Magnus Carlsen would rather trust the numbers than the fairy tale

Lex Fridman opens with football, not chess. Carlsen is a Real Madrid fan, a serious fantasy-soccer player, and unimpressed by the usual GOAT pageant. He finds it hard to make a case for anyone but Lionel Messi. The love of Madrid predates the second Cristiano Ronaldo era; he likes Ronaldo and has been well treated at the Bernabéu, but when the club asked him to name a favorite player on camera they wanted a second take so he would say Ronaldo. He had named Isco. Club loyalty, in other words, is not his scoring system.

What he distrusts is the World Cup as a measuring stick. It is a tiny sample, and titles get worshipped even when luck did a lot of the work. Statistics say Messi is the best finisher of all time; that, plus the less countable parts of his game, is enough. Fridman pushes on the other side of small samples: the Olympics magic of living for one night, one mistake, rise-or-break. Carlsen says he does admire it. He also thinks “rising under pressure” and “breaking under pressure” are usually an oversimplified take on what happened.

Basketball gets a more mischievous answer. Jordan versus LeBron depends on his mood and on whom he is arguing with; he will pick one and fight for it. Statistically he has no doubt LeBron will surpass Jordan. The real argument, he agrees, is quantifiable versus unquantifiable greatness. In general he leans toward the numbers. Asked whether Messi should be punished for taking a merely decent Argentina side to a World Cup final they lost narrowly, he is unmoved by internet score-settling. If he is unsure, he stands with the data. The same habit will organize the rest of the conversation, including the claim in the episode title about the greatest chess player of all time.

Carlsen says he has had two peaks, 2013–14 and 2019, built from different strengths. Early on he mostly tried to avoid opponents’ opening preparation rather than beat them with his own. He was younger, had more energy, and needed that stamina because he could not finish people in the opening. By 2019 opening work had become a real weapon. Through both periods he thinks his intuitive understanding of chess stayed a little ahead of the field, even as everyone, including him, kept learning.

Calculation has always been a strange gap. He is “really, really bad” at solving chess exercises: find the best move, sit with a static puzzle, look deep. He cannot concentrate on them the way training wants. In actual games, with a clock and an opponent, he often finds the right idea anyway. What he knows he is good at is short lines of two to four moves, the little details inside them, and evaluation — judging whether a change in the position is a step in the right direction when the tree is too wide to calculate out. That mix is why some classical giants collapse in blitz: they need time for deep calculation, and blitz does not give it.

Asked to describe the picture in his head, he disappoints anyone waiting for *The Queen’s Gambit*. He visualizes the board. His interpretation is two-dimensional. There are not a lot of colors. He works one position at a time; other boards in a blindfold simul are stored away, and he sometimes has to repeat them because he forgets. The same leak happens in long thinks: he calculates a line, plays a few moves, and realizes he had already seen the position and lost it. Thirty minutes, for him, is usually a waste. It means he does not know what to do and is running into a wall looking for something that is not there. Ten to fifteen minutes in a complicated middlegame is the efficient window. If he has not seen it in ten minutes, he is probably not going to see it at all. Endgames that eat time are more brute-force; middlegames mix calculation, creativity, and evaluation.

The legendary endgame, including the marathon Game 6 of the 2021 world championship against Ian Nepomniachtchi, is less magic than the same skills applied where people expect brute force. With seven or fewer pieces the position is tablebase-solved; he knew that queen versus rook, knight, and two pawns was a theoretical draw and had to stay alert. Many practical endgames are not those rare theoretical dances. Queens come off, many pieces remain, and the work is understanding and evaluation — knowing which plans to go for after a good first assessment, then playing them. He describes the beauty of chess in similar language: arrange pieces and pawns so that small evaluative edges accumulate, then use simple, widely understood technique to suffocate the position.

Psychological intimidation is real, and he uses it. He thinks people would play a lot better against an anonymous version of him. He would love an online event with ten of the world’s best where you do not know the opponent each round. With a decent sample — he guesses about twenty games — style and strength would still leak; AI already recognizes players from their games. In a single game, even he would be unlikely to name the opponent. When he is playing well, people are more timid against him than against each other, sometimes without realizing it. If he senses they will not take all their chances, he takes more risk. He is always hunting that balance.

