Since I’ve already started a threadjack, and I found this really fascinating.
Paywalled article from above -
HAMBURG, Germany— Max Deutsch went through a month of training before he traveled across the ocean, sat down in a regal hotel suite at the appointed hour and waited for the arrival of the world’s greatest chess player.
Max was not very good at chess himself. He’s a 24-year-old entrepreneur who lives in San Francisco and plays the sport occasionally to amuse himself. He was a prototypical amateur. Now he was preparing himself for a match against chess royalty. And he believed he could win.
The unlikely series of events that brought him to this stage began last year, when Max challenged himself to a series of monthly tasks that were ambitious bordering on absurd. He memorized the order of a shuffled deck of cards. He sketched an eerily accurate self-portrait. He solved a Rubik’s Cube in 17 seconds. He developed perfect musical pitch and landed a standing back-flip. He studied enough Hebrew to discuss the future of technology for a half-hour.
Max, a self-diagnosed obsessive learner, wanted his goals to be so lofty that he would fail to reach some. At that, he failed. Max was 11-for-11.
He knew from the beginning of his peculiar year that the hardest challenge would come in October: defeating Magnus Carlsen in a game of chess.
Magnus Carlsen is a 26-year-old world champion from Norway who has become a global celebrity because of chess. He belongs alongside Garry Kasparov and Bobby Fischer in any conversation about the most talented players ever.
Max’s original idea had been to beat a computerized simulation of Magnus. But when The Wall Street Journal stumbled across his “Month to Master” project while reporting another story, it offered to put him in touch with the real-life version. Max was game.
So was Magnus. It was undeniably a stunt, but it was also about something bigger, a grand experiment in human performance. Max’s adventure had implications for children and parents, workers in any industry and really anyone interested in self-improvement. At the heart of their chess match was a question about success: Can we hack our brains in a way that radically accelerates the traditional learning curve?
“Huh,” Magnus said. “Why not?”
A Chess Novice Challenged Magnus Carlsen. He Had One Month to Train.
To understand how Max Deutsch found himself sitting across the chessboard staring at Magnus Carlsen, there are worse places to start than a Brown University dormitory.
Max heard music coming from a room down the hall one night and walked outside to investigate with his friend Cliff Weitzman. They found three people on the floor playing the sitar. Max sat down with them. Weitzman chatted with his hall mates.
Max challenged himself to a series of monthly tasks that were ambitious bordering on absurd.
“But 15 minutes later, I stopped the conversation and started listening to Max,” he said. “He had taught himself sitar in 15 minutes sitting on the floor.”
The most surprising thing about the night was that Weitzman wasn’t surprised.
“Max learns faster than anybody I’ve ever met in my entire life,” he said.
Max has been that way longer than he can remember. His parents say he crawled before his twin sister. Max grew up in the Westchester County, N.Y., suburbs—his father ran a lighting company and his mother was a theater actress before staying home to care for her kids—and he was an inquisitive child with a voracious appetite for learning.
Now he optimizes his days around that interest. He takes a one-hour walk every afternoon to clear his head. He writes out goals for the next day before he goes to bed. And then he sleeps for eight hours. A friend once asked Max what he meant when he claimed that eight hours of sleep was nonnegotiable.
“Do you have a sister?” Max said.
“Yes.”
“Would you ever kiss her?”
“No!”
“Exactly,” Max said.
His first job after school, after writing an employment guide that went viral on Brown’s campus and recording an online lecture about how to negotiate a higher salary after graduating college, was as a product manager for a financial software company in Silicon Valley. It wasn’t long before his personal aspirations got the better of him.
He’d always dreamed of completing a bucket list of seemingly impossible tasks—crazy ideas that would stretch the boundaries of his own performance—and realized last year he didn’t have to wait any longer. So he didn’t. Max came up with a list of goals he believed he could achieve within a month. The only thing they had in common was the underlying motivation.
“To take basic skills,” Max said, “and very rapidly push them to the extreme.”
He told Weitzman about his plan. “Well, this sounds very much like you,” he said. And then he showed Weitzman the list.
“Max, this is absurd,” he said. “You can’t learn things this fast.”
But he could. And he did. Max began every month by considering the process that would lead to his desired result. He concocted an elaborate plan to crack the Rubik’s Cube, for example, that involved memorizing patterns and ordering lubricant to cut seconds off his solving time. He tracked his progress through daily blog posts and video evidence.
He had some familiarity with his tasks. Max had been playing chess since he was young and still messes around on a board with life-size pieces outside Weitzman’s apartment. He’d played Magnus on his Play Magnus app, which is powered by an engine that simulates the Norwegian’s skill and style at different ages from the time he was five years old. But he didn’t expect to play Magnus in person. Not even Max imagined that Magnus would agree to play a novice he’d never met.
