9...a5 and the Spatial Map: An Old Move Returns Against the Bayonet Attack

Samarkand, One in the Morning

I opened the file at one in the morning, Chiang Mai time. The headline described round two of a team event in Samarkand, reporting that the top teams all won despite close calls. The body of the file discussed a chess DVD, specifically a ninth move for Black in the King's Indian Defence. The two texts sat beside each other without sharing a single detail.

I could have called it a splice error, closed the file, and gone to sleep. I did not close it.

On the third reading I understood that I was holding two halves of the same question: who controls the space on the board, and who merely believes they do. One half was a results report, where everything is recorded as a score. The other was a sales pitch for a move, where everything is recorded as a belief. I have spent twenty-three years reading both kinds of documents, and the second kind usually carries more information.

Not the goal, but the gap before the goal appeared. On a chessboard, that gap has a name. It has coordinates. It has a history. And sometimes it returns after twenty years, forgotten in a corner of a database.

This article is about the move 9...a5. It is also about why an old move can be sold as a new invention, and why that is worth our attention rather than our irritation.

---

The King's Indian: The Position of Calculated Concession

The King's Indian begins with an act that looks like an error. Black does not occupy the centre. Black concedes it. While White builds a pawn mass in the middle of the board, Black retreats, places a bishop on a long diagonal, and waits.

This is the hypermodern school, and it is one of the most elegant ideas chess produced in the twentieth century. White is entitled to believe he is better. Black knows White will have to defend that central mass with the very pieces that built it.

The familiar sequence: 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. White has pushed the d-pawn to d5, locking the centre. Black has retreated the knight to e7, preparing ...f5. The position closes. And in closed positions, people assume nothing happens.

Wrong. In closed positions everything happens more slowly, and more heavily. An abandoned square on the queenside is not compensated by a square on the kingside. The spatial map becomes crowded, and in crowded space, error carries more weight.

The spatial map. I have used that phrase since 2026, when I was a mid-level coaching staff member at a Thai League 2 club, writing a blog in my spare time. It began as a way to talk about football. It ended as a way to read any system with control zones, including a board of sixty-four squares.

---

Mar del Plata: Two Wings, Two Clocks

The Mar del Plata Variation is the sharpest version of the Classical King's Indian, sitting in the ECO range E97 to E99. Its name comes from the Argentine coastal city where it was analysed and shaped across several tournaments.

Its structure is plain. Black puts a knight on c6, pushes the e-pawn to e5, then plays ...f5. White pushes the d-pawn to d5 and holds. Then both sides begin a race that analysts call the two-wing race. White attacks the queenside with c4-c5 and b4-b5. Black attacks the kingside with ...f5-f4 and ...g5-g4.

What makes Mar del Plata one of the hardest variations in chess is its clock. Both sides know what the other will do. Both know speed decides. And both know that the side one move slower gets buried under three layers of pawns on its own wing.

Through the 1990s and 2000s, Mar del Plata was a permanent frontline at the highest level. It was also the era when players focused maximally on memorising every branch, every sub-branch, to move twenty-five or thirty. Elite chess became a controlled memory test.

Then computers changed everything.

---

The Bayonet Attack 9.b4: A Blade Into the Queenside

The move 9.b4 has a fine name. It is called the Bayonet Attack. White pushes the b-pawn from b2 to b4, without preparation, without hiding intent.

The idea is crude and effective. White wants to open a second front on the queenside before Black can start on the kingside. The b4-pawn prepares b4-b5, drives the c6-knight away, and undermines Black's supporting structure on that side. If Black loses control of the queenside, White can transfer everything to the kingside before Black opens up.

The Bayonet has an important psychological feature: it does not demand immediate, exact calculation from White. It demands patience. That makes it the perfect weapon for a player who prepares carefully at home and wants to impose a framework from move nine.

The truth is that when 9.b4 became fashionable at the top, Black had to find answers. The orthodox answer is 9...Nh5. The knight jumps to h5, preparing ...Nf4, then ...f5, then ...g5. That is the path that forces calculation. Every move has an immediate consequence, every move can be checked by an engine, every move has already been written into books.

But there is another answer. It is older. It is slower. And it has just been repackaged as a product.

---

9...Nh5: The Path That Forces Calculation

After 9.b4 Nh5, Black attacks the e4-pawn with a minor piece rather than a pawn. This is the signature of Mar del Plata: pieces apply pressure, pawns occupy ground. The knight on h5 has two destinations. It can jump to f4, pressing e2 and g6, or it can support the ...f5 push that opens the kingside.

White usually replies 10.Re1 or 10.c5. Both lead to long, sharp sequences that can be verified by an engine. In those sequences error is measured in centipawns. In those sequences the better prepared player wins.

The most important feature of 9...Nh5 is that it opens the position. When the position opens, engines reach full power. When the position opens, a bad move at move twenty-five can be found in three seconds. And when the position opens, a player's memory becomes an asset you can price.

