The King's Indian Debate: The Politics of the 9...a5 'Antidote' and the Hidden Truth of the Bayonet Attack

You will never see it in a tournament hall: a grandmaster pausing mid-preparation, wondering whether to play the line some DVD is selling. What you see in the hall is plain — the clock, the scoresheet, a glass of water, and sometimes a player's fingers drumming the table without touching anything. But behind the curtain, another market runs — the opening repertoire market. It is the quietest, most overlooked, and perhaps the cleverest corner of the chess world.

The recent discussion around 9...a5 in the King's Indian Defense's Mar del Plata branch is less a technical chess debate than a marketing one. And that is precisely what makes the whole affair interesting.

The Root of the Story: What the Bayonet Attack Is and Why It Matters

First, it is necessary to understand what the Bayonet Attack is. In the Classical Mar del Plata branch of the King's Indian Defense, White's 9.b4 is called the Bayonet Attack. It is not a new move. It has been White's sharpest, most direct, and most difficult plan for decades. White plays b4 to seize space, then presses on Black's kingside with the threat of b5 and c5. It is the move that has cost generations of King's Indian players their sleep.

Reporting on chess since 2026, I have watched how an opening line becomes "fashionable" and how the same line becomes "outdated" a few years later. Calling the Bayonet Attack "fashionable" is technically wrong. It is an established main line, played at world championship level, with thousands of games of data behind it. There is only one reason to call it "fashionable" — to make the reader feel that a new problem has emerged and therefore a new solution is needed.

That is where the real story begins.

9...Nh5 versus 9...a5: Two Paths, One Jungle

In the King's Indian, Black's main answer after 9.b4 has been 9...Nh5. This move is forcing. It compels pieces to attack, its outcome can be checked instantly by computer, and its theory is very deep. That is its strength and also its weakness. A player who plays 9...Nh5 needs 20-25 moves of memory. If the memory is weak, he heads straight for defeat.

By contrast, 9...a5 is a different philosophy. This move does not attack directly. It challenges White's b-pawn claim at its root. Black says — you cannot play b5, because I will first pressure your b-pawn with a5. This is a strategic move, meaning it demands planning, an understanding of piece placement, and most importantly, it does not show an immediate result.

One fact must be made clear here, because it is being suppressed most in this entire discourse: 9...a5 is not a new discovery, and it is not an alternative to 9...Nh5 but a complement to it. Both branches coexist in Black's repertoire. Whoever presents them as rivals is manufacturing an artificial contradiction — so that his product seems necessary.

The Computer Argument: The Weakest Link

The central argument of this entire promotion is that "computers do not know how to handle these complex strategic positions." That argument might have worked in the 2000s. It does not work today.

In the modern neural engine (NNUE) era, computers evaluate closed structures far more reliably. Stockfish, Leela, and other engines give accurate evaluations even in closed positions like the King's Indian. So how can this argument be defended?

There is only one honest version of it: in closed positions, engine evaluations are less stable, and less useful at the human level. Meaning — the engine says "it is equal," but it does not tell a human what to do next. The engine's first-choice move is often one a human cannot practically execute. This is a problem of usability, not of ignorance.

Promotional material blurs the two. It says computers do not know, when the truth is that computers do know, but their knowledge is not directly usable by humans. This is a subtle but very large difference.

I have seen young players in several Delhi tournament halls memorise engine lines and then fail to understand them over the board. This is the same problem — the bridge between engine knowledge and human understanding is broken. But the solution is not to say computers know nothing. The solution is to accept that building this bridge is hard, and that this is the real skill.

"Few Lines, More Strategy" — An Unfalsifiable Promise

The product's second major promise is that it contains "very few concrete lines" and emphasises strategy over memorisation. This promise is internally consistent. But it is also commercially self-undermining.

Why? Because a line without forcing moves cannot be tested by the buyer before purchase. There is no forcing sequence to test against an engine. And after purchase it cannot be "refuted" either. This makes the product's central claim unfalsifiable. It protects the author from criticism but weakens the product's credibility among sophisticated buyers.

One more thing must be understood here. If 9...a5 were objectively strong, it would stand alongside or ahead of 9...Nh5 at elite level. Low theoretical density usually means "playable and unrefuted," not "strong." This is a subtle but decisive difference.

Who Is the Buyer? A Hidden Segmentation

The most interesting thing in this whole affair is that no promotional material names a single player. Not one game. Not one engine evaluation. No date. No publisher. No price.

This is not a coincidence. It is a deliberate strategy.

When you promote a product that cannot be verified, you need an audience that will not demand verification. And in the chess world, such a group exists — players tired of engine-based preparation, who want to understand rather than memorise, who prioritise "thinking for themselves."

This group is real. I have spoken with many such players — club-level, rated 1800 to 2300 — who say they do not enjoy memorising engine lines. They want to understand how a position works. That is a legitimate desire.

But a contradiction hides here. A player who does not want to memorise must work even harder to understand. "Fewer lines" does not mean less work. It means the direction of work changes — from memory to decision-making. And decision-making is always harder than memory.

This is why intermediate players (roughly 2026-2200) are the real target of this product. A lower-rated player will benefit from the reduced memory load but will not fulfil the need for strategic understanding. And a higher-rated player will likely gain little from it.

The Question Nobody Is Asking

Now the part where I must take a somewhat different stance.

One thing is not being said anywhere in this discourse. It is that the chess opening preparation market is fundamentally a trust-based market. You pay for something you cannot test before buying, cannot fully verify after buying, and whose real value is assessed months later in a tournament — when the pain of defeat and the joy of victory both fall to you, not the author.

This is not a moral accusation. It is a structural observation. Opening products have been sold this way in chess for decades. But when a product sells itself as "the computer's undoing," questioning it becomes necessary — because that claim is verifiable, and it is wrong.

I must also say I am not denying the line's technical utility. There is a valid technical logic in 9...a5. White's 9.b4 is a pressure-building move, and Black's queenside pawn move challenges that pressure at its root. This is a legitimate "avoid the machine, play the human" design. But it also means Black permanently accepts a smaller share of the position's dynamic assets — in exchange for strategic clarity. That is a trade, not magic.

What Stays Behind the Curtain

When I am in a tournament hall, I see what the streaming camera never shows. The camera shows the clock, the evaluation bar. It does not show what the coach standing in the corridor is thinking, how the parent in the lobby is checking a bank balance, or how many times a player pauses before signing the scoresheet.

The world of opening preparation is the same. The DVD that sells is the part behind the curtain that never reaches the camera. But the real game happens in the hall — when you play 9...a5, and your opponent thinks for 15 minutes, and then plays the move that was not in your DVD.

In that moment, what do you have left? If you learned strategy, understanding. If you only bought a product, an argument with the clock.

Looking Forward

This whole King's Indian debate actually points to a larger question in chess. In the engine era, what does opening preparation mean? Is it a game of memory, or of understanding? And if it is understanding, how can anyone sell it?

The move 9...a5 is not the answer to these questions. It is a good example of them. It is a valid move, a valid plan, and perhaps a valid product. But it is not the revolution it is being described as.

Next time someone tells you he has found a way to beat the computer, ask one question — have you tested it in a tournament? And if so, against whom, when, and with what result?

If the answer comes with a date, a name, and a result, listen. If only belief comes, understand — you are looking at a product, not a discovery.