Blank Cells on the Scoresheet: Vietnamese Badminton and the Data Story Nobody Has Written

Three Cells Out of Forty-Seven

In September 2026, in an arena in inner Ho Chi Minh City, I opened my laptop just as the semifinal entered the third game. On screen was a spreadsheet with forty-seven columns that I had built over many years: average rally length, net approach rate, efficiency of the third shot, unforced errors from eighteen points onward, performance after the eleven-point interval, average shuttle speed in the deciding game, change-of-direction counts, and forty other things.

When the umpire awarded the final point, I had three cells filled.

Three out of forty-seven. A semifinal in the BWF World Tour system, with international officials, an electronic scoreboard, television cameras, and a full stand. Three lines. They recorded the score of each game, the match duration, and a figure for the longest rally that I could not fully verify because no display confirmed it.

Forty-four cells remained empty. They were empty not because I was lazy. They were empty because no source existed to fill them.

One empty cell is an unanswered question. Forty-four empty cells are a badminton nation that has never been read.

I am writing this to answer a question almost nobody has asked me in thirty-one years of work: if a badminton nation has no data, what happens to it over the next ten years?

A Funnel Called the BWF World Tour

To understand why blank cells appear, you have to understand the tournament structure. The BWF divides professional competition into Super 1000, Super 750, Super 500, Super 300 and Super 100 tiers, with International Challenge, International Series and Future Series below. Each tier has a minimum prize fund and its own ranking points. Under the points table used in the most recent cycle, a Super 1000 winner earns 12,000 points; a Super 750 winner 11,000; Super 500 9,200; Super 300 7,000; Super 100 5,500. Lower down, an International Challenge winner earns about 4,000, an International Series about 2,500, and a Future Series about 1,700. The World Championships and the Olympic Games sit at the top, around 13,000 points for the champion.

Anyone can look those numbers up. What few notice is that match data does not follow the points table. It flows downward like a funnel, and that funnel narrows very fast.

At Super 1000 level, every court has instant-review cameras, a broadcast director, on-screen statistical graphics, and dedicated data operators. At Super 300, usually only a live score feed and a post-match summary remain. By Super 100, most events stop at an electronic scoreboard and a text results file. At International Challenge level, the only reliable artefacts are the umpire's score sheet and a draw sheet taped to a corridor wall.

Vietnam's biggest annual international event, the Vietnam Open, sits at Super 100 level. The second event, the Vietnam International Challenge, sits lower still. Both of Vietnam's main international stages lie exactly in the zone where data disappears.

I have followed matches at this level for years. The feeling is always the same: after each match I know who won, by how much, and how long it took. I do not know why they won. I do not know which rally turned the match. I do not know what percentage of rallies over ten strokes each player won. I do not know who attacked and who lifted safely at eighteen-all. The things that decide results lie outside the record.

This is the fundamental difference between badminton and tennis. Tennis publishes point-by-point data for every major event, and that data becomes a public asset for analysis, debate and coaching. Badminton does not. The BWF publishes results and some aggregate statistics such as error counts, longest rally and fastest smash, but not stroke-level data. Below the top tiers, even aggregate statistics are incomplete.

The result is a paradox: badminton has the highest decision density of any racket sport — a single match can contain more than a hundred rallies, each with dozens of strokes — yet it has the least public data.

And Vietnam sits at the bottom of that funnel.

Nguyen Tien Minh and a Memory Without Tables

To see the consequences, look at the best player we ever had.

Nguyen Tien Minh was born in Ho Chi Minh City and entered the international scene when most rivals barely knew Vietnamese badminton existed. He rose to world No. 5 in 2026 and won a bronze medal at that year's World Championships in Guangzhou, losing to Lin Dan in the semifinal. He competed at four consecutive Olympic Games: Beijing 2026, London 2026, Rio 2026 and Tokyo 2026. That is a career spanning more than two decades at the top, something very few Southeast Asian players have achieved.