It might seem beneath the best player on the planet to entertain the whims of a random amateur for no good reason. But he’s done it before. He seems to enjoy the spectacle.
Magnus agreed to play Bill Gates and limit himself to a severe time handicap. He crushed the billionaire in nine moves. He visited Harvard University to play 10 lawyers at the same time while blindfolded. He beat them anyway. It wasn’t because he felt a duty to evangelize chess or because of any commercial obligations that Magnus Carlsen thought it might be fun to play Max Deutsch.
“It’s just out of genuine curiosity,” he said.
He wanted to know whether someone could become good enough in one month to beat him—in part because Magnus knew better than anyone how difficult that would be.
It was obvious from the time he was a young boy that Magnus possessed the mental aptitude for chess. The first sign of his exceptional recall was that he could memorize world capitals and obscure facts about Norwegian municipalities. He liked puzzles and Legos in the same way that Max liked building houses with cards. But it wasn’t until he was eight years old, which is late for someone of his unique ability, that he showed the requisite interest in chess. And then he became very good very quickly.
Magnus often gets compared to chess greats, but the better analog may be someone in his other favorite sport: basketball. Magnus Carlsen is similar to LeBron James. They were both recognized as prodigies who came of age in an era of unprecedented public scrutiny. They both exceeded the hype.
Magnus became the sport’s youngest grandmaster in 2004 at 13 years old. He ascended to No. 1 in 2010. He won his first world championship in 2013. And he achieved the highest rating in the history of chess in 2014.
Magnus is now an international star and such a Norwegian hero that nearly half the population stayed up past midnight to watch last year’s world championship. What makes him a truly modern champion is not his collection of endorsement deals or his Netflix documentary. It is the way he plays chess.
His style is unpredictable, which makes opponents uncomfortable. He’s less mechanical than previous world champions and far more creative. To play Magnus is a cruel form of chess torture.
“Most schools of chess put a lot of emphasis on the openings to get an edge before the actual fight begins,” said Susan Polgar, the oldest of three renowned chess-playing sisters. “He puts the least emphasis on that. He puts a lot more emphasis on the psychological aspect of the game.”
There is so much chess information available online that anyone can study openings and endgames—even Max Deutsch. Magnus Carlsen’s genius reveals itself with everything that happens in between. It isn’t only innate talent that carried him to unprecedented heights. It’s also thousands of practice hours. Magnus is constantly thinking about chess. He is playing games in his head even when he appears engaged otherwise.
Magnus can look at the pieces of a chessboard and immediately recall what match it was, who was playing, when and where it took place and why it was worth his attention. It’s difficult to appreciate how amazing this is without seeing it for yourself.
The author Malcolm Gladwell popularized the idea that world-class success can be earned through a certain amount of serious practice, which became known as the 10,000-hours rule. There has been a contentious debate over how widely it should be applied, since it was based in large part around a study of elite youth violinists by psychologist K. Anders Ericsson.
In chess, there’s a consensus that expertise comes from years of serious practice. One famous scholarly paper in 1973 concluded: “There are no instant experts in chess—certainly no instant masters or grandmasters.” The academics behind the original research guessed that it would require somewhere between 10,000 and 50,000 hours.
“You can go pretty far in something like a Rubik’s Cube with not a lot of knowledge,” said Florida State University psychologist Neil Charness, who has studied chess for more than four decades. “But with something like chess, when you’re a human being, you cannot get very far unless you have a lot of knowledge.”
Chess experts found themselves in rare agreement: One month of training wouldn’t cut it. Polgar was flabbergasted that Max was even trying.
“What?!” she said. “Will that person have any aid? Like, a computer?”
Nope.
“You mean just his own skill?”
That’s right.
“And no prior tournament experience?”
Correct.
“Well,” she said, “it sounds quite unrealistic.”
Charness was equally blunt about Max’s chances. “If there are still bets available,” he said, “I would like to make a very large bet on Magnus.”
In fact there were bets available. Wynn Las Vegas oddsmaker Johnny Avello said the probability of an upset was 100,000 to 1. No betting house would ever offer those odds. The line that betting house Pinnacle posted, at the Journal’s request, was the most lopsided one that internal regulators would allow.
A $100 wager on Max paid $50,000. A $100 wager on Magnus paid 10 cents.
“I’ve consulted with some of our chess experts,” said Pinnacle sports manager Jelger Wiegersma, “and they all pretty much guaranteed me that Max will have no shot.”
Even those assessments may have been generous. There are calculators that can take the ratings of any players and compute their likelihood of winning any match down to nine decimal points. That number for Max was precisely 0.000000000%.
