Interview with Cai Jiamin: How to Achieve an Annual Income of Hundreds of Millions Using Algorithms? (Part 1)

At the age of 12, he had only HK$20 for lunch, which could only buy the cheapest boxed lunch at the corner store for HK$12. He saved the remaining HK$8, not even daring to dream of the toys he liked. Seeing his classmates' confidence in choosing freely, this poor student was eager to find a shortcut to making money—he studied more than 100 draws of the Hong Kong Mark Six lottery, trying to win the HK$8 million jackpot with HK$5, but only ended up disappointed time and time again.

This "path to alchemy through trading" is fraught with thorns. At 14, he used his brother's account, funded by red envelopes, to enter the market. At 16, his greed for 10x and 20x leverage wiped out his 40,000 yuan capital overnight. At 19, he tried again, saving 150,000 yuan through tutoring, only to have his account wiped out once more due to aggressive trading. Twice severely punished by the market, he finally realized that the emotions and fervor of manual trading ultimately cannot overcome the greed and biases of human nature.

He abandoned his fantasies of getting rich quick and instead embraced the power of data, treating trading as a rigorous scientific experiment. He validated each strategy with historical data, investing only in those that proved effective and iterating on those that didn't. Therefore, quantitative trading may waste time, but it won't waste money. Ultimately, he achieved an annual income of over 100 million yuan through quantitative trading, profiting in both bull and bear market cycles.

At this point, making money became just a number. "Traders who make money love trading itself," he began to share and educate, encouraging more people to do what they truly want to do, rather than doing things they don't want to do for money.

He said that there are no born trading geniuses, only those who study and persevere relentlessly . Maintaining rationality and not being swayed by emotions, correcting cognitive biases promptly and avoiding getting stuck on trivial details, and always maintaining a humble attitude towards learning—these are the core secrets to navigating bull and bear market cycles and building an independent trading system.

This episode of OKX's "Dialogue with Traders" features an in-depth conversation between OKX Mia (@mia_okx) and quantitative trading champion Calvin Tsai, exploring how he built his quantitative trading empire from the ruins of two margin calls, and the purest passion and philosophy for trading behind the cold data.

The following is the full transcript of the dialogue (after editing).

I. The Road to Godhood | From a Million-Dollar Initial Investment to an Annual Income of Hundreds of Millions in One and a Half Years

01 From traditional finance to the crypto industry: Earning 20 times the money in less than 2 years

Mia: Hello everyone, I'm Mia. Welcome to the seventh installment of OKX's "Dialogue with Traders" series. Here, we'll be interviewing renowned traders in the industry to discuss their experiences, from the thrill of their first trade to their trading strategy logic and the ups and downs of market cycles. Each trader has their own unique story and methodology—some suffer overnight financial ruin, others turn the tide; some insist on manual trading, others on quantitative operations. But what they have in common is that they can still find their own path to victory amidst market volatility. Today, we have invited quantitative trading champion Calvin Tsai. He once achieved a peak of 3% of OKX Solana trading volume on our OKX exchange and manages a quantitative fund of 160 million. It's a great honor to have Mr. Tsai here. Please introduce yourself.

Calvin Tsai:

It's a great honor to be interviewed and share my trading journey. I used to work in traditional finance, and after graduating, I joined a hedge fund. We used quantitative trading methods, mainly dealing with different stocks, such as A-shares, Hong Kong stocks, and US stocks, as well as some commodities like gold, silver, copper, and crude oil. Around 2020, I discovered I could try applying traditional strategies to cryptocurrencies, such as Bitcoin. It turned out to be more profitable, so I gradually devoted my time, energy, and capital to it. Now, I've been trading cryptocurrencies for about four and a half to five years.

Mia: Calvin is one of our OKX "Very VVIP" clients, a true billionaire. When did you first get into the crypto industry? And how much initial capital did you have?

Calvin Tsai: I bought my first Bitcoin during the summer of 2017. I remember it clearly; a college classmate invited me to dinner and casually asked if I'd ever heard of Bitcoin. I'd never heard of it before, so I Googled it and found it quite interesting. The more articles and discussion forums I read, the more curious I became, so I decided to buy one Bitcoin. I transferred money from my bank account to a trading platform—there weren't many options back then. My first Bitcoin purchase price was $3,000. After buying it, I just left it there, and soon I saw it rise from $3,000 to $20,000—that was in December 2017. I really felt like I'd struck it rich overnight. Although I didn't buy much, only a few hundred thousand Hong Kong dollars, I thought, "This asset has such explosive potential; I need to keep observing." Unexpectedly, in 2018 it returned to $3,000. I felt like I'd had a dream—it rose sixfold and then came back to square one. Later, I didn't pay much attention to it and continued my work at my hedge fund. 2018 was a bear market, and 2019 saw only modest gains, making it a slow and steady bull market. In 2020, I started noticing news reports about the Bitcoin halving—which occurred in May 2020. That's when I began to study it seriously and discovered that trading volume was gradually increasing and the market size was expanding. It was then that I started applying traditional strategies to the crypto market, testing them with historical data.

