The Forty Games Hikaru Nakamura Still Calls a Mistake

Around 1 a.m. in a Moscow hotel, a yellow lamp picked two faces out of a dark room. It was 2010. Magnus Carlsen and Hikaru Nakamura played forty games of blitz. The score, reconstructed later from a few stray links that began on a subreddit, was 24.5 to 15.5 for Carlsen. There was no ranking at stake, no prize, no broadcast. More than a decade later, Nakamura still calls agreeing to the match one of the biggest mistakes he made in competitive trust, because it gave Carlsen a chance to understand his style.

Nothing about the night was planned. It was the day of the event's final round and closing ceremony. Nakamura had gone to dinner with someone high up at the Internet Chess Club, had a couple of drinks he no longer remembers as wine or beer, and near the end of the meal was told that Carlsen wanted a private match. He is careful to say that the drinks had nothing to do with the yes. In the room were the two of them, Carlsen's father Henrik, and the journalist Macauley Peterson, who filmed parts of it. The clip is still on YouTube. Lex Fridman asks whether, sitting inside that picture, Nakamura can feel the epicness later assigned to it. He has too many memories like it. After he won the 2005 U.S. Championship, held in late 2004 in La Jolla, he played blitz for four or five hours in the hotel lobby, just for the love of the game. Those nights used to happen all the time. Now, he and Carlsen would almost certainly need stakes before they sat down in a dim room again.

Going in, Nakamura's score against Carlsen was not ugly. He thinks Carlsen was up a game or two, and there had been many games where Nakamura had been pressing, close to winning. The forty games changed that. They played all sorts of openings, and Carlsen learned that Nakamura was often not that great in the opening: he would choose slightly dubious sidelines instead of the main lines. The other lesson was crueler. They were playing with increment, and Nakamura kept reaching technically drawn endgames with extra material that Carlsen simply would not lose. A basic example is rook and four pawns versus rook and three, all the pawns on the same side of the board, say f6-g6-h6. That has been known as a theoretical draw for about seventy years, no matter where the pawns sit, for the next fifty moves. Nakamura remembers Carlsen saving five or six of those. The match stayed close until roughly the last ten games, and then Carlsen started winning in spades.

What hurt afterward was not the margin. It was the discovery of how hard it was to beat this person even after you got the advantage. One game stuck: a Sicilian Najdorf, Carlsen playing Bg5 on move six, a line Bobby Fischer and Garry Kasparov both used. Nakamura answered with e5, or Nbd7 and e5, which he calls dubious, not the twenty-move main line. Fridman offers a kinder reading: going outside theory at 1 a.m. could look like chaos, risk, a strength. Nakamura flips it. From then on Carlsen picked lines where you had to play the absolute best moves to equalize, or you were much worse. It was not about that Najdorf. It was a map of the whole repertoire. If Nakamura looks at World Championship aspirations, he still thinks it was a big mistake. His philosophy is that everything happens for a reason, so he does not live in the regret. Competitively, Carlsen exploited the leak for many years.

The leak did not fade quickly. Nakamura says the 2020 Meltwater Champions Chess Tour final, in September, was the first time in years his openings matched Carlsen's. The format was seven matches of four games. Nakamura took the first match 2.5-1.5. It went all the way to a seventh match, and Carlsen won in Armageddon. In the sixth and seventh matches there were Berlin games where Carlsen actually made mistakes, and Nakamura won some critical games. That, he says, made a huge difference. For a long time before that, getting punished out of the opening was one reason the lifetime score sat where it sat.

Fridman asks him to teach the Berlin. It is a Ruy Lopez: e4, e5, Nf3, Nc6, Bb5, Nf6. The line existed in the 1960s and became famous in 2001, when Vladimir Kramnik used it to take the title from Kasparov. Kasparov had been champion for about fifteen years. With White he could not put pressure on the Berlin; many of those games were quick draws. With Black he lost a Grünfeld and a game in Queen's Gambit Declined or Nimzo-Indian territory. Nakamura walks the main line: knight takes e5, d6, d4, trades, knight to d6 hitting the bishop, exchanges, queens come off, and the king recaptures on d8. After twenty or so forced moves you are in an endgame. Opening principles say develop, castle, hide the king. The Berlin throws all of that out. For Kramnik it was brilliant. Kasparov was known for opening preparation and for being a tactical, aggressive player. An endgame from move one took the weapon away.

