“When you start to walk on the way, the way appears.”
— Rumi, 13th century
There’s a question that sounds simple but isn’t: why do some people build lives of extraordinary agency while others, often smarter and better-resourced, drift into comfortable passivity?
The standard answers are unsatisfying. It’s not intelligence. Some of the most inert people you’ll ever meet have pristine academic pedigrees. It’s not resources. Some of the most passive have trust funds. The real answer is more structural. It has to do with how a person relates to uncertainty itself, and whether their default response to it is motion or analysis.
This matters more right now than it ever has. With AGI arguably here and ASI on the horizon, intelligence is about to become the most overrated quality a person can have. More overrated than it already was, and that’s saying something. Intelligence will be as ubiquitous as tap water. You turn on the faucet and it’s there. Not special. Frankly, it never was really that special. The people who built their identities around being the smartest person in the room are about to have a very rude awakening. The new currency is agency. The ability to act, to move, to decide, to execute, to bend reality in your direction. That’s the thing that can’t be automated. That’s the thing that will separate the people who thrive in this new era from the people who get flattened by it. Value will accrue to high-agency people. High-agency people will use these tools to build things that were previously impossible and make themselves extraordinarily wealthy in the process. Low-agency people will sit around debating whether AI is going to take their jobs while high-agency people are already using it to build new ones.
What follows is an attempt to describe an operating system I call Extreme Agency. The individual ideas are not original. But the synthesis of them into a cohesive worldview, and the ways in which they interlock and reinforce one another, is something I think is worth laying out explicitly. These five principles form a system. Remove any one and the others degrade. Together, they compound in ways that look, from the outside, a lot like luck.
They’re not luck. More on that later.
Just do things. Anything really. Motion creates emotion. An inch of movement beats a mile of intention.
This is a foundational belief of Jeff Bezos, and one that’s been instrumental to him literally willing into existence the world’s first online bookstore, then the largest online store of any kind, and now what is essentially a pipeline of stuff into the highest-earning households in America. The guy left a perfectly good job at a hedge fund because he couldn’t stop thinking about the internet, drove across the country, and started selling books out of his garage. He didn’t have a twenty-year plan to build AWS. He had a bias for action and a willingness to figure it out on the move.
Most people, when they don’t know what to do, default to thinking. They research. They consult. They make pro-and-con lists. They wait until they “feel ready.” There is nothing wrong with careful analysis. But there is a failure mode that almost nobody talks about: the cost of hesitation compounds just as reliably as the benefit of action does.
In most real-world situations, the information you need to make a good decision simply doesn’t exist until you act. You can’t know whether you’ll enjoy living in a city until you live there. You can’t know whether a business idea has legs until you put it in front of a real customer. You can’t know if a relationship is right until you’re past the projection phase and dealing with the actual person. The belief that you can figure all of this out in advance, through sheer analysis, is one of the most expensive delusions a person can hold. It feels like prudence, but it's actually just fear.
The math backs this up. Most decisions are reversible. The expected cost of a wrong action is low, because you just course-correct. The expected cost of inaction is enormous, because you never learn, never compound, and you stay exactly where you are. Nassim Taleb would call this convexity: the upside of acting is unbounded, the downside is capped. Your life has that payoff structure whether you like it or not. Treat it accordingly.
And here’s something that doesn’t get said enough: action is the single best antidote to nearly every negative emotional state. Anxiety, sadness, confusion, that gnawing sense of stagnation. They all feed on stillness. They want you to sit there and ruminate. Movement breaks the loop. Not because it’s a distraction, but because it generates new inputs. You were stuck on X, you went and did Y, and now X looks completely different because you’re standing somewhere new.
The answer to “what should I do” is almost never “think harder.” It’s do something. Send the email. Make the call. Book the flight. Start the project. Most people choose inaction and rumination. This is almost always wrong. Just do something. Seriously. The path appears when you move.
Most people move through life as if they’re playing a single-player game. They think about what they want, what they’re good at, what they should do next. This is half the picture at best. Every arena that actually matters, your career, your relationships, your industry, is a multiplayer game. The outcomes you get are not just a function of your moves. They’re a function of everyone else’s moves too. And if you’re not thinking about that, you’re playing blind.
The prisoner’s dilemma is the most famous construct in game theory. Two players, each choosing to cooperate or defect. If both cooperate, both do well. If one defects while the other cooperates, the defector wins big and the cooperator gets destroyed. If both defect, both lose. The insight isn’t really the payoff matrix. It’s the recognition that this structure is literally everywhere, in every interaction you have, all day long, whether you see it or not.