Mia: So, how did you go from a small initial investment to earning hundreds of millions? How much was your initial investment?

Calvin Tsai: My initial investment wasn't much, around HKD 1 to 2 million. That was in 2020 and 2021. 2021 was arguably the most profitable year. Starting in May or June of 2021, I moved my quantitative strategy to the crypto market, opened a new account to test the waters and run the strategy. Then, from May 2021 to January 2023, my account grew from several million HKD to HKD 100 million—about a 20-fold return in a year and a half.

Mia: You earned 100 million in a year and a half. What happened after that? Since you entered the industry in 2017, it has been five or six years now. What has been your average income in the years since then?

Calvin Tsai: The following years were roughly the same, with an average annual profit of about 100 million. This is because our quantitative strategy has an average annual return of over 100%.

02 The secret to outperforming bull and bear markets: Trading is more profitable than holding cash.

Mia: OK, but when we first came into contact with quantitative finance, we might have thought that quantitative finance is a way with low drawdown and stable returns. But can you grow a small amount of capital to a hundred million? Is it all thanks to quantitative finance?

Calvin Tsai: It's all about quantitative trading. Many people think they've made money by simply holding Bitcoin, but actually, holding Bitcoin hasn't resulted in such a huge increase. For example, in 2021, Bitcoin peaked at $60,000, and now it's $120,000, which is only a 100% increase, not a significant one. Even if you bought from the lows of 2021, say $20,000 or $30,000, it's only a 400% increase. So, simply holding Bitcoin, even with leverage, won't generate as much profit as quantitative trading or traditional trading. The key to trading is—you can profit from market fluctuations and pullbacks. Making money with Bitcoin in a bull market is normal. In a bull market, whoever makes the most money has the highest leverage. But in a bear market, the crucial thing is how to avoid pullbacks and prevent being wiped out. In a bear market or during a correction, do you have a "short selling" strategy that tells you: "At this point, I need to short; I don't want to hold Bitcoin; I need to sell my Bitcoin." Trading or quantitative trading allows you to make money in any market condition. For example, we also made money in 2022, with a return of approximately 240%. In a bear market, we can short sell and profit from the decline. So I think this is important—trading will definitely earn you more than just holding Bitcoin.

II. Trading Philosophy | The Underlying Logic of "Code Printing Money"

01 Core Methodology: Mid-to-High Frequency CTA and Risk Management

Mia: Regardless of whether it's a bull or bear market, you're all "printing money"?

Calvin Tsai: We'll try our best. Actually, there are times when we lose money, and strategies can fail at times. For example, some strategies might not make money for three or six months in a row. That's the most painful time for us when developing strategies—we have to think, has this strategy failed? Should we drop it? Or continue running it? Will it one day reach a new high? We have to think and judge whether this strategy can still make money. This is a very, very important point.

Mia: So, during this process, did you experience any significant pullbacks?

Calvin Tsai: Yes. It's not always profitable. For example, a major pain point in quantitative trading is determining whether a strategy can continue to generate profits. Sometimes a strategy might not make money for three or six consecutive months. That's the most crucial point in quantitative trading—determining whether the strategy has become ineffective. If it has, you need to remove it from the portfolio as soon as possible. But if it can still generate profits, then you should keep it. So, a very important judgment point in quantitative trading is: to think about the logic behind a strategy and whether it can keep you "alive" in this market.

Mia: Do you have a specific case to share with us? What was the biggest drawdown you've ever experienced?

