The Board Without a Scorekeeper: Vietnam's Chess Data Gap

A tháng Tám afternoon in Nha Trang, and rain was hammering the corrugated roof of the chess club at the end of a narrow alley. Coach Nguyen Van Bay handed me a clear plastic box before leaving for his home province to recover from illness. Inside were more than four hundred scoresheets, in no order at all: sheets from provincial youth tournaments, from amateur events, from games whose tournament names had faded past reading. The sheet on top was written in blue ink, in the hand of someone around thirteen or fourteen. The game began 1.Nf3 d5 2.c4 e6 3.d4 Nf6. At move nineteen came a rook sacrifice, marked with an exclamation point pressed so hard the paper had nearly torn. At move twenty-seven the page ended. No date. No players' names. No tournament. Only a small line in the margin: "board B, round 7." I kept that sheet in my bag for three months and called twelve people who had organised chess in Khanh Hoa between 2026 and 2026. Four remembered a strong boy from Dien Khanh. Nobody remembered the game. Vietnamese chess does not lack achievements to be proud of. Nguyen Ngoc Truong Son was awarded the international grandmaster title in 2026, at fourteen years old. Le Quang Liem won the World Blitz Championship in 2026 in Khanty-Mansiysk, an event organised by FIDE that June. Those games live in international databases: they carry ECO opening codes, time controls, move-quality indices, and can be replayed move by move. Anyone in Buenos Aires who wants to see Le Quang Liem's final-round win that night needs a few clicks. But the games that made those two people barely exist. Provincial qualifiers. Delta youth championships. A school semi-final for which a coach had to take three bus rides to bring his student. No ECO code. No digitised results table. Nobody recorded how long each player thought on each move, and nobody noted whether, at move twenty-five, one of them had already run short of time. At international level, game archiving became an industry long ago. Commercial databases hold millions of games and update weekly; FIDE publishes its Elo rating lists on a monthly cycle; major tournaments broadcast move by move and file everything automatically. In Vietnam, most chess memory sits in three very different layers, and every layer has its own holes. The first layer is the personal notebook. Most players who came up before 2026 kept a notebook of games they were proud of, usually written only to move twenty and then abandoned, because the rest could not be recalled. The second layer is the organiser's scoresheet. This is the most important layer and the most fragile: carbon paper, cheap pads, a document folder tied with a rubber band, kept in a wooden cupboard in a district sports office. One leaking roof, one office move, one clear-out is enough to erase a generation of games. The third layer is official archives and international databases, which capture only the tip: events registered for rating, events covered by international media. Based on my experience following matches and working with documents over many years, I believe the greatest loss is not the good games. The loss is the ability to ask questions. A complete scoresheet lets you trace back many things memory can never hold. It shows which opening family the game belonged to, who had prepared, who was dragged into a structure they did not know. It shows when the decision point arrived: move eighteen or move twenty-seven. It shows how long a player thought on the hardest move, and how long on the easiest. Those three facts, taken together, say more about a person than any results table. When a game is put through an analysis engine, one measurable number is the average loss per move: the gap between the move actually played and the best move the machine suggests, converted into the smallest unit of material value. The figure is not gospel, but it is a rough yardstick for comparing decision quality across hundreds of games. A young player with a low loss figure in the opening but a spike in the endgame is usually someone drilled in theory but short of real playing hours. A player whose loss figure climbs steadily with move number is usually someone who loses control as the game stretches. Without scoresheets, patterns like these vanish. A coach is left with instinct. Instinct is sometimes very good. But instinct cannot be handed to a successor. Back to the blue-ink sheet. From the twenty-seven legible moves I can rebuild part of the story. White played a knight-opening system; Black answered with a solid queenside structure. At move nineteen, White put a rook on a square controlled by two pieces, trading material balance for a position the engine rates as roughly level in material but nearly one unit better in positional terms. Fully calculated, that sacrifice buys White the initiative for about fifteen moves. If it was a bluff, it collapses within three. What I cannot know: whether White saw the whole line or only the first two moves. Whether Black was short of time. Who was in the room. And where the boy who wrote that sheet is today, what he does, whether he still holds a piece. A game missing its data is not a bad game. It is a game with its portrait cut away. At system level, this gap causes three concrete losses. First, talent identification loses the ability to compare. A fourteen-year-old who wins eight games in a row at a provincial event will be remembered. A fourteen-year-old who plays steadily for three years, winning six games each year with steadily improving move quality, will not be remembered, because nobody holds the data that would reveal the growth