Former Coinbase CPO's lengthy article: I have regrets, but I still firmly believe in Crypto

Author: Sid, former PM at Coinbase

Compiled by: Jiahua, ChainCatcher

I still believe in cryptocurrency and am optimistic about the direction of the industry. As I write these words, I am not someone who has gone bankrupt, sworn off risks, or just returned from a meditative retreat with new insights. I hold cryptocurrencies, stocks, as well as some alternative assets like real estate and gold. And I hold them with great satisfaction.

Having been in this space long enough, I know what I believe in and understand where I have deceived myself in the past. The biggest lie is that better judgment can save me from bad behavior. After spending a decade in this industry (including my recent work at Coinbase and earlier in venture capital), the most honest thing I can say is: In any market—whether it's cryptocurrency, stocks, or any other field—the greatest advantage most people can have is the willingness to do nothing.

For most people, the real risk is not ignorance, but restlessness. It's the inability to hold onto a decent position when the market starts waving around shinier, noisier new toys. It's the extreme craving for more dopamine. This is the core of the entire article.

The Same Regret

If you've been in the market long enough, you'll hear the same phrase in slightly different forms: If only I had just held on.

Not "if I had found the next perfect trade." Not "if I had rotated faster." Not "if I had discovered that 10x coin before everyone else." Just: "If I had just held onto that good thing." The cryptocurrency version of this regret is: selling Bitcoin to chase altcoins. The stock version is: selling Nvidia to speculate on doomsday options, just because you suddenly felt like a volatility expert that week. Different markets. The same regret.

This is not a story of me going bankrupt. I'm doing well. But I've witnessed many smart people slowly destroy considerable wealth simply because they couldn't stand the calm. And the entire suite of products they used—from crypto exchanges to prediction markets to stock trading apps—was built around this human weakness.

If you've ever compulsively checked your portfolio, panicked and sold during a downturn, or made a trade just because you felt bored on a Tuesday afternoon, then this article is for you.

I Entered the Wrong Door to Crypto

In 2015, as a freshman, I stepped into the crypto space and immediately fell in love with what now seems the most embarrassing part: enterprise blockchain. Private chains. Enterprise consortia. "Distributed ledger technology." The entire pitch for "serious applications" in the PPT industrial complex. I thought that was the real deal, while "magical internet money" was just the plaything of libertarians and speculators.

I know, and I feel embarrassed looking back now. I even interned at IBM doing blockchain research, feeling it was a huge recognition. The world's leading tech giant seemed to be patting my head, saying: yes, this is how adults play; you're wise.

The longer I stayed in this field, the more I realized that most of the projects were lifeless. Many enterprise blockchains were just better-packaged databases. Meanwhile, Bitcoin (the largest digital asset) continued to penetrate the mainstream, while Ethereum (a platform allowing developers to build financial applications free from any company control) kept attracting builders. Open systems have real appeal. People flood in without needing any permission. This is a bit embarrassing, mainly because it means those "weirdos" were right.

Eventually, I dropped out in my junior year and moved to San Francisco, right at the early days of DeFi (short for decentralized finance, a wave of applications attempting to rebuild banking and trading without banks). I was fortunate to work at an excellent YC-backed startup, gaining deeper insights into how the industry operates: founders, funds, trading firms, DeFi applications, and everything in between. Ultimately, I led VC investments and joined a proprietary trading firm as a founding team member during the pandemic.

This experience cured my naive belief—I once thought most market participants were rational. They are not rational. Neither am I. One cycle after another.

No One Knows What They Are Buying

In the early days of cryptocurrency, to put it nicely, it was "quirky." Of course, there were smart people around. Libertarians. College dropouts. Anonymous developers. Teenagers making more than professors without knowing how. In Telegram groups, people who had just entered the space four months ago were dispensing garbage advice packaged as insights. At various conferences, those who entered a year before you were visibly getting rich, making you feel that as long as you made the next trade, you could also achieve social mobility.

I bought Bitcoin. I bought ETH. I had real reasons for buying these two.