That is precisely what a certain group of players wants to avoid.

---

9...a5: The Move That Redraws the Map

The move 9...a5 does something very simple: it asks White a question about the b4-pawn.

The a-pawn goes from a7 to a5, attacking b4. White must decide. Trade, hold, or push. Each choice leads to a different position, and none leads to a fully forced one.

This is the crux. Every opening plan is a hypothesis until the pieces touch the board. After 9...a5 the board is no longer a memory test. It becomes an evaluation test.

Black is not trying to win immediately. Black is trying to keep the position manageable. White has queenside space, but that space is challenged at its point of support. White can push b4-b5, but then the b5-pawn becomes a target. White can trade, but then the a-file opens and Black gains a road for the rook.

In other words, 9...a5 shifts the game from a speed race into a negotiation over territory. And in territorial negotiations, the side with better information about the map wins.

---

The Decision Window

In football analysis I always start from the period before the goal. Fourteen seconds — enough to redraw the entire defensive map of the opponent. On a chessboard that window is not measured in seconds but in moves.

After 9.b4 a5, the decision window lasts roughly three to four moves. Inside it, Black must decide his queenside pawn structure, and White must decide whether to keep expanding.

What interests me is not which move is theoretically better. What interests me is the kind of pressure each move applies to the opponent's clock.

After 9...Nh5 the pressure is calculation pressure. White knows he must calculate exactly, and White has usually calculated at home.

After 9...a5 the pressure is evaluation pressure. White must decide for himself what is right in a position where the books say less. For a player used to preparing at home, the second kind of pressure is far more uncomfortable.

That is why an old move can return. Not because it is stronger. Because it is uncomfortable in a different way.

---

Trade-offs: What Is Gained, What Is Lost

Every move in Mar del Plata is a transaction. You trade one thing for another, and no transaction is free.

After 9...a5, Black gains three things. First, a less forced position, meaning less risk of being beaten by a prepared line. Second, a more stable queenside pawn structure, where losing a pawn does not automatically mean collapse. Third, a longer game, where the skill of accumulating small advantages matters more than memory.

Black also loses three things. First, a tempo in the two-wing race. In Mar del Plata, tempo is currency, and 9...a5 is slower than 9...Nh5 at opening the kingside. Second, part of the queenside space, because the a5-pawn cannot come back. Third, some direct attacking options against the white king.

My reading is this. The spatial map never lies — it only exposes what we want to believe. If Black believes he can endure a long positional fight, 9...a5 is a good trade. If Black believes he must generate a fast attack on the white king, 9...a5 is a bad trade.

That is why the same move can be a weapon for one player and a trap for another. No move is good for everyone. There are only moves good for a specific kind of player in a specific kind of position.

---

The Promise of "Very Few Concrete Lines"

The promotional material for this product promises something very specific: very few concrete lines.

According to the description, the package presents 9...a5 as an antidote to the Bayonet Attack, stressing that this variation leads to sound positions with few forced lines to memorise. The chances are claimed to be equal. And the target audience is clearly defined: players who want to think for themselves rather than memorise.

This is a clever marketing argument, and I want to split it in two.

The first half is structurally correct. Mar del Plata is a variation with extremely high theoretical density. Any branch that is less analysed genuinely requires less memorisation. That is simple arithmetic: fewer documents means fewer things to learn by heart.

The second half is inflated. "Few forced lines" does not mean "little calculation." In Mar del Plata the pawn structure remains sharp. The kingside can open at any moment. A well-prepared opponent can still generate an attack for which no pre-written line exists.

In short, this product sells a shift of pressure, not an exemption from pressure.

---

"Computers Don't Understand These Positions" — A Dated Sentence

This is the part where I want to be direct.

The pitch rests on an explicit human-versus-machine contrast. According to it, 9...Nh5 opens the position, forces the lines, and makes computer analysis important again. Meanwhile 9...a5 leads to sound positions that computers often do not know how to handle.

The second sentence belongs to the pre-AlphaZero era. I stress the word era.

Before neural-network engines arrived, chess computers had a widely acknowledged weakness: they played well in tactical positions, where everything reduces to calculation, and less well in long-term positional positions, where concepts like pawn structure and square control cannot be quantified immediately.

That is no longer true in the way it once was. Modern engines handle the long-term positions of the King's Indian far better than the previous generation. They can assess the value of a weak square on the queenside twenty moves ahead. They do not need a forced line to play well.

This does not mean 9...a5 is a bad move. It means the sales argument rests on an outdated premise. And when the sales argument is outdated, the buyer should know he is buying a product, not a truth.

I have seen this many times. A technical claim that is true for ten years, then repeated for the next ten as a mantra. The repeater does not recheck. The listener does not recheck. And a fact becomes a habit.

---

The Problem With an Antidote

An antidote keeps its value only until the opponent finds the antidote to the antidote.