He can be intimidated too, but mostly by the unknown: something he does not fully understand. He sorts elite players into two camps, long lines versus short lines, the tortoise and the hare. He sees himself as the closest thing to a bridge. He can think, to some extent, both rapidly and deeply, which many people cannot do both. The same taxonomy is why blitz and classical sometimes look like different sports wearing the same pieces.

On openings, the job has changed since Kasparov’s day, when White could still expect a real edge from harder work, a strong team, creativity, and early engines. Now neural networks, classical engines, and hybrids are available to almost everyone. You no longer expect to find a White advantage in the old sense. You look for ideas engines miss entirely or miss at low depth, then steer into positions where you have more knowledge. They are often “semi-bluffs”: even if the opponent finds all the right moves you can still draw. Objectively best moves have been analyzed to death, so you have to leave them. Analyze any popular starting position long enough and the engine still says draw; the art is the less obvious path.

At the start of 2019, neural networks had just become a chess thing. Carlsen is careful, but he says you could see players in top events who were not using them, or not using them right. Positions were being evaluated differently by networks and traditional engines because they think about chess differently, and that gap was a huge advantage. His team uses engines; he tries not to look too much himself, because over the board you cannot have them, and imperfect engine knowledge can be worse than none. You are facing a human. You have to think as a human.

If he is not playing the world championship, people say the title may not matter. He said a version of that long ago: the moment he realized he was not the best player in the world, he felt he would have to renounce the title on moral grounds, because it does not mean anything if you are not the best. Ratings, over time, tell a clearer story. He is prouder of being world number one since the summer of 2011 than of the championships. His live rating in the conversation is 2861, which he thinks matches his current level. The 2900 goal would require getting considerably better, which he thinks is hard, or a long run of beating people he is supposed to beat while staying sharp. The number is useful as a horizon more than as a promise.

Twelve or fourteen games is a small sample if you want to know who is better, even in one matchup. More games would need faster time controls, which he also likes: long sessions with deep preparation let you mask deficiencies because you have time to think and defend. Less time makes defensive technique harder and the chess more “pure.” FIDE has floated two games a day at 45 minutes plus 15 or 30 seconds a move, sessions under two hours. The Miami event he is playing — four games a day at 15 minutes plus 10 seconds — would already be more interesting than one classical game. Twelve days, two games a day, 24 games, one black and one white, with rest days, would be a start. Preparing two opening packages a day would also make the teams’ work harder, which he considers a feature.

He first refused the 2011 Candidates because he disliked the format and did not want the pressure; being ranked number one was enough. In 2013 the format changed, he thought it looked interesting, and he tried. It was not a lifelong-dream speech. It was a good challenge that got him in the best shape of his life to that point. Right after winning he proposed that the champion not be seeded into the final. Players and the community mostly told him to keep the privilege. He wanted to have to fight for it every time.

The anxiety around the championship is not sitting at the board. When he arrives it is business; when he is playing well he is not afraid of losing because he trusts his instincts. The fear is off the board: losing is unacceptable, there is no next match for years, and “I am world champion” had become part of his identity. The exception was 2018 against Fabiano Caruana. Their ratings were so close that a loss would have made Caruana number one; every draw left the numbers still. All the classical games were drawn. He had a win he could not convert; Caruana had chances; the last game was a bit better for Carlsen. He enjoyed it because the opponent was, at that moment, his equal. Losing to someone you know you are much better than is what he cannot stand — the draws where he can feel he understands the position more and still cannot get over the hump. That is why 2021 Game 6 felt like a turning point against Nepomniachtchi: once he scored, he believed the match was over unless he messed up royally, the way slightly weaker sides in low-scoring sports can defend forever and then collapse after the first goal.

The hardest loss is not close: Game 8 of the 2016 world championship against Sergey Karjakin in New York. Games 1 and 2 were quiet. In 3 and 4 he was winning and should have converted; Karjakin defended well, but Carlsen mostly messed up. Games 5 through 7 went nowhere. He got impatient. For Game 8 he was ready to take more risk than usual, which he now calls insane, because he knew Karjakin could not beat him unless he beat himself. That knowledge — I am better, I only need one win and the match is over — is what fed the impatience. In 2021, after the first win against Nepomniachtchi, he had the same certainty.

