King's Indian Mar del Plata: 9...a5 and the War Against the Bayonet Attack — What a DVD Does Not Tell You

The Moment the Clock Runs Out

There is a moment every King's Indian player recognises instantly: your clock shows twenty minutes, your opponent's shows forty, and move nine has just appeared on the board — 9.b4. This is the Bayonet Attack. Not 9.Ne1, not 9.Nd2, but a thrust straight into the queenside, a move designed to turn the King's Indian from a strategic war into a race.

After years of watching events large and small, from world championship games to regional team tournaments, one pattern stands out. Most amateur players do not lose because they do not know good moves. They lose because they do not know what kind of position they are in. When 9.b4 lands, Black is no longer allowed to "play chess". He must remember. And memory, at club level, is a more expensive resource than imagination.

Among the hundreds of opening documents I read every season, I came across a product selling itself with exactly that fear. It promised an "antidote" to the Bayonet Attack: 9...a5. What stopped me was not the move. It was the way the move was being sold.

Two Sheets of Paper

On my desk there are two sheets of paper. The first carries a headline about a team event in Samarkand, round two, where "all the top teams won despite close calls". The second is the real content — a blurb for an instructional DVD promoting 9...a5 against the Bayonet Attack in the King's Indian Mar del Plata.

The two sheets have nothing to do with each other. No team, no round, no pairing appears anywhere in the body. It took me nearly an hour to be certain I had not misread. An hour for an editorial error — but also an hour to realise that in chess media, one can still discuss a move without naming a single player.

That is the starting point of this piece. Not to criticise a product, but to re-read a technical argument that has existed for decades under many different covers.

Context: Why the King's Indian Is Always a Gamble

The King's Indian begins with modest moves: 1.d4 Nf6 2.c4 g6 3.Nc3 Bg7, then ...d6. Black concedes almost the entire centre for twenty moves. In return, Black builds a wall on the dark squares and prepares ...e5.

It is one of the most philosophical openings in modern chess. It promises no immediate advantage. It promises a structure. The King's Indian player accepts life in a narrow house, trusting that the wall will eventually open a door toward the enemy king.

The Mar del Plata variation, named after the 2026 tournament in Argentina, pushes this to maximum tension. Black plays ...Nc6 and ...e5, launching the classic two-wing race: White storms with c4-c5, Black storms with ...f5-f4. The 2026 Mar del Plata tournament featured games by David Bronstein, Efim Geller, Miguel Najdorf and Svetozar Gligorić, and those games wrote the system into history.

From the 1950s to the 1990s the King's Indian was flesh and blood in the arsenal of top players. Garry Kasparov used it as a psychological weapon; Soviet players treated it as a way to deny their opponents a comfortable position. Its peak was a belief: if you do not understand the King's Indian, you do not truly understand chess.

Then the engine era arrived, and with it a quiet shift in how the opening was evaluated.

The Bayonet Attack: A Weapon of Impatience

Move 9.b4 in the Classical Mar del Plata is nicknamed the Bayonet Attack. The name says everything. This is not a building move; it is an attacking move, seizing queenside space before Black can complete the kingside plan.

Structurally, 9.b4 forces Black to answer a central question: what to do about speed? White has space, intends c5, wants to lock the queenside, and then turn to the kingside. If Black reacts slowly, he will be crushed literally — square by square, piece by piece.

For decades, the orthodox answer was 9...Nh5. This is the forcing move: the knight goes to the rim, preparing ...f5 and immediately opening a chain of forced variations. It allows White no rest. It creates a position where every mistake is punished concretely.

The problem is that "concrete" has become increasingly expensive for amateurs. After 9...Nh5, both sides must know dozens of lines stretching to move twenty-five or thirty. Forget one move and you lose the game. That is why, in chess clubs from Hanoi to Chengdu, the Bayonet Attack became a hereditary nightmare.

And that is precisely the fertile ground for a product selling 9...a5.

The Technical Core: What 9...a5 Does

Move 9...a5 is known theory. It is not an invention. It is a sideline, a turn to the side, and its value lies in changing the nature of the question on the board.

The mechanism is simple. After 9.b4, White has just pushed the pawn to b4. That pawn needs support — via c5, via a3, via coordinated queenside moves. The move 9...a5 attacks the pillar of that structure directly. It forces White to decide: push c5 immediately, or keep b4 and accept a potential weakness.

The interesting part is the consequence. When Black plays 9...a5, he usually does not rush ...f5 as in the main line. Instead, the game drifts into a more closed position, where both sides manoeuvre, hunt for weaknesses, and wait for mistakes. It is a deliberate trade: giving up forcing character to gain flexibility.

On the board, the difference shows at three levels.