Nakamura later worked with Kasparov and saw the database. There were something like 220 files on the Berlin Defense. Kasparov had learned chess in a world where openings were okay or they were not. The Berlin was considered dubious, so his gut said White was better, and he tried to prove it. Nakamura thinks he spent four or five White games of the match on that endgame and did not come close to an advantage. He kept wasting the White pieces. Near the end he switched, and he came close in one later game, but it was already too late. Fridman asks whether stubbornness is just how chess is supposed to work: if your gut says the position is better, you capitalize. Nakamura calls that an old-school way of thinking. Before computers, human intuition and calculation decided whether a position was better. If an opening was dubious, somebody was better, and White had to prove it.

Computers broke half of that logic. A half-pawn edge, 0.50, still holds with optimal play. In the Berlin starting position an engine agrees with Kasparov that White is very slightly better, because Black has moved the king, cannot castle, and the center is open. Five moves deeper, Nakamura says, it goes to 0.00, a draw. That is a bad thing, because with White you want to put pressure on your opponent. If the engine says 0.00, you are not doing a good job of playing White. The modern default is White tries to press and Black tries to be solid and make a draw. At Kasparov's peak it was the opposite: he tried to win with Black too, in openings like the Sicilian Najdorf, which Fischer popularized and Kasparov played for the last fifteen to twenty years of his career. With modern engines and memorized theory, taking risks with Black almost always backfires. You play almost perfectly just to draw, and you never get winning chances. From a risk-reward standpoint the old fight-with-both-colors approach is nearly extinct.

Psychology, Nakamura says, is a set of choices about what kind of game you are going to have. In the Najdorf, White's sixth move decides whether the game is sharp or positional. Bg5 is aggressive; Be3 can go quiet. Black is always responding, and you have to be extremely precise. He gives a simpler version of the same fork: put a knight on one square and the game is positional; put it on another and White will push f-pawns and maybe g4, and it becomes a fight. Those are the choices you are constantly making. Am I going to be standard, or am I going to be aggressive.

In the 2022 Candidates Tournament in Madrid he had White against Alireza Firouzja, a rising junior born in Iran and playing for France. Firouzja does not normally play the Najdorf. He chose it in this one event, so Nakamura knew he wanted sharp positions he could lose and also win. Nakamura intentionally played a very Asian, very boring line. If Firouzja played it correctly it would be a draw, and he would not be happy. Firouzja did something dubious because he wanted tension and chaos. Fridman calls it the ultimate troll at the highest level. Nakamura says yes. He knew the opponent would be unhappy. He wanted a game he could gamble in, and Nakamura gave him a dull one until he started looking for trouble himself.

The game is worth pulling out of a two-and-a-half-hour conversation because it states the other half of chess. What sits on the board is a position. What is actually being contested is a body of knowledge about the person across from you: what they prepared, what they cannot bear, which opening they never play except when they came to fight. You know what they want, and you do not give it to them. Carlsen spent years picking the one precise line in Nakamura's repertoire. In Madrid, Nakamura spent one White game picking the line Firouzja least wanted to sit in.

Reddit had asked whether losing to Carlsen was a hit on confidence, not only an opening leak. Nakamura says it is not just his problem. It is everybody's problem against Carlsen. When someone saves positions that are much worse, almost miraculously, it starts to affect you. Carlsen has done it to him, to Levon Aronian many times, to Kramnik, to just about everybody. Once is luck. Ten times becomes a psychological fact you do not know how to play around. Every top player who has faced him more than about five times has seen things happen that do not happen against anyone else. Juniors have an advantage here. They do not have the history.

After Moscow it was not the same. Nakamura viewed Carlsen completely differently. After all those saved endgames he started thinking, this guy is superhuman. You cannot have those thoughts when you are playing competitively, but in the back of your mind it is always there. He thinks every top player has that issue. The thing that changed it was not a pep talk. During the pandemic he played Carlsen in online events almost every month, something like twenty or thirty games in a year. He beat him in one event, lost two others, reached a couple of finals. Playing that often, Carlsen did not feel superhuman. Nakamura was doing better, not because he felt he was doing anything differently, but because he was not making the old mistakes.

