9...a5 Against the Bayonet: Reading a King's Indian Theory Product From the Source

The headline reads "Samarkand R2 (open): Despite close calls, all top teams grab wins". The body, from the first line to the last, discusses an opening-training DVD: Black's move 9...a5 against the Bayonet Attack in the King's Indian. Not one team. Not one round. Not one result. Two different texts sit inside the same package, and nobody forced them to match. I read that analysis on an evening in Shanghai, and the first thing I did was go back and check the source — a habit formed over four years of commentating chess for VTC, where a single wrong notation in the scoresheet is enough to ruin an entire broadcast.

The mismatch between headline and body is itself a signal about source quality. The more interesting part, though, sits behind it: the theory product, and the way it is being sold to players.

I have read hundreds of chess reports over thirteen years. This kind of error is not rare: content spliced wrongly, a headline that drifted, a body cut from an entirely different source. For a skimming reader it is harmless. For anyone using it as material, it is a quiet landmine.

The setting of one opening branch

The King's Indian is a hypermodern system: Black concedes the centre early in order to counterattack later. The Mar del Plata branch of the Classical system is the sharpest part of the whole family, where Black plays ...Nc6 and then ...e5, and both sides race to attack opposite wings. By my own classification, this branch sits around the ECO E97–E99 range; that is a personal label, not a fact stated in the source text.

Modern White players choose the Bayonet Attack, 9.b4, to grab queenside space before Black can close the centre. Black's orthodox reply is 9...Nh5, a move that opens the position and forces concrete lines. The product in question proposes something else: 9...a5, challenging the b4-pawn and steering toward a closed, manoeuvring game.

In 2026, while the country poured its attention into the men's World Cup in Russia, I and three friends founded a project called "The Forgotten Queens". That summer in Russia, I went looking for queens whose names were not on any map. From that project I took one principle that works for every kind of text, theory writing included: let the subject speak, and read the label before trusting the content. Here, the label says Samarkand. The content is a DVD.

Samarkand is a genuine chess city in Uzbekistan, where team events and open rounds take place regularly. A round-two report along the lines of "the strong teams all won despite close calls" could easily exist. It simply does not exist in this text. The entire body names no team, no pairing, no result, no standings table.

What is promised and what is left empty

The promise is fairly clear. After 9...Nh5 the position is forced into concrete lines, and engine analysis becomes compulsory. After 9...a5 the game turns strategic, with few concrete lines and little to memorise. The claim that "chances are equal" is a positional judgement, not a claim about a new move. And the product does not claim a novelty either. That is a more honest positioning than I usually see in this product category.

The list of what is not provided is far longer. No engine evaluation. No ACPL figure. No win rate, no draw rate. Not one sample game cited. No author name, no title, no record of the author's results in the Mar del Plata itself. No release date. For a theory product, those are the minimum indicators a buyer needs. Here every one of them is absent, and the absence is systematic.

A theory product does not sell moves. It sells certainty — and certainty is the one thing this text supports with no data at all.

The sales language says more about the format. Phrases like "antidote", "very few concrete lines" and "for players who like to think for themselves" belong to the Fritztrainer-style opening-training catalogue, where value lies in how the author organises and guides, not in raw variations. Raw variations are available free in databases and public studies. If that is true, what the buyer actually pays for is the storytelling: which lines to pick, which to drop, where to put the emphasis.

That storytelling targets a specific group. "For players who like to think for themselves" is identity-based positioning, not variation-based. It tells the buyer they belong to a certain kind of player, unlike the memorising crowd. The approach works well with club players tired of memorising lines, and it is emotionally honest. It simply does not answer the technical question: how strong is 9...a5, really?

On the theoretical map, the Mar del Plata survives at elite level as a surprise weapon rather than a mainstay. It concedes space and invites sharp play, so it only lives as long as the Black player can tolerate calculation pressure. The Bayonet Attack is the fashionable approach White uses to make life difficult. The product sits in the classic position of the theory market: selling an answer to a popular line. That is the most stable and least glamorous revenue stream in chess, and it has never disappeared.

Based on my experience following games, most products of this kind do not produce new moves. They systematise published theory, add some authorial comment, and call the result a solution. To evaluate one seriously, a buyer needs exactly three things: the author's name together with results in this branch, a few sample variations with engine evaluations, and practical statistics drawn from a database. All three can be verified independently. None of the three appears in the source text.

On the industry side, the product belongs to the opening-theory publishing market, one of the steadiest revenue lines in chess. It competes directly with free tools. Its survival depends on authored guidance, not on raw variations — which explains why the "computers don't understand it" argument is useful as marketing even though it is no longer technically true.

For young players, the impact is close to zero in the short term. For online platforms, the impact is also small: public studies and free databases already cover nearly all demand. For the derivative market of courses and training content, the effect is positive but narrow, over a few months. Products like this do not change how chess operates. They add one more layer to an already deep stack of theory.

The counter-current view

The most suspicious claim is the line about computers often being unable to handle these complex strategic positions. That is pre-AlphaZero language. Today's neural-network engines handle long-term King's Indian strategy far better than that sentence implies. Keeping the argument alive signals a text written from an older frame of reference, and it turns a selling point into a technical weakness.

The promise of "very few concrete lines" also needs cooling down. The Mar del Plata remains sharp by nature. No branch removes calculation from this system; at best it reduces the number of lines you must recall before the game starts. The gap between "fewer lines to memorise" and "less to calculate" is enormous, and the product never says which one it is promising.

As for "if Black knows what to do, things quickly become dangerous for White" — that is an unfalsifiable sentence. It is true of almost every move in every opening. It is not wrong, but it cannot be checked. In marketing, that is usually a sign of inflated expectation rather than a technical conclusion.

There is also a logic the theory market always follows: anything sold as an antidote summons a counter-antidote. White will reply. If 9...a5 appears often in practical play, someone will sit down, analyse it again, and find a way to keep a small edge. The real value of 9...a5 most likely lies at club and international-master level, where opponents prepare less, not at the top grandmaster level.

Buyer risk sits on three tiers. The theoretical tier: 9...a5 is a known branch, and a well-prepared White player can neutralise the antidote. The expectation tier: the buyer wants an opening without calculation, but the Mar del Plata stays sharp, and the shock arrives around move twenty. The cost tier: most of the content can be found free in a database before any money changes hands.

The signals worth tracking are concrete. The appearance of 9...a5 in Classical games at top events, in the 2700-plus group, would be the real test of the product's claim. The frequency of the Bayonet Attack in opening trees determines its relevance window. And locating the original Samarkand report would settle the source question for good.

What should be demanded

The problem with this text is not whether 9...a5 is right or wrong. It is that the reader is handed a promise without the tools to test it, plus a headline belonging to an entirely different event. For a theory product, three things ought to be mandatory: the author's name, sample variations with engine evaluations, and practical statistics. With those three, a club player can decide for himself. Without them, what is being bought is a carefully packaged belief.

In the world of chess, the only boundary is the boundary of skill. I have seen a chess generation ready to analyse on its own, and I believe it will never step back — as long as it keeps the habit of checking the source before believing it.