Calvin Tsai: The core of our strategy is CTA (Commodity Trading Advisor). CTA is trend trading, or directional trading. Unlike other funds, some do high-frequency trading, some do arbitrage, and some do low-frequency trading, such as judging major trends over six months or a year. We do mid-to-high-frequency CTA, using hourly charts to judge whether the market is going up or down. If we think it's going up, we go long; if we think it's going down, we go short; profiting directly from the direction. We don't do both long and short simultaneously; we don't "go long on one batch of coins and short on another batch." That's what we call a "long-short strategy," but we are a pure CTA. The biggest difference between pure CTA and other strategies is that we experience relatively large drawdowns. Our drawdown ratio is the highest among different fund types. For example, high-frequency trading usually has a drawdown of no more than 1 point; arbitrage drawdowns are no more than 3 to 5 points; but with CTA, drawdowns can exceed 10 points or even 20 points. We encounter drawdowns exceeding 20 points almost every year. When that point comes, we feel psychological pressure. Investors will ask us: "Is this strategy still working? Can the fund still make money? Do you still have confidence? Have you changed the strategy? Have you reallocated the weights in the portfolio?" They will ask all sorts of questions. At that time, we need to judge from the data and from different perspectives: can this strategy continue to work? This is a very important point.

Therefore, we actually encounter pullbacks of varying degrees every year. For example, some pullbacks can be as large as ten or twenty points. In fact, ten or twenty points is considered quite small.

Mia: Really? You're comparing it to a long position?

Calvin Tsai: Yes, after all, we're not purely proprietary trading, because many of our trades are funded by our own capital. I think proprietary capital can withstand, say, a drawdown of 50 points, or even more. But if you also have investors or clients, the situation is different. For example, if you lose 30-something points, they will definitely ask questions, and might even call in the middle of the night to inquire. So, to make investors feel better, I think a drawdown of 20 or 30 points is already the limit.

Mia: So how do you explain it to these investors every time they call to ask you questions?

Calvin Tsai: When drawdowns are significant, investors need time to adapt. For example, some investors might enter at the peak of our curve and immediately encounter a downtrend, potentially losing 20%. They'll be uneasy, wondering, "Why did I lose 20% right away? Is this a scam?" They'll definitely ask this question. But in my experience, if they stay longer, say more than a year or two, they'll see this pattern: first a 20% or 30% loss, then a rebound to new highs, resulting in a 100% gain. So they'll gradually get used to it, recognizing or understanding what CTA (Commodity Trading Advisor) is, rather than just high-frequency or arbitrage. Many people think that eight out of ten funds in the market are arbitrage or high-frequency, and losing two or three percentage points is common, so why is it 20% here? Therefore, I need to educate them about CTA, what we do, and the logic and foundation we use for judgment. I think this is an educational process that takes time for them to gradually adapt.

Mia: What are the most important metrics you use when doing CTA?

Calvin Tsai: There are two types of important indicators. The first type is what we use to make judgments, such as generating signals, which involves looking at data to determine market direction. Regarding this issue, we don't actually have any single factor with a particularly high weight, or any single factor with excessive emphasis. For example, if you hear about a fund or institution that has a very important indicator, I'll start to wonder: if this indicator fails, will the impact on the fund be very significant? Therefore, I prefer to distribute the weight of each indicator and each factor as evenly as possible. This way, if one factor fails, it won't have a significant impact on the entire portfolio. So, we shouldn't let any one factor have an excessively large weight. I've tried this before: if a factor is very profitable, I gradually increase its weight until, at some point, it might account for more than half of the entire portfolio. For example, out of 100 yuan, more than 50 yuan is invested in this factor. It might be very profitable next month or next week, making the overall portfolio's profit ratio very high, but if it fails next month or next quarter, it will cause the entire portfolio to fluctuate greatly. Therefore, we prefer to even out the weight of each factor.

The second approach involves determining which metrics we use to judge whether a factor can generate profits. We prioritize its risk-reward ratio, such as the Sharpe ratio, which is a crucial metric; a higher Sharpe ratio is generally better. We also consider the Calma ratio, which is the annual return divided by the maximum drawdown. We examine these different figures to assess whether a factor is a good indicator.

Mia : In this process, if you can achieve a return of over 200% even in a bear market, do you have any unique methods or strategies? And how do you control risk?

Calvin Tsai : This is a very difficult question to answer. Simply put, it's about how to build a truly successful quantitative model, right? I think you need to do each step well. For example, you need more data, you need to be more precise in analyzing the data, and when building a quantitative model, you need to test different models to see which factor is more useful. At the same time, you need a very rigorous methodology to determine whether a factor is truly useful. What is "false usefulness"? False usefulness means that something looks very profitable in the database over the past three to five years, but when you run it in a live trading account, you find it loses money. In other words, you think it's useful during testing, but it's not useful in actual operation. We used to encounter this situation frequently, so we had to have a strict method for screening. Only after confirming that it's truly useful would I slowly add it to the live trading account, adding it in small increments, one dollar or two dollars at a time. Another point is that you need to judge from the underlying logic whether a factor can truly make money, which is very difficult. Many people find that factors that seem good don't make money when used, or even cause losses. Therefore, every step must be done meticulously, rigorously, and thoroughly.