He chose an innocuous White setup, more a system than a concrete opening: pieces on decent squares, no early raid into the opponent’s half, get him out of book, meet in a middlegame of as much “pure chess” as possible. After exchanges and maneuvering he started taking too many risks. He was still roughly fine, then realized he had gone too far and had to be careful. Then he froze. Mentally he filled with I might lose this, what have I done, why so much risk, I could have drawn at any moment. One move he had missed showed the game might not go his way. He made more mistakes. Karjakin, to his credit, understood it was his one chance and took it.

It is the worst Carlsen has ever felt after a game. He thought he was going to lose the title to someone not close to his level, mostly through his own stupidity. On the rest day after Game 8 he did something he never did at any other world championship: he got drunk with his team, just to forget. Game 9 he almost lost again. Game 10 he was tense, got a good opening, then made a mistake that could have forced a draw; the negativity flooded back, including how he would ever win with Black the next day. Karjakin missed the line. Carlsen ground him down in an endgame he eventually played well. After that, the weight lifted. He did not think about losing the match again. He looks at 2016 as a match that went very wrong for him — and that he still won, which he takes as evidence he was a bigger favorite than the 55–45 crowd thought.

He used to think about quitting every time he lost a classical game, especially a stupid one: if I am going to do things I know are wrong, I might as well stop. That happened a bunch of times. He has gotten more carefree about losing, which he is not sure is good, because hatred of losing is what kept the losses rare. Over the last ten years his performance after classical losses sits over 2900. The way he dealt with them is not particularly healthy. It has worked. Asked whether politics enters the board, he notes that FIDE’s then-reelected president is Russian and connected to the Kremlin, which still wants the crown back. On Karjakin’s ban after comments on the invasion of Ukraine, Carlsen has no strong view on the specific case, disagrees with the comments, and as a principle would rather ban no Russians or all of them than police “wrong opinions,” even reprehensible ones. After games, even rivals talk. Kasparov, asked why he was speaking to Karpov, whom he hated, said he was the only one who understood him.

If a game is at three, he wakes around eleven, walks, maybe listens to a podcast or watches last night’s NBA, and stays off chess. He comes back for a big lunch, usually a large omelette with a lot of salad, goes to the board, and in the fantasy version wins a clean game. What makes him happiest at tournaments is a good routine and feeling well. He does not like too much happening around him. The Chess Olympiad he had just played was the opposite: Norway’s team did horribly, he did okay, Uzbekistan won with a wave of young players, and the Norwegian camp still ate together, played football, swam. The camaraderie was fine. He still could not explain the result, and he was glad to be back on his own with a tiny team. At home in Norway he often goes out with big headphones, not as a disguise, just to stay in his own world. People are generally nice; he would rather not notice them turning around.

He tends to eat twice a day at events and often at home. He does not play fasted. He avoids a heavy meal before a game and avoids sugary food so blood sugar stays stable. Energy does not need to be high. It needs not to drop. Food poisoning at tournaments — sleep-deprived, no energy — has made him very calm and, he found, super creative. Sleep deprivation in general, he thinks, can do that. The first thing that goes is the ability to do simple things precisely. That is the only part of fasting that worries him. Creativity he would trade for; precision he will not.

On rare occasions he has played after a drink. He would not recommend it for classical chess. In blitz you live on short calculation and intuition, and a little alcohol can help those by making you think less and feel more confident — the same way people believe they speak a foreign language better after a drink, because barriers move. At the 2012 World Blitz Championship he had food poisoning, could barely play for three days, and sat around 20th at the last break. As a last gasp he hit the minibar. He came back relaxed, fast, and confident. He was not playing as well as he thought. He still won his remaining eight games and finished second; one more round and he thinks he would have won. Numbing the mind a bit, as a last resort when you already have the ability, can be a good thing. He does not drink coffee as a habit. He likes the taste. He has never had a job, so he has never needed to wake up early. If he is tired, he is tired, and he will work it out. He does not want his brain used to caffeine. If you see him drinking coffee, he is probably massively hungover and trying anything to make his brain work.

Exercise follows the same refusal to become a poster boy. In 2013 he was in great shape from football, tennis, cycling, whatever was fun. He never liked weights much. Fabiano Caruana actually lives in the gym; Carlsen just likes sports. He has always felt slightly undeserving of the “fit chess player” headline. Sport is an escape from the obsession of the main thing. Even juggling a football before a game makes him feel better: a skilled activity he is worse at, where he can still feel improvement. After the pandemic he was out of shape and trying to rebuild habits, still more for fun than for a spreadsheet of gains.