Then the ICO craze hit, and many of us became even more foolish. For those fortunate enough to have missed that era, ICOs were basically unregulated IPOs filled with more nonsense. Back then, almost anything could be turned into a token. A website, a PDF white paper, a few grand declarations about reshaping a massive industry, and suddenly, you could go public.

People easily confused whimsical ambitions with actual value. It was like a Kickstarter campaign that was 90% emotional atmosphere and 10% product, but with a money-printing feature.

This video provides a more detailed timeline of when and where these companies received investments: YouTube link

So yes, I bought Cardano because it claimed to be the academic version of Ethereum. I bought SpankChain because someone pitched it to me as a huge but broken market for adult payments, and this token would become its payment layer. Readers, it did not. I also bought IOTA, and to this day, even if you held a gun to my head, I couldn't explain what it is. In 2017, this was normal behavior. That's the point. This was not marginal stupidity; it was mainstream stupidity in the crypto space.

The infrastructure at that time was also very rudimentary: Bitfinex, Bittrex, Poloniex, EtherDelta. Each platform made you feel like it could be hacked, freeze withdrawals, get sued by the SEC, or disappear overnight, sometimes all three at once. Everything was held together by duct tape and optimism, but as long as the market was rising, everyone felt they could survive.

And most of us had no idea what we were buying.

The most common cognitive bias is also the dumbest: People confuse "low price" with "cheap." A three-cent token feels "cheap," just like penny stocks on Robinhood feel cheap. In the face of a $10,000 Bitcoin, it looks like a freebie. Almost no one checks the total supply of circulating tokens. Almost no one cares about dilution. Denomination bias plays a part.

Despite the chaos of the ICO era, Ethereum itself proved to be one of the best generational investments of a lifetime. If you had bought ETH at around $0.31 during the crowdfunding in 2014 and held on, your return at the peak would have been about 15,000 times. Ironically, most people in the ecosystem at that time traded away that level of wealth in pursuit of those now non-existent tokens. By early 2018, most of these tokens had lost 80% to 95% of their value. It turned out that SpankChain did not revolutionize the adult entertainment payment tech stack.

And what about the Bitcoin I sold to fund these adventures? It's still Bitcoin. Still scarce. Still thriving. This mistake is not only painful but particularly insulting. Not because its operation was complex and failed, but because it was both foolish and entirely avoidable.

Bitcoin's price trajectory from 2017 to now proves this point. Those who weathered every crash and held firm saw returns in the thousands of percentage points.

Early DeFi Made Greed Sound Academic

If 2017 was immature speculation, then the summer of 2020's DeFi was complex speculation. It was more dangerous precisely because it felt smart.

DeFi applications offered absurdly high interest rates on deposited funds, sometimes with annualized yields of 500% or even 1000%, paid not in dollars but in newly minted tokens. This was the "genius" of the DeFi summer. In hindsight, ICOs were crude. DeFi felt sleek and innovative. Its terminology was hardcore enough that participating felt like doing financial engineering, masking most of its essence: chasing unsustainable yields that would evaporate the moment the incentives dried up.

• You were no longer gambling. You were "providing liquidity."

• You were no longer chasing token inflation. You were "participating in protocol growth."

• You were no longer chasing absurd yields. You were "putting idle assets to work."

Here’s a good explanatory article: CoinGecko link

A lot of what happened during that time was just the same old greed translated into more sophisticated terms. Then came NFTs. Then new Layer 1 blockchains. Then perpetual contracts. Then prediction markets. Then Memecoins made a comeback, stripped down to their purest form: noise with trading code.

In the crypto space, your counterparties are often: teams that know exactly when tokens unlock, funds that build positions at a fraction of your buy price, and market participants who understand position management better than you. The closer you get to professional investors and traders, the more the experience of ordinary retail investors looks grim.

Most people, whether on Robinhood or Coinbase, trade with the least information and the most emotion. Various narratives fill the information vacuum and are presented to retail investors as high-certainty bets. "This is the next Solana." "This is the next Nvidia." "This coin hasn't pumped yet." Usually, you are just someone else's exit liquidity.

Markets keep changing their outfits, but behavior patterns remain the same

People love to act as if the later cycles have become more sophisticated. They have not. The packaging has become more sophisticated. But the behavior patterns still feel familiar.

