9...a5 Against the Bayonet: An Old Answer and a Data Gap

On an autumn afternoon in 2026, the tournament bulletin was thin enough for the desk lamp to shine through both sides. The front page carried a headline about round seven of an open in Samarkand, Uzbekistan: the hosts were still leading, and the gap had widened. The back page held a small advertisement for a booklet on opening theory, and someone had circled four characters in pencil: 9...a5.

The person who circled it left no name. I like to think it was a seventeen-year-old boy, and that when a seventeen-year-old prodigy does not yet know he is about to become a legend, the only thing he does is draw a pencil ring around a move.

I have kept that bulletin in a drawer. Years later, when someone asks me why I believe chess is the sport of small memories, it is the first thing I think of.

To understand why those four characters deserved a circle, you have to go back to the position that produced them.

The King's Indian Defence is a system in which Black willingly concedes the centre and then comes back to attack it. In the Mar del Plata branch, White pushes his pawn to d5 and locks the centre, while Black gathers his pieces on the kingside. The two sides fight on different fields, and whoever arrives first wins.

White has a way of breaking the structure: the move 9.b4. Analysts call it the Bayonet Attack. White drives the bayonet into the queenside, demanding space, files and every small white square. There, every step backwards is a step lost.

The familiar answer is numbered 9...Nh5. The black knight jumps to the rim of the board, eyes the f4 square and clears the way for the f5 pawn. That is the main road, dissected move by move. It is also why the road is tiring: after 9...Nh5 the game is nearly forced. Black has to remember a long, tight chain of variations with no room for intuition; one move off and the whole wing is gone.

The booklet in that advertisement proposed another road: 9...a5.

The move does not attack, does not threaten mate, does not produce a beautiful combination. It touches White's b4 pawn and asks one plain question: do you take, or do you push?

The real value of 9...a5 lies in the time it buys by cutting down what must be memorised — trading depth for breadth. If White takes, the a-file opens for Black's rook. If White pushes to b5, the queenside closes and Black is free to do his work on the other side of the board. Both branches lead out of the book, into ground where understanding matters more than memory.

That is why club players like it. It is also why it rarely appears on the big boards.

From my hours of note-taking in tournament halls, one thing stood out: amateurs learn openings in the opposite way to grandmasters. They do not study to win on move fifteen. They study not to lose on move fifteen. A move like 9...a5 answers exactly that fear, and that is where its appeal comes from — the appeal of a small path that runs around the crowded market.

But here the booklet stops exactly where it most needs to keep talking.

It states that in the position after 9...a5, objectively the chances are equal. It includes no sample game. No table of wins and draws. Not one line saying how many games were tried, who had which colour, how they ended. A single claim standing alone in the middle of the page.

My trade taught me that a claim with no data behind it can still be true. It simply cannot be checked. In chess, those two things are very far apart.

Then comes the part where I want to linger.

The front and the back of that bulletin tell two stories that have nothing to do with each other: on one side, round seven of a tournament in Samarkand, where the Uzbeks are leading; on the other, an invitation to buy a booklet about 9...a5. A reader skimming past would assume the two belong together. They do not belong together. A bulletin is only paper; the position was signed in sweat.

I do not raise this to nitpick an advertisement. I raise it because it reminds me of a bad habit among readers: believing that whatever is printed close together is connected. Paper knows nothing about its own contents.

About 9...a5, there are things the advertisement does not say.

A surprise move is only a surprise once. Once it is common in club chess, it stops startling anyone and becomes the opponent's homework.

Equality does not mean ease. An equal but unfamiliar position still forces the player to decide every move himself; it does not think on anyone's behalf.

And 9...a5 is not a solution to the Bayonet. It is a calculated postponement. White still holds the bayonet; Black has only turned its direction.

The most romantic mistake of mine was believing that chess pieces can write poetry. I still believe it. I have only learned that the poetry of pieces is not in the good moves, but in the moves people hesitate over longest before playing them.

Samarkand, round seven. Uzbekistan leading. And in Tashkent, not far away, the country had just lost the first of its sons to be awarded the international grandmaster title — Georgy Agzamov, who died this summer in a climbing accident. He never played 9...a5. Perhaps he never needed to.

An empty stadium still keeps the applause of the dead. So does a tournament hall. The moves people never got to play stay on the board, waiting for someone to pick them up.

So what I carry away is not the question of whether 9...a5 is good or bad. What I carry away is this: how much is a player willing to learn in exchange for freedom? Whoever pays in memory buys safety. Whoever pays in understanding buys peace of mind. Few get both.