Vietnamese Table Tennis and the Data Void: Reading the SEA Games 31 Gold Medal Through Numbers That Were Never Recorded

In May 2026, at the sports hall in Hai Duong province, Dinh Anh Hoang and Tran Mai Ngoc ended the mixed doubles final of the 31st SEA Games with a short ball on the left side of the table. The stands erupted. I was standing in the technical area, notebook still in hand, and the first thing I did after leaving the arena was open my phone to look for match data.

Three metrics mattered to me: the pair's win rate on their own service sequence, the average rally length, and the win rate in the first three shots. None had been published. That evening's coverage had photographs, commentary, tears, and exactly one form of data: the set score. A gold medal Vietnamese table tennis had waited years for, and not one line of data was preserved so that anyone could learn from it later.

I am not writing this to assign blame to a person or a federation. It comes from a professional habit: after every major match I ask myself the same question — if I could keep only three numbers from this match for the next generation, which three would I choose. For most table tennis matches in Vietnam over the past several years, the answer is that there is nothing to keep, because nothing beyond the score was ever recorded.

Based on my experience watching matches at the national championship, at the quadrennial national sports congress, and at several rounds of the WTT circuit, I treat this gap as a structural problem rather than a minor technical one. It determines how far the country can go.

Table tennis has the shortest feedback loop of any mainstream sport, so it requires the highest data resolution, not the lowest. A match lasts 25 to 40 minutes, consists of three or four winning sets, and each set runs to 11 points. The total number of points in a match typically falls between 150 and 300. Compared with a 90-minute football match containing thousands of passes, the number of events here is far smaller, but the weight of each event is far greater. A point at 9-9 is not equivalent to a point at 2-0.

Because the number of events is so small, match variance runs high. Matches end 3-0 on sets with a total point count of 33-31. Read only the set score and you describe that match as comfortable, when in reality it was decided by four balls that missed a line by millimetres. That is the first reason a table tennis nation serious about improving must measure more than the scoreline.

The timing of this piece coincides with the registration period in which clubs finalise squads for the domestic season and players lock in WTT entries. Table tennis has no transfer window in the football sense, but it has an equivalent in consequence: the entry list determines who gets on the table, who accumulates ranking points, and who drops out of the points cycle. That is where the real story sits, not in commentary about form.

Before the numbers, the international system needs reconstructing, because every analysis of Vietnamese table tennis is conditioned by it.

The ITTF world ranking has operated on a new mechanism since 2026, tied to the WTT circuit created by the ITTF. Ranking points are drawn from a player's best results within a rolling 12-month window, and older points expire when their cycle ends. This creates what analysts call points-defence pressure: a player with a strong result last year must repeat or exceed it this year or lose the points already banked.

The consequence for Vietnamese players is direct. A player who competes in four to six international events a year has at most four to six results available for selection. A player from a country with dense competitive infrastructure, competing in 12 to 15 events, has a much larger pool to choose the best from. In a system built on best results, whoever has more access to events holds a structural advantage. The world ranking of a Vietnamese player therefore reflects both ability and access, blended in a way that is very hard to separate.

Ranking under the current system is a compound index: it measures ability and it measures logistics at the same time. Domestic coverage almost never mentions this when comparing Vietnamese players with regional rivals.

Now the core question: if a federation wanted to build data infrastructure from zero, what should it measure. I split it into three layers, each with a one-sentence explanation so anyone can read it.

The first layer is point structure. The two most important metrics are the win rate on own service and the win rate on receive. In plain terms: out of ten service points, how many does the player win, and out of ten receiving points, how many does the player take. At elite level the gap between the two is usually narrow, and that gap is the measure of whether a player owns a genuine service weapon or is merely living off luck in later exchanges.

At the domestic Vietnamese events I have observed, that gap tends to be wider than the international average. Players with a pronounced service advantage win quickly at home but struggle abroad, because international opponents read spin better and handle short balls better. This is a domestic data trap: strong numbers at home do not convert into strong numbers away.

The second layer is rally shape. The two metrics to record are average rally length and the share of topspin-to-topspin exchanges in total points. In plain terms: how many strokes an average point lasts, and whether most points in a match are settled in open exchange or in short play. In modern table tennis, average rally length for men's singles typically sits around three to five strokes depending on level and equipment. Matches with long average rallies are those where both players want the exchange; matches with short averages are decided in the first three shots.

