When a Chess Record Comes Back Empty: The Data Discipline of a Sports Writer

In November 2026, in an editing room on the fourth floor of an old house on Lach Tray Street in Hai Phong, I opened a hard drive holding footage from an international chess tournament. Seven hours of recording. Four folders. The metadata field showed one state only: empty. No player names. No opening codes. Not a single line of PGN. Not one Elo figure saved. I sat still for a long while under the ceiling fan. My job is to rewind tape looking for what everyone else forgot — I did exactly that with SEA Games 29, counting a runner's stride rhythm in the final lap to rescue an entire film from a box-office disaster. This time what I found on the tape was total silence. I trust footage more than testimony, because tape does not lie — even when what it says is that our process has broken. Chess is the most thoroughly documented sport humanity has ever produced. Since the 19th century, games have been notated for print. Today, every game at any FIDE-rated event leaves behind a PGN file recording every move, the thinking time for each move, and usually the engine's evaluation. FIDE adopted the Elo system in 2026. No other sport digitises competitive behaviour in such detail, to the point where each individual decision leaves a trace. Which means that when a chess record comes back empty, it says more about the record-keeper than about the board. The board always leaves traces. The keeper is the one who can lose them. The Vietnamese context makes this heavier. We have a generation of players the world knows by name. Nguyen Ngoc Truong Son of Can Tho was once the youngest grandmaster in the history of Vietnamese chess, at fourteen. Le Quang Liem of Ho Chi Minh City won the World Blitz Championship in 2026 and the Aeroflot Open three times, in 2026, 2026 and 2026. On the women's side, names like Pham Le Thao Nguyen and Nguyen Thi Mai Hung keep the national team from being blank at SEA Games. A country like that has more than enough raw data to tell stories at international level. It is also a period when Vietnamese audiences are swept up in chess and in stories more than ever. In a major-event cycle, readers do not lack emotion. They lack verification. That is why I treat the empty record as a professional event, not a technical accident. Let me start at the simplest place: Elo. The system does not rank the beauty of a move. It is a probabilistic statement — two players separated by a certain number of points will have a certain expected score over a sufficiently long series of games. The larger the gap, the more the expectation tilts. But the word "expected" is the most forgotten term in every report. A game is a sample. A sample does not confirm a distribution. When a player two hundred Elo points above an opponent loses, that is not automatically earth-shattering; it may simply be the tail of a distribution any statistician must accept. Inside a major event, people also track live rating, updated immediately after each game. Live rating turns round numbers into news. Crossing 2700, touching 2800, dropping out of the leading group — each of those becomes a headline. But the live-rating line is a noise line. It swings with each game and with the strength of the opponent in each round. Reading it as a long-term trend is a basic methodological error, and I have seen that error appear steadily in sports pages, including very serious ones. One level deeper sits the family of process metrics. ACPL, average centipawn loss per move, shows how far a player drifted from the best available move. Engine match rate shows how often a player's choice coincided with the engine's first pick. Both measure how someone thinks, not whether they won. A player can produce a very clean ACPL game and still lose on a single endgame move, and a player can produce a sloppy game and win because the opponent blundered last. If a report prints only the word "won" and skips those two numbers, readers will misread the very player they love. Another layer of data audiences almost never see is the volume of preparation. At the top level, each player has a team of seconds whose job is to build pre-analysed branches for a specific opponent. Most of that work never appears on the board, because if it appears it has already failed. It is the kind of data that can only be judged through indirect results: a player moving unusually fast inside the first twenty moves is usually the better-prepared one. The biggest structural feature of the current chess world also sits exactly here, where data and titles have come apart. The world number one by rating and the world champion have not been the same person for several years. Magnus Carlsen of Norway holds the top rating but chose not to defend the world title. Ding Liren of China became world champion in 2026. Then in December 2026, in Singapore, Gukesh Dommaraju of India, aged eighteen, beat Ding Liren in the fourteenth game to become the youngest world champion in the history of the game. Gukesh did not enter that match as the highest-rated player in the world. He entered it as the world champion. This is the point any chess analysis has to state plainly, because it governs how everything else is read. In many