Samarkand in the Headline, a DVD in the Body: Where Is the Receipt for the Bayonet Attack?

One.

When I read anything about chess, my first question is never what the writer intends to say. My first question is: what paper sits under this text? Last month a piece came to me whose headline read — Samarkand Open, round seven, Uzbekistan extending its lead. I opened the body and found no round, no pairings, no standings, no player names, not even a rating number. What I found was a DVD, an opening, and one sentence: "objectively, chances are equal."

In May 2026 I published a four-thousand-word reconstruction of Spartak Moscow's title run. Sixty league goals against a shot-quality model that said 47.3 — a +12.7 overperformance. That was not structure; that was finishing luck. The first public xG model was not a prophecy. It was a receipt. In chess I still look for that receipt, and in most instructional content I do not find it.

Two.

The opening in question is familiar ground. King's Indian Defense, Classical branch, Mar del Plata character: Black plays ...e5 to strike the centre, both sides build attacks on opposite wings, and White plays 9.b4 to take queenside space. That is the Bayonet Attack. Black has historically had two roads: 9...Nh5 — sharp, forcing, dense with computer memorisation — and 9...a5, quiet, strategic, where plans speak louder than tactics.

The advertisement sells the second road as an "antidote." The argument is simple: the first line creates memorisation pressure, the second does not. From Moscow I have watched Russian chess schools for years — the pain is real there too, but the answer is given differently: memorisation is reduced through deeper coach education, not by selling an alternative line.

Three.

The pitch makes three claims. One, 9...a5 has "few" concrete lines. Two, "computers often do not know how to handle the arising complex strategic positions." Three, "objectively, chances are equal."

The first is verifiable, and the verification is easy — count the database: how many games, at what level, in which years, in whose hands. The text supplies not a single number. My rule does not change: a transfer rumour is a story; a transfer fee is a number. I wait for the number.

The second claim carries a time stamp, and the stamp reads 2026. AlphaZero showed that closed, strategic positions can be learned by a neural network. From 2026 onward, NNUE architectures began evaluating closed-structure positions reliably. In the quiet branch of the Bayonet, "the engine won't understand" is now archive language, quotation language, old advertising copy. Any teaching material still repeating it likely has a source note sitting in a pre-neural-network file.

The third claim is the heaviest. "Chances are equal" is not a feeling; it is a short verdict. Verdicts come after evidence. Evidence means two things — a current engine evaluation of the main 9...a5 lines, and a database sample showing what happened across games, rating bands, and decades. Neither is present. What is present is an introduction, not a conclusion.

Four.

Now the other side. Suppose 9...a5 is practical. Suppose elite White preparation has not yet refuted it — possible. A confusion remains anyway: a strategic branch is not an easy branch for a weaker opponent. When engine verification arrives, a kind of disbelief settles into many players, and over-reliance on "memorisation-free" tools lets them dodge the real responsibility of the position. The result is not a quick loss in a forcing line, but a slow, quiet decline full of bad decisions, caught on the analysis screen far too late.

My deeper doubt is methodological. When everyone lists the same argument for the same line, the question becomes: who, at what level, actually handled the position after 9...a5? There are many narrow antidotes; a formula does not work everywhere. The verification has not been done.

One headline matter belongs here too. The headline says Samarkand, Uzbekistan, a lead; the body says DVD. That is not a rules violation, it is an editorial problem — and a harmful one, because a reader may cite the headline as fact. An open tournament has minimum structure: place, date, pairings, tiebreaks. Not one of them appears. Uzbekistan's rise is real and I respect it — but this article is not its evidence; the Uzbekistan story needs its own receipt.

Five.

So everything is zero? No. A null result is not a failure; it is also a receipt.

The demand for a "no-memorisation repertoire" is real and growing at club level. Memorisation fatigue is genuine; sitting beside boards in Dhaka club chess I watched juniors memorise twenty-five moves and then lose. That is where a repertoire product has value — not in its claim to refute engines.

But a quiet line and an equal line are two different things, and no bridge joins them. Quiet lines still demand correct handling, and under rapid or tiebreak time controls it gets harder. A product that reduces forced lines buys time; it does not add strength. In the transfer market I apply the same rule — not the language of the package, but the numbers in the ledger.

I raise Bangladesh not for ornament but for a base. The count is this: five grandmasters — Niaz Murshed 2026, then Ziaur Rahman 2026, Reefat Bin-Sattar 2026, Abdullah Al Rakib 2026, Enamul Hossain Razib 2026; nothing since. Home norm tournaments stopped after 2026, the school pipeline switched off in 2026, the federation budget sits around eleven lakh taka a year. Into that market foreign DVDs arrive while the local input base is close to zero. That is not a lament; it is a denominator.

What this analysis cannot see I leave open: who the author is, what title they hold, whether the DVD contains any complete games, and how recent the lines are. I will not cover the unknown with a definition.

Three things I will track next. One, practical proof for 9...a5 — its presence or absence at elite level in a live database. Two, the original Samarkand document — a source report or official standings, if one exists at all. Three, the distance between engine choice and human choice in closed-structure King's Indian positions in the NNUE era.

When does a line earn the name antidote? The day games accumulate on its side — not the day advertising says so.