Mia : So, did you do your own strategy development?

Calvin Tsai : Yes, there are several people on our team. I'm mainly responsible for strategy development. Other colleagues are responsible for system development and machine learning research.

02 Strategy Iteration: From Inspiration to Implementation, a Rigorous Three-Step Approach

Mia : For example, if you have a strategic idea, what is the entire iterative process from strategy development to implementation? Can you share it?

Calvin Tsai : Initially, you need to have a set of rules, just like with lot trading. Lot trading also involves certain judgment points. For example, some people look at charts, some look at prices, some look at volume, some look at news, some look at KOLs (Key Opinion Leaders), or information provided by others; everyone's judgment points are different. At the beginning, you need to clearly define which factors or indicators you will consider for your entry strategy. Using the simplest example of price, for instance, some people look at moving averages, specifically the 20-day moving average. If the price breaks through the 20-day moving average, I buy. So the first step is to determine which factors to use.

The second and most important step in quantitative trading is to validate the strategy's profitability using historical data. Therefore, we search for price data from the past three to five years, whether online or from exchanges. Then we determine the trading frequency—whether to look at minute, hourly, or daily prices—and input this data into the computer to begin programming. For example, we might establish a 20-day moving average logic: buy if the price is above the 20-day moving average; sell if the price falls back below it. Then we run the strategy on the computer, which will tell us the average annual profit over the past five years, which months resulted in losses, the profit/loss ratio, Sharpe ratio, Calmar ratio, etc., to determine if the strategy is up to standard. You can set criteria, such as an annual profit exceeding 50% and a drawdown of less than 20% to pass. If it passes, we move to the next step; if it fails, we adjust the parameters. For example, the 20-day moving average can be changed to 10 days, 30 days, or even tried up to 100 days to find the optimal parameters. Ultimately, we might find that a 50-day moving average is the best. Then, use the 50-day moving average strategy in a demo account for one or two weeks or a month to see if the system runs stably and if the signals are sent to your computer in real time, telling you how much Bitcoin to buy or how many contracts to sell. After completing the demo trading, move on to live trading. In live trading, gradually add to your position, starting with $100, then $1000, then $10,000, and finally $10,000 until you reach your target position.

The final step is risk management, checking whether the strategy has caused significant losses in the portfolio or whether it has failed. If everything is normal, let the strategy gradually generate profits.

Mia : How often do you iterate on your strategy?

Calvin Tsai : It depends on the frequency of the strategy. If it's high-frequency, such as at the second or minute level, we adjust it more frequently, perhaps once a week or every few days. But if it's slower, such as at the hour level, we might only adjust it once a month or every few months.

Mia : With so many quantitative teams developing strategies now, how do you maintain your leading position in the industry while also maintaining very high returns?

Calvin Tsai : I think it's related to a point I just mentioned—trying to be as meticulous as possible in every step. It also relies on data. The foundation of quantitative trading is having a large amount of data, and the data you're looking at needs to be different from others. To earn money that others can't, you need to look at things others haven't looked at, or things others have overlooked. I've noticed some teams that didn't look at on-chain data before, so I started looking at on-chain data. I found some teams didn't pay attention to the sentiment in discussion forums, so I started paying attention to sentiment. Many teams are looking at charts and prices, but I don't look at charts or prices; I try to do what others don't do. Only by doing what others don't do can you earn money that others haven't earned.

Mia : So, if you encounter some extreme market conditions, such as the LUNA collapse, how would you adjust your strategy or hedge to ensure that you don't suffer significant losses?

Calvin Tsai : LUNA collapsed in May 2022. At that time, our quantitative trading portfolio didn't trade LUNA; we mainly traded large-cap coins like BTC and ETH. LUNA was traded in my spot trading account, but most of the funds were in the quantitative portfolio. At the beginning of 2022, I bought some LUNA. By April, I noticed it was offering about 20% interest annually, which I felt was unsustainable. In the long run, such high interest rates were difficult for the protocol to support. I looked at its reserves and found that the funds weren't enough to support several months of high interest payments. So I shorted one of the interest-paying protocols, called Anchor. At the time, I was long on LUNA and short on Anchor, with a roughly 1:1 ratio. When LUNA crashed in May, my spot trading account was both losing and making money, ultimately breaking even. As for the quantitative portfolio, we continued trend trading. The market was very volatile when LUNA collapsed; initially, we might have gotten the direction wrong, but if the market moved in a certain direction, we could still make money.