First, pawn structure. After 9...a5, the queenside is often locked or semi-locked. Pieces no longer move freely along the ranks as in open positions. Every move must be calculated in a narrower space. For amateurs this is good news: fewer lines, fewer traps, less chance of being punished by a bolt from the sky.

Second, the tempo of the two-wing race. In the main line 9...Nh5, the race runs at the speed of an explosion: White plays c5, Black plays ...f5, both charge. After 9...a5, the race slows. Black has time to choose the right moment for ...f5, or even postpone it to shore up the queenside first. That favours thinkers over memorisers.

Third, psychology. And this is the level marketers understand best.

What the Product Promises and What Theory Says

The promotional text is built on a clear opposition. On one side, 9...Nh5 "opens the position, forces the lines, and makes computer analysis important again". On the other, 9...a5 leads to "sound positions with very few concrete lines".

The phrasing is skilful. It turns a sideline into a philosophical choice. It turns lack of preparation into a virtue. It tells the buyer: you are not lazy, you are independent.

But read the technical part again. The product claims to offer "very few concrete lines". What does that mean in practice? It means the buyer gets fewer memorised variations and more guiding principles. For some, that is a bargain. For others, it is ambiguity repackaged.

Theoretically, both assessments have merit. Move 9...a5 is sound, playable, and it genuinely leads to balanced positions. But "balanced" does not mean "safe". It only means both sides still have work to do. And in the King's Indian, "work to do" always means "room to err".

I spent many evenings reconstructing positions after 9...a5 on my analysis board. What I found is that the branch is not easy as the blurb suggests. It is merely different. Instead of memorising a chain of forced moves, the player must understand a set of ideas. The required skill shifts from memory to judgement. For some, that is liberation. For those who only know how to memorise, it is a new trap.

The Counterintuitive Angle: Engines Are Not Naive

This is where I want to linger.

The blurb implies that computers often "do not know how to handle" such complex strategic positions. That was once true. It was true in the pre-AlphaZero era, when search-based engines with handcrafted evaluation struggled with positions demanding long-term plans.

That era is over. Modern neural-network engines handle closed King's Indian positions far better than the sentence suggests. They do not need to "understand" the way humans do; they only need to evaluate correctly. And they do.

So why does the old argument survive? Because it is a perfect marketing tool. It creates a common enemy — the soulless machine — and turns the buyer into a rebel. It says: the world has been conquered by engines, but you can escape. Just buy this disc.

I do not believe in rebellion sold by the box. Based on my experience following matches, I have seen too many closed positions broken by amateurs who believed "computers do not understand" — and then made basic strategic errors any engine points out in half a second.

The real counterintuitive point lies elsewhere. The value of 9...a5 is not that it fools computers. It is that it fools human opponents. An opponent prepared for 9...Nh5 is pushed out of his comfort zone when he meets 9...a5. He must think for himself. And in club chess, forcing your opponent to think is a far stronger weapon than forcing him to remember.

That is the real technical argument. It needs no story about naive computers.

The Data Problem: Not a Single Number

One thing troubled me throughout this document: the complete absence of data.

No win rate. No draw rate. No average centipawn loss. No sample games. No player names. No author. No credentials, no titles, no tournament results.

In any other field, an advertisement without data would be considered unserious. In chess it survives, because buyers purchase with belief rather than proof.

I understand commercial pressure. I once wrote analysis for no fee, then saw it republished a month later by a large sports site. This market is not generous. But the absence of data is still a signal to be read correctly: it tells you the product wants you to believe, not to verify.

In my notes I always ask one question: what is the data hiding? Here the answer is simple. It hides how common 9...a5 actually is at high level. It hides whether strong players truly struggle against it. And it hides whether a well-prepared opponent can neutralise the whole idea with one evening of study.

Everything on the board is data waiting to be read — if you are willing to sit down. This document refused to sit down.

The Biggest Risk: Theoretical Obsolescence

There is a law in opening theory I have observed for years: every "antidote" has a limited lifespan.

When a sideline is widely promoted, it gets analysed. When it is analysed, it gets refuted. When it is refuted, it returns to being a sideline — but this time it is marked. Well-prepared White players will know what to do. A move that once confused opponents becomes a test they already have the answers to.

With 9...a5 the lifespan is even shorter, because this is a known branch, not an invention. A known but rarely played branch is a good weapon. A known and loudly advertised branch is a marked target.

This is the paradox of the entire instructional-DVD market. The more commercially successful the product, the faster its technical value erodes. It sells surprise, then destroys that surprise with its own hands.

