based16z - The Hyperliquid Trade, Early Lore, & How to Make It
Summary
Hyperliquid was based16z’s defining trade because he recognized a superior product and exceptional founder before there was an investable token. After his fund offered Jeff a term sheet at a roughly $1 billion pre-revenue valuation, Jeff declined because he wanted the “Bitcoin immaculate conception”; based16z responded by farming full-time for a year. The opportunity combined product need, difficult access, founder quality, and a market that had not yet priced any of it.
His biggest execution mistake was treating a potentially permanent asset like another high-variance trade. Once he recovered keys from an old laptop, the airdrop was already up from roughly $3 to $8; at $10 it represented 90% of his portfolio, and he bought more before selling some after Shoku advised him to change his life. He now thinks selling the remainder around $30–$40 was probably impatience: “What is a forever hold asset? How do you realize tail outcomes?”
The repeatable strategy is not constant activity but waiting for a handful of radically asymmetric setups. Weekly Bitcoin calls around the major Bitcoin headline offered up to roughly 300x notional leverage, while Robinhood and American Eagle paired limited perceived downside with narrative upside. His stated win rate is only 25%–30%, yet he says he has been profitable every crypto year because “good things only come up once in a while that you can really crush.”
Sizing has often increased theoretical returns while sabotaging realized returns and personal stability. He put about 80% of his book into American Eagle, sold most after a 40% rise because he was glued to the tape, and then watched the stock triple. Snap cost him about 10%; under his former approach, he says he could have put roughly 1.5x his net worth into a position. He now walls off roughly 70% in short-term Treasuries so he can still take 40% positions inside the remaining trading book without risking his livelihood.
His largest recurring losses come when leverage substitutes for genuine edge. Hyperliquid trained him to crave 40x perpetual-futures exposure despite admitting “little to no edge, or maybe negative” on low-timeframe price action; repeated SPX puts reflected the same refusal to stop when a thesis failed. The correction is blunt: “If at first you don’t succeed, actually stop trading.”
Twitter is simultaneously his research terminal, conviction engine, distribution channel, and ethical hazard. DeepSeek showed how news could sit dormant for weeks before a viral post drove narrative and price into a parabola, while based16z admits his own posts probably accelerated HOOD and smaller trades. He uses bullish or bearish “chants and mantras” to hold positions, but dislikes trades where his post creates the very liquidity from which he profits.
The deeper transition is from zero-sum trading toward equity, products, and positive-sum work. Meme coins taught him short-term reflexes—“there’s not a lot of compounding in playing poker all day”—while Whop represents an attempt to help people earn from what they genuinely want to do. His closing advice is to “spam a bunch of shit,” find the activity that produces both early traction and healthy addiction, then “press, press, press” until it compounds.
Deep dive
1. Hyperliquid began as a founder bet before it became a token trade
Thread Guy dates the call to late 2023, when based16z had few market takes but insisted the unfamiliar Hyperliquid perpetuals venue was “pretty sick” and worth farming. The host now calls it “the most profitable airdrop of all time” and the trade most associated with his guest.
The setup began with practical frustration. To establish a 100-BTC position on-chain, based16z might split half across dYdX, a quarter on Vertex, and the rest elsewhere because one venue could not absorb the order; Hyperliquid immediately felt “insanely good” against that fragmented experience.
Rollbit was an earlier win. It combined a popular gambling product, a cheap coin, and a revenue story; buying it on Solana was accessible to based16z because of his experience trading Solana NFTs in 2021.
Meeting Jeff converted product enthusiasm into founder conviction. Based16z told his fund to invest around a $1 billion valuation before revenue; Jeff declined its term sheet because he wanted the “Bitcoin immaculate conception,” so based16z made farming his full-time job and clicked through roughly $3 billion of Bitcoin perp volume at 40x during the farming period.
2. The airdrop nearly vanished with one dead MacBook
Five days before the drop, the old MacBook holding every farming wallet stopped working; there was no seed-phrase backup. He tried the Apple Store and a specialized repair shop, and one technician said the machine would have to be sent “to the IDF,” while based16z understood that it contained “my ticket.”
The rescue came from an old East Village roommate who still had an older laptop. Asked to wait a few weeks, based16z replied that it would be “really chill” if he could arrive within two hours; the older machine contained the keys.
By the time he recovered access, the token had already run from roughly $3 to $8. At about $10, the allocation was already 90% of his portfolio, yet his first action after logging in was to buy more and only afterward ask Shoku whether that was sensible.
Shoku—while buying some of those coins—recommended selling, putting substantial proceeds into the S&P, and changing his life. Based16z is glad he sold a portion, but calls exiting the remainder around $30–$40 probably a mistake born of impatience rather than a changed Hyperliquid thesis.
