9...a5 Against the Bayonet: An Opening Solution That Has Never Been Timed

On the board, a black pawn is quietly placed on a5. A dry, short move, almost soundless. If you were sitting in the spectator seats at a major tournament, you would not turn your head for it. But to a sports documentary screenwriter like me, moves that raise no echo are the most chewable kind of data. I trust footage more than testimony, because footage does not know how to lie.

In the Mar del Plata system of the King's Indian Defense, the move 9...a5 has just been pushed to center stage by a commercial opening product, with the promise that it is an antidote to the Bayonet Attack (9.b4). What caught my attention was not the move. What caught my attention was that someone is selling a solution that no one has ever timed.

Context: a system that lives on two choices

The King's Indian Defense is a story of calculated concession. Black lets White occupy the center with pawns, then builds the Black wall with a bishop on g7 to counterattack that very center. When White pushes the pawn to d5, the position locks into a two-wing war: White attacks the queenside, Black attacks the kingside. Whoever gets there first wins.

The Mar del Plata is the sharpest version of this system. After move eight, Black develops the knight to c6, then retreats the other knight to e7 to clear the way for the e5 pawn push. A knight on c6, a knight on e7, pawns on e5 and d6, a bishop on g7 — that is the framework of a long-term counterattack in which Black accepts being squeezed for space.

The problem with the Mar del Plata in the computer era is that it has been theorized too deeply. White has too many ways to play, and each way leads to a different forest of variations. One of the most annoying is called the Bayonet Attack: once Black has completed the setup, White pushes 9.b4. A short pawn move, but it opens a spearhead toward the queenside, threatens b4-b5 to chase the knight, and turns the queenside into a giant press.

The Bayonet is what keeps Black players who love the King's Indian awake at night — not because it is objectively strong, but because it forces Black to memorize an enormous amount of theory to avoid being pushed into a passive position. For those who play the King's Indian because they love strategic thinking more than cramming variations, the Bayonet is a nightmare.

Black has two paths of resistance. The main line is 9...Nh5 — the knight jumps to the rim of the board, opens the position, and forces the game into concrete calculating variations. The second path, less often mentioned, is 9...a5 — the black pawn strikes directly at the base of the b4 pawn, inviting a closed, slow game in which strategy overrides memory. The opening product I am reading centers on the second path, and turns it into a commercial promise.

What is sold, and what is promised

The first thing a documentary maker like me notices is how the product positions itself. It does not say "this is a new move." It says "this is an antidote." An antidote is a marketing category, not a theoretical one. In chess theory, people talk about novelties, engine evaluations, and win-draw-loss rates in databases. In opening marketing, people talk about the buyer's "pain" and the "dose" that solves it.

The pain here is the Bayonet Attack. The dose is 9...a5. The core promise is a change in the character of the game: from the concrete, forcing variations of 9...Nh5 to a strategic struggle, with little to memorize, after 9...a5. The product describes itself as suited to "people who like to think for themselves" rather than "people who cram variations." This is a sales angle based on the buyer's identity, not on technical evidence.

The second thing I noticed is the role of the computer in the argument. The product argues that 9...Nh5 "opens the position, forces the variations, and makes computer analysis important again," while 9...a5 "leads to healthy positions with very few concrete lines." And somewhere in the argument is a sentence claiming that computers often do not know how to handle such complex strategic positions.

That is the sentence that made me stop the longest.

What is the data actually saying?

When an opening product is sold, an analyst like me has to ask: what evidence exists, and what evidence is missing?

What the product provides: an alternative move (9...a5) to a main move (9...Nh5); a claim about the character of the position (more strategic, fewer variations); a claim about objective balance ("chances are equal"); and an argument about humans versus computers.

What the product does not provide: not a single sample game is cited; not a single engine evaluation; not a single statistic — win rate, draw rate, average centipawn loss are all absent; no author name, no title, no record in this opening system; no release date.

For someone who has written chess documentary scripts since 2026, this is a familiar sign. I learned that every script must carry its own data appendix, so that when challenged, I can just flip it open without arguing. This product does not have that appendix. It has a promise.

This does not mean 9...a5 is a bad move. In the Mar del Plata, 9...a5 is a branch known for a long time — not an invention, but a valid alternative. It strikes at the b4 pawn, forcing White to decide whether to keep the pawn or keep expanding. The resulting positions are usually more closed, with fewer forcing variations, and they do demand more strategic thinking than memory. In that respect, the product's argument is not wrong.

But "not wrong" does not mean "evidenced." A branch judged to be positionally balanced does not automatically become an antidote. Balance means White still has a position, still has a plan, still has play. A true antidote — if one exists — must be a path that leaves the opponent out of ideas. Such a claim needs data: it needs top-level games, it needs verified refutations, it needs an author with a record in this very opening system.

And here is the key point my experience has taught me: in the opening, the value of a branch lies in its ability to withstand refutation. A branch is only credible once people have tried to knock it down many times and it still stands. Without data on how many times it has been knocked down, a promise is only a promise.

When the computer is no longer an excuse

The claim that "computers often do not know how to handle complex strategic positions" belongs to a bygone era. It is the language of the pre-AlphaZero era — a time when people believed computers were strong in calculation but weak in long-term strategy. In the age of neural networks, that understanding is outdated. Modern engines handle long-term King's Indian positions far better than the sentence implies.

