A 2026 Gold Medal, a 2026 DVD: Capablanca's Name Is Selling a King's Indian Repertoire

At the Buenos Aires Olympiad he played first board for Cuba, scored eight and a half from eleven, and took individual gold. It was September 2026. War headlines were arriving from Europe, and the hall at the Park Central Hotel was watching the last generation that ever saw Capablanca play live. This is the same player from whom Alekhine took the world title in 2026 with six wins, three losses and twenty-five draws.

Now scroll under that headline. There is no Capablanca story below it. Below it is an advertisement for a training DVD on the Mar del Plata branch of the King's Indian Defence, recommending 9...a5 for Black. Three paragraphs, and not one sentence mentions Capablanca, Buenos Aires, or 2026. There are two readings of this mess, and both are uncomfortable: either a multi-headline newsletter page attached the wrong header to a product block, or Capablanca's name was planted there to stop the thumb. In three decades of watching chess media, I have learned that this kind of mess is usually a habit rather than an accident.

The technical part matters, because that is where the real argument sits. 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 is the Classical King's Indian, and after 7...Nc6 it becomes the Mar del Plata, named for the 2026 Argentine tournament where it was first played at scale. Black's plan is blunt: ...f5, ...f4, ...g5, a pawn storm on the kingside, mate threats against White's king. White's answer is space on the queenside: c4-c5, b4-b5, and a clamp on Black's centre. It is chess's oldest trade — Black buys time, White buys ground.

9.b4 is the Bayonet Attack. Kramnik and Topalov pushed it into the mainstream in the 1990s, and it has since filtered down to club level. Its purpose is to slow the storm and clear the ground for c5. Black has two routes. 9...Nh5 is the main line: knight toward f4, then ...f5 and ...g5, and a forcing sequence over the next ten moves that every database has chewed through thousands of times. The alternative is 9...a5: an immediate question to b4, space on the queenside, and the same old King's Indian plan continuing. It is not a novelty. It is legitimate theory. The only thing that is marketing rather than theory is calling it an antidote.

Here is the first thing no advertisement will tell you: the real value of 9...a5 is not equality, it is unpreparedness. A well-prepared White will carry thirty moves of engine-checked lines against 9...Nh5, because that line has been played thousands of times in every database on earth. Against 9...a5 he will have two or three routine replies and then he has to think. That is the genuine difference in preparation economics, and it is entirely valid.

Second, and this is what the product is actually selling: the DVD is not selling you moves, it is selling you a choice between memorising and understanding. 9...Nh5 carries more forcing content, which means a heavier memory load and fewer decisions. 9...a5 carries less forcing content, which means fewer lines to memorise and a positional judgement call at every turn. Most club players will pick the second option, and it is not a wrong choice. For someone who can sit and think for two hours in a Sunday league game, less theory means less fear.

Third, and this is where the advertisement shoots itself: the page claims computers often cannot handle the complex strategic positions that arise. That sentence was true until about 2026. After NNUE-class engines and neural networks (Stockfish 12, September 2026; Leela Zero, 2026), the claim that engines get lost in these positions became marketing rather than technical fact. Anyone in 2026 building a product on engine weakness in closed King's Indian structures is selling old tape. Engines do not hand you a plan in these positions; they hand you a score. That distinction is real, and the advertisement blurs it on purpose.

Fourth, and this is a journalism question: the page offers no engine match rate, no ACPL, no database win-draw percentage, and not one cited game. Without a database score and without engine verification, the word equality belongs to advertising, not analysis. In 2026, at ChessBase India's makeshift studio at the National Premier in Delhi, I timed 23 minutes of dead air in a single broadcast and cut a three-minute video about it. That clip drew 40,000 views, more than my previous six columns combined. Since that day I test every claim on a public scoreboard first. Here the scoreboard is simple: pull the database score for 9...a5 and publish it. A product that will not publish it is asking for trust instead of supplying evidence.

Fifth, in transfer-window language: an opening line's release clause is its novelty window, and its wage bill is the hours you spend maintaining it. The day 9...a5 lands in an opponent's file, its value collapses toward zero, because the entire advantage rests on the assumption that White has not prepared. That is not a moral failing; it is the structure of the repertoire business. The money is equally legible. It goes to the publisher, not the player. Even in the franchise money swirling around Gukesh's 2026 Singapore crown and the Global Chess League, a share rarely reaches the person who spends the hours building lines. The economics of an opening DVD is a small, honest map of that same system.

Sixth, and this matters most: the DVD is weakest exactly where amateurs lose the most points — fast time controls. In 15+10 rapid and 3+2 blitz, a judgement-heavy position is dangerous, because the right move has to be the fast move. The whole 9...a5 model is think-for-yourself, and the clock does not permit thinking. It can work in a 90-minute league game. It will invert in an online blitz arena. A product that sells understanding owes the buyer a note on which clock that understanding ripens under and which clock rots it.

I learned that lesson during the 2026 Online Olympiad, when India and Russia were declared joint winners after server failures in the final. Most coverage was selling a feel-good story; I pulled the match log, counted the disconnects, and wrote that India's first global team title arrived with an asterisk made of Wi-Fi. Federation officials replied publicly for the first time that week. The lesson holds: one uncomfortable fact per segment, and the rest is noise.

Now the part where I cut myself. I cannot fully rule out the deliberate-clickbait reading, but I lean toward carelessness, because the product page reads like aggressive selling rather than deception. Second, the computers-don't-understand claim is not entirely empty. In closed King's Indian structures where Black offers long-term positional sacrifice, engine evaluation can stay shallow and human planning is still required. Rephrase it away from marketing and into the sentence engines give you a score, not a route, and it stands in 2026. The problem is not the claim. It is the packaging.

Third, 9...a5 may genuinely convert at club level. Between 1600 and 2026, players do not memorise lines; they play positions. Face an opponent carrying twenty-five moves of 9...Nh5 preparation and 9...a5 wipes his file out, even if the engine says you are slightly worse. In opening theory, slightly worse but unprepared routinely scores better than equal but known. Kolkata didn't shake hands, remember? At the 2026 Tata Steel Chess India, Anand and Nakamura were trading 15+10 games while the hall sat half-full and the broadcast treated silence as reverence. I wrote then that a city of fourteen million had no street poster for the event, and that this was a marketing crime rather than a chess problem. Same here. The content of this DVD may be good. Its headline made it suspect.

My prediction is testable. If 9...a5 starts appearing in open tournaments and online leagues over the next twelve months, its practical score will drift toward the 9...Nh5 baseline and White's files will fill up. Check the database score yourself; I do not need to tell you. The question is what a repertoire sells on the day the opponent has prepared, when its entire pitch was that he had not. And the gold Capablanca won in 2026 was not minted for a DVD cover. The day chess marketing starts reading its own sport, headlines like this one stop getting written.