It was the reigning world champion’s right to set the rules of the match. His camp decided it would be rapid-format chess in which each player had 20 minutes to make all his moves. The date was set for Nov. 9 in Hamburg, where Magnus was already scheduled to host a promotional event.
Max’s year of monthly challenges had already been more successful than he could have imagined. He’d been contacted by students in a Belgian school who started their own projects after discovering his blog. Max, too, had been inspired by “Month to Master.” He left his job in August, raised money and started a company, Openmind, to guide people through the learning process.
Max hadn’t started thinking about chess at the end of September. He was still learning how to freestyle rap. “I don’t have a plan until the month begins,” he said. It was fairly conventional at first. He played Magni of different ages on the Play Magnus app.
He also played real people online, but only after lying about his meager chess rating to make himself appear better than he really was. Max figured he could only improve by playing better competition—that he would have to lose as much as possible to learn as much as possible.
He took advantage when Magnus offered access to his own youth coach, Norwegian grandmaster Torbjørn Ringdal Hansen, and they discussed chess principles before settling on two potential styles of play: conservative or aggressive. “Both are low-probability, but I think I’m going with the second option,” he said. “There’s no reason to play this safe.”
He was in New York visiting family one day midway through October when he agreed to take on the chess regulars who congregate every afternoon in Bryant Park. On the way there, Max passed a kiosk with other board games.
“If I could play Magnus in Boggle,” he said, “he would get wrecked.”
Max played three matches that day. He lost all three. The only sign his month of preparation might not be an epic waste of time was that one of his opponents happened to be wearing jeans made by G-Star—the same G-Star that once sponsored Magnus Carlsen.
Max realized that he would have to be more inventive in his approach to learning chess.
“If I can’t play like a human,” he said, “then how can I play?”
Max figured he would have to play like a computer.
He thought about memorizing every configuration of the chessboard. But he calculated that would take approximately one trillion trillion trillion years. Max didn’t have that kind of time.
He went hunting for shortcuts that would allow him to automate Magnus’s intuition. Max guessed that Magnus would play a certain opening, and he downloaded thousands of games with that opening to build a computer model that would distinguish good moves from bad moves. He would use techniques of machine learning to identify patterns—the patterns that Magnus has internalized—and devise an algorithm that computed whether a move was good or bad.
His formula would assign a value to every piece and every square, and Max would do the math in his head by deploying tricks he’d acquired through previous challenges to memorize about 30,000 numbers. If a move was good, he would play it. If a move was bad, he would try again.
His technology was less sophisticated than something like IBM’s chess-playing computer Deep Blue, he acknowledged, but it had to be to have any hope of making it work. He was relying on his own brain to process the information. The goal was to absorb enough of the computer’s objective analysis beforehand to compensate for his lack of intuition.
Max wasn’t delusional. “At least I don’t think I’m delusional,” he said.
He began to doubt himself two weeks before the match. He didn’t have the algorithm even after buying extra computing power to expedite the number-crunching. And he admitted he wasn’t sure he could pull off the mental gymnastics in 20 minutes. By the time he made it to Hamburg, the algorithm was churning away on his laptop, but it wasn’t ready. There were no numbers to memorize and no time even if there were.
His attempt to build himself into a computer had failed. Max Deutsch would have to beat Magnus Carlsen as a human being.
Max was anxious. He tried to relax by listening to funk music and fiddling with his Rubik’s Cube, but the setting wasn’t ideal. The room on the ground floor of the Hotel Atlantic Kempinski was so chilly Max had to wear a fleece North Face jacket he nearly forgot to pack. He was more tired than he would’ve preferred. The jet lag of flying to Europe from California had cut into his precious eight hours of sleep.
Magnus commanded attention from the second he sat down. He looked slick in a tailored suit, and he maintained a steely demeanor behind thick-rimmed glasses. He was taking the match seriously enough that he barely exchanged pleasantries beforehand. He didn’t try to make small talk, either. Max looked intimidated.
Magnus wasn’t invincible. His peak rating is higher than that of anyone else who has ever played chess, but his career winning percentage in competition is only 62.5%. He lost several days earlier to someone online whose name he couldn’t recall. Magnus didn’t want to lose again, and he didn’t think he would.
“But I’ve been surprised before,” he said.
Max moved his white pawn to e4. Magnus moved his black pawn to e5. And they were off.
Max had been right about the opening. If his algorithm had worked, he would’ve been in a solid position. But he was anyway. After eight moves, using his own limited chess ability, the unthinkable was occurring: Max was winning.
His skill wasn’t lost on Play Magnus’s chief executive officer, Kate Murphy, and head of communications Arne Horvei as they watched silently from a distance. “It’s lasting much longer than I expected,” Horvei whispered.
