9...a5 in the King's Indian: an escape from the Bayonet and a lesson from a mismatched headline

Nha Trang, a June evening. A chess cafe on Nguyen Thien Thuat Street, the third table from the door, a young man of about twenty playing Black in a King's Indian. The clock on his phone drops from 41 minutes to 14 for exactly one move — move nine. The board stays flat: no piece lost, no position lost. Yet he sits motionless as if working out an arithmetic problem. I am behind him and I know precisely what is happening: he is trying to remember. A sequence someone taught him, and the sequence just snapped at the most important point.

The scene repeats often enough that I treat it as a technical problem rather than a private memory failure. It is also why I read very carefully a promotional text for a King's Indian opening DVD that proposes 9...a5 as an antidote to the Bayonet Attack.

Background for newcomers. The King's Indian is a hypermodern system: Black willingly grants White a large centre, then attacks that very centre with ...e5, ...c5 and flank thrusts. The Mar del Plata Variation — named after the 2026 tournament in Argentina — is its most violent version: after 7...Nc6 and usually 8.d5 Ne7, the two sides race on opposite wings. Black rolls out ...f5-f4, ...g5-g4, ...h5; White breaks through on the queenside with b4 and c5. It is one of the few structures where both sides can be objectively losing at the same time.

The Bayonet Attack, 9.b4, is White's answer: seize queenside space before Black closes the net, prepare b5 and a knight to c4, and take away squares Black needs. Since the early 1990s it has been the weapon of players who want to avoid long theoretical arguments — ironically, it created a mountain of new theory itself.

Black has two directions. 9...Nh5 is the traditional main line: the knight goes to the rim to support ...f5 and aim at f4, the position opens up, forced sequences dominate, and the computer becomes referee again. 9...a5 is the DVD's recommendation: hit b4 directly, prepare ...a4 or ...axb4, and force White to declare intentions on the queenside before Black commits on the kingside.

The first thing worth noting is how the author limits himself. The text states plainly that 9...a5 leads to a position that is objectively equal. An opening product willing to say that is rare; most would shout that this is a refutation. What is being sold is practical difficulty for the opponent, not an objective advantage — and that honesty is the product's greatest strength.

The second point is the repertoire argument. The author promises no novelty and no knockout. He draws a route from the tactical world into the strategic world: fewer forced lines, less memorisation, more room to think. For a club player that is an appealing offer, because their real fear is usually losing to forgotten theory at move fourteen, not losing to poor calculation.

The third point is where I want to linger: the spatial piece on a chessboard is not the square a piece occupies, but the square it will occupy three moves from now. 9...a5 works because it does not argue with White about attacking speed; it argues about the right to be left alone. White wants b5, wants the knight on c4, wants a quiet queenside while pieces swing to the kingside. Black plays ...a5, and the queenside suddenly makes noise. Black's pressing problem is not one of speed but of redrawing the map — breaking the silence exactly where White needs it.

Based on my experience tracking games at domestic youth events and thousands of games on online servers, a fairly durable rule holds: Vietnamese players rated 1800-2200 usually lose because the idea arrives after the clock has run out, not because the idea is missing. An opening system that reduces memorisation frees exactly the mental capacity they lack. Esports showed me that a tactical space needs no pitch to make the heart race — and an online board is the clearest example. In that sense, 9...a5 has a reason to exist independent of whether it is strong.

But the weak parts deserve plain speech. The text claims that computers often cannot handle these complex strategic positions. That was true in an earlier engine era and is false now. Modern engines handle Mar del Plata structures very well, often producing the plan within seconds. The defensible version is narrower: humans, without guidance, will get lost in that strategic maze — perfectly reasonable, but a pedagogical argument rather than a technical one.

The second problem is heavier: there is no data at all. The source contains 13 information points, and none of them include ACPL, engine match rate, database win-draw-loss figures, or reference games. In other words, every technical assertion sits at the status of an unverified hypothesis. A researcher has to read it that way rather than nodding at claims offered without samples.

The third problem is a time-control paradox. Fewer forced lines is an advantage with 90 minutes for 40 moves and a liability with three minutes for the whole game. Strategic positions reward thinking; blitz rewards nobody's thinking. The product sells an advantage that depends on the format, and never says so. A blitz buyer pays for a gift he cannot use.

And finally, the headline. The document I read carries the line "A farewell triumph: Capablanca wins gold in Buenos Aires 2026", while the entire body is an advertisement for a King's Indian DVD. Capablanca and Buenos Aires 2026 appear in none of the 13 information points. A glamorous historical headline attached to purely technical content. For readers the lesson is small but expensive: when the opening and the body do not match, trust the body.

9...a5 deserves a place in any King's Indian player's repertoire as a surprise weapon, not a pillar. Before trusting it, verify it yourself with an engine and your own database — the only way to turn an advertisement into a decision. Technique is only complete when it is told in a language players dare to believe. And that belief, at the board as on the page, comes only from what we have measured ourselves.