The mechanism is simple and a little bleak. If you play someone three or four times in classical chess, lose one and draw three, you never get the glimpse. You feel negative. If you play twenty or thirty games, you get opportunities even when you fail to convert them, and you start to feel you have chances. All of them, he says, are capable of beating Carlsen. All of them have very bad scores against him. People underestimate how much that plays. In the 2020 online events he also felt there was nothing to lose, which ties into everything else that happened during the pandemic. He does not claim he expects to beat Carlsen. He felt, for the first time in years, that he was playing freely.

Asked about mental preparation, Nakamura says he feels like two different people. Until the pandemic he was a professional chess player, and all of his income came from the game. From the pandemic on he is a different person, because that is no longer where the money comes from, and the psychological profile changed with it. The phrase that became a meme, "I literally don't care," is a way of being. He is competitive. He wants to do well. If he loses a game or a tournament, it is not the end of the world in the way it used to be, when the pressure of needing to always perform was very high. Before, the closest thing he had to a hack was short-term memory loss: a bad game, a walk that evening, a cleared mind. Fridman mentions Charles Bukowski's tombstone, "Don't Try." Nakamura agrees they point at something similar from different angles. Bukowski's version lives in art. His own version is for someone who used to eat from results.

Only the top twenty, maybe thirty players in the world make a living from chess. The living can be good, and he is not diminishing it. It is not secure. Invitations to the top events, with prize funds from maybe a hundred thousand up to half a million dollars, are the difference between earning two or three hundred thousand dollars a year and earning something like fifty thousand. That instability affected him, and not in a good way. Once he could make money elsewhere, the pressure came off. He could take risks and play the pure game. If he loses, it is not the end-all be-all. Looking back at the periods when he improved, one of them was 2008. He dropped out of college around 2650 Elo, roughly top 100 in the world. For the first half of 2008 he played almost nothing. He moved to Vancouver, lived on his own for the first time, and did not study much. He hiked mountains and went kayaking, things he had not done for years. Fridman had imagined a dark room, drinking, video games. After that stretch he started playing again, improved very quickly, and broke 2700. It had nothing to do with chess.

He has said many times that he is not a naturally talented chess player. His brother was. The brother was one of the top-ranked players in his age group in the United States. Nakamura's account of how he got to the top anyway is stubborn and small. If he does not get something, he tries to figure out why, what he is doing wrong, over and over. He uses Geometry Dash, a silly phone game, as an example. He played for an hour, was terrible, forgot it for a week, then came back and spent maybe a hundred hours over the following month tapping the same levels. He is not world class at it. He got pretty good. Chess was the same refusal to accept that he could not be good. He started poorly. His parents pulled him out for about six months. They relented. In the first tournament back, at seven or eight, he won three games against other kids and lost the fourth to a fool's mate. He stuck with it, played as much as he could, online, against his brother, devouring diagrams in books. He was not studying openings cover to cover. For anyone below master level, serious opening work barely existed until the early 2000s. The first leaps were tactics: forks, discovered attacks, pins, batteries. The more of those patterns you see, the quicker you catch them.

At the board, Nakamura says he is doing something very basic that a lot of people skip. He looks at piece placement. Are the pieces on optimal squares, or are there better ones. Only after that does he start calculating direct moves. He walks Fridman through a beginner's position: bring the bishop out, castle, then develop the other bishop and knight, put a rook on the c-file, and only then look for a plan. Pure calculation five or ten moves ahead is not realistic for amateurs. Asking whether each piece is in its happy place is. Strong players also do this to the opponent's pieces, but he thinks the upper levels have drifted toward treating the game as raw calculation and losing the human element. He and Carlsen both learned without engines at the start. Carlsen, Nakamura thinks, did not use a computer until around eleven. The next generation will grow up with evaluations after every game. They may well be stronger. The approach will be different.

The old post-mortem is mostly gone. Before engines became too strong, you analyzed with your opponent, two peers inventing human ideas. Now you go back to the room and the machine tells you the move. Some of those moves you cannot reverse-engineer. In a Najdorf he shows h4, a move humans now play. Twenty years ago, he says, if you showed that to Kasparov he would laugh and call you an idiot: you are pushing a pawn on the edge of the board in the opening, it does nothing. Ask Nakamura or any top grandmaster why it makes sense and they cannot really say. The computer says it is fine. With perfect play everything is probably equal, and if Black is careless, White can be better. Kids will play it nonchalantly because they know the engine signed off. It still feels wrong to him.

