Vietnamese Table Tennis and the Lesson of an Empty Data Sheet

In September 2026, I opened a spreadsheet with 42 columns and not a single number. It was the dataset for a domestic table tennis tournament that a client in Nha Trang had asked me to analyze. No point-win rate after serve, no first-three-shots index, no head-to-head data. All I had was a list of player names and the raw results of seven rounds. After 48 years of watching sport, this was not the first time I had met an empty dataset. But only now did I realize that most table tennis breakdowns circulating in Vietnam begin from exactly that void, and end with conclusions assembled systematically out of nothing. Score is only the verdict. The first-three-shots index is the testimony. Table tennis differs from football at one fatal point: every point is a closed event lasting a few seconds, recordable on the phone in your pocket. That is why many people assume the sport is easy to analyze. The reality is the opposite. A football match offers roughly 1,000 passes to build a model from; a table tennis match offers only 60 to 90 points, and each point is driven by dozens of micro-variables: spin, placement, ball speed into the table, footwork rhythm, and psychology at the final moment. With a sample that small, a wrong model fails faster and is harder to detect than any football model. I came to table tennis not out of pure passion, but because of data. In 2026 I shook Vietnamese football by publishing the first xG article for the V-League, showing that possession does not decide victory. Eight years later, I realized that principle holds even more strongly in table tennis, where people routinely confuse playing beautifully with playing to win. A perfect topspin loop can send the ball off the table; a safe push can win the point. If you cannot separate those two things with numbers, every judgment is just sentiment in makeup. What level is Vietnamese table tennis data infrastructure at today? Watching domestic tournaments and the WTT system over the past two years, I see three clear tiers. The first tier is raw results: game-by-game scores, who beat whom, available everywhere. The second tier is semi-advanced data: point-win rate on serve, point-win rate on receive, unforced-error rate, recorded in only a few places. The third tier is genuine tactical indices: the pressure equivalent in table tennis, the number of touches an opponent is allowed before you seize initiative; point-win rate on the third ball; and pressure-endurance index at decisive points. The third tier is almost empty. For a data writer, emptiness is not a catastrophe. Data does not forgive emotion. And that is why I converted. The only question is whether we dare to say there is not enough information to conclude. Most do not dare. They fill the gap with words. When I analyze a table tennis match, I start with the first three shots: serve, receive, and the third-ball attack. This is the decisive zone, accounting for roughly 55 to 65 percent of points in elite matches. If a player wins 70 percent of points across these three shots, the rest of the match is almost a formality. With Vietnamese players such as Nguyen Anh Tu and Dinh Quang Linh, I have counted back through video myself to verify rather than trusting ready-made tables. That discipline comes from the fact-checking lesson I learned in 2026: if you do not count yourself, you are counting someone else's errors for them. Alongside the first three shots are fitness and scheduling. This is where I have held the same position for years: match density is the single biggest culprit behind injury. A WTT system with back-to-back events across 52 weeks creates enormous pressure to defend ranking points. Players do not rest for fear of losing points, and the body pays. No medical team can save a player forced to play two matches a day for three consecutive days across three different time zones. Now let us talk about rules. Since 2026, the ban on hiding the ball during the serve has completely changed the structure of the first shot. The server lost the advantage of concealing spin direction, and the receiver gained time to read the ball. Any model that uses pre-2026 and post-2026 data without separating them is a garbage model. This is the textbook case of what I call a systematically wrong model: it is not wrong in one prediction, it is wrong in its entire foundation. Then there is the tournament system and Olympic qualification. Olympic berths are allocated through world ranking and regional qualifying events. This turns participation strategy into more than a purely technical matter: a player may have to choose between resting to recover or entering a small event to accumulate points. That decision is not made on the table; it is made on the coaching staff's spreadsheet. And it is precisely here that high-quality data creates the difference between an Olympic berth and a seat at home. The competitive landscape of Asian table tennis can be divided into four tiers. The dominant tier is China, where squad depth means even the tenth-ranked player domestically can reach the semifinals of an international event. The second tier includes Japan, South Korea, and Chinese Taipei with systematically trained next generations. The third tier is rising Southeast Asian table tennis nations, where Vietnam holds a notable position in certain disciplines. The fourth tier is the rest, where the gap in data infrastructure and physical conditioning keeps widening. If I had to name the biggest threat to Vietnamese table tennis, I would not choose a specific country. I would choose our own data void. Opponents can measure; we are guessing. This is where I must argue against myself. For years I believed that with enough data, everything could be predicted. An empty spreadsheet taught me the opposite: some things cannot be inferred from numbers, and trying to infer them is a form of arrogance. Correlation is not causation. A player who wins many points on the third ball is not necessarily strong there; it may simply be that opponents receive too poorly. If I merge those two into one index, I have created a systematically wrong model, and it will fail consistently, subtly enough that I will not notice until I lose money. The second trap is template imposition. I was born in Japan, where organization and hierarchy are revered, and I once tended to import European or Japanese models wholesale into the Vietnamese context. A mistake. A 17-year-old Vietnamese player training three sessions a week cannot be judged by the same index as a full-time professional in China. Before comparing, I must dedicate a separate original data analysis to the local context. Otherwise, every comparison is just lazy copying. I fear a wrong model more than a wrong judgment, because it is wrong systematically. And the first discipline against it is daring to say I do not yet know. When Vietnamese table tennis lacked data, I realized my model had taken root in a false context. The only way to fix it is to start again from what can truly be measured: raw results, point-by-point video, and the times I counted by hand rather than trusting prepared tables. So what are the signals for the next cycle? If you follow Vietnamese table tennis next season, watch three things. First, whether domestic tournament organizers begin recording first-three-shots data, a sign the infrastructure is shifting. Second, whether national teams adjust their tournament strategy under points-defense pressure, or still play every event as before. Third, whether a cohort of 16-to-19-year-olds emerges with a pressure-endurance index at decisive points good enough to step onto the Asian stage. As for me, I still keep that empty spreadsheet on my machine. Not out of regret, but as a reminder. An honest data writer is defined by what he refuses to write, not only by what he dares to write.