Vietnam's table tennis data void: every match leaves behind only a scoresheet

On a Friday night I sat in the seventh row of a domestic arena, opened my laptop and started logging every point by hand. Seventy minutes later I had 214 points, sorted by server, receiver, rally length and how each point ended. That was the entire dataset for the match, assembled by a man in the stands, because nobody publishes it. The scoreboard above the table showed set scores. The story inside each point had nowhere to be written down. A week earlier I had fed a table tennis event into the nine-dimension analysis routine I use in my consultancy work. The system returned a report with every heading in place, every table framed, every field empty. No player names, no match date, no detail to hold on to. The domain classifier ran correctly: table tennis. Nothing else ran. Looking at that screen, I recognised a fairly accurate metaphor for this sport in Vietnam. The gap with football is measurable. A single English Premier League match generates thousands of data events with coordinates, timestamps and attribution; the passing count alone is enough to rebuild a team's structure. A high-level table tennis match generates roughly 150 to 250 points, and each point can be split into seven or eight variables: serve type, placement, spin, point winner, rally length, ending type, score state. Multiplied out, that is more than a thousand observations in under an hour of play. Football has mined that seam almost dry. Table tennis has barely opened the box. Based on my experience tracking matches over 12 years, from the Table Tennis World Cup broadcasts I once anchored to regional junior events, most matches leave behind three things: the score, the ranking, and a few selected video clips. Ranking is the only universal number, and it is a lagging one — it reports what already happened, not why. The international ranking system runs on a rolling 52-week deduction: old points expire, new ones replace them, and a player who sits out three months slides down the list without losing a single match. That mechanism puts genuine pressure on scheduling, but it measures only the final outcome. It cannot tell you why a player won five straight and then lost three at the same event — a minor injury, a blade change, or an opponent who altered tactics. Serve-point win rate is the first metric I want at every domestic event. It answers exactly one question: what percentage of points does this player take when holding the initiative? At Asian level, 55 percent is generally treated as the floor for a solid player; below 48 percent and everything behind it has to be covered by defence. Receive-point win rate sits beside it, and the spread between the two tells you where a player is strong and where he is exposed. Without those two figures, every remark of the 'he has a good serve' variety is just the memory of someone sitting close to the table. The second cluster is first-three-shot structure. Modern table tennis is decided largely in the first two or three beats: serve, push return, and the finishing shot. A player who wins 60 percent of points on the third beat but loses 55 percent in long rallies is a completely different profile from one with the reverse numbers. Two such players can share a ranking, share a head-to-head record, and require two entirely different training plans. Rally-length distribution answers the question the naked eye misses: where is the match being dragged? A counter-hitting specialist will try to shorten rallies; a heavy-spin defender will try to extend them. Count that distribution and a coach knows which tempo the opponent is trying to impose, instead of guessing from a feeling that 'he looked slow today'. Then the pressure cluster. At 9-9 and 10-10, behavioural patterns shift sharply: many players choose a safer serve, surrendering initiative to avoid conceding directly. Win rate at level scores, comeback rate from behind, and unforced-error count in decisive rallies are three variables entirely separate from overall win rate. A player who wins 70 percent of matches but loses 60 percent of decisive points has a psychological problem, and it only becomes visible with point-level data. Without those numbers, selection runs on eyes and memory. A SEA Games place or a national team spot is usually settled by coaching observation plus results at two or three recent events. Small sample, large variance, and no way to audit the decision two years later. At youth level the gap is wider still: with no data to compare a 15-year-old in Hanoi against a 15-year-old in Da Nang, admission to a training programme often depends on who recommended whom. The counter-argument has to start here. More data does not automatically produce better decisions. Across 40 serves in one match, the standard error on win rate is roughly seven to eight percentage points; a five-point gap between two players is, in effect, noise. I have seen enough three-match stat sheets used to pronounce on an entire career. Numbers never lie; only the people reading them do. And this is where I have to talk about myself. When the system returned an empty report, my first reflex was to fill it in. Nine sections were framed, headings were ready, only the content was missing — and the writing engine was perfectly willing to produce polished judgements about players who do not exist. I stopped, but the lesson is clear: the greatest risk in this trade is confidence in the absence of data. Data is the confession of those who once trusted feeling, and a confession is only worth something when you know what you are confessing to. There is another layer rarely discussed. The table tennis data that does exist on international platforms mostly serves the broadcast product: who is leading, which shot was spectacular, which player is trending. It answers the viewer's question, not the coach's. Taking that dataset into tactical analysis is like using a weather bulletin to design a roof. Form is an illusion; only the sequence of numbers is real current — but that sequence has to be measured in the right place. A minimum proposal, small enough to start immediately. Six metrics per domestic match, published within 48 hours: serve-point win rate, receive-point win rate, first-three-shot win rate, rally-length distribution, win rate at level scores, and unforced errors in decisive rallies. No expensive camera system, no foreign consultants; two loggers, one standard template, and one publication rule. Those six numbers are enough for a coach to review a season with something other than memory. Reputation can be lost in a night; a data model stays. Vietnam's table tennis problem has never been a shortage of players or a shortage of shots; it is a shortage of a language to describe those shots in a way that can be checked. If next season every domestic match published just six figures, who would be the first to change a decision because of them?