Will chess be solved from the opening position? Without major advances in quantum computing he highly doubts a hard solve. If it ever is solved, he is sure everything is a draw after move one. Engine games look decisive because the openings are set for them. Symmetric replies, e4 e5, d4 d5, c4 c5, will draw. The Petrov, e4 e5 Nf3 Nf6, is his candidate for the optimal line. Fridman wonders whether White can walk a thin line of permanent asymmetry. Nakamura does not think that exists. Demis Hassabis's line about creative tension between bishop and knight, he says, is partly poetry and also true. Fischer was especially good in bishop-versus-knight endgames, and the interesting positions tend to live in those imbalances. The beauty Nakamura finds now is often in the engine after a game: a move he would never consider, a depth that did not exist ten years ago. It is humbling. He and Carlsen have spent twenty or thirty years on this game, and then something just does not click until the machine shows it.

Nakamura was a very aggressive player until about 2700, in love with the King's Indian and the Najdorf, trying to fight with both colors every game. As opponents got better prepared, those openings had diminishing returns. He still plays them. About three weeks before the interview he won a King's Indian in St. Louis against an American junior. Play them every game, though, and the losses pile up. On the C-Squared Podcast he described getting blown off the board in the Najdorf by Fabiano Caruana four games in a row. Caruana had done poorly against that opening, worked until he solved the weakness, and punished the repetition. At some point you cannot keep doing it.

Ego, he thinks, is a mix. You have to be very confident to get to the top. Some players, like him and Kasparov, are expressive. Viswanathan Anand kept it buttoned up in public; a recent book described him angry in his room, banging walls or chairs. Every top player has that edge. If you do not get upset when you lose, you will not get near the top, in chess or probably in any field. For Nakamura it was never quite "I am the best in the world." It was walking into this game, against this person, knowing he could beat them. Stay too long at the top and the same openings produce drawish endgames and a kind of jade. As a teenager on the internet he got angry at opponents instead of at himself, and he insulted a lot of people. He wishes he had channeled it differently. He also is not sure he would have gotten as good without a me-against-the-world attitude. That anger, he says, really helped him improve.

Fridman asks him to empathize with Hans Niemann, another player with a me-against-the-world streak, at the start of the over-the-board cheating controversy. As of early October 2022, Nakamura will not say Niemann cheated over the board. He does not think there will ever be hard evidence unless someone is caught in the act with a phone or an earpiece. There is a lot of circumstantial evidence. Some people find the statistics convincing, some think they are nonsense. He is undecided. He thinks the next six to twelve months of over-the-board results will make it clear, and that Niemann is already at least 2650 no matter what, which is very high. A Bobby Fischer-level genius is absolutely possible. What he never saw, in a career of watching juniors rise, including Carlsen, was that early brilliant game against someone much higher rated. It is still possible something clicked. He just has not seen the pattern he is used to seeing.

Engine-correlation analysis looked legitimate to him at first. He later concluded anyone can upload the data, so he no longer treats it as statistical evidence. Some 2020 games still look suspicious given Niemann's rating at the time. That call, he says, is FIDE's, not his or Chess.com's. Carlsen has called cheating an existential threat. Nakamura agrees it is a serious problem. For five to ten years he played top events with no security at all. Earlier scandals, Sébastien Feller, Igors Rausis, Borislav Ivanov, never sat at the absolute elite, which is why organizers stayed lax. Doubt itself is a weapon. If you think you might be playing a computer, you close the position, you start questioning weird moves, you play worse. In his heart of hearts he thinks Carlsen could have handled the forfeit better. Having heard the rumors for two years, including one he heard in real time about Niemann playing better when games were broadcast, he is not sure there was a better practical option. Ken Rogoff wrote in the Boston Globe that they should just play a match under security. Nakamura doubts Carlsen would entertain playing someone he believes cheated.