sports, the best player and the title-holder are one. In chess, the two concepts are measured by two different systems: one measures average strength over a large sample, the other measures the ability to win one specific match. Neither system is wrong. Blending them together is always wrong. Turning to Vietnam, Le Quang Liem's record is the cleanest illustration of what long-horizon data is worth. Three Aeroflot Open titles in four years, plus a world blitz title, form a stable distribution rather than a single spike. He is not a man who produced one moment and then vanished from the news. He is someone who stayed near the top for many consecutive years, across several generations of rivals. For a writer, that stability is harder to sell than a shock, but it is the only thing that can serve as the foundation of a documentary without regret two years later. People routinely underestimate the weight of small samples. A tournament lasting a few days is a small sample. The performance rating computed from it is a small sample. A single game with a very low ACPL is a small sample. In small samples, noise always exceeds signal, and the writer's job is to tell noise apart from signal. I have had to remind myself of that many times while standing over a beautiful shot: a beautiful shot proves nothing about an athlete's ability. While waiting for the source material, I reread the statistics tables for international women's events. There is a pattern I have tracked for years. Women's events are mentioned most by the media at precisely the moment a sponsor needs a corporate-social-responsibility statement, and least through the rest of the calendar. Looking at broadcast hours, games promoted to the front page, and tactical analyses devoted to women's games, the gap is not small. For a writer, this kind of data matters more than any declaration about equality, because it can be counted. Then comes the hardest part, the part worth more discussion than any number. When a record comes back empty, two distinct failures occur, and they belong to two sides. The first belongs to the system: it returned a valid but empty structure, and stayed silent. It did not shout that it had read nothing. Technically it did half the job correctly — it refused to invent. But it skipped the other half, the more important half: raising the alarm. The second failure belongs to people, and it is the one I meet more often in my own trade. When data is missing, the writer's instinct is to fill the gap with emotion. No ACPL, so write about fighting spirit. No decisive move, so write about a historic moment. No player names, so write about the atmosphere in the stands. All of those sentences read beautifully. They share one property: they cannot be verified, and therefore cannot be refuted. A sentence that cannot be refuted is a useless sentence in this profession. That temptation is stronger for documentary makers, because we sell emotion. But I once built a film about a women's football club whose sponsorship had been cut, training in an old indoor hall, with not a single spectator in the stands for the entire shoot. When there was nothing left to film, I filmed with my ears. The sound of eleven players' studs on a wooden floor formed a rhythm I had never heard in any sports film before. An empty stadium, and I turned the sound of studs into the heartbeat of the film. That film was still selected for an Asian festival. It did not survive on emotion substituting for data. It survived on a different kind of data — audio data — that nobody had bothered to record before. I should also say something honest about my own habit. Colleagues label me as the one who rewinds tape to find the forgotten detail. That label has a dark side. Not every forgotten detail deserves a place in a film. Some details are factually accurate but narratively meaningless, and forcing them in only dilutes the signal. My own test is simple: does this detail change how the result is understood? If not, it belongs in the data appendix, not in the film. And the same standard has to apply to my own counter-arguments. I spent years pushing back against claims like "girls shouldn't run hurdles," and the biggest lesson did not come from the other side. It came from this: once you are used to arguing against something, you become prone to arguing automatically. That reflex is exactly as dangerous as nodding automatically. A contrarian point is only worth making when it survives the same level of verification you demand from others. One small task remained, and I did it that same afternoon. I stuck a note on the hard drive: record empty, unverified, do not publish before sourcing. Three lines, no emotion, no imagery. Three months later, when the original material was re-downloaded and the record contained player names, opening codes and every move, I began writing. The film came out seven months behind schedule. It lost nothing. Every documentary is a track: the audience sees the finish line, I live at every starting block. In chess, the starting block is a data file that must contain a human name. I kept that empty record from that year. I did not delete it. It is the only evidence that a process broke before anyone managed to invent a beautiful story out of it.