9...a5, the Bayonet Attack and the Promise of 'Few Concrete Lines': Notes on a Chess Antidote

The Club at Eleven at Night

On the second floor of an old house on Nguyen Thi Minh Khai Street in Saigon sits a chess club. Eleven at night, dim fluorescent light, a ceiling fan turning slowly. At board three, a man of about forty places his hand on a knight and leaves it there. Black's ninth move has not yet been made.

Before him is a textbook Mar del Plata: 1.d4 Nf6 2.c4 g6 3.Nc3 Bg7 4.e4 d6 5.Nf3 0-0 6.Be2 e5 7.0-0 Nc6 8.d5 Ne7, and White has just pushed 9.b4. His opponent, a second-year student with a newly acquired second-hand laptop, taps the edge of the table. "Play anything, just make it quick." The man does not answer.

Three months earlier, he had spent the equivalent of two weeks of meals on an online course. The course taught exactly one thing: after 9.b4, play 9...a5. Not 9...Nh5 the way everyone else plays. 9...a5, a quiet retreat on the queenside, and according to the advertisement, one with "very few concrete lines."

He plays 9...a5. The student frowns and pushes a piece in a direction that appears in no book he has ever read. In the silence that follows, what I hear is not the sound of pieces on the board, but the sound of a different battle — the battle between memorized memory and the ability to think for oneself.

Why a Ninth Move Is Worth Writing About

To understand why 9...a5 can make a man sit still for ten minutes in the middle of the night, we have to go back to the structure of an entire opening system.

The King's Indian Defense begins with 1.d4 Nf6 2.c4 g6, followed by ...Bg7, ...d6, ...0-0. Its philosophy is hypermodern: Black deliberately concedes the centre for the first ten moves, builds a wall of pieces on the kingside, then counterattacks the centre with ...e5 and ...f5. Black accepts a space disadvantage in exchange for a position in which every piece has a job.

The sharpest branch of this system is the Mar del Plata, named after the tournament held in Mar del Plata, Argentina, in 2026, where it was tried and popularised. In the Mar del Plata, Black plays ...Nc6 and ...e5, locks the centre, then launches a kingside attack with ...f5. White responds by sealing the queenside and pushing pawns. The result is two sides fighting on two different wings, and whoever arrives first wins. In the ECO classification, this branch sits roughly between E97 and E99.

The move that made the Mar del Plata hot for decades is the Bayonet Attack, 9.b4. White pushes the b-pawn to gain queenside space, prepares bxa5 or b5, and opens the a3 square for the bishop and b1 for the rook. It is the most concrete approach White can take: instead of dreaming of a kingside attack, White drills straight into Black's pawn structure.

The orthodox answer has always been 9...Nh5. The knight jumps to the rim, opens the way for ...f5, and turns the game into a concrete calculation race. After 9...Nh5, almost every move is constrained by theory that has been analysed to move twenty-five or thirty. That is why professionals both love and fear this branch: it gives Black a lively position, but it charges Black hundreds of hours in front of a database.

9...a5 sits on the opposite side of the same story. Black does not jump the knight to the rim, does not rush ...f5. Black pushes the a-pawn one square, hits b4 directly, and forces White to decide immediately: either push b5 to keep space and accept an isolated pawn, or trade on a5 and accept a completely different pawn structure. Both roads lead to positions where no move is "forced" in the style of forcing theory. Black concedes a little initiative in return for the right to decide.

And that is precisely the selling point.

A Retreating Move Is Not a Weak Move

In chess, a move that retreats to one's own seventh rank is rarely celebrated. Spectators remember advances. Commentators like advances. Videos with sensational titles like advances. But chess history is written by a great many retreating moves — moves that create no moment, only possibility.

9...a5 is such a move. It threatens no mate. It opens no tactical blow. It does exactly one thing: it puts a question to White's pawn structure and forces the opponent to answer with understanding rather than recall.

