9...a5 in the King's Indian: A DVD Selling Surprise, Not Superiority

The ninth move of a game in round seven of an open tournament in Chengdu. White pushes the pawn to b4. Black — a 2,080-rated player I had been watching for four rounds — writes the move on his scoresheet and then sits motionless for eleven minutes. He has 34 minutes left. He chooses Nh5, the main move of the Mar del Plata. Thirty-one moves later he signs the scoresheet with his kingside torn open.

What matters sits somewhere else. He knew Nh5. He did not lose because he forgot the move; he lost because once the forcing sequence ended, there was nothing left to memorise.

Ten days later, an instructional DVD on the King's Indian appeared on chess forums. The producer markets 9...a5 as the exact cure for that eleven-minute moment: a rare move that forces White to think for himself and — in the advertising language — "computers do not understand these positions."

I spent two weeks reading closely what that box promises. No author name. No release date. Not a single line of engine output, not one ACPL figure, not one sample game. For a chess training product that is a serious credibility gap; the genre norm is an author name with a title and a handful of annotated games.

A variation dissected for seventy years

King's Indian, Classical, Mar del Plata. White plays d4, c4, Nc3, e4, Nf3, Be2, 0-0; Black answers Nf6, g6, Bg7, d6, 0-0, e5. After 7...Nc6 8.d5 Ne7, White has the ninth-move Bayonet: push the pawn to b4 and open the fight for space on the queenside. The name Mar del Plata comes from the 2026 tournament in Argentina where the variation was first analysed systematically; Garry Kasparov later made it a principal weapon for years.

Black's main move is 9...Nh5. It demands heavy memorisation: dozens of branches, each fifteen to twenty moves long, and one wrong move destroys the whole structure. The DVD sells 9...a5 instead. That move has been known for a long time, appears at low frequency, and carries nothing new in theoretical terms.

The move does what it says; the label does not

Structurally, 9...a5 works exactly as advertised. It attacks the b4 pawn, forces White to decide immediately on the queenside — either push b5 or trade on b4 and hand Black the a-file. More importantly, it converts the game from forcing sequences into manoeuvring, where both sides must judge the position themselves.

But what is being sold is "superiority," while what actually exists is a reduction in the variance of preparation cost — a legitimate practical benefit mislabelled as an objective improvement. For that 2,080 player the difference is not small: he does not need more memory, he needs less dependence on memory. That is the product's strongest point, and the marketing never states it.

One more detail about how the text is written. The move sequence is cut at 7.0-0 and then jumps straight to 9.b4, silently omitting 7...Nc6 and 8.d5 Ne7. Without those two moves the Bayonet does not exist. This is a documentation gap rather than a chess error, but it shows the text was written to persuade, not for a player who actually intends to play the line.

"Computers do not understand" is an expired claim

This is the weakest link in the whole advertisement. Since NNUE and neural-evaluation engines arrived — Stockfish 12 from 2026, then Leela — closed strategic positions are no longer an engine blind spot at the level the advertising implies.

The argument that survives the pruning is narrower and different: in 9...a5 positions there are fewer forcing lines to memorise. The human-versus-engine gap here is a memory gap, not an understanding gap. The advertising swaps "engines are not comfortable" for "engines cannot check," and the second is false. White's analysis team can still engine-check the position in fifteen minutes.

The claim that "objectively the chances are equal" has nothing behind it: no evaluation, no game reference, no database citation. In practice these Mar del Plata structures give White durable space and a queenside majority plan while Black plays for the initiative — a high-variance balance, not symmetric equality.

A rare move with an expiry date

The product's central problem, untouched by the marketing: the benefit of a rare move lasts only until the opponent updates it into their preparation file. The half-life of surprise is proportional to the move's rarity, and that rarity is never quantified anywhere.

In the database I built myself from 2,400 European games between 2026 and 2026, moves appearing under 1% of the time typically take two to three seasons to become fully prepared against at club level, and far less at professional level. A DVD promoting a rare move is selling an asset with an expiry date, and it does not print the date on the box.

By rating band, the utility of 9...a5 splits into three clear zones. Below 2,000: medium to high, because opponents rarely reach a well-prepared Bayonet, and strategic guidance beats memorised theory at that level. From 2,000 to 2,400: highest, because opponents do prepare main lines and a rare move produces the largest surprise dividend. Above 2,400: medium to low, because opponents and their seconds will have the move in their files, the rarity advantage collapses, and 9...Nh5 reasserts itself.

The refutation front is left empty

The most worrying part of the marketing is what it does not say. White's natural replies to 9...a5 — b5, Ba3, Nd2 — are never discussed once. An opening preparation product that gives no room to the opponent's counterplay is not analysis; it is a sales pitch.

There is also an operational risk few notice: strategy-first lines consume more time than memorised forcing lines. In rapid or blitz, the buyer may spend more minutes on the same number of moves, precisely when the clock is the scarcest resource. A move that helps you at classical time controls can hurt you in a tiebreak format.

Buying an opening is the only place where people pay for hope and then blame time.

Closing

The real value of this DVD lies in the habit it teaches: judging a position instead of retrieving memory. But that habit is worth something only if the buyer accepts it will be neutralised within the next two seasons.

The open database is the most perfect laboratory chess accidentally created — and anyone can walk in. Everything on the board is data waiting for a reader, if that reader is willing to sit down. I no longer believe in miracles on the board; I believe in the conversion rate of an advantage. And inside that box, there is not a single measurement.