In the opening market this cycle is short. A line is published, sold, played in club and international master events, then analysed back. Afterwards it returns to its original state: a known branch with a known answer.

9...a5 is not an invention. It is a branch with a history. It appeared in games before 9...Nh5 took the leading role. It is known to well-prepared White players. That means its competitive value lies in surprise, and surprise has an expiry date.

I do not say this to belittle the product. I say it because this is how the opening market works. Half the value of an opening course lies in what it teaches. The other half lies in the fact that your opponent has not read it.

And that is an investment with a very clear downside risk.

---

Selling to Thinkers

There is one detail in the advertisement I want to dwell on. The product is positioned for players who like to think for themselves, not for memorisers.

This is identity-based selling, and it works extremely well.

Here is how it operates. Club-level players carry a mild inferiority complex about opening theory. They know that at the top people remember thousands of lines. They know they lack the time and the memory to do the same. So when a product says "you don't need to memorise, you only need to understand," that product is not selling a course. It is selling permission.

That permission can be genuinely valuable. Many players improve significantly when they stop trying to remember everything and start trying to understand principles. But that permission can also be a trap. If you buy a course on a Mar del Plata variation hoping it removes the need to calculate, you will be disappointed. Mar del Plata does not allow that.

In the 120-page report I wrote during the pandemic, while coding 380 Thai League matches, I found the thing the season never records: repetition. Patterns return. Gaps return. And in chess, pawn structures return in exactly the same way.

That means learning principles, not lines, is the higher-yield investment. A good opening product teaches you principles. A bad one teaches you fifteen moves and leaves you alone on move sixteen.

---

The Loan With an Obligation to Buy

I have a professional bias I do not hide. In football, I argue that loans with obligations to buy distort the finances of small clubs. They develop players for others, carry the cost of development, then hand the finished product to a big club at a pre-set price.

I recognise the same structure in the opening-course market.

When you buy a course teaching a specific branch, you are not only buying information. You commit to building your entire opening system around that branch. You learn it, you train it, you play it. For months, you become a specialist in a line most of your opponents do not play.

Then one day your opponent buys a different course. And you start again.

The obligation to buy in the opening market is the time you have already invested. You cannot withdraw it. You can only hope your branch outlives theirs.

That is why I always tell young players: buy structures, not lines. Structures do not expire. Lines do.

---

The Human Variable

There is one thing every opening analysis ignores, and it usually decides the game.

I call it the fourteen-second gap inside the player's head.

On the board, after 9...a5, White has a short window to decide whether to continue with his original plan. That window is recorded in no book. It depends on how much White slept, how long he prepared for this opponent, and whether he trusts his own assessment.

A well-prepared player's first reflex is to search for a familiar line. When he cannot find one, he loses time. In modern chess, losing time on move ten can mean losing on move forty.

This is the real value of 9...a5, and it appears in no evaluation table. Its value lies in creating a cognitive gap in the opponent's head, and inside that gap a chess player often errs not because he does not understand the position, but because he does not understand himself.

Coaching does not create identical players; we create different paths. That is true in football. It is true in chess.

---

The Mismatched File

Back to the Samarkand file.

The mismatch between headline and body is an information-integrity problem. If I cite this document without checking its origin, I transmit an error. If another journalist cites me, the error doubles.

In twenty-three years of working with sports data I have learned one rule: check the headline first, check the body second, and check whether they belong together. Most mistakes in sports journalism do not come from reporting something false. They come from placing correct data in the wrong slot.

I have not yet located the original Samarkand tournament report. I have not yet verified the identity of the opening product's author. Both are gaps I will keep tracking, because an analysis is only trustworthy when the writer states clearly what he does not know.

---

What to Track

I offer no prediction on whether 9...a5 will succeed or fail. I offer a list of things to observe, because that is the only way to turn an opinion into a testable hypothesis.

First, the frequency of 9...a5 in classical games at grandmaster level. If it appears only in opens and club games, its theoretical value stays low. If it appears in high-rated events, the story changes.

Second, the frequency of the Bayonet Attack 9.b4 itself. If White keeps playing it often, the market for answers survives. If White shifts to other systems, the product loses its reason to exist.

Third, the response from the White side. Every newly published line usually generates a counter-cycle lasting several months. Tracking that cycle is the best way to measure an idea's lifespan.

Fourth, the author's identity and record. An opening course written by a player with practical results in that very variation has a different value from one assembled from a database.

---

Open Ending

I began this article with a file error, and I want to end with a question about whether that error was random.

A headline about strong teams winning hard games. A body about a modest move on move nine. Both speak to the same thing: the gap between gaining an advantage and keeping it.

In Mar del Plata that gap is measured by one square. One square on the queenside. One square the a5-pawn has just entered. One square most players will not see, because they are looking at the kingside, where everything is louder.

The next game I watch, I will start my clock on move nine. Not to measure thinking time. To measure the time before the map changes. That is the interval every opening book skips, and it is usually the interval that decides.