03 The AI Wave: Opportunity and Threat

Mia : I remember you mentioned the AI trading crisis in a previous interview. Do you think you'll use AI now, given the current AI wave? Will AI pose a threat to quantitative trading?

Calvin Tsai : AI has indeed been helpful to our quantitative trading system. We have about two or three layers of strategies that use AI to generate signals. We feed in different factors and different data, train them using time-series machine learning models, and then it generates signals, such as whether to go long or short. In live trading, these strategies are profitable, so we have two or three layers of strategy combinations that rely on AI to generate signals. In addition, AI is also a great help in programming. Previously, writing a piece of code might take ten hours, but now, using tools like ChatGPT and DeepSeek, you can complete it in just five or ten minutes. The same function is much more efficient and saves a lot of time. Of course, it sounds like a huge benefit, but other institutions or teams can also use the same AI tools and machine learning models to do quantitative trading and help them improve their efficiency. So AI has both opportunities and risks. It can help you, but it can also make your competitors stronger. The key in the next five or ten years is how to make good use of AI tools. Everyone uses AI differently. Some people may not be able to produce useful signals using the same machine learning model, but you might be able to produce effective signals. Every detail matters. If you can use AI more meticulously and efficiently, this should be a key focus over the next five to ten years.

Mia : You just mentioned that AI can be used to generate some useful signals. How long did it take you to complete this training process?

Calvin Tsai : We actually started testing it back in 2021. At that time, we didn't find anything particularly useful, so the early results weren't significant. In 2022, we revisited whether machine learning could generate revenue, but it wasn't very profitable that year either. For the first year or two, we didn't truly integrate machine learning into our strategies. Then, in 2023, with the rise of the AI wave, many people around the world started discussing AI tools, and we tested it again, finding that it was starting to generate revenue. I think there's an effect we call "self-fulfilling prophecy." It means that as more and more people use this tool, it becomes increasingly useful. From 2023 to 2024 and 2025, each year's results were better than the previous year. More and more people are using it, so it gradually went from "useless" to "useful."

Mia : I've noticed you're a very perceptive trader. For example, you can spot what others aren't doing and optimize your own strategies accordingly. For instance, you've been working on machine learning and AI since 2021. So, how did you cultivate this acumen?

Calvin Tsai : I think part of the reason is my experience working in a traditional hedge fund. Back then, a company might have six teams, each competing independently. We'd see the performance of different teams each month; for example, one team might be profitable for several months in a row, while other teams weren't making money in the market. At that time, I would go to those teams to have meals, learn, and exchange ideas with the traders. I think this is very different from trading at home. Trading at home is a closed environment; there's no one to talk to, no one to comfort you when you lose money, and no one to share your joy when you make money. In an institution, the advantage of a team is that you can communicate and share. I also learned how to trade other asset classes from different teams. For example, I wasn't trading forex at the time, but other teams were, and they were willing to share their experience. Although we traded different instruments, I could learn the logic and methodology of their strategies from their forex or other asset trading approaches. So, I think communicating with different traders is very helpful in improving your acumen.

III. Two Setbacks to Zero: The Making of a Trader

01 A Child and Trading: Starting with Studying Mark Six at Age 12

Mia : Calvin just mentioned his past experiences in traditional finance, so I think we should start by talking about how you first learned to trade and how you became such a legendary trader. I remember you mentioning that you started learning to trade with your first red envelope money when you were 12 years old? At 12, I didn't even know what I was doing, was I just playing with clay? (laughs)

Calvin Tsai : I think I was in middle school then, in my first year. My family didn't have much money back then; I was quite poor. I remember having only about 20 Hong Kong dollars in my hand when I had lunch—really, just 20 Hong Kong dollars.

Mia : When I was 12, 20 Hong Kong dollars was still a good amount, right?

Calvin Tsai : I remember every day at school, lunch was expensive. Some meals cost around 25 yuan, which I couldn't afford; others were cheaper, around 15 yuan. I could only choose the cheaper option. I also remember there was a sports field near my middle school with a small shop where food was incredibly cheap, only 12 yuan. Every day I would go to that shop, spend 20 yuan on 12 yuan worth of food, and have 8 yuan left over. I was really poor back then. I didn't have money to buy toys, and extracurricular activities at school cost money. I saw other students and envied them. Why were their families so rich? Why could they freely choose what they liked? At that time, I wondered, what methods could I use to make money?

