9...a5: An Escape Route Through the Theory Maze of the King's Indian Defence

On an October afternoon, after the final round's report had gone to press, I stayed behind alone with a database left open all night. What lingered before my eyes was a small silence buried deep inside the Mar del Plata branch of the King's Indian Defence: the move 9...a5. Fifteen years ago, I scribbled that move into my notebook while covering an international tournament in Hanoi, then let it fall asleep among hundreds of other variations. Now it has returned, carrying a question I have never fully answered: in an opening that computers have shredded into thousands of branches, is there still room for human slowness?

A young player once asked me the same thing. I did not answer at once, because the answer lies in the forgotten sidelines - where people stop counting moves and begin to think about the position. World champion Garry Kasparov made this very structure his weapon for two decades, and every one of his games reminds us that the King's Indian is the opening of those who dare to think long.

The King's Indian Defence, particularly the Mar del Plata variation with its 7...Nc6 and 8...Ne7 structure, was once the battlefield of an entire generation. When the centre closes, both sides open fronts on opposite wings: White floods the queenside with c5 and b4, Black counterattacks on the kingside with f5 and g5. This is a war of distant armies, where the loser usually loses by exactly one tempo.

Then, in the late 1990s and early 2000s, the move 9.b4 - the Bayonet Attack - began appearing regularly on top boards. White grabs queenside space, disrupts Black's standard plans, and forces the opponent to face a chain of forcing variations. Many nights I sat reviewing those games, and what struck me lay in the way players began to memorise rather than think. When an opening becomes too familiar, players no longer choose moves; they choose memories.

I remember a commentary session on VTC years ago. I told the audience that opening play is a contest of who understands the structure more deeply. An older colleague laughed and said: "You can say that, but try sitting alone with a database." His words have followed me ever since, a reminder that theory is a double-edged sword.

Black's main move for decades has been 9...Nh5, bringing the knight to the centre to restrain White's queenside advance. But the price is a long, narrow theoretical tunnel. Afterwards, the lines become forcing, and by the admission of those who promote it, "computer analysis becomes important again". Taking that road does not free you from the maze - you only trade one corner for another.

An opening DVD I had the chance to consult proposes a different direction: 9...a5! instead of 9...Nh5. This is a pragmatic substitution rather than a theoretical revolution. The real value of 9...a5 lies in opening a strategic region with few forcing lines - where a player can think for himself rather than follow memory, not in being stronger than the main move. After 9...a5, White has more space, but Black gains a stable position and a long-term kingside plan without needing to memorise hundreds of sub-variations.

I understand why this direction is loved. It suits players with limited time, club players who want a compact defensive weapon. But looking closer, I noticed something incomplete in the way it is presented.

The most notable claim in the product's introduction is this: "Computers often do not know how to handle the arising complex strategic positions." This sentence belongs to the pre-AlphaZero era. Today's neural-network engines, together with modern Stockfish builds, evaluate the closed positions of the King's Indian far more accurately than the 2010s generation. The irony is that the technically weakest statement is the one carrying the sales pitch. Engines remain weak at long-horizon strategic concessions, but that is a problem of evaluation horizon, not of complex positions as such. This is a familiar feature of promotional writing: the strongest claim rests on the thinnest evidence.

Throughout the material, I found no illustrative game, no database frequency, and no concrete engine evaluation. That is a significant evidentiary gap for technical advice, especially when a buyer has to pay for it. Based on my experience following games, a practically valuable opening usually leaves traces in top-level games - and those traces are entirely absent here.

To be fair, I do not think 9...a5 is a wrong or weak move. Its compilers do not claim it is better than 9...Nh5; they only argue it is sound and strategically rich. That is honest framing within the limits of a commercial product. But I still notice a small tension I cannot overlook: the product is marketed both as a solution for those who "like to think for themselves instead of memorising", and sold on the strength of having "very few concrete lines". These belong to two different buyer profiles - the independent thinker and the time-poor player. Blending them blurs the product's true portrait.

Meanwhile, the medium itself is shifting. The opening DVD - once the premium format of professional chess instruction - is steadily displaced by subscription platforms and interactive courses. The content may remain sound, but the distribution channel has fallen behind.

And of course, I must remind myself of one thing about the nature of every escape route: a sideline retains surprise value only as long as it stays rare. The more people buy the DVD, the faster 9...a5 enters public databases - and your opponent will be able to prepare. That is the harsh law every pragmatic weapon must obey.

I closed the analysis session when the street outside had grown late. What I carry with me is a familiar feeling, more than a new opening variation: opening theory never stops changing; only the way we face it changes first. The treadmill of theory does not produce better moves; it produces more tired players, and that very tiredness sustains a whole market of pragmatic escape routes.

Perhaps what I learned tonight lies beyond the board. A good chess player is one who knows when to stop remembering and start thinking.