Every professional interaction is a version of this. Share credit or hoard it? Invest in the relationship or extract from it? The right move depends entirely on whether you’re in an iterated game (you’ll deal with this person again and again) or a one-shot game (you won’t). In iterated games, cooperation dominates. In one-shot games, it doesn’t. Most people don’t think about this consciously, and so they either cooperate when they should be protecting themselves, or compete when they should be building alliances. Understanding this deeply is a genuine cheat code for navigating any competitive environment.
The people who build real, durable careers in competitive industries are not the nicest people and they’re not the most ruthless. They’re the ones who read the structure of each interaction correctly. They know when to be generous and when to be firm, when to invest in someone and when to walk away. Game theory doesn’t make you cynical. It makes you precise. There’s a difference.
Robert Axelrod ran a famous computer tournament in the early 1980s where he invited academics to submit strategies for an iterated prisoner’s dilemma. The winner was the simplest strategy in the whole tournament: Tit for Tat. Cooperate on the first move. Then just mirror whatever the other player did last. Be nice, be retaliatory, be forgiving, be clear. It is close to a complete philosophy for professional relationships. I’m not even exaggerating.
An interesting proxy for whether someone has really internalized the prisoner’s dilemma is to ask them their views on AI in white collar work. It’s a popular debate right now, and most people default to a version of the same lazy argument: “well, a lawyer does X, and if the machines can do X, then we don’t need the lawyer anymore.” It’s the Luddite talk track, and it sounds logical enough if you don’t think about it for more than thirty seconds. But it completely ignores the second-order effects. Law firms and investment banks are multiplayer games. If Law Firm A says “great, we automated all our junior associates’ work and fired them, here’s a lower price,” while Law Firm B says “we automated that same work but redeployed those people onto new tasks we couldn’t previously offer you, and we’re passing along the cost savings so you’re getting more services for the same price,” which firm wins the business? It’s obviously B. The game doesn’t end when you make your move. It ends when everyone has made theirs. People who think about AI as a single-player optimization problem are missing the actual game entirely.
Your strategy should be a function of the game’s structure, not your feelings about the game. Wanting to cooperate with a known defector isn’t noble. It’s bad strategy. Wanting to compete with a reliable long-term partner isn’t tough. It’s self-sabotage. Read the board. Play accordingly.
This is the one that sounds insane when you say it out loud.
The most successful people I’ve ever met share a trait that has nothing to do with intelligence or education or pedigree. They wake up every single morning with a level of self-belief that is, by any empirical standard, completely unjustified. Not confidence. Confidence is calibrated to evidence. This is something else entirely. It’s a deep, almost religious conviction that the world is set up for them. That things will work out. That they are exactly the right person in exactly the right place at exactly the right time.
Some of these people are genuinely brilliant. Most, honestly, are shockingly not that smart. Not in the traditional sense. What they all share is the operating assumption. They start from “I will figure this out” and work backward to the how. Everyone else starts from “can I figure this out?” and usually talks themselves out of it before they’ve even begun.
And here’s the part that’s hard to articulate without sounding like a description of a personality disorder. These people’s self-awareness is either shockingly absent by default, or they’ve learned to suppress it so deeply that it simply doesn’t make an appearance. There is no inner critic. There is no voice saying “who are you to try this?” They have somehow either never developed that voice or strangled it so thoroughly that it’s functionally dead. These people will wake up every day, step out of bed, see themselves in the mirror and think “I’m awesome, and by the way you should give me a million dollars.” And then they go about their day acting on that assumption with such conviction and such relentless energy that people actually do give them the million dollars. And then they do it again the next day.
This is, strictly speaking, irrational. But it is instrumentally correct. The delusion generates the effort. The effort generates the results. The results validate the delusion. It becomes a flywheel. And the end result, on a long enough time horizon, is that people tend to end up with the lives they think they deserve. That cuts in both directions. The person who wakes up every morning thinking they’re awesome and the world owes them something will, through sheer repetition and energy, bend their circumstances toward that belief. But the person who wakes up without that conviction, who doesn’t really think they’re special or deserving, will be delusionally less optimistic, and they’ll end up somewhere they’d probably rather not be. Same mechanism, opposite direction. This is an uncomfortable truth but I think it’s true.
There’s a practical application of this that goes beyond individual psychology: act as if. Act as if you’re already a big deal and then bend the universe to make it true. Self-fulfilling prophecies run the world, and once you start looking for this pattern you see it absolutely everywhere. When Google was a tiny startup operating out of a garage, they labeled their first real office space “Building 40” so that visitors and potential partners would assume there were thirty-nine other buildings. There weren’t. It was just Building 40. But the perception of scale created the reality of scale. Every company that has ever achieved unicorn status has used some version of this principle, whether consciously or not. You project the thing you want to become, and then you work like hell to close the gap between the projection and reality. The projection isn’t dishonesty. It’s a forcing function. It creates expectations, both from others and from yourself, that you then have no choice but to meet.