3. Knowing what to hold forever remains the unsolved execution problem
Based16z usually owns positions “really big or not at all,” which makes even a correct asset mentally expensive. A concentrated holding occupies attention, competes with new ideas, and can become “a weight on you” when the trader is also trying to invest or build elsewhere.
His unresolved questions are therefore psychological: “What is a forever hold asset? How do you realize tail outcomes?” He has not yet learned how to mentally remove himself from a position and accept that some capital may belong in a multi-year or permanent allocation.
American Eagle later exposed the same problem from another angle: a position can be correct under a Kelly-style payoff calculation yet too large for the trader to hold properly. The relevant sizing constraint is not only drawdown probability but whether size will force premature management.
4. Life-changing money delivered freedom, then an identity vacuum
His childhood mixed upper-middle-class security, a mother influenced by meditation and Buddhist ideas about desire, and intense personal ambition. School, math competitions, elite college, and the “best job” became successive scoreboards, while proximity to extreme wealth made ordinary financial success feel strangely small.
After three years of spending roughly 16 hours a day at a computer, the Hyperliquid outcome felt like reaching “level 1 off level 0,” not completion. He spent several months acting retired—partying, doing “stupid things,” and surrounding himself with “low-vibrational people”—without feeling especially good.
The harder question was what followed the trade that defined him: “Where does my identity go next? Is this all there was?” He eventually returned to playing difficult games, explored earlier-stage companies, and tried to develop skills beyond trading, including programming and writing.
5. His real edge is rare asymmetry, not frequent trading
The origin story starts with roughly $5,000 from tutoring and online poker. He full-ported his first NFT mint because Blipmaps were fun, watched them rise from about 0.1 ETH to 30 ETH, and suddenly had roughly $200,000—enough to convince him that early digital markets were the game to pursue.
NFTs, Terraforms, and early Solana taught a simple filter: “Am I early or not?” and “Is this legit or not?” As crypto matured, he shifted toward narrative and positioning inflections where conviction can support an aggressive option-like bet.
One 2026 example was buying short-dated weekly Bitcoin calls around the major Bitcoin headline, with up to roughly 300x leverage on notional. The headline aligned with an approaching Trump conference, traders anticipating an upside break, and option specialists warning that call sellers were offside—producing the desired “get me to the screen” moment.
He began the year down roughly 15%–20%, then says his portfolio rose about 2.5x in a month through Bitcoin, HOOD, and GoPro trades. That streak required being “hyper online” across news, price action, and smart people, creating a direct trade-off with longer-term company work.
6. Leverage becomes destructive whenever conviction is merely activity
Perpetual futures are his clearest negative-edge arena. Hyperliquid was so engaging that it encouraged all-day clicking, while farming conditioned him to want 40x leverage despite having “little to no edge, or maybe negative” on very short-term price action; fees and repeated small errors made perps his largest historical loss category.
Repeated SPX puts provided another failure pattern: “I just kept wanting something to work that wasn’t.” His revised maxim is that in trading, unlike motivational folklore, “if at first you don’t succeed, actually stop.”
He also accepts a structural objection to shorting the index: if the dollar is losing purchasing power, betting against equities means asking to receive more depreciating dollars. Quoting Good Alexander’s framing, “You’re trying to short the stock market, like you’re trying to get more USD?”
Because he once wagered huge percentages of his book, a 3%–4% adverse move often meant immediate exit: “Goodbye, next idea.” He suspects that reflex—learned from 40x perps—is sometimes too rigid, but it prevented conviction from quietly turning into a martingale.
7. Snap demonstrated how a clever thesis can omit the decisive variable
The Snap thesis combined improving subscription revenue, a possible turn to free-cash-flow positivity, a hardware or new-form-factor option, and evidence from his younger brother and online culture that Snapchat was becoming ubiquitous again. It looked like a neglected turnaround with several independent catalysts.
The missing variable was management. Based16z concluded that Evan Spiegel was “out to lunch,” Snap could not enter the S&P under its structure, management lacked a compelling incentive to promote the stock, and persistent stock issuance meant operations needed to improve enough to overcome dilution too.
He had bought at the top and says he probably drew down about 10%. Under his former trading approach, he says he could have put roughly 1.5x his net worth into Snap, which he now calls “way too much” and “not smart at all.” He contrasts that with his broader record: a 25%–30% win rate, public wrong calls, and positive annual crypto results produced by the winners’ asymmetry.
8. Meme coins taught reflexes that work in narratives but fail at compounding
Based16z treats meme coins as an extension of NFTs: assets could rise dramatically and then go back to zero. He rejected the contemporaneous thesis that they were becoming a broadly permanent asset class, while allowing that a tiny minority of holders may own a favored token forever.
