, one of which is a PhD from physics and the other is a PhD from computer science. We wanted to write a paper that does this new methodology of making predictions, what you mentioned, right? So using wisdom of crowds and network analysis to try to make better inferences than regular polls and other prediction methods. However, when— and we tested this on elections, we tested on students first, but then on elections. And once it was really, really good with predicting Brexit and Trump back in 2016, we decided not to write the paper, but to go with the company instead. And so we founded the company. So it became part of our academic endeavors, essentially. It was kind of a Right, it morphed into it. And that was a company that did mostly market research and elections and stuff from 2016. And then after 2020, after we made the Biden prediction, so another one that was an election that was called really, really accurately this time for a lot of paying clients. Then we decided, let's try to see if this works on markets, on predicting like weekly market moves. So, you know, the S&P 500, is it going up or down on a weekly basis? And it did. I took 2 years of testing. I took my own money, took $20,000 of my own money. Brought it up to 54,000 over 2 years, did it all in the newsletter. So it was public. People were, you know, saying, all right, this is, you know, I was posting screenshots of like—

So you, you were not, so you were transparently testing the methodology.

Absolutely. Yeah. And showing people what— And gaining a crowd, gaining a, so on one hand, gaining a crowd of people that were participating in the, you know, in, in our methodology and in the surveys that we're using. And on the other hand, also potential investors, actually like half of our initial investors Half of the money, at least when we started the fund, came from people who, who saw us on the newsletter. And that's what inspired it.

So stay there for a minute. So not only were you, you doing this transparently, but then you're putting your own money where your mouth was, right? People are watching, 'cause it's one thing to have a theory, but it doesn't get more applied than when, when you're putting your own money up. And, and what I love about the conversation you and I had a couple weeks ago was that was really attractive to investors, right? Because picking a political market is interesting and there is a business model there., but picking the direction of, uh, whatever, it's the S&P or NASDAQ or whatever, there's a lot more money, uh, that can be made and a lot more at stake there.

So, uh, those, those are logic persists to, to the letter.

Absolutely.

Yeah.

Yeah. Exactly. Exactly. So you also mentioned, and I, I wanted to go back a half a step, Wisdom of Crowds. And I remember when I think it was James Surowiecki came out with that book, had to be almost 20 years ago.

I had it somewhere here. Yeah.

Yeah, I've got it sitting on my desk. But, but the, the, and the concept, and maybe you could describe it better than I could, of wisdom of crowd is that the crowd is always smarter than any individual in the crowd, right? Is it, is that kind of, let me give us the definition of wisdom of crowds.

Yes. However, under certain conditions, right? So, so that is true. So that's kind of the, in a nutshell, actually, even, even like 100 years before Suravietsky came out with that, it was a guy, Francis Galton. From England, right? A professor, he was a chemist, I think, a professor. And he was at this county fair where he was trying to, you know, where people were competing, trying to guess the weight of an ox, right? And if you guess the accurate weight of an ox, you get to take it home, which was a big deal back then, right? And so he was writing down everyone's individual predictions and he noticed that no single person was accurate individually, but as a group, when you average it out, they were incredibly close, right? So, you know, like it was like 1 pound was the difference. In weight. So, so he applied this later and there was a lot of bunch of papers doing the same where you have a decentralized crowd where people are not influencing each other's opinions, right? So you're not, so for example, if I ask a crowd and one person says one opinion and then everyone else kind of goes with him, that's not a, you know, that's a biased crowd, right? So you need to have a decentralized.

So they're not influenced by the trend of influence.

Exactly.

Right. So they're making their predictions independent of.

Independently.

Exactly.

So that's what I'm, yeah. So independent or uncorrelated predictions. Decentralized opinion making and an aggregation mechanism, which is like a survey. So you need those conditions and then you can get to the wisdom of crowds. But with us, it was not just the wisdom of crowds, right? So that was one part, right? If we just use wisdom of crowds for the Trump election 2016 or Brexit, it would not have been enough, right? What really helped us, because as I mentioned, right? So sometimes the crowds can be very biased if, yes, they lean in one direction. So what we use in addition was a network analysis, right? So where we can figure out if some people are in kind of bubbles, right? So you have like a conservative bubble and a liberal bubble, and then those people that are in those like echo chambers, they tend to have, they tend to be worse predictors, which makes sense, right? So you, you, you, you don't have a good view of how, you know, the other group thinks if you're only surrounded by like-minded individuals. And when we all are, we all are to a certain extent surrounded by like-minded individuals., but what you really want is heterogeneous groups, right? So think of elections, right? Some of your friends are left-wing, some of them are right-wing, some of them are centrist. So you have a higher probability of getting it right, whereas as opposed to someone who is only surrounded with people who are left-wing Jews or right-wing Jews, that's the idea. And when you apply that methodology, you kind of, you get different weights for people in different levels of network. And that's how you get a more accurate signal. At least this is what we found.