The part almost nobody wants to demonstrate is how easy cheating is. At a weekend tournament with a hundred-dollar first prize, Nakamura says it can already be done. Even St. Louis's nonlinear junction device cost about eleven thousand dollars, which a local club does not have. Fridman asks a technical question: how little information would help at grandmaster level. Morse code of two squares is the naive version. Nakamura's answer is smaller. One buzz means the position is great; two means it is equal or nothing special. In a critical position he usually has two or three candidate moves. If he knows there is something, he can find it. He has that feeling in online blitz all the time, and the engine usually confirms he was right. In the last round of the Candidates against Ding Liren there was one last move that would have drawn easily. Every other move was significantly worse. He missed it and lost. A buzz then would have been enough. He would probably beat Niemann in blitz and rapid; classical is a different question. Niemann was adamant he had never cheated against Nakamura, and according to the report that part was truthful. If there is blame, Nakamura puts it on Chess.com for not releasing the online cheating, which he thinks happened at least four or five times, much earlier. He is an independent contractor. They do not feed him lists. He still does not understand why Carlsen stayed silent after the report came out, when he could simply have said he would not play someone who cheated online in a hundred games, a proven fact, without speculating over the board.

Asked for the greatest of all time, Nakamura will not put himself in the conversation, including the side conversation about the greatest never to play a world championship match. He thinks Levon Aronian tops that list by a big margin: number two in the world for a very long time, and he never even got to the match. Among champions he takes Carlsen, because he is playing the best chess by a bigger margin and has the highest Elo of all time. You can also make the case for Kasparov, number one for twenty-plus years, still living in opening systems people play today. Fischer shot up, was champion for a short window, and quit. Nakamura does not put him in the GOAT argument for lack of longevity. If Carlsen stays number one another three or four years, he thinks the debate ends.

What makes Carlsen so good is that there is no defined weakness. Reach an endgame and he is not going to blunder. Players out-prepare him in the opening, then start thinking for themselves and make mistakes. Maybe he does not like being attacked; king safety is not something you can steer a game toward the way you can steer a positional player into tactics. Kasparov had clear weaknesses: stubborn belief that his openings were infallible, and he was not a great defender. Kramnik understood him. They had worked together in the late 1990s, and Kasparov had helped Kramnik onto a Russian Olympiad team in the mid-90s. Kramnik beat him in the King's Indian four or five games in a row, and Kasparov gave it up for the Grünfeld. Nakamura's own breakdown is simpler. He is a universal player. He can play pretty much any opening. He does not make many blunders unless he is under a lot of pressure. That is more dramatic in blitz, where he is currently world number one. He guesses he has played over 300,000 games online. Down to ten seconds there is no calculation. There is a feel. He looks at a move, checks it is not a blunder, and plays it. He keeps games going until both players are low on time, and in those last ten or fifteen seconds he almost always wins.

The games he is proudest of both sacrifice the queen. In 2010 against Boris Gelfand, a King's Indian, he offered a queen sacrifice on something like five consecutive moves. Gelfand could not take it because of mate. Nakamura calls it probably the best game he ever played. In Barcelona in 2007 against Michał Krasenkow, in a Catalan, he sacrificed the queen for a pawn to drag the king into the middle of the board and checkmated. In those spots you usually know there is at least a draw by perpetual, which gives you a margin. Sacrifices of major pieces are rare at the top because both players have to cooperate to reach a double-edged game, and most of them do not want that. Engines, meanwhile, do not care about pawns. They will give one, two, three if the pieces are active, and in Fischer Random they happily sack rooks for bishops. Humans see the idea, do not know the follow-up, and lose the material in the long run. Paul Morphy, a Reddit request, was the best of his time by a bigger margin than anyone in modern history. Nakamura would put him around 2400 today against peers at 2000 or 2100, a 300-point gap, versus Fischer's roughly 170 over Boris Spassky. He would be a strong IM now. He quit because there was no one to beat, spent his last fifteen or twenty years doing nothing, and Nakamura has seen the grave in New Orleans. Not the GOAT. Still top ten if the criterion is how far you stood above your peers.

