The Mar del Plata Branch: The 9...a5 Move and the Promise of a Battle Without Memorization

On the same page, two chess games sit side by side without ever knowing each other. The headline speaks of round two of a team event in Samarkand, where the strongest teams all won after close calls. The body text speaks of a chess DVD teaching how to meet the Bayonet Attack with the ninth move 9...a5 in the King's Indian Defence. Two texts, two worlds, stitched together by an editorial error nobody bothered to fix.

I read those fourteen information points three times. Not a single tournament, not a single player, not a single concrete game appeared beneath the headline. Only a product, a promise, a label.

And I remembered a winter afternoon in Shanghai, sitting in the fourth row of a small chess club on the second floor of an old apartment block. In front of me sat a fifteen-year-old boy studying the board. His ninth move was b4. Opposite him, a grey-haired man who had played a national championship in his youth did not think long. He pushed his a-pawn two squares.

The boy looked up. The old man did not explain. He said only one sentence, which I translated roughly: "Now you have to think for yourself."

I wrote that sentence in my notebook. Years later, reading a document about a DVD advertising this very move, I still had the same page.

When the board stops being a board

The King's Indian Defence is one of the strangest systems chess produced in the twentieth century. Black lets White take the entire centre, lets White build a massive pawn block, then waits patiently for the moment to counterattack. The name "Mar del Plata" attaches to a tournament held in that Argentine coastal city in 2026, where the structure with ...Nc6 and ...e5 was first shown at the highest level and quickly became one of the most complex variations in all of opening theory.

That structure later became a sequence of parallel counterattacks. Black pushes ...f5, White pushes c5, Black pushes ...f4, White pushes b5, and the two sides hurl themselves at each other like armies marching around the flank to strike the enemy's back. There is no room for hesitation. There is no room for neutral moves. One small error on move twenty-one can lead to defeat on move thirty-five, and sometimes the player never knows when the loss actually began.

For many decades this branch was one of the most thoroughly analysed battlegrounds. The great names of world chess all passed through it; some loved it, some hated it, but none dared dismiss it. The price of choosing it was time. The reward was surprise.

That is the context into which the ninth move ...a5 stepped.

But before speaking of ...a5, I need to speak of what it was designed to fight.

The Bayonet Attack: White's blade

The ninth move b4 in the King's Indian, Mar del Plata branch, carries a memorable nickname: the Bayonet Attack. The bayonet. The name is no idle metaphor. White pushes the b-pawn to the fourth rank to gain space on the queenside, prepares to push b5 and c5, opens a diagonal for the bishop to a3, and turns the queenside into a grinding machine.

That is a modernising approach. At a time when the old Mar del Plata patterns had been analysed to the bottom, the Bayonet Attack emerged as a way for White to avoid the worn paths and force the opponent to solve new problems. It does not attack the king first. It attacks the strategic order of the position.

In a sense, the Bayonet Attack is a product of the computer age. It demands extremely precise move sequences, defensive ideas calculated dozens of moves deep, and an understanding of pawn structures that only diligent students possess. Black does not have many pleasant choices. The main reply, considered standard for years, is ...Nh5.

When Black plays ...Nh5, the game is immediately pulled onto a specific track. Black threatens ...f5, a knight jump to f4, and tearing open White's kingside. White must react quickly, usually pushing c5 and opening the bishop's diagonal. The variations become forced, linear, and memorisable in stages. This is the crux: ...Nh5 makes the variations concrete, and concreteness always favours the better-prepared side.

In the world of professional players, memorisation is a weapon. In the world of amateur players, memorisation is a burden.

That is precisely the gap that 9...a5 aims at.

A gift with no instructions attached

The DVD described in the document I read presents 9...a5 as an "antidote" to the Bayonet Attack. That framing is very familiar to me. In more than thirty years watching the opening-theory market, I have seen hundreds of such antidotes. Every year another move is called the answer to a popular system. Every year another headline promises that your problem has been solved.