Got it. So your monetization strategy from a thought leadership standpoint is really the hedge fund, right? You've got this unique perspective, you've got this methodology, you've got these frameworks as a, let's call it a marketing launch for lack of a better term. You went transparent on this portfolio, people were intrigued, and then you said, okay, you know what? I need more money, right? I want to do this as a hedge fund. And the people that were following you became fans of the thought leadership. So it's an interesting parallel with a different twist to what many people do, right? You might have an aspiring manager that fire— that, that follows someone who's got great management content and they apply their ideas and they say, wow, they really work. Maybe I'll bring them in or I'll license their content, et cetera. So what other things or ways are you thinking about using productizing, getting your thought leadership out there to achieve your business objectives?

So, I mean, as you mentioned, that was the main one. So we, but it was not the first one that came to mind, right? Initially, once we did this, we went through elections, right? And market research. So that was kind of the obvious point because we were, our claim was, you know, we can do it better than the polls, right? Yeah, we did, you know, back in 2016 and in 2020. And we did French elections, we did a bunch of stuff as well, which was always more accurate than the polls, but there wasn't a lot of I mean, disrupting that industry is very difficult because it's typically linked to either, you know, party or whatever, or certain news media, and it's very hard to get into that. So that's why we did a lot of market research projects on the side as well. But, you know, with this methodology, I mean, it's great with market research when you have like very polar opposite opinions, but it's not good for like, you know, picking an optimal toothpaste or something. The methodology doesn't add value. It's just like a regular survey., but it really adds value when you have like differing opinions. And that's when I said, you know, we should do this in markets. But it took us a while to get there because we needed to get more clients and more money in order to fund this on our own. And that's why it, it took a while, even though the idea was always there. But yeah, it took a while.

If you're enjoying this episode of Leveraging Thought Leadership, please make sure to subscribe. If you'd like to help spread the word about our podcast, Please leave a 5-star review at ratethispodcast.com/LTL and share it with your friends. We're available on Apple Podcasts and on all major listening apps, as well as at thoughtleadershipleverage.com/podcast. So if you think about, let's, let's just apply this to the S&P 500 index to make it simple, right? So the S&P 500 index, you've got your methodology and the methodology's gonna tell you it's gonna go up this week or down this week or whatever the case may be. And then you're gonna make some bets, right? You're gonna make some trades or whatever based on that. At what point does things like regression to the mean and all that come into place? Like, can you beat the market? More than the average that the market will return, I guess is my question.

Yes. I mean, so, so we have thus, thus far over 4 years of data where we're doing this, right? And we, we've noticed that the methodology itself is, let's say, accurate enough. It predicts the market correctly in over 60% of the time. So it's never been less than 60%, it's never been over 70. So somewhere between 60 and 70%.

So if you, so if you can be right 60 to 70% of the time.

Exactly.

Okay. So you're, you're—

doesn't necessarily mean that we make money because we're buying or 60% of the time you buy, maybe, maybe a little bit less, but it's all about like staying consistent, right? And that's the point of, that's the where the trading skill comes in, right? So you can be accurate, you know, as much as you want, but if you're not being consistent in following your methodology, you're probably not gonna make a lot of money. That's true for every, every trading out there. So you need to have, you know, the, the system that works, the signal that works, but you also need to have a consistent trading strategy. What we came up with was just We do a very simple option strategy where we buy, we risk only 2% of the portfolio every week. So that's it. So we take the 2% of the portfolio to buy options premiums, meaning buying either calls, markets go up, or puts, markets go down. And if we're right, we make money, right? If we're wrong, we lose only the 2%, nothing, not more than that. That's, that's the whole idea. So, you know, we're gonna lose money 40% of the time, but in those 60% of the time, we're gonna make much, much more., and grow.

Got it. Got it. So people can understand that because even though there might be some complexity in the algorithms that you're applying, the way that you just explained it is fairly simple, right? And you're also not making an outlandish claim of, you know, we're gonna beat the market 99 times out of 100, right? Because that's not, you just need to be a fair amount better than a coin toss, right?

Yeah, exactly. And, and that's the point. I don't know if you heard that. There was a speech by Roger Federer. That's a great thought leadership piece, right? When he, when he said, like, it was like a Duke University or something, he said, You know, I won 80% of all my matches, but I only won 54% of all my points throughout my career. 54%, right? It's like just a little bit over a coin toss, but that's the point. He knows like when you're in the zone, when you're in the game, you know, if you lose a point, doesn't matter. You go on to the next one and that's how you have to think of this, right? So every week for us is a new thing, right? Every week is whatever happened the week before doesn't matter. Every week we start over and I need to be in in the market every single week because I don't know which week is gonna be a good one, which one is gonna be a bad one, right? If I knew that, obviously.