Nakamura still thinks there is merit in a world championship. The game has changed around it. Rapid and blitz events at the top now feel close to fifty-fifty with classical, which used to be the only thing that mattered. The championship's biggest issue is draws. Players have six months or more to prepare. The Candidates he had just played ended around July 5; the title match would be in February or March, about nine months later. With that much time, neither side will have opening weaknesses, both will be solid, and it may come down to tiebreaks. Carlsen had to win rapid tiebreaks against both Sergey Karjakin and Caruana. Nakamura thinks Carlsen does not feel the format lets him play creative chess, which is why he pushed so hard to change it. If you keep classical, Nakamura's suggestion is to hold the match about a month after the Candidates, long enough to prepare, not long enough to sandpaper every idea away. Ordinary events like St. Louis or the U.S. Championship give players a week or two. They still play exciting games.

Are he and Carlsen friends, enemies, frenemies? Rivalries get hyped as hatred. They want to beat each other. It is a small world. The top twenty to thirty can make a living, and there is a brotherhood. Frenemies, he says, is the simplest word. What he loves about Carlsen as a person is that everything seems to come easily, the way Andy Roddick used to talk about Roger Federer. The Niemann situation rolled off Carlsen's back and he went to the European Club Cup and gained rating as if nothing had happened. Nakamura knows the stress is probably there. Spectators never see it. He relates more to Rafael Nadal, because nothing in chess came easily to him either. In tennis, after eight years of lessons and hitting two or three times a week, he thinks he could beat Carlsen seven or ten times out of ten. In chess boxing, Carlsen is taller and has more reach.

Karjakin's disqualification from the Candidates over his support for Russia's invasion of Ukraine does not, Nakamura says, seep into the games. What it did was make visible how hard it is for Russian players to compete. In St. Louis, Ian Nepomniachtchi had to bring cash because Russia was cut off from SWIFT. Many have fled the country to keep a career going. Peter Svidler told him, "Don't get me started," and would not talk about it. He thinks most of them are not supportive of what is happening. Over the board he treats players as individuals. The one time he felt national tension, when he was younger, was going out of his way to beat Chinese opponents. The beauty of chess, looking back, is not a combination. It is the travel, and the fact that a billionaire and a nine-year-old from the inner city can talk about the same game on the same level. In 2003 he played the FIDE World Cup in Tripoli under Gaddafi and found people far friendlier than the picture he had arrived with. He still regrets not seeing Leptis Magna. He reads Herodotus. He likes getting away from technology enough to remember that humanity flourished without it.

A perfect chess day, which almost never happens, would be nine hours of sleep, preparation from nine to noon, lunch, then a walk or a nap with no chess in his head until a two o'clock round. He tries to listen to birds the way meditation asks you to hear your breath. He does not sleep well at tournaments. Most players eat light before a round; he has heard that Kasparov would eat a big steak and be fine. Focus depends on the game. Quiet positions let the mind drift until a critical moment. Complicated games require blocking noise, and he has never been great at that. The worst drift is a winning position against Carlsen: the mind jumps to rating points and what happens after you have won, and you have not won yet. A long classical day might run from two until seven or eight, then dinner, a thirty-minute reset, and two hours of work on the next game, win, lose, or draw. That is why he does not play more classical events. If he cannot give his best shot, he will not take a spot from someone who will. He did give the Candidates everything he had.

A perfect day off the clock is getting up at three or four and driving an hour to climb. A perfectly productive ordinary day starts at 7:30, Bloomberg or CNBC, then the markets. He has traded and invested for at least a decade, including real estate, and he watches Fed minutes for patterns. In 2015 or 2016 he noticed gold ETFs often rose when minutes came out. In 2016, shortly before a Candidates, he held a 3x long oil ETF as prices crashed, finally took the loss near the bottom, and decided he would never put himself in that daily stress again. He does not straight-short. In 2021, as Tesla rose, he sold puts and did well. He tries not to get hard-set in a conviction. He almost only takes long positions because shorting feels like betting on collapse, and the most successful people he has met stay optimistic. Retail apps that throw confetti after a trade worry him more than the legality of shorting. Chess excellence, he is clear, does not make him a genius at everything else. That is something he admires in Carlsen, who does not assume the title transfers.

