The 9...a5 Move and the Theory Arms Race Leaving Club Players Behind

Samarkand, round 10, women's section. The headline says Kazakhstan is closing the gap. I opened the body copy looking for the standings, the player names, the board-by-board results.

No standings. No player names. No Samarkand, no Kazakhstan, no boards at all.

The ten information points describe a completely different object: an opening-instruction DVD devoted to the King's Indian Defence, Mar del Plata variation, and a move marketed as an antidote — 9...a5. The two objects do not share a single detail. The mismatch sits at the level of object classification, well beyond a difference in editorial nuance.

I trust footage over testimony, because footage does not know how to lie.

In 2026, when I proposed a technical study of Nguyễn Thị Oanh and was waved away with a single sentence about women and running, I did not argue. I rewound the SEA Games 29 tapes to find what people had left behind, counting her cadence over the final lap. The habit followed me into pieces that have nothing to do with a running track. A headline about a women's team event in Uzbekistan; a body about an opening product. Someone will assume the headline is correct. I open both and compare.

The board before the numbers

To understand why Black's ninth move deserves an article, the position has to be rebuilt from the start.

The game follows the King's Indian Defence, Classical System: 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. ECO codes place this family at E97–E99. The centre is locked solid at d5 on top of e4, and both sides tacitly accept a race on opposite wings: White pushes c5 and b4 to break through on the queenside, Black pushes f5 and g5 against the king. Whoever arrives first wins. There is no second line of defence.

The move 9.b4 is the Bayonet Attack. It is not new, but it became a fashionable weapon at the top level from the late 1990s, when Garry Kasparov was still at the board. The idea is simple: take queenside space before Black can open the kingside, and disrupt the rhythm of Black's standard development.

Black's standard reply is 9...Nh5. The knight goes to the rim, preparing ...f5. This is the most demanding test, and it is also the branch that has been analysed almost to exhaustion. After 9...Nh5 the lines become forced, and engine analysis becomes important again.

The DVD in the article proposes a different route: 9...a5. Same objective — preparing ...f5 — different road. Black does not send the knight to the rim immediately, keeps the structure flexible, and in exchange gives White a freer game with fewer concrete lines.

The product advertises itself with two phrases: “very few concrete lines,” and “strategy, not tactics.” It targets players who prefer to think for themselves rather than memorise.

That is the entirety of the technically verifiable material. The rest is assertion.

What stands behind a move that is sold

The first thing to separate: 9...a5 belongs to the family of revived sidelines, not to the family of novelties. In analytical language, a novelty is a move never previously played in a known line — genuinely new theory. 9...a5 has sat in the databases for a long time, merely appearing rarely. The product's value lies in repertoire practicality, not in breaking theory open.

That leads to an important judgment: Black is trading criticality for preparation cost. The move 9...a5 does not pose White as hard a test as 9...Nh5. It merely makes that test cheaper for the player holding the black pieces. The real value of the product does not lie in 9...a5; it lies in the pricing of the buyer's time.

Second: the claim that “objectively the chances are equal” is an assertion, not a measurement. Across the ten information points there is not a single ACPL figure, not one quoted engine evaluation, no database frequency, no illustrative game, no named practitioner who has used the line. For a technical product, that is a substantial evidentiary gap. The buyer cannot assess the soundness of the variation from the promotional material itself.

Third, and the weakest item in the whole package: “computers often do not know how to handle the arising complex strategic positions.” That phrasing belongs to the pre-AlphaZero era. Neural-network engines today, Leela-class included, and contemporary Stockfish, evaluate closed, manoeuvring King's Indian positions far more reliably than the 2010s generation did. The claim is partly true in spirit — engines remain weak at long-horizon strategic concession judgments — but false as a blanket statement. And it is the sales argument. The weakest evidence is propping up the strongest claim. That is the classic signature of promotional technical writing.

Fourth: the two advertising phrases do not address the same buyer. “Don't memorise, think for yourself” speaks to the independent player who accepts risk to keep an identity. “Very few concrete lines” speaks to the time-poor player who needs safety. Two different portraits, merged into one. Merged, the product looks suitable for everyone, and everyone is easier to sell to.

Fifth: there is an internal tension between two sentences. The article's author concludes that the chances are equal. The promotional copy calls the variation “dangerous for White.” Equal means a drawing or survival weapon. Dangerous means a winning weapon. Buyers will read it the second way.

Sixth, and this is the point I consider most valuable: the very existence of an antidote is itself a market signal. People manufacture an antidote only when someone is being poisoned. Had the Bayonet Attack not been scoring well for White at precisely the level where this product's buyers play, there would be no demand for a substitute sideline. Put differently, the rise of 9.b4 created the market for 9...a5.

Seventh: this is an asset with diminishing returns. Opening knowledge excludes no one. Once a sideline is in print, opponents and their analytical seconds can prepare against it. Surprise value runs inversely to popularity — the better the DVD sells, the faster the edge evaporates.

Eighth concerns distribution rather than content. The product exists in DVD format, the traditional premium line of chess publishing. That channel is being displaced by subscription course platforms, by material integrated into streaming, and by preparation tools wired directly into databases. The content may be sound. The channel is structurally disadvantaged.

On the buyer file, my estimate is that the product targets club players and improving adults, roughly 1600 to 2200 Elo. Titled players will treat 9...a5 as an occasional surprise for a few games a year, not as a main line of defence.

And here is where I want to stop for a moment.

Based on my experience following games, most amateur chess games are never recorded. No scoresheet, no database entry, no one keeping time. I have run hurdles all my life, only I was never timed. The theory arms race is calibrated on games that the people paying for it will never play.

Every documentary is a running track: the audience sees the finish, I live at every starting line. Here too. The buyer sees a move. I see a time problem.

The contrarian angle: the most notable thing sits off the board

If I had to pick the single most memorable detail in the package, I would not pick 9...a5. I would pick the mismatch between headline and body.

A headline about a women's team event in Samarkand, a body about an opening DVD. Two objects that do not intersect. And nobody in the processing chain caught it. For a documentary maker, the loudest warning is not wrong data — it is correct data attached to the wrong place. A fake scoreboard gets caught immediately. A drifting headline travels straight into every aggregation downstream without making a sound.

The second contrarian angle concerns the advice itself. The variation is advertised as “strategy, not tactics.” But the King's Indian Defence, at exactly the decisive moments, demands very high tactical sharpness — the article itself concedes as much. An all-strategy diet can atrophy precisely the muscle the position requires. Players should pair this sideline with separate tactical training rather than treat it as a substitute.

And the third: “computers don't understand” is not merely outdated, it encourages a harmful epistemic habit — dismissing engine evaluation instead of understanding it. The more honest formulation would be that engines are weak at long-horizon judgment of strategic concessions, not weak at complexity as such.

What is worth keeping

A move on the ninth ply will not make anyone rich, and it breaks no doctrine. But the way it is sold — through a contradiction at headline level, through the weakest evidence placed beneath the strongest claim, through two buyer portraits merged into one — deserves to be remembered by chess readers longer than the opening variation itself.

Next time a product promises you will no longer need to memorise, ask one question: what is the frequency of this variation in the database. If the answer does not exist, you are buying time, not theory. And time always has an expiry date.