Well, so, so going into a, a different channel of thought leadership, when you talk about something like black swan, right? If you miss a week, that could be, you know, if it, and I guess in your case, the down week, your downside is capped at 2%, but an up week could be working up 20% or something, or, or something really extreme.

Or, or, or vice versa. So if it's, um, so if we're short the market, And there's a black swan event that sends anything, everything down, then you can make, you know, 10, 15% a week. It can happen.

Yeah. Interesting. What other ways are you going to continue to get the thought leadership out there to attract investors and such? Any other, 'cause you, you've gotten it in a certain way, right? Where they're following your, your trends, speaking, writing, like what other formats and modalities you're thinking to use to put the thought leadership in? To amplify the message that it's got?

So, great question. I mean, on various things, and typically the way that we developed this was from my own outreach on LinkedIn and Twitter, right? So I have, I do a lot of things that are trying to explain concepts in economics or in politics, political econ, which kind of my background. One thing that I'm using to that extent is my book that recently came out, was published by Oxford University Press. Now this has nothing to do with the fund. It was my PhD thesis from Oxford where the thesis was, I was looking at the collusion between politics and the corporate world and how it affects top income inequality, right? And I wrote a whole book around it, some history, some kind of, you know, a unified theory and it came out and yeah, it's called Elite Networks: The Political Economy of Inequality. It's available on Amazon, Barnes Noble, et cetera. And this has kind of been, something that I'm, you know, pushing forward into, you know, presenting myself out there as, as an author, as someone who, you know, knows stuff about, you know, economics and, and politics, et cetera, and happens to have a, a fund lying around based on, so it's, so the network theory is underlying, uh, the underlying mechanism behind both, right? So I'm using network theory to explain the collusion between elites on, on, on the top level of, uh, of society. And I'm also using networks to try to understand You know, social media bubbles.

So I want to go back to sort of the beginning of that story where you said you've got the book out and it's based on your PhD thesis. Now I would argue that you didn't just take your thesis, hand it over to a publisher and say, publish that as a book. How do you change the, not the methodology, but let's say the language, you know, because most, with all due respect, most PhD theses are unreadable to the average human.

Extremely boring. I agree.

Yeah. Yeah. How do you move that? It's almost a translation issue. How do you move it to a book? So it's, it is a readable, uh, it's a place I want to spend 6 or 8 hours with.

Right. Exactly. I mean, so it's, it's not easy. Obviously you have to kind of, um, people like to say dumb down. I don't like to use that word. I like promoting it to a general audience. Right. So talking about things, that's why I introduced a lot of history, a lot of storytelling, right?

Yeah.

Because on one hand, because I know my findings, I know my research. It's not, you know, I don't have to talk about the methodology, the research. But I can talk about stories that explain what the research is about, right? So I can talk about things like, so how does these elite network mechanism work in different countries? How, how it worked in the US during the 19th century and doing Tammany Hall in New York, for example, right? How it worked in countries like, you know, Tunisia or, or, uh, or Indonesia or whatever, right? Uh, Russia, Venezuela.

So it's the connection of the storytelling. So, so if I were to sort of simplify what you said, in a thesis, the data is king, right?

Yes.

The data has to, has to stand up and the storytelling probably is secondary. Secondary, if not tertiary, right? Doesn't even matter. That's true. When, when you're going to put a book in general format, the storytelling has to rise to the top and it has to use the data and connect the dots there because most people need the, you know, humans need stories to bring something to life. Numbers don't typically do it.

So, exactly. That, that's, that's absolutely the idea. And then if you throw in a little bit of history, a little bit of, you know, yeah, famous examples. So I talk about, you know, Sam Bankman-Fried and Elizabeth Holmes, those kind of things, right?

Yeah. Yeah. Yes.

But the collusion element, and that makes it interesting, right? That makes the whole, uh, the whole theory applicable, right? And, and finally, again, so, so someone would say, but this is not, you know, these are just anecdotes. Of course they're anecdotes, right? But there's a theory behind it and I'm using the anecdotes to make the story more appealing, to make it more attractive so that people can understand what I'm talking about.

Love it. Well, this has been great, Vuk. I appreciate your time and best of luck. Thank you.

Absolutely. Enjoy.

To learn more about Thought Leadership Leverage, please visit our website at thoughtleadershipleverage.com. To reach me directly, feel free to email me at peter@thoughtleadershipleverage.com. And please subscribe to Leveraging Thought Leadership on iTunes or your favorite podcast app to get your weekly episode automatically.