At home he does very little deliberate practice. He has never been the guy who can force himself to sit down and work. For many grandmasters that work is puzzles and opening analysis. Studying games, for Carlsen, is closer to reading an article or a book. He loves chess books — openings, strategy, history — and will read almost anything interesting. A better question than how many hours a board floats in his head is how many hours it does not. It is often running in parallel with something else. He looks at a random game on his phone or in a book, his brain keeps going, often all the way to an endgame, and then he nudges the pieces into a more interesting setup and plays on from there. If a position is too hard, he goes on to something else.

His first coach gave up on take-home exercises because Carlsen would not do them and did not enjoy them. The school had a massive chess library, so the coach told him to pick any books he wanted and bring them back next time. He raided the shelves. At the next tournament he would try an opening from whatever he had been reading. Nobody assigned a curriculum. That path is also why Kasparov-as-coach did not take. The first homework was to analyze three or four of his worst losses and send thoughts. The losses were not too painful. He simply did not enjoy the task, and the whole structured frame felt like a hassle from the start. He loved the idea of picking Kasparov’s brain. He could not see himself enjoying the rest. Then and always, he has played for fun.

Fridman hears a bold refusal of one of the greatest players alive. Carlsen does not remember thinking of it as his one opportunity. It was not for him; try another way. Advice to other players makes him uneasy for the same reason. The easiest line is love chess and be obsessed, which does not work for everybody. His way needs some mix of talent and obsession. He was lucky to have hands-off coaches who let him do his thing. Give him a bunch of work and it becomes a chore. Give him none and he will spend the hours anyway without calling it improvement. He names YouTube channels he has seen — agadmator, Gotham Chess, Botez Live, the St. Louis Chess Club, Daniel Naroditsky, John Bartholomew — and then undercuts the guru pose: he has never watched an agadmator video start to finish. His father has. Henrik Carlsen is about 2000 strength; agadmator is around 1900. It is the only chess YouTube video his father sat through. That, Magnus says, is the best recommendation he can give.

Fridman wants the romance of the 13-year-old who held Kasparov. Carlsen cools it down. They played three games; he lost two and drew one. In the draw he had a better position and could not convert. It was a rapid event, pairings set by a blitz tournament the day before, which had gone badly, including a beating by Garry. He slept twelve hours, woke up, searched ChessBase for Kasparov, and realized how many openings in his own repertoire had been Kasparov’s ideas. He was intimidated. Kasparov arrived late. The situation was psychologically awkward for the champion: it would be embarrassing not to win both games. Carlsen spent far too much time double-checking because it was Kasparov; at that age he normally played fast. He calculated better at one point, Kasparov missed a detail, and the youngster could not finish. The second game he lost badly. He was a very strong international master with grandmaster strength. Players of that level draw Garry once in a while. He was 13, he gets why it was a story, and he still felt it was more gimmick than prophecy.

The training games from their 2009 collaboration, ending in early 2010, mean more to him. Kasparov was still very strong. The score was fairly equal: Garry outplayed him quite a bit, Carlsen fought well. He appreciates those games more than a random 13-year-old draw. He is skeptical of anyone who claims they knew, from that game, that he would take Kasparov’s place. In 2018 an Uzbek kid beat him badly in the World Rapid; the kid later sat on a gold-medal Olympiad team without being the crucial board. Carlsen did not conclude the boy would be world champion. It was easy to see he would be very strong. Seeing the best in the world, or one of the best ever, is hard. We forget the projections that fail and remember the ones that land.

He can make a GOAT case for three names. Bobby Fischer, 1970–72: a peak ahead of its time, 20 games in a row, powerful chess with few mistakes, and the largest gap between first and second in history. Garry Kasparov: a brutal era, several generations beaten, consensus best for almost 20 years, longevity nobody in modern times has matched, a peak gap similar to Carlsen’s or a little larger, short of Fischer’s. Himself: unbeaten as champion in five matches, world number one for 11 years straight in an even more competitive engine-leveled era, highest rating ever, longest unbeaten streak. He has said before, and has not changed his mind, that Kasparov generally edges it on longevity in a competitive era. There are arguments. He is not interested in ranking style as part of greatness. Messi is best because his overall impact on winning is highest, not because the highlights are prettier. If he has to name a favorite player or style, it is young Kasparov before the title: an overflow of energy, extremely aggressive dynamic chess, appealing precisely because those are things Carlsen cannot do as well.