There is a real tension with game theory here that is worth naming. Game theory says be rational, read the structure, play the optimal move. Delusional optimism says forget the probabilities, believe you’ll win, swing for the fences. The way to reconcile them is this: delusional optimism tells you which game to play. It’s the thing that makes you look at the highly rewarding, highly competitive arena and say “yeah, I’ll figure it out, because I’m me.” Without it you’d never enter the room. Game theory tells you how to play once you’re in there. It’s the part that says ok, you’re in the arena now, here are the actual dynamics, here’s who cooperates and who defects, here’s where the leverage is, play skillfully. Conviction without strategy produces wasted energy. Strategy without conviction means you never pick the ambitious game in the first place. The combination of both is lethal.
There’s a famous story about Masayoshi Son and Adam Neumann. It was 2017 and Son had just spent about twelve minutes at WeWork’s headquarters before inviting Neumann to ride with him to his next meeting. In the car, Son turned to Neumann and asked: “In a fight, who wins? The smart guy or the crazy guy?” Neumann said the crazy guy. Son told him he was correct, but that he and his co-founder were not crazy enough.
It’s a great anecdote. Son poured billions into WeWork on the strength of that thesis, and for awhile the crazy guy narrative looked like genius. Then WeWork imploded, filed for bankruptcy, and Neumann walked away with hundreds of millions while thousands of employees got wiped out. The crazy guy lost. Spectacularly.
But Son was asking the wrong question. It’s not smart guy versus crazy guy. There’s a third option they both missed: the obsessed guy.
The crazy guy is erratic. He moves fast and breaks things and occasionally gets lucky because nobody sees him coming. But crazy is not a strategy. Crazy is chaos with good PR. The smart guy is careful. He analyzes and optimizes and hedges. He’s right a lot of the time but he’s slow, and he talks himself out of the big swings because the expected value doesn’t look clean enough.
The obsessed guy is different from both, and the reason is deceptively simple: he’s having fun.
Naval Ravikant has this idea that you should find the thing that feels like play to you but looks like work to everyone else. I think that might be the single deepest insight in this entire essay. The obsessed guy isn’t grinding. He’s not white-knuckling through the pain. He is absolutely, genuinely having a great time. He studies the fight because he loves studying the fight. He watches tape at midnight because watching tape is his idea of entertainment. He iterates and iterates and iterates not because he has superhuman discipline but because to him the iteration itself is the reward. The preparation is the fun part.
And that’s why he’s nearly impossible to compete against. You cannot out-work someone who isn’t working. Think about that for a second. The guy who’s forcing himself to put in the hours will eventually burn out, take a vacation, get distracted, lose motivation. The obsessed guy doesn’t have that failure mode. He’s not depleting willpower. There’s no tank to empty. He’s playing his favorite game, and he will play it all day and all night and wake up wanting to play it again. How do you beat that? You basically can’t.
There is a gear beyond discipline and work ethic where the work stops feeling like obligation and becomes something closer to compulsion, except compulsion sounds like suffering and this isn’t suffering. It’s joy. Waking up at 4 a.m. with an idea and being at the laptop before being fully conscious, not tired but excited. That’s not discipline. Discipline gets you to the gym four days a week. This is something else entirely.
The distinction that matters, and this is where Neumann went wrong, is whether you’re genuinely obsessed or performing obsession. The crazy guy performs intensity for an audience. Neumann was performing disruption for an audience of one very rich investor. The workaholic grinding eighty hours at a job they secretly resent is performing dedication for an audience of senior partners. Neither of them is having fun. They’re captured, running someone else’s race, and deep down they know it.
Carl Jung drew a distinction between the ego, the constructed self that handles daily life, and the Self, the deeper and more complete personality you’re growing toward. He called the process of aligning the two “individuation.” The feeling of play, the feeling that the work is fun, is actually the signal that you’ve found your alignment. When the ego and the Self are pointing in the same direction, the whole personality mobilizes. There’s no internal friction. No ambivalence. No energy spent forcing yourself to care about something you don’t actually care about.
Most people never experience this, and I think the reason is brutally simple: most people are obsessed with the wrong things. Status, approval, security, the metrics their environment taught them to care about. They’re running someone else’s objective function and wondering why everything feels like a grind. Individuation is, at its core, the process of finding your own objective function. And once you do, the obsession follows naturally, because the work becomes play. You stop pushing the boulder uphill. You start running downhill.
Ray Kroc was fifty-two years old, and by any reasonable standard, his career was behind him.
He’d spent over a decade selling Multimixer milkshake machines out of his car, crisscrossing the country, calling on drive-in restaurants. Before the Multimixer it was paper cups. Before paper cups it was piano playing. Before piano playing it was a brief stint in real estate. A lifetime of modest salesmanship. No real wealth. No recognition. Nothing.
