9...a5 and the War Against Memory in the King's Indian Defense

In forty-one years sitting at the chessboard, I have counted exactly three players of the Black pieces who admitted to me that they prefer 9...a5 over 9...Nh5 in the King's Indian Defense. The first was a silver-haired Hungarian grandmaster I met at an open tournament in Budapest in 2026, a man who always said "chess is the sport of those who know how to forget." The second was a young coach in Da Nang, who taught his students with the line "don't study variations, understand the position." The third was an amateur player in Can Tho, who told me plainly that his memory was not good enough to keep up with 9...Nh5, so he chose another path. Three men, three generations, three completely different levels. But the reason was identical: they wanted to escape the prison of variations they were forced to memorize. Today, that reason has been packaged into a commercial product. And I think this is the right moment to discuss it seriously, because behind the glossy advertising lies a much larger question: in an era when computers understand chess more deeply than humans, does the strategy of "escaping memory" still have a place? The King's Indian Defense is one of the classic hypermodern systems, beginning with 1.d4 Nf6 2.c4 g6, with Black's bishop going to g7. Its philosophy runs counter to intuition: Black concedes the center to White in the early phase, then counterattacks that center with ...e5 and ...c5 after completing development. It is one of the most classical systems in modern chess, played at every level from club tournaments to world championship matches. The Mar del Plata variation — named after the Argentine city where it gained prominence at a 2026 tournament — is the sharpest version of this system. Black plays ...Nc6 and ...e5, leading to positions where both wings launch attacks, both kings come under fire, and often a single small mistake is enough to end the game. In ECO classification, this branch falls between E97 and E99. Players such as Gligoric and Najdorf, along with many generations of grandmasters who followed, contributed to building the theory of this branch, turning it into one of the most complex systems in classical chess. For decades, White's most popular way of meeting the Mar del Plata has been the Bayonet Attack — the move 9.b4. White pushes the b-pawn, gains space on the queenside, and forces Black to choose a response immediately. And the orthodox response for decades has been 9...Nh5: Black's knight leaps to the rim, opens the position, forces White into concrete calculation, and turns the game into a contest of who has memorized more. 9...Nh5 opens the position and forces the variations to pour out. It turns the King's Indian into a contest in which the computer's voice carries decisive weight. 9...a5 is a different road. Instead of pushing the knight up, Black pushes the a-pawn to a5, attacks White's b4-pawn, and steers the game toward a more closed position with fewer forced variations. This is not a new move. It has existed in theory for a long time, as a side branch off the main road. The product being advertised is not selling a novelty; it is selling a repackaging of an old idea. In Vietnam, the King's Indian Defense is not a popular opening among amateurs. Most Vietnamese players are more familiar with systems such as the Sicilian Defense or the French Defense, where plans are clearer and less dependent on memorizing long variations. That makes products like this a niche market — but also an opportunity for Vietnamese players to access a system they have had few chances to study properly. And that, precisely, is the point that needs careful analysis. The most notable thing is the argument's structure. The promotional material places the two moves on opposite sides of a scale: 9...Nh5 "opens the position, forces the lines, and makes computer analysis important again," while 9...a5 "leads to sound positions with very few concrete lines." The core promise is a change in the character of the game: from concrete struggle that demands memorization to strategic struggle that demands little study. Formally, this is an appealing argument. It strikes a very real pain of the amateur chess player: they spend hundreds of hours studying variations, then sit down at the board and meet a move that is not in the book. The feeling of being betrayed by one's own preparation is one of the most painful experiences in this sport. But the argument lacks data entirely. There is no ACPL figure. There is no win rate, no draw rate. No sample games are cited. There is no engine evaluation for any line. There is no author name, no record of the author in the King's Indian itself. The material says the two sides have "equal chances," but that is a positional claim, not a theoretical proof. In an era when every line can be checked with a single click, a theoretical claim without data is only a marketing claim. So why does 9...a5 still deserve discussion? Because its true value does not lie in being stronger than 9...Nh5. It lies in being different from 9...Nh5 psychologically and practically. In modern chess, when every player above 2500 has a vast database and a strong engine in their pocket, forcing the opponent off a familiar path is worth far more than gaining a small theoretical edge. 9...a5 does that: it does not try to prove Black is better, it only tries to make White think from scratch. The difference between the two moves can be described with a simple image. After 9...Nh5, the board is like a boxing