Pump.fun is a platform on Solana (one of the largest blockchains, which can be seen as a faster, cheaper alternative to Ethereum) where anyone can create a tradable new crypto token in seconds. It strips the speculation of Memecoins down to its most efficient form: quick issuance, quick trading, quick losses, rinse and repeat. It industrializes impulse control disorders and exacerbates FOMO.

When Pump launched in January 2024, the trading price of Solana's native token, SOL, was around $84. According to Dune Analytics, out of approximately 32.8 million wallet addresses on Pump.fun, only about 139,000 (around 0.4%) had realized profits of over $10,000. About 55% of traders were purely losing money, and over 90% either lost money or made less than $1,000. In 2025, research from Solidus Labs marked 98% of the tokens on the platform as scams or exhibiting fraudulent activity.

Data source: Dune Analytics link

The next part might make you want to throw your phone into the sea. If ordinary participants on Pump.fun had taken the principal they burned on Memecoins (even just $500, which is the most common loss range) and simply bought SOL in January 2024, they could have held it from $84 to nearly $295 by January 2025. By doing nothing, they could have achieved about 3.5 times the return. Even at today's price of around $84, they would at least break even, which is much better than losing everything chasing a token named after a frog wearing sunglasses.

The ugly truth is not that someone won. Of course, some people will win. The ugly part is how concentrated the winnings of these winners are, and how many people wasted time and money chasing a rise that simply does not exist for them. This is a rough calculation, not an exact financial audit. But directionally, for millions of wallets, buying Solana and then going for a walk outdoors would have turned out much better.

On Polymarket (which can be seen as a stock market for predicting events, where you buy and sell shares based on your predictions of whether real-world events will happen), about 70% of over 1.7 million addresses are in a losing position, while less than 0.04% of addresses captured over 70% of the profits on the entire platform.

I personally have not traded on Pump.fun or Polymarket. But I have watched people burn considerable principal on both. Those who truly believe they have an edge ultimately find themselves just exit liquidity for those who are faster and better informed.

This pattern is not unique to the crypto space. Classic studies by Barber and Odean found that the most active traders had an annualized return of about 11%, while the market return during the same period was close to 18%; but more recent data is equally devastating. A 2024 study by Dalbar Inc. found that the average retail investor underperformed the S&P 500 by 5.5% in 2023, marking the third-largest gap in retail returns over the past decade, and the gap often widens during bull markets as people sell during downturns and miss the rebounds.

Data source: Crews Bank link

According to FINRA's (Financial Industry Regulatory Authority) 2025 day trading statistics, 72% of day traders experienced financial losses by the end of the year, only 13% managed to maintain stable profits within six months, and only 1% succeeded over five years. Bloomberg found that 80% of day traders quit within the first two years. If you've ever stared at your Robinhood or Fidelity account, wondering why it would have been better to just buy the S&P 500 index and forget about it, this is the reason.

For leveraged products like options or futures (where borrowed money amplifies your bets, and tiny price fluctuations can lead to liquidation), the data becomes even worse. India's market regulator SEBI reported that in recent years, 70% to 91% of retail investors frequently trading derivatives lost money. This is the case wherever you give ordinary people a flashy interface, combined with enough jargon to make them mistake gambling for an edge.

I’ve watched people trade away the chips that could have made them wealthy

In every cycle, I’ve seen the same thing happen. The veterans who survived the last crash slowly trade away the positions that made them wealthy, chasing any new narrative the market throws out. Then they quietly disappear. Not due to some dramatic blow-up, but drained of blood by thousands of small rotations. Then, the next wave of retail investors arrives, wide-eyed and confident, and everything starts over again.

This is not how to build a lasting economy. This is strangling the very people who should be pushing it forward.

People love to fantasize that wealth comes from catching every new wave. Sometimes it does. But more often, wealth comes from catching a real wave and not blindly paddling away every time the water splashes around.

Trading is not the enemy. But it is likely not your friend either.

I want to clarify one thing: trading itself is not bad. There are indeed skilled traders. They often have extremely strict risk management, objective detachment from emotions, deep knowledge of market structure, discipline in cutting losses immediately, and years of pattern recognition experience. Most of them view it as a full-time job that requires years of apprenticeship.