The first thing that came to mind, since I didn't have any capital at the time, was to buy lottery tickets, specifically the Hong Kong Mark Six lottery. A ticket only cost 5 dollars, and if you won the grand prize, it would be about 8 million Hong Kong dollars—the method with the highest leverage ratio—turning 5 dollars into the possibility of millions. After seeing this possibility, around the summer after my first year of middle school, I started researching how to predict the next draw's results. The local Mark Six lottery requires guessing 6 numbers plus a special number, choosing which 7 out of more than 40 numbers will be drawn. I backtested the data from the past 100+ draws to see if I could predict, for example, how many times number 1 would appear consecutively, how many times number 2 would appear consecutively, and then predict what numbers would be drawn tonight. I spent about two months on this, but failed every time. Later, I gave up on this method, feeling it was completely random.

The second thing that came to mind was trading. Working was illegal at 12; you had to be 16 or 18. Besides, there was no leverage, and you only earned 40, 50, or 60 yuan an hour—no way to get rich quickly. So I thought of trading stocks. However, for a 12-year-old like me, stock trading was still a distant dream. Without parental guidance, it would be difficult to achieve. At that time, my parents didn't trade stocks either, but I had an older brother, eight years my senior, who had just entered university and opened a stock account to look at stocks. Seeing him look at moving averages and charts sparked my interest. I thought, if I could successfully predict whether stocks would rise or fall tomorrow, I could make money.

Mia : Did that older brother make money through stocks back then?

Calvin Tsai : Sometimes I made money, sometimes I lost money. Anyway, later I started reading books and doing my own research. When I was about 14 years old, I really studied for a whole year. I used my high school textbooks to write down my predictions about whether a stock would go up or down the next day. Then the next day I would go to my computer and see if it had really gone up or down, just like checking answers—OK, this one was right, this one was wrong, this one was right again.

Mia : Like a demo disk?

Calvin Tsai : Yes. For a whole year, my "win rate" was pretty high. So I decided to open a live trading account. I wasn't 18 yet, so I asked my brother to open a securities account for me at the bank downstairs from our house using his ID. I told him—this is my red envelope money, I'll set the password myself, don't touch the money in there.

Mia : We also need to guard against him?

Calvin Tsai : Yes. I'll use your identity to open the account, but the money is mine, the password is mine, and you don't need to worry about what stocks I buy. He said OK, no problem. I remember making money in the first year, earning about 30%. When I turned 16, I thought—since I'm making money and have a decent eye for opportunities, I might as well try high leverage. So I started using derivatives. At that time, Hong Kong stocks didn't have leverage, but there were bull/bear contracts and warrants, which could achieve 10x or 20x leverage. I transferred my money from stocks to bull/bear contracts and warrants to try higher-risk ventures.

02 Lessons from Margin Calls: Two brushes with zero at ages 16 and 19

Mia : Starting to use high leverage?

Calvin Tsai : Yes, I started using high leverage. I remember one day I turned on my computer and saw that my principal had dropped from over HK$40,000 to just over HK$200, or even just over HK$100. I thought I had logged into the wrong account—why was there so little left?

Mia : Did you think it was stolen?

Calvin Tsai : I thought it had been stolen.

Mia : Did you think your brother took it?

Calvin Tsai : Yes. Later I found out it wasn't stolen by a person, but "stolen" by the market. At that time, I had no idea how to face it. Seeing that there were only a few hundred dollars left in my account, I didn't dare tell my family or my brother. I had no idea what to do. I remember that night, I went for a walk and chat in the park with my best friend from high school. I asked him, because his dad was involved in stocks, "If it were you, how would you feel if you lost money in the stock market?" At that time, I didn't know what money was, nor how to deal with losses. For me, making money and losing money were like playing a game. The first time I saw myself lose money, I had no idea how to face it or what emotions to have. He thought for a moment and said, "I wouldn't feel anything even if I lost everything." I said, "Are you stupid? Why wouldn't you feel anything if you lost 100%?" Then he told me, "Think about it, you can still eat tomorrow, go to school as usual tomorrow, sleep in your own bed tonight, and life will go on as usual tomorrow. You don't have children to raise or a family to take care of, so whether there are 30,000 or 40,000 dollars in your bank account or 300 or 400 dollars, it doesn't make any difference to you." I thought about it seriously and realized he was right. That moment had a huge impact on me. I realized—it really is a good time to start trading when you're young. If you lose a lot of money when you're thirty or forty, with a family and children, it would be terrible. So at that time, I decided—to learn more about trading while I'm young and use more leverage. Leverage is something only young people can play with; that was my biggest feeling and thought at the time. After my first account blowout, I continued to work hard to learn how to trade. When I was in college, around nineteen, I had my second account blowout—and lost even more. I was a freshman then, and I was also tutoring high school students, so I quickly accumulated some capital to trade. At nineteen, I went back to leverage, using options and futures, and once again lost everything from about HKD 150,000.