match: the two sides exchange punches practiced a thousand times, and the winner is usually the one with better reflexes. After 9...a5, the board is like a game of Go: the two sides build territory, and the winner is usually the one who understands the position more deeply. Both are top-level chess, but they demand two different kinds of intelligence. I once watched such a game at an open tournament in Europe years ago. White, a confident young grandmaster, had prepared very carefully for 9...Nh5. When his opponent played 9...a5, he sat still for nearly twenty minutes, then played a series of inaccurate moves. The game lasted sixty moves and ended in a win for Black, but what I remember is not the result. What I remember is the silence. The young man sat there, not knowing what to do, because everything he had memorized was no longer usable. That is the strength of 9...a5: not the strength of a move, but the strength of a silence. In other words, this product sells something that cannot be measured in centipawns: mental self-reliance. The Black player does not need to remember more than the opponent; they only need to understand the position well enough not to be dragged into a fight the opponent has prepared better. This is a psychological strategy more than a technical one. And in chess, psychological strategy has a very long history. I remember a story about an aging grandmaster who throughout his career always chose the least-played openings, not because he believed they were stronger, but because he knew the opponent would have to think more. He told me: "I do not need to win in the opening. I only need to make the opponent feel lonely." 9...a5 shares that spirit. But here, I must say something those selling this product will not like. The argument that "computers often do not know how to handle these complex strategic positions" is a product of the pre-AlphaZero era. In the age of neural-network engines, computers handle long-term King's Indian positions far better than that sentence suggests. A modern engine does not need to see mate in ten moves to correctly assess a closed position; it evaluates based on millions of games and a neural network that has learned strategic patterns the old Stockfish generation could not. The claim "computers don't understand" sounds appealing to amateur players, but it is an outdated frame of thinking. Moreover, 9...a5 is not a secret. It is a famous side branch. Any White player prepared decently knows it exists and knows how to handle it. Selling an "antidote" for a line the opponent already knows is a gamble — and at the top level, such gambles usually survive only one short theoretical cycle. Once the move appears in a few major games, it enters the database, gets analyzed by engines, and is neutralized. On top of that, there is a problem of provenance. This material is packaged with a title about a team event in Samarkand, while the entire content inside is about an opening DVD. The two are unrelated. That is a red flag for information quality: if those behind it cannot keep the title and the content aligned, then the technical claims inside deserve a similar dose of skepticism. Failure is not a wrong move; it is when we see the plan clearly and still choose another path. But I do not want to end this section with a denial. Because there is another truth that cannot be denied. The opening-theory market is one of the most stable — and least glamorous — revenue streams in chess. It survives not because it offers the strongest lines, but because it offers structure, narrative, and guidance. A good book does not just hand us moves; it hands us a way of thinking. And 9...a5, though not new, remains a way of thinking worth studying. What I always tell my students: do not study a move because it is strong. Study it because it teaches you something about the board. 9...a5 teaches one very concrete lesson: sometimes, the best way to counter a well-prepared opponent is to refuse the game they have prepared. What I want to keep from this story is not a specific move. From keyboard to chessboard, the speed of words is never as fast as the speed of the pieces. An opening product may be written over many months, but it survives only until a strong enough player finds a refutation. And in an age when everyone has an engine, that window grows shorter every year. But the bigger question is: why do we study chess? If we study to win, then 9...a5 is only a temporary tool, one that will expire on some not-too-distant day. If we study to understand, then 9...a5 is a lesson in choice. It teaches us that sometimes the best path is not the strongest path, but the path that forces the opponent out of their comfort zone. I wonder whether in ten years anyone will still write an opening DVD, or whether all of it will be replaced by online courses personalized by artificial intelligence. But no matter how the format changes, the player's need remains the same: they want a path through the theoretical forest without getting lost. In forty-one years sitting at the chessboard, I have learned one thing: the best players are not those who have memorized the most variations. They are the ones who know when to forget. The three men I mentioned at the start — the Hungarian grandmaster, the Da Nang coach, the Can Tho amateur — have probably never met. But they share one philosophy: sometimes, the strongest way to counter an opponent's memory is to refuse to play the memory game. And that, perhaps, is the only thing in a commercial product I am willing to call genuine value.