That is not something the vast majority can do. The blunt truth is that the vast majority of ordinary investors would accumulate more wealth if they could buy a small amount of high-certainty assets and then do the hardest thing in finance—not touch them.

Trading products are becoming slot machines

So why is it so difficult to just sit still? Because the design of the products themselves makes it difficult.

Think about what brokers or exchanges are, whether it's a crypto exchange or a stock trading app. They make money when you trade. Not when you profit. Every trade incurs fees and spreads (the small difference between the buy and sell prices that the platform earns). In the crypto space, you also have to pay funding rates for leveraged positions; and when your bet fails, the platform will force liquidate and take the remaining money. This is called liquidation.

The platforms do not care whether your trades are good or bad. They only care that you trade.

When you are calm, brokers do not make money. When you learn to exercise restraint, exchanges do not celebrate you. These businesses make money off your actions. Off your clicks, your views, and your reactions to the urgency they create. Therefore, their incentive structures are naturally built around "action" to encourage you to trade as frequently as possible.

Price alerts. Bright greens. Blinding reds. Hot trend lists. Confetti falling across the screen. Various nudges suggest: something is happening, and maybe you shouldn't be the last to react. The same behavioral science principles that keep you endlessly scrolling through TikTok short videos drive you back to trading. Variable reward mechanisms: sometimes you win, sometimes you lose; it’s this unpredictability that makes you addicted. Continuous stimulation. Endless dopamine.

This is not accidental design. This is a polished business model. It is turning into a slot machine with candlestick charts. Even VIP programs targeting high-frequency traders have made this clear: the more you trade, the lower your fees, unlocking higher status, making you feel like a big shot. Casinos do the same thing. The house rewards not your skill. The house rewards your utility.

Dopamine itself ultimately is worthless. Those hours spent refreshing screens and checking prices before getting out of bed generate no lasting value. What remains is anxiety and a restlessness that seeps into everything you do. A patient holder generates zero revenue for the platform. A compulsive trader generates revenue every day. The entire design of the product experience is to ensure you never become that patient holder. So, of course, the products will continually cater to and nourish that restless part of you.

This is not to say these platforms are bad. They are robust, profitable businesses with a valid existence. They allow people to express market views without friction, which is great. It’s just that better guardrails are needed, which should be established by you and the platforms together.

What I Truly Believe In

I believe that cryptocurrency (and of course AI) is one of the most important technological and financial transformations of our generation. I also believe that the lessons here apply far beyond the crypto space. Whether you are trading options on Robinhood, buying and selling leveraged ETFs, or compulsively checking your 401(k) retirement account, the psychological mechanisms are the same. The design logic of the products is the same. And for most people, the outcome is the same: the more you trade, the less wealth you retain.

I hold a relatively diversified portfolio that includes cryptocurrencies, stocks, and some hard assets. The reason I can hold these is not because I made some brilliant trades, but because I ultimately learned to stop making unnecessary trades.

Compared to the past, I now believe more in the power of calm. As I age, trading for the sake of trading becomes less and less compelling to me. Especially in the market, trading always has a way of disguising itself as "getting things done." You feel engaged. Sharp. In control. Then, over time, you realize that much of it is just self-interference. Every time you sell a quality asset to buy a speculative one, you are betting that your timing and judgment can outperform the long-term trajectory of the asset you originally held. For most people, this gamble will lose. Not in a dramatic way. Not all at once. But slowly and persistently, only revealing itself when you stretch the timeline to years.

In any market, the hardest skill is not finding the next opportunity. It is being able to remain still when everything around you (apps, notification pop-ups, group chats, timelines, financial media) is screaming at you that "doing nothing is equivalent to falling behind." It is not.

So, if I had to distill all of this into one sentence, it would be: You do not need more trading. You need less temptation.

Do your best to hold (HODL),

Sid

Acknowledgments: Thanks to Dan Elitzer, Roy Learner, and Jesse Walden for proofreading. The experience of going through my past tax returns inspired this article, and during the writing process, I had Claude assist with editing through prompts.