The most satisfying tournament victory of his career is not a world championship. It is the Norwegian under-11 championship in 2000. He started late, first tournament at eight and a half, while rivals had two to four years on him. In 1999 he finished a little above mid-pack among kids he had never played. A year of daily books and a board after school, plus his two weekly hours of internet spent only on chess, and he arrived knowing he was finally level with the best of his age, vibrating with excitement and nerves. He gifted a draw to a weaker player, drew the favorite, thought it was over, then the favorite lost the next day. Carlsen watched that board while winning the rest of his games and finished half a point ahead. First time he was best in his age group. He did not think world number one was possible. Best in Norway, maybe. The other kids treated chess like football practice. He had a passion they did not. He was never the fastest to grasp a thing. Once he started, he got hooked and never stopped learning.

Demis Hassabis, CEO of DeepMind, asked Fridman to ask what Carlsen felt watching AlphaZero — fear that the machine was taking over, or inspiration. Carlsen was hugely inspired. Man versus machine was lost before he reached top-level chess, and he never liked playing computers anyway. What amazed him was seeing them “think” about chess in a different way, in a way you could mistake for creativity. Hassabis himself, he notes, does not play like an AI; he plays in a very human way. AlphaZero will sacrifice a knight, sometimes two or three pawns, hunting positional domination that is hard to understand, then wait. In 2019 Carlsen was sacrificing a lot of pawns and it was a great joy. People have since found more solid openings that do not allow it as often. He is still trying to learn the art.

When he sacrifices, he needs something tangible a few moves later: material back, pressure on the king, or a concrete positional plus such as the bishop pair for a pawn. AlphaZero’s delayed, almost idle compensation is among the hardest things for his style to copy. Harmony, for him, is pieces protecting each other, nothing stuck on a bad square. He looks in overlapping groups, more toward the middle; control the center and you can attack both sides. The 2018 world-championship game that made him most nervous was the opposite of harmony: he abandoned the queenside to attack the king, knew that if the attack failed he would lose, and rode a wave of adrenaline into mutual time trouble before the position settled toward a draw.

Hassabis’s poetic line — chess stays fun because of the creative tension between bishop and knight — sounds good to him. What annoyed him in a little shogi was how weak most pieces felt besides the rook and bishop. Chess’s attraction includes how powerful the queen is. Knight and bishop ended up almost equal with wildly different qualities, a balance no designer could have fully planned. He wants more long Chess960 / Fischer Random: 960 shuffled back ranks, bishops on opposite colors, king between the rooks so you can still castle, even from odd squares. Engines have an even bigger edge there. He wishes for four-to-seven-hour classical 960, because you need that time on the first moves. You get the position 15 minutes before the game. Some setups punish you if you do not start symmetrically. The traditional starting position is still, he thinks, as close to ideal as we have; no variant he has seen is better than the game played for a thousand years. No-castling chess, promoted by Vladimir Kramnik, probably makes the opening slower and the whole game less drawish. There are weirder rules: pawns that also go diagonally, self-capture for clearance. Interesting to look at. Not replacements.

Henrik Carlsen has played his whole life around 2000 strength and never knew many openings. Magnus taught him two as Black: the Sveshnikov Sicilian, which Magnus used in the 2018 world championship and in 2019, and the Ragozin. The family’s Black repertoire is comically narrow; two of his sisters play the same lines. Watching his father in an online blitz game, Magnus saw a White setup he had never encountered. How did you come up with that? Henrik had just played the Ragozin with colors reversed: if the knight stood on the other square, it was the same position. Magnus’s reaction was that he was supposed to be one of the best players in the world and had never thought of it. He started playing the line as White with decent results. It became somewhat popular. Anyone who asked was told it was the Henry Carlsen. He would not necessarily explain why. He hopes the name sticks in the books.

What he learned from his parents is less opening theory than a refusal to turn a child into a project. They wanted a good education and a job. His father loved chess and still had no plan for a professional. For a long time they fought over school as Magnus cared less and less. At some point they gave up, coinciding with real tournament money. After that they put no pressure and no demands on him. They did want curiosity: a general education, not necessarily from school, knowledge of the world, history, some interest in society. He thinks they did well at that.