The product's marketing rests on a very clear comparison. The move ...Nh5 opens the position, forces the lines onto a predetermined track, and pushes the game back toward computer analysis — where the better-prepared side wins. The move ...a5 leads to sound positions with very few concrete lines, where the player must think for himself instead of consulting a database.

The real value of this promise lies in psychology, not in theory: 9...a5 is not objectively stronger than 9...Nh5, it merely shifts the game from a memory contest to a long-term calculation contest.

That is an appealing offer, and I understand why it appeals. Club-level players spend most of their free time memorising sequences they will forget in two weeks. The feeling of being dragged into a line the opponent knows better is the worst feeling in chess. Anyone promising release from that feeling has a ready market.

But does the promise come with evidence?

The document I read supplies not a single figure on average move accuracy. It cites no sample games. There is no author name, no grandmaster title, no personal record in this branch. There are no win rates, draw rates, or any other quantitative indicator.

An opening product without data is a product selling belief.

In my trade, belief is good material for an article and bad material for a purchasing decision. I learned this the expensive way, on a summer afternoon when I misnamed a player on live broadcast and was reminded of it for months. That mistake taught me that what we say with the greatest confidence is often what we have verified the least.

An old move dressed as an invention

There is one detail in the document I consider more important than all the rest, and it is mentioned too lightly. That is the admission that 9...a5 is a known sideline, an alternative that has existed in theory rather than a new invention.

This changes the whole evaluation of the product.

When a known move is sold as an invention, the buyer is paying for three things: a reorganisation of old knowledge, an easy-to-learn classification system, and the reassurance that comes with having bought an answer. None of these three is worthless. But all three can be found free of charge in open databases, where millions of games are stored and searched with a single click.

The buyer is not paying for the move. They are paying for the story.

Well-prepared White players do not stand still either. Any move ever analysed, even modestly, has a ready answer in chess's shared archive. An antidote is valuable only until it is used enough, and in the modern chess world that threshold is very low. A few games at the top level, a few forum analyses, and the psychological edge evaporates.

I do not want to be understood as denying the value of 9...a5. It is a legitimate move. It leads to positions where Black can hold. But legitimate does not mean advantageous, and holding does not mean easy.

An argument past its expiry date

One sentence in the document made me pause longest: the claim that computers often do not know how to handle these complex strategic positions.

If that sentence was written in 2026, it was reasonable. If it was written in 2026, it is a relic.

Before the neural-network era, chess engines evaluated positions relatively crudely. They excelled at calculating forced sequences and struggled to assess positions whose value only emerges twenty moves later. The King's Indian, with its closed structures and slow manoeuvres, was once territory where humans could outplay machines in orientation.

The arrival of neural-network engines erased that boundary. Modern machines assess long-term positions better than most grandmasters, even in the most complex closed structures. A closed Mar del Plata position is no longer a machine-free zone. It is merely a position the machine evaluates slightly differently.

The "computers don't understand" argument therefore expired years ago. Its persistence in marketing copy reveals something about the market: old arguments can outlive the truth they describe, as long as they still sell.

Buyers should read this promise as a marketing claim, not a technical assessment, because it rests on a premise proven obsolete — and once the premise collapses, the whole argument built on it wobbles.

That is why I remember the old man's sentence in the Shanghai club. "Now you have to think for yourself." That advice had value because it was true. It did not promise thinking would be easy. It only said thinking was the boy's job.

The document I read, by contrast, promised that the thinking had been done for you.

An empty hall and unarchived memories

An empty stadium still holds the applause of the dead. I first wrote that sentence in the summer of 2026, when football returned after the pandemic and I took a recorder into a silent tournament hall, hearing clearly the click of pieces on the board and the referee's sigh. What I had assumed was the sound of emptiness was in fact the sound of memory trying not to be erased.

The opening-theory market works the same way.