I have spent many evenings reconstructing old games and annotating move by move, and what I have come to realise over the years is this: the real value of a move lies not in how strong it is, but in how long it forces the opponent to think. A move like 9...Nh5 forces White to remember. A move like 9...a5 forces White to think. These are two entirely different kinds of pressure, and the second kind never appears on the scoresheet.

Based on my experience following games in domestic tournaments and in analysis sessions with young coaches, I keep seeing one repeated behavioural pattern: Vietnamese club-level players often lose not because they lack knowledge, but because their knowledge is too narrow and too deep. They know the first fifteen moves of one single line by heart. When the opponent branches off at move nine, they collapse within the next ten minutes. A course that teaches "a different ninth move" therefore touches a real wound.

But this is where a question appears that the promotional material never answers: after playing 9...a5, what does Black actually do next?

"Computers Do Not Understand" — An Expired Sentence

The most seductive part of this product's sales pitch is the claim that 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" — with the implication that computers often do not know how to handle such complex strategic positions.

That sentence was once true. It is no longer true.

Before the neural-network era, chess engines worked by tree search and evaluation based on material count plus a set of geometric factors. For that kind of engine, a closed Mar del Plata where Black concedes the centre and waits for twenty moves really is a hard problem. An engine has no concept of a "long-term plan." It has only a score.

AlphaZero, then Leela Chess Zero, then the neural-network generation of Stockfish, changed the picture entirely. These engines evaluate positions using neural networks trained on millions of self-play games, and they handle long-term strategic positions better than humans do. In closed King's Indian structures, modern engines are no longer "confused" in the way the promotional copy describes. They assess such positions calmly, and sometimes they are the very source that discovers plans humans never considered.

None of this means 9...a5 is a bad move. It only means the reason given to justify it has gone stale. This is a form of marketing language that might be called "pre-AlphaZero heritage": using an expired premise to create the feeling of a secret advantage. Buyers have no way to verify it, because the material offers no sample game, no engine evaluation, no ACPL figure, no win or draw rate from any database.

An opening-theory product without accompanying data can still be good, but it cannot be assessed. And something that cannot be assessed should not be trusted absolutely.

The Market of Antidotes

To place 9...a5 in its proper context, we need to look at the market it belongs to.

The opening-theory industry is one of the most stable and least glamorous revenue streams in chess. It sells no tickets, no television rights, no shirts printed with sponsor logos. It sells exactly one thing: answers to the questions players face on the board.

The most common format for this product line over several decades has been the Fritztrainer-style instructional DVD published by ChessBase from Germany, later expanded into subscription-based online course platforms. The general structure is very consistent: a grandmaster or an established international master picks a line that is causing trouble, presents its history, proposes an alternative, and closes with the promise that this alternative solves the problem.

Competition in this market does not come from other paid products. It comes from what is free. Enormous databases, powerful analysis tools, community studies on online platforms, thousands of free analysis videos — all available, all updated faster than any DVD. A paid product that wants to survive must sell what raw data cannot: a story, a roadmap, an identity.

And that is exactly what 9...a5 is wrapped in. The product does not promise to beat the Bayonet Attack theoretically. It promises to free the buyer from the burden of memorisation. That promise targets a very specific segment: players who see themselves as people who prefer independent thinking over rote learning. In marketing this is called identity selling. You are not buying a move. You are buying an image of yourself.

In Vietnam, this segment is real and growing. The number of amateur chess players has risen quickly in recent years, bringing with it a demand for structured opening study. But most of them have no private coach, no paid database, and no one to point out that a course teaching a ninth move cannot replace an understanding of pawn structures. They buy an antidote because they have no antibiotic.

When the Headline and the Body No Longer Look the Same Way

There is one detail in the source material I cannot overlook, and because fact-checking is my trade, I am obliged to raise it.

The title of the material is a team-round report: "Samarkand R2 (open): Despite close calls, all top teams grab wins" — a report about team matches, in Samarkand, with team names, results, and strong squads. But the entire body instead discusses a chess DVD teaching an opening in the King's Indian Defence. No teams. No rounds. No results.

