Capablanca's 2026 and the Bayonet's Present: The Seam Between a Chess Product's Headline and Its Content

August 2026. The eighth Chess Olympiad opens in Buenos Aires on 24 August. While the tournament is still running, on 1 September, war is declared in Europe; some teams sail home mid-event, some players are stranded where they sit. In that tournament, Cuba's first board is occupied by José Raúl Capablanca, and it becomes his last major international performance. He will be dead by 2026, three years later.

That is the text I found at the top of a chess product page. Directly beneath it, without a transition, begins an advertisement for a training DVD on the King's Indian Defence, Mar del Plata Variation and the Bayonet Attack. Nowhere in the advertisement is Capablanca. Nowhere is Buenos Aires. Nowhere is 2026. Nowhere is any player at all. Not one sentence builds a bridge between the headline and the body.

I went looking for the score and found a page arguing with its own memory.

Only two readings survive. Either this is a bundled item lifted from a multi-headline newsletter page, and the wrong header got attached to product copy. Or the header is a deliberate evergreen frame, borrowing historical weight to stop a scrolling reader. In both cases this is not a technical question. It is a content-integrity flag. And that flag is the most interesting thing on the page.

Clear the board first. The largest thing missing from this headline is Elo. The rating system was introduced in 2026; no result from 2026 can be converted into a rating coordinate. What survives is the score, and it comes from 2026: the World Championship match against Alekhine, 34 games, a 6–3 margin, 25 draws. That is the reference point for Capablanca's late-career ceiling, and Buenos Aires 2026 is the last light after that ceiling.

The headline is not false at the level of fact. The problem is not the fact; it is the seam. A sentence that belongs to the night between Cuba's first board and a European war has been placed on top of 1.d4 Nf6 2.c4 g6, and nobody flinched.

From here, the part that can actually be checked.

The Mar del Plata Variation of the King's Indian takes its name from the 2026 Mar del Plata tournament, where the structure established its identity. The base move order runs 1.d4 Nf6 2.c4 g6 3.Nc3 Bg7 4.e4 d6 5.Nf3 0-0 6.Be2 e5 7.0-0 Nc6 8.d5 Ne7. Black's entire bet rests on playing ...f5, pressing White's e-pawn, and collecting dynamic compensation on the kingside for a cramped position. White's historical answer is to keep the centre locked and open a second front on the queenside.

The Bayonet Attack, 9.b4, is the most direct declaration of that second front. White says: attack my king, and I will open your queenside. After that, Black has two distinct philosophies, and this DVD is selling the choice between them.

First, 9...Nh5, the main line. It is forcing, it contains concrete moves, and it is among the most deeply engine-mapped branches in the entire King's Indian. Second, 9...a5, a structural move. It puts a question mark under White's b4-pawn, slows the queenside expansion, and buys Black a tempo of preparation for ...f5.

The real insight here is that this is not a choice of strength but a trade between memory and judgment. 9...Nh5 demands less understanding and more recall; 9...a5 demands less recall and more positional judgment. Both are true on their own ledger, and that is the most honest thing in the product.

The second honest thing is preparation economics. Against a well-prepared White, the odds of deep files on 9...Nh5 are very high, because that line has been scouted for decades. The odds of equally deep files on 9...a5 are lower. The move's real strength therefore sits not on the board but in the density of the opponent's preparation. That is a defensible argument, and it could have been made without any mystique.

Where the advertisement slips is here: the claim that "computers often do not know how to handle the arising complex strategic positions." That sentence was true, up to a date. With the arrival of neural engines — NNUE-class and Leela-style networks, roughly between 2026 and 2026 — engines began evaluating closed King's Indian structures and long-term positional sacrifices far more reliably. "Engines get lost here" is no longer a technical fact; it is marketing framing.

In the twelve years I have spent watching King's Indian games from beside the board, one pattern keeps returning: at club level this branch beats players who work from recall, and loses to players who work from understanding. Which means the product is selling precisely the skill its buyer segment most commonly lacks — and that contradiction appears nowhere in the advertisement.

The third risk is the time control. A low-forcing, structural position means every move requires a decision, and decisions are paid for on the clock. Amateur players shed most of their points in fast time controls — exactly where the premise of "think for yourself" is weakest. The whole promise breaks at the speed its buyer actually plays.

The fourth risk is the fastest one. The practical value of this line begins to decay the day it enters an opponent's files. No unpublished branch is a permanent asset; it is borrowed capital.

Now the part where I disagree most sharply.

The consensus, stated in its strongest form, is this: the page is junk. A wrong header, an advertisement, and no argument in between. Ignore it, because there is nothing in it worth analysing. I respect that case — it is clean, it is checkable, and most editors would sign it.

My disagreement is that the mismatch is itself the data. Some games are not played on the field; they are played in the space between headlines. A sentence from a 2026 Olympiad and the blurb of a modern DVD stand on the same page with no seam, and no reader files a complaint. That is the real finding: the commercial economy of chess treats history as decoration. Capablanca is a brand weight, not a causal agent. His 2026 presence opens no argument; it merely lends the page dignity.

And a second disagreement, which matters more to me: the expiry date of the "engines don't understand" claim is the actual news. Behind that sentence lies a genuine history — through the 1990s and the early 2000s, engines really were shaky in closed King's Indian positions. The moment that truth ended, every marketing sentence resting on it ended too. The product is still selling the old date, and the market is buying it without a question.

I trust the pattern that survives the noise, not the noise that survives the pattern. Here the pattern is plain: a legitimate strategic argument, an expired technical claim, and a historical headline with no relation to its own content.

So the question is not whether 9...a5 is a good move. The question is what remains for Black's dynamic compensation when the buyer sits down against a prepared White who has kept the queenside closed and chosen to play on space. "Few concrete lines" is one face of the coin. The other face is that the same feature lowers Black's winning chances against a strategically competent opponent.

Capablanca left his last light in 2026, filed with a date attached. The DVD's claim is running without one. The next time a product tells you engines do not understand a position, ask one question: until which year did they not? If no answer arrives, you are not reading chess. You are reading an advertisement.