The third layer is the opening sequence. The central metric is the win rate inside the first three shots, counted from the moment the ball leaves the server's hand to the completion of the third stroke. This is the single most important metric in modern table tennis and the hardest to collect, because it requires a human to classify each rally rather than merely record the score. Derived metrics follow: third-ball attack rate, receive-by-attack rate, and error rate inside the first two strokes.

Those three layers form a toolkit sufficient to describe almost an entire table tennis match. The problem in Vietnam is that none of them is recorded systematically.

To see the distance, look at how established table tennis nations operate. Japan built a national training centre tied to video capture and opponent analysis over many years. China runs a closed internal data system in which every internal training match is recorded and classified. Europe, particularly Germany and France, benefits from professional broadcast club leagues, meaning every domestic match automatically generates data. Generations such as Truls Moregard or the Lebrun brothers grew up in an environment where match data was the default rather than something to be requested.

In Vietnam, table tennis data exists almost exclusively as scores. The national championship produces results, player lists and standings, but no rally data. The quadrennial national sports congress, which gathers nearly the entire domestic talent pool, generates an enormous volume of matches that no institution mines for a long-term database. This is the most wasted resource in Vietnamese table tennis, more significant than any training budget shortfall.

A sport that does not archive its own data will always restart from zero each cycle. Every generation of players leaves and takes with it the accumulated understanding of technique, opponents and pressure handling, leaving behind no inheritable record.

Look at the national squad structure and generational risk becomes visible. Vietnam's core players over the past decade include familiar names such as Nguyen Anh Tu, Doan Ba Tuan Anh, Mai Hoang My Trang and Nguyen Khoa Dieu Khanh, with the succeeding cohort tied to the SEA Games 31 gold generation in Dinh Anh Hoang and Tran Mai Ngoc. The numerical gap between the core group and the next group is measurable, and on the index I usually reference — the VangBong.vn Player Depth Index, which counts players aged 18 to 24 with international results in the past 24 months — Vietnamese depth is markedly thinner than regional peers with functioning junior circuits.

This carries a competition-structure consequence. Qualification for the Los Angeles 2028 Olympics will open progressively through the 2026-2027 cycle, with places allocated mainly through world ranking and qualifying events. A player who wants to be there must accumulate points inside a narrow window, which requires an international calendar designed months in advance. That is a logistics and data problem, not purely a talent problem.

At this point I have to say what I always say at the end of a long analysis: data is not a cure-all.

The correlation between having more data and winning more medals is not causal. China dominated world table tennis before sports analytics existed, through selection systems and training intensity. India and Brazil have invested in analysis in recent years and still run into limits of squad depth. Japan has good data infrastructure but still waited years for a men's generation capable of competing at the top. Data is a necessary condition in some problems and never a sufficient one for results.

There is another risk I call data theatre. When a federation decides to invest in analysis, the pressure to show results often leads to buying foreign models and applying them to domestic data without calibration. A model built on WTT-level data, where service quality and match speed sit on a very different plane, will produce wrong conclusions when applied to a Vietnamese domestic event. I have seen analysis tables that looked highly professional, full of English acronyms, where checking the input source revealed everything had been inferred from set scores. That is decorative data, not analysis.

Form is an illusion; only a numeric series is a real current. But to have a real series, someone must sit and log every rally, and that is work without glory.

Finally, what the numbers cannot say.

Data cannot measure the moment a young player realises they will never reach world class, after years of training at an intensity most spectators cannot imagine. Data cannot measure the pressure on a family when a child chooses professional table tennis in a country where education and sport are weighed on an unbalanced scale. Nor can data measure the feeling of walking to the table at 9-9, when every career-long statistic about service points becomes meaningless in front of one specific ball resting in the palm.

I keep an old line: "Numbers never lie; only the people reading them fool themselves." I also keep a line on the opposite side: data is the penance of those who once trusted feeling, and that penance has its own limits. Let data replace human judgement entirely and we simply move from trusting intuition to trusting a model we do not understand. Both are different ways of refusing to think.

When the world shouts emotion, I choose to listen to the chart. But whoever listens to the chart must answer for what they chose to draw.

So which signals should be tracked going into the next cycle?

Three. First, the progress of Vietnamese players in the WTT points system: specifically the number of events entered and the number of results counted inside the 12-month window, because that index reveals access. Second, whether any institution publishes an open rally dataset for a national championship or national sports congress, even in its simplest form. Third, the speed at which the young cohort enters small international events, because that is the only early indicator of next-generation depth.

Reputation can vanish in a night, but a data model stays. Vietnamese table tennis has already had its famous night in Hai Duong. The remaining question is whether we want to turn such nights into something countable, so the next generation does not have to start again from zero.