Two different texts have been spliced into one package.

For an ordinary reader, this is a small error. For someone in the trade, it is an important signal about source quality. When a document's title does not match its content, every claim inside it must be cross-checked against the original, especially technical claims. If the passage describing the ninth move was pasted into the wrong place, there is no reason to believe that passage is technically accurate.

I raise this not to diminish the value of 9...a5. That move still stands on the board, however it is advertised. But how a move is sold and what the move actually is are two different matters. A young writer needs to be able to tell the two apart before spending money.

The empty stand still has an echo — I still use this phrase to talk about Vietnamese chess.

Domestic chess tournaments are held in air-conditioned halls with rows of folding chairs and a team of arbiters who walk very quietly. Sometimes a few parents sit in the last row. Outside spectators are rare. There is no cheering, no drums, no banners. There is only the sound of pieces touching the board and the clock ticking.

And inside that silence, a small market still operates. People study at home. People buy courses at eleven at night. They have no stand to perform on, so they buy a weapon to reassure themselves.

I thought about this while watching the man at board three lay his knight down on a5. He was not playing for an audience. He was playing for himself, and what he needed was not a resounding victory but the feeling that he was still in control of the board. A course that sells that feeling costs more than a course that sells a line.

The empty stand still has an echo — we only need to be quiet enough to hear it. And that echo usually speaks of things that appear in no theory book.

The Contrarian Angle: What the Promise of "Few Lines" Really Is

At this point I want to say plainly what most opening-product promotions avoid.

The whole chess world is arguing about whether 9...a5 genuinely neutralises the Bayonet Attack. That argument is fun, but it misses the centre. 9...a5 is a line that has been known for a long time, that already exists in databases, and that any seriously prepared White player has met at least once. Its value is practical and psychological, not theoretical superiority. That is entirely fine — but it should not be sold as an invention.

The real issue lies elsewhere: the promise of "few concrete lines" is in fact a promise to shift the burden. It does not make chess easier. It only changes the kind of difficulty. After 9...Nh5, the player must be good at memorising. After 9...a5, the player must be good at evaluating positions. If you are not good at memorising, this product helps you. If you are also not good at evaluating positions, this product only helps you lose more slowly and lose more beautifully.

The blind spot of collective memory sits here. We remember the games in which a retreating move turned the tide. We do not remember the thousands who bought a course, played the taught move correctly, and still lost forty moves later. The true death of a golden generation is when we stop remembering them. The true death of an opening line works the same way: it does not die when it is theoretically refuted, it dies when nobody bothers to analyse it any more.

And here is what I believe: in an era when every move can be checked by machine in three seconds, the only thing left to contest between two human beings is the capacity to endure uncertainty. No course can sell that. It has to be forged by sitting still and looking at the board, like the man at board three that night.

What Remains After the Piece Is Set Down

The club game ended close to one in the morning. The man won after forty-seven moves, in an endgame where he had an extra bishop but weaker pawns. The student lost without anger. He asked: "That 9...a5 move, which book teaches it?" The man smiled, put the pieces back in the box, and answered: "Books teaching it is one thing. Playing it at the right moment is another."

A goal can fade, but the applause in memory never grows old. In chess, the applause is usually not real — only the sound of pieces and the sound of breathing. But that breathing is enough to measure the distance between a player who understands what he is doing and one who is merely repeating what someone else has done.

If there is one thing I want to keep after following and cross-checking this whole story, it is this.

Generations will pass, but the poem about them keeps being written. Opening lines are the same. The Bayonet Attack will one day fall out of fashion. 9...a5 will one day be thoroughly decoded by a generation of players raised alongside neural networks, people who do not believe in the promise of "few lines" because they are used to verifying everything themselves. What remains is not a move, but the habit of standing before the board in silence and taking responsibility for one's own choice.

The question left for the reader, especially for young players weighing whether to spend money on an opening antidote: are you looking for a move to escape thinking, or a way to think better? Those two things carry different price tags, and on the board, they produce different results.