Mia : From age fourteen to nineteen, you didn't trade much, right? Were you mostly studying?

Calvin Tsai : I am studying. I also read books sometimes, especially investment books—I read different kinds, such as books on fundamental analysis, technical analysis, and charts.

Mia : You are very patient. From the very beginning of your trading journey, you studied from the age of twelve to fourteen, and then you traded. After your account was wiped out, you continued to study for another five years.

Calvin Tsai : Yes, that's right.

Mia : I've been waiting, and then I'll strike again at nineteen.

Calvin Tsai : Yes. Because when I was eighteen—I just mentioned that I had accumulated some money from tutoring—I had the funds to go back to the battlefield.

So I entered the market, back when I was trading Hong Kong stocks. Then I also encountered options, futures, and different high-leverage products. By the time I turned nineteen, I had lost all of my 150,000.

03 Quantitative Insight: Abandon Manual Operations, Embrace Data

Mia : Yes, that feeling is really unpleasant. Because this is the second time you've encountered this situation, and you've already tried your best.

Calvin Tsai : At that time, I thought, I've been reading the news, I've looked at different charts—second-level, minute-level, hourly, daily charts—and different technical indicators. I've also heard different people talk about fundamentals, whether a company is good or not, and its future prospects. I felt I had analyzed it thoroughly, so why was I still losing money? I couldn't figure out why, so I seriously considered—was there something wrong with my method? I felt that my current approach was incorrect, which led to those two major losses. Then I wondered if there were other trading methods that could allow me to consistently make money in the market again. It certainly wasn't my current method. Then I searched online and read different books, and I came across the words "quantitative trading." I realized that what I had been doing all along was ignoring one point—I hadn't used historical data to verify whether my method could make money. I had always been listening to what others said. If someone said that the 20-day moving average could make money, I thought it really could; if someone said that a stock was good, I thought it was good; if someone said that an industry had great future growth potential, I thought it must be good. Theoretically, you should verify your method. For example, were what this person said previously correct? Are the KOLs' statements accurate? If we had used the 20-day moving average to buy and sell ten or five years ago, would it really have made money? I realized that I had overlooked a crucial point in my previous manual trading: I hadn't used historical data to verify the effectiveness of my methods. Quantitative trading is about using historical data to see if your method can make money. Only when it proves profitable should you start investing. Therefore, quantitative trading may waste time, but it won't waste money.

Mia : You experienced a huge shock like a margin call when you were very young. Did it affect your personality and subsequent risk control?

Calvin Tsai : I think it's helped me gradually become less emotionally affected by profits and losses. Before, I'd be happy to see profits and unhappy to see losses; my emotions were completely tied to the market. When the market went up, my emotions went up; when the market went down, my emotions went down. Later, I gradually developed the ability to not feel much difference whether my account was showing a profit or loss. I think this has made me more rational.

04 Choosing Pharmacy as a Major: Keeping a backup plan, but not giving up on your passion.

Mia : I understand. I remember you studied pharmacy in university. Normally, after studying trading for so many years, you would have chosen finance or a related subject. Why did you choose pharmacy?

Calvin Tsai : Back when I was choosing my courses in high school, I was already thinking about what subjects I should take in college. I also asked for different people's opinions, and they said, "Take subjects that require a license to work in this industry." For example, doctors, lawyers, and pharmacists all need to study those subjects in college to get licenses and work in the industry. But for things like finance, stocks, and trading, you don't need any licenses. So many people suggested—"Take subjects that require you to get a license in college."

Mia : Leave yourself a way out?

Calvin Tsai : Yes. Back then, for example, in Hong Kong, I heard that the medical industry had relatively stable salaries. So I thought: OK, I need to find a relatively stable job to help me with trading. Even if I lose money trading, I'll still have a stable job to balance things out and have cash flow to support my assets and money. That's what I was thinking at the time. Later, in my junior year of university, I participated in several trading competitions and was fortunate enough to win some awards. At that time, a friend said to me, "Have you ever heard of proprietary trading firms?" I asked, "What are proprietary trading firms?" My friend said, "These companies give you a sum of money and don't care what strategies you use; you decide for yourself. If you make money, you share the profits." I thought this model was pretty good. Anyway, I was trading myself, and my initial capital wasn't large. If I joined a company, they would give me a larger sum of money, and I could use other people's money to make money and share the profits. So I became very interested in this, searched online, found a proprietary trading firm, and went for an interview. During the interview, I showed them the awards I had won in the competitions.