Magnus had reason to believe his opponent was better than he actually was. He was aware of Max’s algorithm, but Max hadn’t informed the enemy it wasn’t done. Max had his full attention because Magnus didn’t know he was bluffing. At one point, Magnus’s hands were shaking, not unlike his first world championship, when he was so nervous that he dropped his pencil.
“This is not going to be easy,” Magnus thought.
Max knew the probability of him winning. But even while being highly rational, he’d allowed himself some irrational thoughts. A small part of him believed he could win. He’d fantasized about how it would happen.
It was on the ninth move—the same point in the game that Magnus checkmated Bill Gates—that Max showed vulnerability. Every move he’d made until then had been the right one. And yet he knew immediately that he’d done something wrong, even if he didn’t know what it was. He could see it on Magnus’s face.
“You twitched,” Max said afterward.
Max dragged his knight to the middle of the board. It wasn’t technically a mistake. It was more a waste of a move that didn’t advance a larger strategy. If he were playing Weitzman back home, he might have recovered. But he couldn’t against Magnus. It was the myopia of an amateur—someone who wasn’t seeing the game several moves in advance. It also was an opportunity for Magnus to attack.
“Having the world champion attacking you can be a bit uncomfortable,” said Hansen, his youth coach.
They remained statistically tied until Max picked up his queen and jumped her two spots diagonally to the right for his 12th move. He could’ve kept his slight edge for at least another four turns by repositioning a pawn instead. But he didn’t have Magnus’s experience to foresee he was leaving his knight exposed, and he didn’t have his proprietary algorithm to let him know that moving his queen was foolish.
“This is a typical mistake for an amateur,” Polgar said, “not recognizing the potential threat.”
“It takes years,” Magnus said.
Max was in trouble. It only got worse from there. Two moves later, instead of taking Magnus’s knight with his pawn, Max used his queen. It was a horrible mistake. Magnus made him pay.
“When you moved your queen over here,” Magnus said as he remade the board from memory, “what was the idea?”
Max didn’t have a convincing explanation. There was none. It was the type of error his opponent had methodically drilled himself to avoid, and Magnus pounced once he identified the precise moment that probability had swung decisively in his favor. He knew he was not going to lose from that point on. He was right.
Magnus’s body language shifted. He barely thought about his moves anymore. Max deliberated for minutes; Magnus swiped his pieces in seconds. He felt the board shrinking. Max was beginning to see he couldn’t escape. The situation was as unpleasant as the chess intelligentsia had cautioned. At one point, Max accidentally toppled his king. Not long afterward, he was officially checkmated. The match had lasted 39 moves each over 22 minutes and 21 seconds. Magnus stuck out his hand. Max shook it. Only then did Magnus finally unfurl a smile.
Max’s year of monthly challenges was over. But he refused to take his loss as anything but a victory. He’d wanted his ambitions to be ambitious enough that he fell short. He said in a postgame interview that attempting to beat the most unbeatable chess player had introduced him to new lines of thinking. He was smarter about machine learning. There was also nothing stopping him memorizing those tens of thousands of numbers when his algorithm was finished. Maybe there would be a rematch.
And then something funny happened. It became clear Magnus wasn’t ready to leave. His previously blank face brightened. Now he was ebullient. He zipped pieces around the board and mumbled how he would’ve handled certain situations if he were Max. He recalled the exact chronological order of all 39 moves and scribbled them in a notebook. He looked disappointed when Max revealed that his original plan to write an algorithm had been foiled.
That encouraged Max to keep trying. Less than a week later, when he’d returned home and his algorithm was nearly done, Max tested its accuracy by checking how it would have played Magnus. He plugged in the queen move that Magnus had exploited. “Bad move,” the model said.
Max was delighted. This was proof his algorithm could have worked. Right after the match, Max hadn’t been sure. He figured he might as well ask the expert across from him for advice.
“If you had a month starting from scratch with chess and you had to get as good as possible,” Max said, “how would you think about it?”
“It’s very hard for me to answer that question,” Magnus said. “I haven’t been doing much else but chess for 20 years.”
His handlers looked at their watches. Magnus was late. They had panicked earlier that afternoon when he was one minute behind schedule, because Magnus always has somewhere else to be. In a few weeks he’ll fly to London for the final stage of the Grand Chess Tour as the heavy favorite to win the sport’s prestigious annual circuit, and he will almost certainly end this year the same way he’s ended the last six years: as the No. 1-ranked chess player.
So why was he still lingering with this stranger? It turned out that Magnus Carlsen was envious of Max Deutsch. He still had the whole game ahead of him.
“I hope you at least keep an interest in the game, because it’s very interesting,” Magnus said. “I wish I could learn it new.”
Appeared in the November 18, 2017, print edition as ‘An Unlikely Chess Match Tests Limits of Self-Improvement.’