In other words, the argument "computers cannot understand this position" was once a reasonable selling point. Now it is a stale one. And more notably: that very claim may be a differentiation strategy against free tools. If you cannot compete with free databases and engines on data, you compete on story. You tell the buyer that those tools cannot understand the strategic beauty that only humans can.

That is a clever marketing angle. But it is not a technical argument.

A problem at the source level: the title and the body do not match

There is one detail in the original document itself that I must point out, because it concerns information integrity. The title given to this document is a team round report from Samarkand — round two, open format, suggesting that despite some close calls, all the top teams won. But the entire body is about a chess opening DVD selling 9...a5 against the Bayonet. The two texts have nothing to do with each other.

For a documentary maker, this is a red flag. It means the editing or aggregation stage mistakenly spliced two sources together. One headline talks about teams and a round; the body talks about an opening move. When a source has an error at the metadata level like that, every other piece of information in it must be doubted until independently verified.

This is why I never cite a source just because it reads well. I cite when I have verified it myself. In this case, I could not verify the author, could not verify the data, could not verify the release date. I could only verify one thing: 9...a5 is a real move, it really exists, and the argument about it is real.

Looking back at the opening-theory market

This product lives in a very specific market: the market for selling solutions to currently fashionable opening variations. It is a stable but unglamorous revenue stream in chess. People do not sell dramatic wins here; they sell the peace of mind of an amateur player facing a variation they fear.

At the top level, the King's Indian Mar del Plata survives mainly as a surprise weapon. It is not a mainstream choice in the engine era, because it concedes space and invites sharp positions that are easy to theorize. But precisely because it is the opening system of the loyal — people who play for love of the position rather than for a win rate — it still sustains a small, loyal market of its own.

In that market, the real differentiation is not theoretical superiority but practical steerability. That is what this product aims at: not to prove 9...a5 is stronger than 9...Nh5, but to prove it is easier to play, easier to remember, easier to feel confident with for the player holding the black pieces.

That is a reasonable goal. But reasonable does not make it an evidenced fact.

I once witnessed something similar on the track. In 2026, while making a documentary about SEA Games 29 in Kuala Lumpur, I chose to focus on the technique of athlete Nguyen Thi Oanh — who won three gold medals in the 1500m, 5000m, and 3000m steeplechase in just five days. The director brushed my proposal aside and told me to just write by feeling. I did not argue. I borrowed all the footage, sat down and counted Oanh's stride rhythm, recording 47 strides per minute in the final lap. That two-minute technical segment was what saved the whole film from a box-office disaster.

The lesson I drew was not "emotion is wrong." The lesson was: emotion without data does not stand. An opening product is the same. No matter how good the story is, it must come with a data appendix.

Who actually buys an opening antidote?

The buyer of this product is not a grandmaster preparing for a top tournament. A grandmaster has an analysis team, a private database, the strongest engines. The buyer is a club player, an amateur looking for a weapon so they no longer tremble every time the opponent pushes 9.b4.

And here is the tragedy of that market: what the buyer needs most is exactly what the product provides least. They need data, they need sample games, they need evidence that this branch has withstood testing. But what they receive is a story. A good story, perhaps. But still only a story.

This does not mean the product is useless. For an amateur, a sufficiently good promise can generate confidence, and confidence has real value at the board. But that value is psychological, not theoretical. And an opening product should be honest about what it is selling.

The contrarian angle

I must apply the same data standard to my own contrarian argument. If I demand that the product have data, I must also admit my own limits: I have no top-level game proving that 9...a5 has been neutralized. I only have a reasoning: a secondary branch like 9...a5, in a system analyzed down to the bone, is almost certainly already in the preparation of serious Bayonet players.

What I actually doubt is not the move, but the nature of the promise. The promise of "few variations, little memory" is a promise about experience, not about result. The buyer may find it true — they may feel more comfortable playing 9...a5. But that comfort does not equal an advantage, and still less does it equal the opponent running out of play.

Another trap: the absolutizing promise. Sentences like "if Black knows what to do, the position quickly becomes dangerous for White" are the kind of language that inflates expectations. It can hardly be refuted, because it depends entirely on White's level of preparation. A claim that cannot be wrong is not a technical claim; it is a marketing claim.

In sports documentary work, I have learned to distinguish two kinds of sentences. One kind says: "This athlete ran the 400m hurdles in 55.71 seconds, reaching the Tokyo Olympic semifinal." That is a verifiable sentence. Another kind says: "She is the embodiment of the never-give-up spirit." That is an unverifiable sentence. Both have a place in a film. But only the first can serve as a foundation.

What is worth keeping

I have run hurdles all my life, I just have never been timed. That is true of me on the track, and it is also true of moves that never make the front page. 9...a5 is such a move. It does not need a medal. It needs a data appendix.

What is worth keeping from this product is not the promise, but the question it raises: does a genuine strategic path exist for Black in the Mar del Plata, a path that does not demand the memory of a machine, but the patience of a human? That question is better than the answer. And in chess, as on the track, a good question usually outlives whoever answers it.

Every documentary is a race: the audience sees the finish line, I live at every starting line. With 9...a5, the starting line has been drawn. As for the finish, no one has timed it yet.