Fridman offers the cliché that the world is lonely and family is who has your back. Carlsen does not dodge it. Everyone you surround yourself with can help you, and to some extent they are also helping themselves. Only family has his interests at heart. That is why his father is the only person who has been constantly on the team. Whatever happens, he knows that man is on his side. Chess as a metaphor for geopolitics annoys him when leaders make moves they should know are wrong. He is never trying to get reelected at the board. Sometimes the right play is to do nothing and let the opponent lose — Petrosian’s joke about not stopping someone who wants to play the Dutch. A bad plan is not better than no plan. That slogan, he says, is absolute nonsense.

If he is not playing, does that tell everyone the title is not so important? He has thought so for a long time. After the last match he told family, friends, and his chess team it was likely the last. He had planned to announce immediately that he was done. Then Alireza Firouzja, 18, shot up to world number two and qualified for the Candidates. The thought of playing him lit a fire. Carlsen said publicly he was unlikely to play the next championship, partly to hype the Candidates — you might be playing for more than first — and to motivate himself and Firouzja. Hikaru Nakamura, among others, did not believe him and said so loudly. Nakamura then lost the last round and did not qualify. Even if he had, Carlsen says the decision was already made. Internet copium about a Nakamura match would not have dragged him back.

Does it break his heart to walk away from a historic event? He frames 2013 as a cool tournament he decided to try, not a destiny he is now betraying. Ding Liren versus Nepomniachtchi, the pairing without him, is close. He would give Ding a slight edge in overall strength: excellent on dynamics and imbalances, bishops versus knights, material versus time. Nepomniachtchi is an even faster short-line calculator than Carlsen, a monster, sometimes lacking a bit of depth, recently more solid in the openings. Their head-to-head being 4–1 for Nepo before Carlsen’s match did not mean much; it was a tiny sample, about the size of the match itself. Large samples, Carlsen says, are always good for him.

*The Queen’s Gambit* impressed him less as prestige television than as a chess source. They did the chess accurately and found games and positions he had never seen; he would lose the plot trying to solve them. Beth Harmon starts too aggressive and becomes universal, able to play any style. He sees a version of that in himself: aggressive, then too technical and not dynamic enough, then better at dynamics, now, he would say, the most universal stylistic player. Poker is a newer sandbox. The transferable skill is not letting past decisions dictate future thinking. He bluffs too much, enjoys the thrill more than the theory of bet-sizing, and notes you can be one of the best in the world and lose two or three years in a row without it being a wild outlier. Trash talk is easy when you are the best at the thing and can back it up; people expect a robot, so anything human looks like color. He is not sure he is actually good at it.

Does he get lonely? Yes. A chess player’s life is lonely by definition, because when you lose you have nobody to blame but yourself. Comfort is hard to find elsewhere. It is in your own mind. You versus yourself. Any sport that is just you and your brain is lonely by structure, not by personality. He is not worried it will destroy him as long as he is aware of it. He does not think that professional loneliness wrecks the rest of his life in a major way.

Love is not a hunt he is always on. Sometimes he is looking, sometimes not; he is trying to find his way. Love of the game comes and goes but never hits zero. Romantic love has come and gone and is not there now, and he is not worried. Stressing over love — looking or not, coupled or married — is a blueprint for being unhappy. He is more interested in being a good version of himself. Not always the best version. At least good. Asked the meaning of life, he does not buffer the answer. There is obviously no meaning. He thinks we are here by accident. It ends at some point. It is still a great thing. You can have fun, pursue goals, and stop treating the search for a special meaning as if it made sense. Life is both meaningless and meaningful just by being here, finding what makes you happy in the short term and the long term.

Fridman’s “simple” rules question — how does the knight move — gets a straight answer and a preference. The knight moves in an L, unlike in shogi it goes backward as well as forward, it jumps, and it is less happy in open positions where it must race from side to side. Carlsen is generally more of a bishops guy: quality over intangibles. He can still appreciate a good knight. At the end he admits he had been apprehensive about the podcast. Fridman is smart, the audience is smart, and Carlsen had a bit of imposter syndrome. He waited until after the conversation to say it. Fridman calls that beautiful even at the very top. The host closes with Bobby Fischer: chess is a war over the board, the object is to crush the opponent’s mind. Set next to Carlsen’s register — fun, mechanisms, small samples, a father on the team — it is a different religion using the same pieces.