Then he walked into a hamburger stand in San Bernardino and watched Dick and Mac McDonald run an operation so brutally efficient it stopped him in his tracks. Eight Multimixers running simultaneously. Limited menu. Self-service. Speed. Consistency. He knew, right there, that he was looking at something that could be replicated across the whole country.
He was fifty-two.
What happened next is well-documented. What’s less discussed is the nature of the bet. Kroc was not a young man with optionality and decades of runway. He mortgaged his house to finance the first franchises. He made almost no money for years while building the system. He fought with the McDonald brothers over every detail. He eventually maneuvered them out of their own company and spent the rest of his life taking credit for their innovations.
The story is not clean. Kroc was relentless, and he was also ruthless. Reading his autobiography you get the full picture. The vision and the vindictiveness, the persistence and the pettiness. But the mechanism is undeniable. Kroc succeeded because he absolutely, categorically refused to stop. He had an operating assumption, “this will work, and I am the person who will make it work,” and he maintained it through a decade of failure and a lifetime of false starts. That’s it. That’s the whole trick.
Most people learn persistence not from a book or a podcast but from a situation where quitting isn’t really an option. And what you discover, if you stay in it long enough, is that persistence doesn’t look the way you’d expect. It’s not heroic. It’s almost comical. You just keep doing it, and doing it, and doing it, and doing it, and doing it, and doing it. An absurd number of iterations. Until it starts to feel pointless. Ridiculous even. Like you’re an idiot for still being here. But if you keep subtly iterating, tweaking, tinkering, something eventually shifts. Not dramatically. Not with trumpets and fireworks. It just quietly starts to work, and you look back and realize the only reason it worked is that you didn’t stop.
The math of persistence is more interesting than the motivation. Most of the really valuable outcomes in life have fat-tailed distributions. The payoff from the hundredth attempt can make the cumulative cost of the first ninety-nine look trivial. But most people quit at attempt fifteen, because they evaluate the process on a short time horizon and conclude it doesn’t work. They’re not wrong about the data they have. They’re wrong about the timeline.
Persistence and delusional optimism form a ratchet. The delusion sustains the persistence: this will work. The persistence eventually validates the delusion: it worked. Without the delusion, you quit too early. Without the persistence, the delusion is just fantasy. Together they create a one-way mechanism. Each iteration either succeeds or generates information that makes the next one better. On a long enough timeline, assuming you’re playing the right game, you just can’t lose.
And that’s where game theory closes the loop. Persistence in the wrong arena is just stubbornness. You need the strategic clarity to know where to persist. And you need the bias for action to actually begin.
People look at someone whose life has gone well and call it luck. He got lucky with that job. She got lucky meeting the right investor. They were in the right place at the right time.
This is, with respect, almost always wrong.
What looks like luck is almost always the compound product of bias for action and persistence. Every action you take is a lottery ticket. Most of them don’t pay out. But each one expands your surface area to chance. You meet someone at a conference you almost didn’t attend. A cold email you almost didn’t send gets forwarded to the right person. A side project you almost abandoned turns out to be the thing that gets you the interview. None of these feel like luck from the inside. They feel like the natural, almost boring result of having been in motion for a long time.
The “lucky” person and the “unlucky” person are often equally talented. The difference is that the lucky person bought a thousand lottery tickets through sheer volume of action and repetition, while the unlucky person bought ten and concluded the game was rigged. Luck is just a function of persistence and bias for action. The more you do, the more surface area you create. The longer you keep doing it, the more inevitable it becomes that you’ll be standing in the right place when something breaks your way. Call it luck if you want. It’s math.
Seneca said that luck is what happens when preparation meets opportunity. But even that framing is too passive. The preparation creates the opportunity. The bias for action puts you in the room. The persistence keeps you there long enough for something to happen. The game theory tells you which room to be in. The delusional optimism makes you believe you belong there. And the obsession means you’re having such a good time that you never leave.
These five principles are not a menu to pick from. They’re a system, and the system only works when all five are running together.
Bias for action without game theory is blind hustle. You move fast in the wrong direction. Game theory without delusional optimism is cold strategy. You identify the best move but never make it because the odds look unfavorable. Delusional optimism without obsession is daydreaming. You believe you’ll win but you don’t do the actual work. Obsession without persistence is a sprint. You burn white-hot for six months and then you flame out. And all four without a bias for action never leave the starting line.
What holds all of it together is individuation. Jung argued that it’s the central task of human life. Not success, not happiness, but becoming who you actually are. Most people avoid this because it requires confronting uncomfortable things. Your shadow. Your real desires. The gap between the life you’re living and the life you’d choose if you weren’t afraid.
The world is full of people waiting. Waiting for the right moment, the right information, the right feeling. Waiting for permission. Waiting for certainty that never comes.
The path appears when you walk.