Streaming taught him how to run a business. A Mary Buffett book on Warren Buffett's hands-off style is close to how he tries to manage the people around him. He attributes the streaming success first to already being a strong player with a following, then to the fans: people with cancer, people going through divorces, someone in a factory in the middle of the country. If a kid learns critical thinking from chess and becomes a scientist, that is what he hopes comes out of it. He does not feel the usual creator lows, because he considers himself too fortunate to complain. On the chess boom, he gives ChessBae, a Reddit moderator, real credit for pushing Twitch chess forward even if people hated the methods. After April 2021, collaborations among streamers broke apart. He has watched chess boom and bust before, Intel and IBM in the 1990s, Deep Blue, a Seattle U.S. Championship, the Fischer boom, and he thinks the busts come from a refusal to treat what is good for one as good for all. He talks publicly about his and Levy Rozman's income because he wants FIDE to feel the success and change. It has not happened, and that disheartens him. Eric Hansen, he says, is charismatic and owns his brand. The historical figure who would have streamed best is Mikhail Tal, who left a hospital in 1989 or 1991 to play a Moscow blitz event and beat Kasparov. Nakamura's own streaming path, looked at backward, includes commenting games in text on ICC, listening to Yankees radio because he did not have cable, and watching a Great Courses lecturer who dressed up as a medieval orator while he was homeschooled.

Fridman rents, on purpose, because owning nothing is a kind of freedom. Nakamura is a minimalist about clothes and cars, but security has always been the thing he chased, and he thinks it may have been a hiccup on the way to a title. He grew up with a single mother, an elementary school teacher, for the first six years. After she remarried, a lot of his stepfather's money went to flying a kid to tournaments. Once Nakamura made it as a player, he wanted to get to a point where that insecurity stopped. Owning property brings HOA fees and illegal-parking tickets, but it is the one tie he accepts. As a pro you can make a couple hundred thousand pre-tax, then do poorly for a year and never get back. Those thoughts, about prizes and results rather than only the chess, were always on his mind. Streaming in October 2018 did not feel like a startup that could lose everything. In hindsight it was a career risk. 2019 was a bad year over the board. He last played in India in November, planned a six-month break until the 2020 U.S. Championship, and used the time to stream. He never expected it to become a living. He thought it was fun, a way for fans to have access.

Chess is a singular pursuit. Results dictate everything. He has been lonely and depressed. The worst stretch was 2005 into 2006. He had reached the top 100 and stagnated, so he went to college, partly to be around people after being homeschooled. College was a disappointment. Chess had mixed ages and backgrounds; campus felt like some people mattered more than others, an Animal Farm line he could not live with. He spent the semester on PokerStars. Near the end he played a tournament in Philadelphia with almost no preparation, won it against strong grandmasters, and decided that if he was ever going to take the chance it had to be now. Four years of political science would have left him with a what-if he could never test. If he had not been a player, college-era Nakamura might have done law. Younger Nakamura loved ancient history, archaeology, and languages. More recently, finance. In thirty years he wants a beach house in Malibu, not a yacht, and not a life where people recognize him in the supermarket. Inside chess he is famous. Outside it he can disappear. He does not want TMZ down the street.

Advice for young people: follow your heart. His mother was a strong violinist, maybe All-State in California, not close to Juilliard, and there are a lot of starving musicians. His stepfather, of Sri Lankan descent, comes from a family of lawyers, including an uncle at the International Court of Justice. He went to Southampton to study law. His parents took away the chess board and the books and told him he would become a lawyer. Nakamura feels lucky his own parents, having lived those stories, supported whatever he did. You might pursue a passion and fail. He thinks you have to pursue it. Experiences matter more than the what-ifs. Too many people follow the standard route, college and two hundred thousand dollars of debt, because that is what you are supposed to do. At some point you make hard decisions if you are not making it. Time and experience are the things you cannot put a price on. When he dropped out there was no guarantee he would make it as a professional. It worked out. He would rather fight a horse-sized duck than a hundred duck-sized horses, one clear enemy. For a day he would be Elon Musk in the years the rockets were blowing up. If he were the last person on earth he would walk, bicycle, plant trees. He used to hate losing more than he loved winning. Now it is the reverse. Love, he says, can be the greatest thing in the world and, when it falls apart, a real low. He has been through both. For some chess players it takes the edge off and they play worse. For him it is still better to have loved and lost. Fridman teases him: I thought you did not care. A tiny bit, Nakamura says. A very, very tiny bit.