Nine Sections of N/A: Vietnam's March 2036 Badminton Data Report

2:47 a.m., March 14, 2036, Nagoya. A nine-section file lands in my inbox. I read all nine sections before putting the kettle on, keeping to a habit I have held for twenty years: read everything, then speak. The technical and tactical section returns N/A. The form and player-data section returns N/A. Tournament system, world landscape, competition rules, coaching staff, risk surface, public narrative, badminton industry transmission — nine out of nine, N/A.

Not a single serve-error rate. Not a single average rally length. Not a single player name. Not a single match date. The source field is blank.

The final line of the file says only: “Please analyse it for me.”

I sat down at 3:10 a.m. and closed the file at 3:50 a.m. Those forty minutes produced exactly what the source file contained: a page stating that there is nothing yet to state. It sounds like failure. In my trade, it is a result.

Context: the real cost of recording

By 2036, sports-data infrastructure has passed through mass automation. A good enough vision model can extract scores, movement directions and shuttle trajectories from a singles match in minutes. At the domestic Vietnamese level, the gap between technical capability and actual deployment remains wide. Most matches in the youth system, domestic qualifiers and invitational events have nobody coding them, and nobody checking the machine's output either.

The real cost of recording sits with people, not devices. An experienced coder needs three to four hours for a five-game singles match if shuttle trajectory, spin direction and landing point are to be captured. A six-day tournament across five events yields roughly two hundred matches. Sponsors pay for fifteen-second highlight clips; few pay for a data table nobody clicks.

Based on my experience watching matches in domestic arenas, I have drawn one simple rule: a badminton match that is not coded within seven days will not exist in any queryable form three months later. Footage is deleted to free up drives. Paper records go into storage. The match still happened; it simply left behind no usable trace.

In 2026 I sat in the broadcast booth for the Sudirman Cup and learned that a tournament can be beamed across Asia while producing not one line of data. In 2026 I went on air for the World Cup with Japan's PPDA of 6.8 against Colombia; Japan won 2-1, and the switchboard filled with calls complaining that I spoke in nothing but strange jargon. “PPDA 6,8 là con số, còn tôi chỉ là kẻ chép lại hiện thực.” In 2026, when every competition stopped, I rewatched 547 J-League matches from 2026 to 2026 to find out what happens to a team that leads at minute 70 and then drops deep. “Đêm 547 trận đấu đã dạy tôi: bóng đá đóng băng, nhưng con số không đóng băng.” In this March 2036 file, even the old numbers are missing.

Three kinds of void and one threshold

The gaps in this file fall into three different categories, and each demands different handling.

Voids because nobody measured are the most common. Nobody logged the serve-error rate in the third game of a domestic qualifier, nobody counted net approaches after a short serve, nobody measured average rally length once a player was pushed to the backhand corner. The data never existed, so there is no retrieval path. Filling it means going back to original footage — and the original footage is usually long deleted.

Voids because something is withheld sit on the subject's side. Injury status, selection decisions, ranking points that must be defended within a cycle — these exist, they simply never leave the meeting room. This is the most uncomfortable category, because it dresses absence in a very reasonable-sounding justification.

Voids because something has not happened yet are the only kind that effort cannot fix. Two players who have never met have no head-to-head record to analyse. A draw that has not been made has no bracket to read. No budget buys the data of a match not yet played.

Above those three categories sits a threshold I set for myself: three independent sources, or ten matches of footage watched end to end, before I allow myself to write down a judgement. Below that threshold, the correct answer is N/A, not a guess dressed up with footnotes.

To grasp the volume this file would require, walk through what each section demands. The technical section needs a minimum of four badminton metrics: serve-error rate in the deciding game, front-court point-win rate, average rally length after being pushed to the backhand corner, and net-approach rate after a short serve. The form section needs the last ten results with opponent quality, schedule density over thirty days, and games played per match. The tournament section needs the event name, its tier, and its position in the points cycle. The source file has none of those cells.

An empty dataset is still a datum. It points precisely to where Vietnam's badminton record-keeping is holed, and it names each hole. A blank technical section means nobody coded the match. A blank rules section means no document was cross-checked. A blank industry-transmission section means no contract was public enough to analyse. The emptiness itself is a map.

That is also why a bare N/A is useless while an annotated N/A is a product. The annotation must answer three things: what is missing, who is holding it, and how long it would take to fill. “Nagoya không đọc báo cáo của tôi, nhưng dữ liệu thì không cần người đọc.” This March 2036 file has nothing to wait for, so I recorded plainly that it has nothing.

The trap of the fully filled report

The market does not pay for accuracy. It pays for output. An automated system fills all nine sections in four seconds; the complete version reads more smoothly, spreads faster, and in most cases nobody checks it. The paradox of that night lies here: the empty version costs more than the full one, because it burns a night of labour to produce nothing. Four seconds against four hours. Nobody pays for four hours.

The deeper trap sits in the errors. Wrong still reads as smoothly as right, and a complete report that is wrong is far harder to detect than a blank space. A blank space incriminates itself. A carefully filled table does not.

In 2026 I submitted a fourteen-page report in Nagoya, concluding that a young striker was being forced to play away from his strength of hunting the ball inside the box. The head coach dismissed it. By season's end the player had moved to Belgium for 1.2 million euros and scored 12 goals. I had the data and still lost the argument. In 2036 I have no data, and am still asked to perform certainty. Two different failures in nature, identical in one respect: people want an answer more than they want a correct answer.

In 2026, when Saudi Arabia beat Argentina, I counted five successful offside traps in the first half and was accused of stripping the magic from the match. Saying that what I saw on tape was choreographed tactics, and that the magic lay on the spectators' side, was not what anyone wanted to hear. Since then I close every report with a section on what data cannot measure: belief, passion, the crowd's fury. In this file, that section is blank too. There is nothing to limit.

The signal for the next cycle

The signal will sit in the metadata, where nobody reads. Which organisation dares to write N/A, how many times, and in which section — that will be the real measure of data capability, not the number of tables they publish. Over the next twelve months, I expect some data department to publish a void index alongside its ranking list, and that index will be read more than the ranking.

From today, every report I send will open with one line: the number of independently verified sources. On March 14, 2036, that line reads zero. “Dữ liệu không bao giờ vội. Nó đợi tôi đủ kiên nhẫn để hiểu.”

If an empty report is more useful than a full one, what exactly are we paying for?