9...a5 in the Mar del Plata: a repertoire DVD and what the data has not yet confirmed

On my shelf sits a product titled A night unlike any other. The title evokes a bayonet assault in the dark — the literal sense of the word Bayonet. Inside, the author presents a plan for Black in the King's Indian Defence, Classical system, Mar del Plata Variation, after White's ninth move: 9.b4. White pushes the queenside pawn forward like a spear, and Black is invited to answer with 9...a5. I opened the disc with an empty data sheet beside me. No player names. No game counts. No engine evaluations. No publication date. No author credentials. A single-source commercial product, and by definition, every technical claim inside it is pending verification.

The interesting part is that the title functions as a military joke rather than a piece of information. Bayonet Attack has been an established name in chess theory for decades; pairing it with a night raid is packaging, not description. Someone buying an opening disc needs three things: what Black's ninth move is, which main line it replaces, and how many hours of preparation it actually costs. The title answers none of them.

Context first. In the Classical King's Indian, the Mar del Plata arises from a fairly fixed 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 locks the centre with d5 pawn, Black accepts being pushed back on both wings and then counterattacks with ...f5 and ...f4. The name comes from tournaments held in the Argentine city in the early 1950s, where this pawn structure was first explored systematically. Since then it has become one of the most theory-dense battlefields in elite chess.

The move 9.b4 is the bayonet. White does not hide the intent; the attack starts before Black can launch ...f5. White's ideas fit in a few lines: push b4-b5 to freeze the sector, use Bd2 and c5 to break Black's central pawn structure, and exploit the b-file and the a3-f8 diagonal. If c5 lands under favourable conditions, White trades queenside pawns for space and opens a route for the b1 rook.

Black's main answer for decades has been 9...Nh5, heading for ...f5, ...Nf4 and a race on opposite wings. It is highly forcing: subsequent moves are tightly constrained, one tempo lost means a lost position, and both sides must memorise long sequences. That is why it dominates at the top level, and why it is hard for club players.

This is where chess theory and the product market intersect. The transfer market is not a chess game; it is a synchronised performance by thousands of algorithms — a line I wrote about football, and it fits the opening-disc market exactly. Sellers need a differentiator to justify a price. Buyers need a reason to believe they are saving time. When the main line 9...Nh5 is saturated, commercial value lies in repackaging sidelines as alternatives. 9...a5 sits precisely in that slot.

On the move itself. After 9.b4, Black plays 9...a5. Strategically, this is counterattack in the sector where you are inferior. Black has less space on the queenside, so instead of waiting, Black strikes at the base of the spear: the a5 pawn hits b4 and forces White to decide whether to keep or open the a-file. If White trades, the a-file opens and Black's a8 rook gains a highway. If White maintains the tension, Black can trade at a better moment and target the remaining queenside squares with ...Na6, ...Nc5 and ...Bd7.

The core point deserves to be stated plainly: 9...a5 is an established alternative, not a refutation. The promotional material calls it an antidote, yet in the conclusion the author concedes chances are equal. When the seller himself says equal, the product's real value shifts from objective superiority to preparation asymmetry — Black wins because the opponent is unfamiliar, not because the position is better. That is a real value, but it has an expiry date.

What happens on the board? The structure stops being a kingside race. Black delays ...f5, keeps the e5 and d6 pawns intact longer, and shifts to slow manoeuvring. The a5 pawn creates a queenside spike White must factor into every plan involving c5 or b5. If White plays 10.bxa5 and Black recaptures with a piece, the a-file opens and Black gains an active line where previously there was only endurance. If White holds with something like 10.Ba3, then after the pawn trade White's bishop on b4 looks loose, and Black gains time to develop ...Na6 towards c5.

The subtlety is tempo. In 9...Nh5 every move carries forcing weight; in 9...a5 many moves simply improve the position. Such positions reward plan-based understanding and punish memory-based play. Once White leaves theory and starts thinking, preparation asymmetry appears — but it appears for both sides, not only for the one who bought the disc.

The highest-value claim in the product is also the least verifiable: that this branch contains very few concrete lines and therefore lightens the memory load. A Black repertoire built on 9...a5 still requires at least two other knowledge blocks. The first is White's alternatives to 9.b4: 9.Ne1, 9.Nd2, 9.Bg5 and similar moves lead to entirely different structures with entirely different plans. The second is the main line 9...Nh5 itself, because after 9...a5 White can adjust and steer back into familiar structures. What is sold as a shortcut is in fact a side branch attached to a much larger theoretical block.

One more item: the scope of the product is undefined. Does it cover only the 9.b4 branch, or the whole King's Indian complex after 7...Nc6? That question is not small. If it is only 9.b4, the buyer must handle the rest alone and the real preparation cost is far higher than it first appears. If it covers the complex, the question shifts to quality of coverage. In the source I have, neither possibility is confirmed.

Then comes the most technically contentious claim: that computers often do not know how to handle the complex strategic positions arising from this branch. That argument held for the engine generation before neural networks, when closed-position evaluation was noisy. It is dated against the post-2026 landscape, when neural and NNUE architectures evaluate closed manoeuvring positions with far greater reliability. A sideline is no longer protected by computers failing to understand it; it is protected only by opponents failing to prepare. Those are different conditions, and the second can be erased by one database update.