That environment encouraged bad portfolio habits—being insufficiently long beta, excessively short-term, and trained for player-versus-player exits. “There’s not a lot of compounding in playing poker all day,” he says, even when the game is profitable.
The reflex remains useful in equities whose prices increasingly respond to memes and narratives. He points to trading the meme story on Meta or “SaaSpocalypse” reactions to Anthropic blog posts as similarly short-duration, zero-sum arenas.
His counterweight is the old Naval framework: own equity or tokens, develop specific knowledge, build distribution, automate replaceable work, and play long-term games with people one could still work beside in 20 years. He considers that advice cliché now but still foundationally true.
9. American Eagle was the “Megachurch” trade—and size cost most of the upside
The setup accumulated before the viral moment. Sleepy Sol had spent roughly three months describing American Eagle as a cheap turnaround; when the Sydney Sweeney campaign appeared and Wagey Capital predicted it would be huge, the background thesis and cultural catalyst instantly fused.
Based16z’s call was direct: Sydney Sweeney was “the biggest person on Earth,” the advertisement would go maximally viral, and the underlying stock was already inexpensive. He put about 80% of the book into spot American Eagle, reasoning that it was a decent consumer stock unlikely to go to zero immediately while virality opened substantial upside.
The stock rose roughly 40%, and he sold most of the position for a strong portfolio gain; it ultimately tripled. Oversizing had glued him to the tick chart, so the nominally aggressive sizing reduced the equity he actually realized from being right.
His lesson is behavioral rather than purely statistical: smaller sizing would have let him hold longer. A mathematically defensible position is still wrong if its size changes how the trader manages it.
10. Twitter now converts dormant information into sudden price action
Twitter supplies individual ideas through posts and group chats, but its unique value is the ambient “general feeling” of a narrative. Based16z admires Ansem’s ability to discover tiny, high-quality accounts early—the social-media equivalent of venture sourcing.
DeepSeek illustrated narrative latency: relevant material had existed for roughly three weeks while NVIDIA barely reacted, then attention intensified until a Good Alexander post was expected to approach five million views. Based16z saw that post as an accelerant or inflection point before the price action arrived the following Monday.
He admits his own posting probably affected HOOD and some smaller names. The ethical boundary became obvious when he bought an “American shitcoin” on a tracked wallet and immediately sold it as a joke: the ripple itself created liquidity, leaving him uncomfortable profiting directly from his influence.
His preferred role is to accelerate a story that eventually becomes larger than his account. Yet he sees the whole market adapting to X: Jensen Huang and Alexandr Wang build presences, executives seek attention, and capital allocators with ten hours of screen time absorb the same “slop” whether they work at pods or long-only funds.
11. “Loot and Steal” diagnoses incentives rather than endorsing them
Asked whether young people should treat looting the system as a patriotic obligation, based16z says no: it is commentary, not “a very empowering primary ethic.” He acknowledges that four years in crypto meant playing largely zero-sum games and sees elite talent choosing quant trading as a softer version of the same allocation problem.
His model is a game-theoretic divergence between what benefits society and what rewards the individual. Building may be the better collective action, yet the local Nash equilibrium can be “just start skimming off the top”—a pattern he even detects in admired figures who simultaneously discuss stock prices or promote Dogecoin.
On where this goes toward 2028, his answer is explicitly uncertain: “I have no idea.” He frames anti-rich politics, nationalism, antisemitism, and renewed communism as familiar reactions to a higher-volatility society, probably “for the worse,” rather than uniquely modern phenomena.
A friend’s distinction sharpens the diagnosis: classical Marxism centers production and control of production, while current status centers consumption. The resulting politics becomes, “I deserve to consume too and you are consuming too much,” making the billionaire “the focal point of my anger.”
12. The proposed exit is experimentation, ownership, and healthier incentives
He does not use Instagram or TikTok and keeps YouTube off his phone because he expects to lose three hours if the apps are available. Consumer software is “an adversarial game between you and the person who owns it,” progressively monetizing captured attention once user-growth utility no longer dominates.
Whop is his largest investment and the project occupying most of his time because he sees a chance to align internet mechanics with user benefit. His preferred post-AGI branch is not becoming a “slopped-out, doom-scrolling vegetable machine with a UBI,” but helping people earn income from activities they genuinely want to pursue.
For a 20-year-old, he would not prescribe crypto trading. He recommends internet-native work that builds equity, skill, or specific knowledge; is new enough to be “hard to explain to old people”; and may currently mean anything from building Roblox games to an opportunity he is already too old to notice.
The closing rule is discovery before commitment: “spam a bunch of shit,” maximize shots on goal, and watch for either early success or a healthy form of addiction. He tried music, math research, blogging, poker, venture, and NFTs; once enjoyment and traction converged, the response was “press, press, press”—then compound rather than keep restarting.