Every move ever played, ever analysed, ever debated on forums leaves a trace. That trace does not vanish when a new product claims to reinvent it. It sits there, in databases, in old magazines, in the memory of people who have spent a lifetime following the branch. What is called "new" is often just the old rearranged into a more sellable order.

I have nothing against rearrangement. Rearrangement is the work of writers, teachers, guides to beginners. But an honest arranger must say clearly what is being arranged.

Here, the document I read failed notably, not in its chess content, but in placing itself in the wrong frame. A headline about a team event in Samarkand sat above an article about an opening DVD. Two texts, one page. No tournament. No team. No round two.

In the news trade, that error has a name. It is called broken metadata.

In a very concrete sense, it is worse than a technical slip. It means anyone who reads that page and cites it carries forward a false belief: that there was a tournament in Samarkand, that there were strong teams, that there were close calls. Those false beliefs will live on in follow-up articles, in social posts, in automated summaries. They will reproduce.

What I saw that the document did not

In more than thirty years reading and writing about chess, I have learned one thing about opening products: their best content is not in what they claim, but in what they inadvertently reveal.

This product reveals three things.

First, it shows that the amateur market is tired of memorisation. This is a real signal, and it is bigger than one DVD. Club players, junior coaches, amateurs with day jobs — all are looking for a way out from an ever-heavier opening-theory burden. Any product that touches that fatigue has a place.

Second, it shows that the distinction between theory for machines and theory for humans has become an important sales axis. This is a notable development in chess culture. Twenty years ago, people sold accuracy. Now they sell comfort.

Third, it shows that even in a field with perfect open databases like chess, people will still pay for a storyteller. Chess can retrieve every game ever played with a single click, and there is still a market for guides.

None of these three things appears in the fourteen information points I read. They live in the gaps between the lines.

And it is those gaps I wanted to write about.

The golden second of an ordinary move

When a seventeen-year-old prodigy does not yet know he is about to become a legend, he tends to play moves nobody notices. I have seen this many times. A boy places a knight on a square that looks meaningless. A girl pushes a pawn one square no one is contesting. Nobody applauds. Nobody records it. Ten years later, people open the database and realise that move changed the understanding of an entire system.

The move 9...a5 is not a golden second. It is an ordinary move that has existed in theory for decades, played by players history does not remember, in games nobody bothered to save. That ordinariness is its most interesting feature.

Modern chess is built from millions of ordinary moves. The greatest inventions are just ordinary moves placed in a context never seen before. So when a product sells us an ordinary move as though it were an invention, the point is not the move but us — how much we need a story to be told.

My most romantic mistake was believing the ball knows how to write poetry. I wrote that about football, but it holds for chess at a deeper level. Pieces do not write poetry. Pieces do exactly what geometry permits. The poetry lies in the viewer, in the teller, in the one who chooses to call an ordinary move a golden second.

That does not disappoint me. It only makes me more careful about who is telling the story and for whose benefit.

Those who are never named

In the story of an opening move there is always one absent character: the first player who tried it.

Nobody knows who first pushed the a-pawn two squares on move nine in a Mar del Plata game. Perhaps an unknown player at a small tournament. Perhaps a correspondence player in an empty room. Perhaps a fourteen-year-old somewhere, looking for a way out of a line he had memorised but never understood.

That person tried a move out of curiosity, or despair, or a reason nobody remembers. He lost. Or won. The game was not saved. Nobody wrote about it. Nobody paid for it.

And decades later, a DVD sells for the price of a restaurant dinner, advertising that move as a solution.

I write this not to belittle the product. I write it to place it in its proper spot within a much longer current. Chess is the only sport where everything ever played can be found, and also the only sport where most of what was ever played has been forgotten.

We remember the winners. We remember the inventions. We do not remember the first person who tried an ordinary move and got nothing but an unnamed game in a vast database.

If there is one thing worth writing about 9...a5, it is that it belongs to those unnamed people.