The data gaps should be listed because they determine how far any conclusion above can be pushed. No reference games. No occurrence counts for 9...a5 in major databases. No average centipawn loss. No named elite player who has used it. No publication date. No author information. For an opening product, that is close to zero evidence.

A personal note explains why I read this disc differently than I would have a decade ago. In 2026 I predicted Germany would defend their World Cup title, based on possession data and passing accuracy in qualifying. They went out in the group stage. The numbers were not wrong arithmetically; they were the wrong numbers. I had ignored pressure-conversion and wide-attack speed. After that shock I spent three weeks rewatching the entire group stage and built my own index for underrated teams. It took me three months to learn that a beautiful chart is no substitute for a correct process.

That lesson maps directly onto an opening disc. A neatly packaged product with graphics and a promise of low memory load is the beautiful chart. The correct process starts with questions about sample, sample size and representativeness. Where has this branch been played, by whom, with what results, in what period, under what time control? If the material cannot answer those, it can still work as a conceptual guide, but it cannot serve as the foundation of a competitive repertoire.

Here I will counter my own reading, as I try to do with my own data. I understand why the author chose 9...a5 as the selling point: dependence on correlation. Few players use the branch, so early adopters win more, so it looks stronger than it is. Correlation is not causation. A high win rate in a rare branch usually reflects opponent surprise, not objective value. Once the branch spreads, the rate converges on the true value — and according to the author's own words, the true value is equality.

There is a deeper layer, and it is uncomfortable for club players. The marketing argument says the branch suits those who prefer thinking for themselves over memorising long variations. That sounds reasonable but reverses the nature of the problem. Closed manoeuvring positions demand more understanding, not less. In a forcing theoretical sequence, memory carries you for dozens of moves. In a strategic position nothing carries you; every move is an independent judgement about pawn structure, weak squares and priorities between wings. A newcomer to this branch does not escape study; they simply switch to a harder kind of study.

So where does the product's genuine value lie? In the fact that it suggests a real plan: counterattack the base of the advance, accept less space in exchange for an open file and a distinct tempo. That is a strategic lesson worth having, and it can teach more than a conventional opening disc. Its commercial packaging, however, should be read through a filter: the punning title, the low-memory promise and the engine claim all fail verification.

Data is a mirror; only those who dare face themselves see the truth in it. Buyers of this disc will see one of two people. One is a player with strategic foundations who reads the branch as an addition to an arsenal, used as a surprise in a few games a season. The other is a player looking to reduce memorisation volume, who buys it thinking it is light, then discovers they have walked into the hardest type of position with no one to lean on.

It took me considerable time to understand that everything in the sports-data industry works this way. A Chinese club taught me that data is not the destination but a walking stick. I once presented to a club's leadership that their striker scored above expectation while actual efficiency fell below it, because he depended heavily on set pieces. One number was enough to change a plan. But that number only mattered because it was tied to a specific verification process — not because it looked good.

With an opening disc, the process is simpler than it appears. Count the games in the database. Check at what level and frequency the branch is played, and which side the results favour. Verify engine evaluation at the critical positions. Then ask: if my opponent knows I have prepared this branch, what do I have left? If the answer is nothing, I bought a time-limited surprise, not a weapon.

A time-limited surprise still has value if it is managed properly. Any sideline, once published, will be answered. The cycle repeats endlessly in chess: a rare move appears in a few games, causes disruption, is absorbed, and becomes an ordinary line in the database. The wise player does not buy a branch in order to believe in it forever; they buy it to use while it remains unabsorbed, and they keep a retreat to the main line ready.

Read this disc as a proposal, then, not a verdict. It proposes that in a position where theory is saturated, there is still room for plan-based thought. It does not prove that 9...a5 is better than 9...Nh5, and does not try to — the author himself says chances are equal. It proves something more modest: that Black can choose a different kind of game, slower, less forcing, and demanding more understanding.

After 2026 I stopped trusting predictions. I trust only early-warning systems. For an opening product, mine has four items: whether the data source is traceable, whether the product's scope is clearly defined, whether the true preparation burden is stated honestly, and whether the technical claims still hold against current engines. A night unlike any other passes the first weakly, is unclear on the middle two, and fails the last.

When the data does not lie, we are the ones lying to ourselves. This product does not lie in the sense of fabrication; it is simply silent exactly where silence benefits a sale. That silence is not a crime, but the buyer must ask the questions in its place.

For a player building a repertoire for next season, I suggest a different route. Take 9...a5 as a thinking exercise: set up the position after 9.b4, analyse for thirty minutes without an engine, write down the plans you find, and only then open the database and engine to compare. The comparison will show how deeply you understand the structure and how difficult the branch really is. Then decide whether to add it to your arsenal. At least the decision will belong to a process you own rather than a promise you bought.

The signal I am watching for the next cycle is not whether 9...a5 becomes fashionable. I am watching whether anyone at professional level uses it in a serious game, how the opponent responds, and which branch enters the database first. As the game count rises, the asymmetry value falls. If by then the branch retreats from competitive practice while the strategic plan inside it still teaches something, the product has completed its most valuable work: not providing a move, but creating a way of seeing.

Otherwise it is just a surprise package that expires with the next database refresh. A bayonet in the dark is only worth something until dawn.