Selling an Antidote to the Bayonet With No Numbers Attached: Inside the King's Indian Mar del Plata Pitch

There were six words on the box and one of them was wrong.

The cover prints 'King's Indian Attack' in the largest type available. The variation being taught inside is not the King's Indian Attack. It is the Mar del Plata branch of the King's Indian Defence — the game that begins 1.d4 Nf6 2.c4 g6 3.Nc3 Bg7 4.e4 d6, in which Black concedes the centre early and then counter-punches with ...e5 and ...f5. The two names look like relatives. In chess they are separate organisms: one is a White system, the other a Black defence. That single wrong word on the cover is the cleanest clue to the reporting problem that follows — a product whose first promise is "strategy, not memorisation" has misnamed its own subject in its opening line.

I have three racks of such discs, bought from 2026 onwards. When a new one arrives I do not read the blurb. I look for two things first: the index of model games, and an engine evaluation printed beside any position at all. This box had neither. The third thing — the presenter's name, title, rating — was missing too. Three blanks in one package. The first clue was the contents list nobody thought worth keeping.

You do not need to sit at a board to grasp the Mar del Plata. Black lightens his centre and says: spread west, and I will break through east. White opens files on the queenside, Black pushes ...f5 in front of his own king, and the race on opposite wings reduces to a single question — whose attack arrives first. It is chess's oldest drama, and that is precisely its strength: the result is decided not by one move but by thirty moves of patience.

White's most fashionable weapon in this branch is the Bayonet Attack, 9.b4. The pawn advances, takes space on the queenside, and slows Black's ...f5–...f4 programme. For years it has been White's main language. Black is credited with two answers. The first is 9...Nh5, the main line, where the game drops quickly into forced sequences. The second is 9...a5, the road less travelled, where the board does not open so much as tighten. This product recommends the second.

Take the sales copy in order, because its logic is coherent. One: the Mar del Plata produces long strategic positions for both sides. Two: computers do not understand these positions properly. Three: therefore the human player benefits. Four: but the Bayonet disrupts that comfort. Five: and the main line 9...Nh5 collapses back into forced, computer-friendly sequences. Six: so take 9...a5 and stay in the human domain. Six steps, one straight line. The argument is internally consistent — the trouble is not that the story is wrong, but that not one number stands behind it.

Step two, the claim that computers cannot handle these positions, is the weakest link today. That sentence belongs to the 2026–2026 decade, when engines were fast in open counting and blind in closed, manoeuvring positions. In September 2026, Stockfish 12 brought NNUE evaluation, and engine comprehension of closed, plan-based positions shifted at a stroke. In a branch like the Mar del Plata, engine judgement is now far more reliable than it was. An old heuristic is being polished and resold to a new market.

Stranger still, the copy cuts itself off in its own next sentence. Where 9...a5 is praised, we are told computers cannot understand these positions. Where 9...Nh5 is criticised, we are told that in that line "computer analysis becomes important again". Both cannot hold. If engine analysis is the norm, the recommendation's entire basis becomes an attempt to escape it rather than to outplay it.

The real craft sits in the third line: "objectively equal, but dangerous for White." The pairing is an old and legitimate device in opening literature. "Equal" soothes the buyer's fear — this line has not been refuted. "Dangerous for White" creates expectation — points will follow. Both can be true at once. Here neither is supported by anything: no engine evaluation, no results split, no named model game. A claim that asks me for money should at least offer me a table. Ordinary consumer-advertising standards agree: a claim that can be verified must be substantiated.

The scope is narrower still. This product answers exactly one White branch — 9.b4. In the Mar del Plata, White retains 9.Ne1 and 9.Nd2 as separate, living options at move nine, plus earlier escapes such as 8.Be3 and 8.Re1. A Black player adopting the Mar del Plata must find answers to all of them elsewhere. An antidote to the Bayonet and a Mar del Plata repertoire are two different things, and the cover makes them look alike. Selling one key for a house with four doors is not wrong, provided the box says which door it opens.

From four decades of notebooks I can say this much: at club and state level the Bayonet is very common, because it is the natural choice of an attacking player. At 2600+ the share held by 9.Ne1 remains large. That is the product's real market boundary — it is being sold to players who already play the Mar del Plata and who meet the Bayonet often. For a newcomer it is not a gateway. It is a supplement.

Using the Olympiad as a source is sensible. The biennial team championship brings more than 180 federations, includes much of the world's top hundred, and every game is public. No publisher finds cheaper or more prestigious raw material. But a selection bias hides inside it, and I watched it from the press seats at Bled in 2026. In a team match a draw is often enough for the team; the player plays to the result, takes no risk, and the lines produced that way are blunter than those from a Candidates or a pure round-robin. Olympiad games are good material; they are not evidence. They should be cross-checked against individual elite events. Thirty years later the notebook still knows the sound — the captain's half-whispered instruction in the team hall: "a draw is fine."

Then comes the costliest promise of all: "few concrete lines, strategy-first." It sounds like ease. On the board it inverts. Forced sequences narrow the room for error; where nothing is forced, every move demands a decision. For a player rated 1700–2026 that is exactly where advantage leaks away. The copy itself says these positions require long-term planning and tactical sharpness — more judgement, less comfort. A product that promises ease and delivers difficulty generates complaints in the first month, not the twelfth.

The cover is silent about the presenter. The series title carries 'Svitlana', but no surname, no title, no rating, and no statement of whether she has ever played 9...a5 in a serious classical game. For a repertoire product that is a serious gap. I went looking for a result and found a whole life — I first wrote that line in 2026, after Humpy's father sent me a letter saying he was glad I had got her middle-game choices right. That day I understood that chess analysis without a name stays unfinished. Here there is no name at all. And it is no accident that in an industry which gave a nineteen-year-old two paragraphs for becoming the youngest woman grandmaster in history, a woman presenter's rating was not thought worth printing on the box. In that Ahmedabad hall in 2026 there were forty spectators and one photographer; I filed nine hundred words, the desk ran a hundred and fifty and cut the games. The archive began with 150 words, but it was never about the word count. It was about whose name got printed and whose did not.

My argument turns here. The easy reading is that the problem is missing data. The harder reading is that the missing data is not the wound. The entire value of "few concrete lines" depends on the opponent not knowing the line — and a line with a DVD made about it is the first thing everyone knows. The antidote eats itself. From the day the series became popular, White acquired a concrete reason to prepare. This is not a permanent weapon. It is a surprise, and surprises last a season.

The second inversion is that the product's real asset is not the move 9...a5. The move is public already; nobody discovered it. The asset is the audience the copy identifies so cleanly: players who would rather think than memorise. That segment exists, and the big publishers' repertoire lists serve it poorly. Engines no longer fear closed positions; the fear stayed with the human — shortage of time, limits of memory. This should have been sold as a human's weapon, not an engine's antidote.

What to watch next is plain. Watch whether 9...a5 returns in a 2650+ classical game and whether Black holds — that would be real certification, not a press release. Watch whether the publisher corrects 'Attack' to 'Defence' in its catalogue; a one-word fix tells you whether the product is being maintained. And watch whether the model games are named — because a line built from public Olympiad games has no reason to hide its sources. The question is not whether 9...a5 can be played. The question is whether the next teenager who rebuilds his preparation on the strength of a box will find at least one number printed on it.