The line between studying and buying

Club-level players face a problem professionals do not: very little time and very many options.

A DVD promising to save time is a DVD that understands the buyer's pain. But saving time is not the same as absorbing knowledge. In chess there is a large difference between knowing a move and understanding a position.

Buyers of this product will learn some move sequences. They will know that after 9...a5, White usually continues with certain moves, and Black usually replies in certain ways. That knowledge has practical value. It will help them in some club games.

But when the opponent steps outside those sequences — and that will happen, because the opponent did not read the same book — the player will have to think for himself. And that is exactly what the product promised to spare him.

This is the paradox of every opening product that sells "no memorisation". To use a no-memorisation system well, the player must understand it more deeply than a memorisation system. You can survive concrete variations without understanding anything. You cannot survive open strategic positions without understanding anything, because those positions demand judgement, and judgement is not sold on discs.

The value of an opening product lies not in helping you avoid thinking, but in giving you a frame of thought good enough to think faster than your opponent.

If this product does that, it is worth the money. If it only does the first, it is worth no more than an afternoon of free database search.

A little about Samarkand

I cannot end this piece without addressing the headline.

Samarkand is a city in Uzbekistan, one of Asia's fastest-growing chess centres in recent years. If there really was a team event there, it deserves its own article, its own analysis, and its own stories about the strong teams that won after close calls.

But there is no data about that event in the document I read. No team names. No scores. No standings. No games.

I write about this absence because it matters more than a typo. In a world where content is aggregated automatically, where headlines are assigned by algorithm and bodies are assembled from fragments, such mismatches will occur more often, not less. Readers will find it harder and harder to tell an article from a patchwork of several.

That is why I keep my habit of visiting chess clubs every Wednesday, sitting in the fourth row, noting what I see with my own eyes. No algorithm can assign a wrong headline to an afternoon like that.

And that is also why I still believe chess reportage has a future. When anyone can have every game with one click, the writer's only remaining value is in what cannot be looked up: the look on a boy's face when he realises he is in a position he has never seen, the sigh of an old man who knows he has erred, and the silence of a room when an ordinary move is placed on the board and nobody there realises it will be discussed years later.

What I still do not know

I do not yet know the name of the DVD's author. I do not know whether he is a grandmaster specialising in the King's Indian or simply a strong player with a good database library. I do not know whether the variations come with engine evaluations, and if so, whether they have been updated since the neural-network era.

Nor do I know whether 9...a5 is currently being used at the highest level. That is an answerable question, and the answer lies in open databases. If the move appears in classical games by players rated above 2700, that will be a notable signal. If it is entirely absent, that is a signal too.

After three decades holding a pen, I realise I have never written about sport. I write about people trying to understand something within a limited span of time. Chess is only the purest form of that, because in chess nothing can hide the fact that we never have enough time to understand everything.

A DVD promising to shorten that time is a very human promise. I do not mock it. I only want the buyer to know what he is buying: a shorter road, not a free one.

What remains after the box is closed

I write about chess to find the self I was twenty years ago. That person used to sit before a board with an opening book open, trying to memorise every move and always failing at move seventeen. He used to believe that if he read enough, every game would become easy.

He was wrong, but wrong beautifully.

And if there is one thing I want to tell my readers about 9...a5, it is this: the move matters less than your reason for choosing it. If you choose it because you are tired of memorising and want a position where you must think, you are heading the right way. If you choose it because a DVD says it will make your opponent dangerous quickly, you are buying a belief, and belief is worth nothing in a game that lasts sixty moves.

The board does not care who sold you the idea. The board only cares whether you understand it.

And if one day, in an empty hall, a young player pushes the a-pawn two squares on move nine without knowing that a DVD is trying to sell that move as an invention, then perhaps that game is the one worth recording. He bought nothing. He simply thought.

Games like that are not saved to databases. They remain only in the memory of whoever sat there.

And perhaps, after everything, that is the only database money cannot buy.