King's Indian Defense: The 9...a5 Escape from the Bayonet Maze

I once sat for a long while beside a chessboard in a small Saint Petersburg club, just watching the player handling Black grope for a path through the Mar del Plata structure. The board was silent, but inside his head was an entire maze. That night I understood why the King's Indian Defense is both the pride and the nightmare of those who choose it. People come to this system through faith, and many leave it through exhaustion. The King's Indian Defense is one of the most distinctive systems Black can choose after 1.d4. The familiar opening moves - 1.d4 Nf6 2.c4 g6 3.Nc3 Bg7 - allow White to build a broad center while Black accepts conceding space in order to prepare a counterattack. That philosophy is called hypermodernism, and it has accompanied countless generations of players. The Mar del Plata branch is the heart of the system. After ...e5 and ...Nc6, the center locks shut and both sides race to attack on opposite wings. White pushes pawns toward the kingside; Black drills into the queenside. This was the chess of the 1960s, when Soviet players turned it into a national weapon and believed the center was only something to seize and then destroy. In the 1990s, Garry Kasparov brought the King's Indian Defense back to the summit. He played it like a manifesto, turning positions that seemed dry into stages for lightning counterattacks. A whole generation of young players followed, and the system entered a golden age with thousands of games at every level. But then, at move nine, a variation appeared and changed everything: the Bayonet Attack, White's 9.b4. Instead of waiting, White immediately grabs queenside space, preventing Black from setting up counterplay before it is too late. That move is not beautiful in form, but it is as sharp as a bayonet blade. And the name Bayonet was born from it. Facing the Bayonet, Black has one main route: 9...Nh5. That is the move of top players, analyzed to the quarter-move, to the point where many positions are almost entirely forced. Both sides must memorize dozens of exact moves. A single small error is enough to pay with the whole game. That is precisely its weakness. In the opening-product market, someone is selling a different turn. A chess DVD presents 9...a5 as an antidote to the Bayonet Attack. Instead of plunging into forced variations like 9...Nh5, Black steps back a tempo, reinforces the queenside with a modest pawn move, and turns the game into a battle of strategic understanding rather than memory. The core the product sells is simplification: 9...a5 promises sound positions, few concrete lines, and balanced chances for both sides. On the surface, that is an appealing offer to anyone exhausted by memorization. But I have followed chess for more than forty years, long enough to know that every promise of simplification must be verified. The first question I ask is not whether 9...a5 is good or bad, but how many games prove it. And the answer, in the promotional material, is none. The source provides no sample game, no win-draw rate, no computer evaluation, and no ACPL - average centipawn loss. All we have is a qualitative claim: computers often do not know how to handle these positions. That is an easy marketing line, not evidence. I have said that turning data into fate is the instinct of this trade. But I also learned one thing: when no number exists, the silence of the data is more trustworthy than any promise. Here, that silence is rather loud. Interestingly, technically, 9...a5 is not unreasonable at all. In the King's Indian Defense, Black often seeks to break the queenside with ...c5 or ...a5 to open lines for pawns and weaken White's structure. The move 9...a5 anticipates that trend, preparing ...a4 and then ...a3, or simply creating a long-term anchor. It fits the whole system's counterattacking philosophy. With 9...Nh5, Black plays the chess of technique and obligation; with 9...a5, Black plays the chess of planning and patience. Club players, who do not have all day to memorize, will find the second side far more agreeable. But this is where I want to linger longer. The promise of objectively equal chances sounds beautiful. The problem is that the word equal in advertising is often misunderstood as easy. Equal does not mean Black only needs a few moves to be done. It means Black must understand the queenside structure, know when to strike with ...c5, when to regroup pieces, and when to accept a cramped center. Otherwise, Black will hold an equal position that is uncomfortable - the kind of equality easily lost. And here is the blind spot that advertising rarely mentions: White players who specialize against the King's Indian will soon prepare for 9...a5. At high level, their support teams have enormous databases and can neutralize an uncommon variation within hours of study. This means the surprise value of 9...a5 will fade over time, just like any bayonet once it is discovered. A variation is only useful while it is new; once everyone knows it, it becomes an ordinary path, and the cost falls back on the one who believed in it. I remember a line I once wrote in a column: The transfer market is where people sell the future to buy hope. The opening-product market is the same. A DVD sells you hope of escaping the maze, and in return you pay with money and expectation. The future of that position lies in your study room, not on the disc. Then there is one more point worth mentioning, and this time about source quality. The material I hold carries a title about a tournament in Samarkand, about Uzbekistan continuing to dominate, about round seven. But the actual content inside is a promotion for a chess DVD. The title and the content do not match. This is no small matter. If a source is muddled like this, every conclusion drawn from it must be held in reserve. I will say nothing about Samarkand, I will speculate nothing about the standings, because the material supplies not a single game or line of result. The confusion itself is a signal. What is sad is that in an age of automated information aggregation, such errors grow more frequent. A sidebar advertisement gets pulled together with a tournament report, and out comes a title that does not belong to the content. A clear-headed reader must separate the two, or else will think a chess DVD relates to the results of an ongoing event. So, should 9...a5 become your main weapon? My answer is not yet. Start by cross-checking the databases of grandmaster games. Use an engine to see whether White has a clear refutation. Test practical games with friends who play the Bayonet as White. If, after all of that, you find the position after 9...a5 comfortable, keep it as a side skill rather than turning it into an entire school of thought. I still remember the feeling of sitting in the tournament hall watching the player with Black grope about. Some games are not won by memory, but by how deeply one understands the position. 9...a5 is one of the doors that opens that kind of understanding. When the stadium has no ball left, memory begins to score. The empty stadium - memory remains. If there is one thing I have drawn from forty-nine years of watching this industry, it is this: every opening variation eventually fades, but the way a person understands a position does not. 9...a5 does not save you from having to think. It only changes where you think - from the computer's calculation to your own mind. And for many players handling Black, that alone is a liberation worth having.