curve. Second, coaches lose the ability to prepare against domestic opponents. In stronger chess nations, a coach can pull up a national-level opponent's entire opening history in minutes. Third, arbiter decisions become impossible to contest. When a game is decided under the rules with no preserved scoresheet, no recording, no clock record, spectators can only choose to believe or not to believe. There is almost no mechanism for explaining a ruling at the board, and transparency remains a slogan. I once watched a rapid game at a youth event lost on a ruling about a miswritten scoresheet, while both players insisted they had written it correctly. Nobody won that argument. In 2026, when indoor events stopped, Vietnamese chess moved heavily onto online platforms. The stadiums were empty, but the applause of 2026 still echoes inside every old record. What was interesting is that this period produced a leap in data: every online game was logged automatically, with a game code, per-move times, and a standard downloadable format. For the first time, a large volume of games by Vietnamese players across every age group was archived without anyone copying by hand. But automatically generated data carries its own trap. An online game cannot measure a shaking hand, the noise of a hall, or what it means for a fifteen-year-old to sit at a real board opposite an adult two heads taller. To understand why a talent fails to grow, you need both kinds of data: the number and the circumstance. We have just acquired the first, and are steadily losing the second. Here I want to say something that may irritate people. Romanticising the era without data is a convenient kind of romance. We like to say chess used to be "purer", that players of that time calculated with their heads and had no machines. Technically, that is true. It is also true that we lost the names of hundreds of children who played well, simply because nobody bothered to keep the results of a district tournament. Every stadium I walk through has a secret door, and the entrance is always a person. In chess, that door is usually a sheet of paper forgotten in a drawer. The opposite view also deserves a hearing. Many believe that digitising every game will automatically advance Vietnamese chess. I do not fully believe it. There is a paradox in the engine age: the more completely games are recorded, the more alike they become, because everyone consults the same databases and avoids the same lines. The match rate with engines in the opening rises, while the share of games that leave a personal mark falls. If we copy someone else's data model without building our own analytical culture, we will end up with a large archive and a chess nation smaller than that archive. The second concern is the reflex to fill gaps with ready-made stories. With no data on a generation of players, a writer easily assumes that generation was weak, or conversely that it was more glorious than today. Both are speculation. In documentary work I learned a dry but necessary rule: where there is no evidence, leave it blank rather than filling it with a plausible-sounding hypothesis. A record never ages; it waits for someone who knows how to listen. But it can only speak if it still exists. There is one detail I have not told. In Coach Bay's box I found seventeen scoresheets in the same handwriting. Same notation, same habit of marking good moves with a plus sign in the left margin, same habit of noting the clock in the top right corner. Seventeen games, scattered across about four years, none of them named. I grouped them by opening style and found a fairly clear growth curve: the opening repertoire narrowed, the number of moves before the decisive moment dropped from twenty-seven to nineteen, the frequency of sacrifices rose. A player learning to play more sharply. That is the entire biography we have of one human being. With those seventeen sheets I could write a professional report convincing enough that any coach would say this child needed two more years of tracking. But the report arrives twenty-eight years late. That is the real cost of the data gap: it does not erase the past, it merely stops the past from reaching the present in time. This is not unique to chess. Any sport that depends on a piece of paper lives with the same risk: athletics with hand-timed sheets at local meets, swimming with electronic boards in pools that do not archive, combat sports with judges' scorecards. The same mechanism repeats: the technical part is recorded, the human part is dropped. And once the human part has been dropped long enough, people start calling the dropping fate. If there is one thing to do this season, I do not think it is hiring more experts or buying more software. It is imposing a minimum rule: every chess tournament at provincial level and above must keep scoresheets on a standard form, with the date, the time control, both players' names and birth years, and scanned copies filed to a shared archive within thirty days of the event's end. The cost of that rule is smaller than one meal allowance for a delegation. Its value may be measured in decades. The blue-ink sheet sits in my drawer, next to a notebook with a worn spine. Whoever wrote it sometime in the mid-1990s may have left the board long ago, or may be teaching a child the first opening moves. I do not know the face. But I know that at move nineteen, that person put a rook into danger, and nobody wrote it down. Perhaps the next move is being written right now, in blue ballpoint, in a playing hall with no camera. Our task is simple and very hard: keep that sheet, before it turns into a story that can no longer be verified.