Mia : What awards did you receive at that time?

Calvin Tsai : At the time, I participated in some manual trading competitions and also quantitative trading competitions. Quantitative trading competitions are quite different from manual trading competitions—quantitative trading requires you to program your own strategy. After you write the strategy, they will use historical data to test whether the strategy is profitable. If it is profitable, you are the champion.

05 Joining a hedge fund: Analyze data more thoroughly and comprehensively, and learn to manage client expectations.

Mia : So back then, did you also teach yourself programming when you were doing quantitative analysis?

Calvin Tsai : Yes, I taught myself in college. My major didn't include any programming courses. When I went for interviews, I showed them my monthly statements, which showed I'd been making money for several consecutive months. They asked me, "How did you make these trades? Why did you make these trades?" I clearly explained: "OK, I looked at the data, did backtesting, and saw that this strategy consistently made money, so I used it." And that's how I successfully entered the industry and got my first quantitative trading job.

Mia : How much money did he give you to do it?

Calvin Tsai : Back then, it started with demo trading. He would observe you for a few months to see if you could make money in the demo account. After you made money, he would gradually understand the essence of your strategy and risk control, and then decide how much money he would give you. It might start with a few million Hong Kong dollars, and if you performed well, it might gradually increase to tens of millions of Hong Kong dollars.

Mia : OK, so how long did you use the demo account before you started making a profit?

Calvin Tsai : At that company? It took about three to six months at first to prove—OK, I have this strategy, and it can make money consistently.

Mia : I understand. So after you lost everything at fourteen or nineteen, you started getting involved in quantitative trading, participating in various competitions, and then joined this company. Have you stayed with this company ever since? Or have things changed?

Calvin Tsai : Things have changed. When I was a junior in college, around twenty years old, I interned at this company for a year and a half. After graduation, I moved to another hedge fund, where I worked for about five years.

Mia : Did you learn any trading concepts at that hedge fund that you might be able to use on Crypto?

Calvin Tsai : Yes. In fact, most strategies can be replicated across different asset classes. For example, the simplest trend trading, like moving average strategies—selling above a certain price level and buying below a certain price level—can be applied to different assets. However, some strategies differ between traditional and crypto markets. Traditional markets have opening and closing times, and are closed at night, resulting in overnight gaps—upward and downward gaps. Cryptocurrencies trade 24/7, without the concept of opening and closing times, so some strategies based on opening and closing times cannot be directly applied. Of course, the crypto market also has unique strategies that traditional markets lack. For example, on-chain data allows us to judge market fluctuations through various information on the blockchain. In traditional markets, there isn't such transparent data for quantitative analysis. Therefore, the two markets have both common strategies and their own unique strategies.

Mia : So what are the most important things you learned at that traditional fund that have been crucial to your career?

Calvin Tsai : First, you need to look at the data more closely and more extensively. Second, you absolutely cannot learn this if you're trading at home—it's about managing clients and managing expectations. For example, how to communicate with clients when you make or lose money, helping them understand the whole situation. I think this is one of the biggest lessons I learned in those five years. For instance, if you can earn 50% annually in live trading, and a client asks you, "If I invest with you, how much can I earn in a year?" If you directly say 50%, you haven't done a good job of managing expectations. Usually, I'll discount it, saying, "You can expect to earn 20% or 30%." By the end of the last year, you might only have earned 40%, less than the previous year, but because you initially told them 20%, they'll feel good and happy. They'll be willing to continue keeping their money with you, and even add more capital. So, I think mastering "expectation management" is a very important lesson I learned in those five years.

The fascinating dialogue continues; more will be presented in the next installment…

Disclaimer

This article is for informational purposes only. The views expressed are solely those of the author and do not represent the position of OKX. This article is not intended to provide (i) investment advice or recommendations; (ii) an offer or solicitation to buy, sell, or hold digital assets; or (iii) financial, accounting, legal, or tax advice. We do not guarantee the accuracy, completeness, or usefulness of such information. Holding digital assets (including stablecoins and NFTs) involves high risk and may result in significant volatility. Past performance is not indicative of future results. OKX assumes no liability for any potential losses. You should carefully consider whether trading or holding digital assets is suitable for you based on your financial situation. Consult your legal/tax/investment professional regarding your specific circumstances. Not all products and services are available in all regions, and products and services may be restricted or unavailable in some regions. You are solely responsible for understanding and complying with applicable local laws and regulations.