The document also makes an absolute claim: if Black knows what he is doing, "things quickly become dangerous for White". This is the classic expectation-inflating sentence of any advertisement. It describes an outcome as inevitable, when in reality it depends entirely on the opponent's level of preparation. If White is better prepared, the sentence reverses. No data tells us who will be dangerous to whom.

I have watched "safe" sidelines shattered within a single season. And I have watched sidelines dismissed as weak rise to become main lines. Theory does not stand still. It only waits for people to forget it.

An Information-Integrity Problem

Back to the two sheets of paper.

A headline about a team event in Samarkand being spliced onto a chess-opening DVD blurb is not just a technical error. It is a problem of the trust chain. If a document can be mis-titled, its technical claims should also be re-checked against the original source.

In my own work I always add a methodology note: how I collected data, from stopwatch to tracking software. This makes me hard to attack and gives readers the right to judge for themselves. A document without methodology, without provenance, without an author, cannot structurally be ranked alongside serious analysis.

Here every link is empty. The title speaks of an event. The body speaks of a move. Nothing connects them.

Who Actually Buys a Product Like This

The target buyer is not a grandmaster. At 2700 level, every sideline is processed by an analysis team within days. They do not buy safety.

The target buyer is a club player, rated roughly 1500 to 2200, looking for a weapon he can play without three sleepless nights in a database. Someone overwhelmed by the explosion of theory, seeking a reasonable escape.

For them, 9...a5 is a reasonable choice. But the promise "you no longer need to memorise" is a dangerous one. You do not need to memorise, but you need deep understanding. You do not need to recall thirty moves, but you need to judge ten structures accurately. Liberation from memory is only a transfer of responsibility to judgement — and judgement is far harder to train.

Identity-based advertising always works. It does not say "this product is good". It says "you are the kind of person who understands this". For players who see themselves as independent thinkers rather than memorisation machines, that hits something deep.

I understand that desire. I once wanted to escape memory too. But after nearly half a century of watching, I have learned that no shortcut replaces understanding. There are only shortcuts that make understanding more expensive later.

The Chess-Theory Market

This product lives in a very stable but unglamorous ecosystem: the market for "answers to popular lines".

Its flow is clear. Upstream is the author and their opening expertise. Midstream is the publishing product — DVD, online course, book. Downstream is the study habits of players, coaching, and Thursday-night club games.

Its impact on each segment is small but real. Neutral for youth development. A grain of sand for online platforms. A small, steady revenue stream for niche commerce. Almost no ripple for the public image of chess.

One point is worth noting: this product competes directly with free databases and free engines. It cannot win on the quality of its lines, because every line exists elsewhere. It can only win on the story it tells. And its best story is "computers do not understand this".

That is why I call that argument a marketing asset, not a technical truth.

What I Am Still Waiting to Measure

I do not announce predictions. I only track observable signals.

The first signal is how often 9...a5 appears in elite classical games. If it appears regularly among 2700+ players, the product's premise is partly validated. If it fades within months, it is refuted.

The second is the frequency of the Bayonet Attack itself. If 9.b4 remains fashionable, the product's window stays open. If another branch replaces it, the product becomes a historical footnote.

The third is White's response. In every theory cycle, after an antidote comes an antidote to the antidote. The question is not whether it arrives, but when.

And there is a fourth signal I always wait for, though it rarely comes: the author's name. When a technical product hides its author, it places itself outside evaluation. The day we know the author and their results in this very system, that is the day we can assess it fully.

My Method for This Piece

I used no motion-tracking software here, because the subject is an opening document, not a game. Instead I reconstructed the 9...a5 branch on my analysis board, compared it with how 9...Nh5 operates, and noted three levels of difference: pawn structure, tempo of the two-wing race, and player psychology.

I read the source twice — once to grasp the technical content, once to check integrity. The second reading is where I found the mismatch between title and body. I did not fix it. I recorded it, because such an error belongs to the document itself, and readers have a right to know.

I keep my predictive conclusions to myself. In the article I present only what can be verified.

A Thought to Carry Away

I once wrote a line I still stand by: the King's Indian is not a tactic, it is how one side agrees to concede the centre in order to converse with the position.

Move 9...a5 is part of that conversation. It is no miracle. It is a choice — trading memory for judgement, forcing character for flexibility, certainty for freedom. For a player who knows what he is choosing, that is a beautiful choice. For a player who believes he has found an escape from study, it may be a trap with a red carpet.

And there is one thing I still believe at sixty-four: chess does not reward those who avoid work. It rewards those who choose the right kind of work to do.

If you are Black in the King's Indian and your opponent places 9.b4 on the board, the real question is not "which move should I remember". The real question is "what kind of player do I want to be". Every DVD in the world can only answer the first question. The second you must answer yourself, move by move, game by game, season by season.