Today, if you want to study those four Olympic campaigns seriously, what do you have? You have match results. You have a few video clips. You have articles recounting emotion. You have no rally data. No stroke distribution per game. No correlation between rally length and point-win rate. No error profile by scoreline.

In other words, we know he won and lost, but we do not know how he played in any measurable sense.

That is a greater loss than it appears. A sport that cannot transmit method transmits only anecdote. When a player reaches the top without data, the next generation inherits admiration but not technique. They know the predecessor played a style built on stamina, rally tolerance and counter-attacking defence — but that is a description, not a curriculum. Nobody measured how much of it there was, at which scorelines it worked, or when it broke down.

When the whole world shouts, I read the spreadsheet again. This time the spreadsheet had nothing to read. That is the problem.

Old data is not wrong; it only tells the story of a dead era. In Vietnamese badminton, even old data never existed to be wrong.

Nguyen Thuy Linh, the Points Ladder and a Budget Problem

On the women's side the story arrives later but with clearer numbers.

Nguyen Thuy Linh, born in 2026 in Phu Tho, became the first Vietnamese woman to reach the world top twenty in the BWF rankings. Under the way the system works, that is a verifiable milestone.

But to read it correctly you have to look at the scoring machinery.

The BWF world ranking takes a player's best ten results from the past fifty-two weeks. A player's total is built from ten standout results, not from every match played. This reduces the influence of weak events, but it creates an entirely different game on the calculator: scheduling.

To sit inside the top twenty, a player needs roughly 45,000 to 55,000 points, depending on the time of year. Divided across ten counting events, that means averaging about 4,500 to 5,500 points per event. Against the points table: a Super 100 title is worth 5,500 points, a Super 300 runner-up about 5,950, a Super 300 semifinal about 4,900, a Super 500 quarterfinal about 5,040. In short, holding a top-twenty place requires consistently reaching quarterfinals or semifinals at Super 300 level and above, all year, across many destinations.

This is where a sporting problem becomes a budget problem — and where data becomes useful in the least expected way.

A season at that level means twenty to twenty-five international flights: Malaysia, Indonesia, India, Thailand, China, South Korea, Japan, Denmark, France, Germany, England, Spain. A player travels with a coach. Add flights, hotels, meals, local transport, entry fees, equipment, and training costs at home. The total I keep in my own sheet lands between sixty and eighty thousand US dollars a year for one player competing densely enough to chase points.

Against the prize fund of a Super 100 event at the low end of the system, that investment almost certainly cannot be recovered through prize money. It can only be recovered through three other sources: personal sponsorship, local government budgets, and media value generated by results. In other words, a top-twenty ranking is not only a ranking. It is an expenditure.

I do not believe in sentiment; I believe in time series. And the time series here says one thing clearly: every badminton nation that holds a top-twenty place has multi-year funding pathways rather than event-by-event funding. Vietnam mostly still works event by event.

But there is a data trap I want to dwell on longer, because it is rarely discussed.

The Points Bubble and the Comfort of the Top Twenty

When the scoring system allows accumulation across many tiers, it opens a shortcut up the ladder: play often, play widely, play where points are cheap and the field is thin. A player can move from world No. 100 to No. 30 by winning a few International Challenges, going deep at Super 100s, and scoring in early rounds at Super 300s. Arithmetically, the ranking climbs beautifully. Professionally, it can be a bubble.

I call it phantom points. Phantom points are not cheating; they are a legitimate output of a legitimate system. But when a player is ranked twenty-fifth in the world without ever having beaten a top-twenty opponent, the gap between ranking and level becomes a dangerous void. It is dangerous in three ways.

First, coaches misread the map. A number in a ranking table does not say that a player has learned how to beat elite opponents. When that player enters a Super 750 where everyone is inside the top thirty, the ranking provides entry but not experience of handling shuttle quality at that level.

Second, sponsorship is mispriced. While building a player-valuation model for a sports data company in Shanghai, I found a fairly durable rule: young players with high chance-creation metrics tend to be valued about thirty percent above their true worth. The cause lies not in the data but in how people read it — rankings and highlights are more visible than the quality of decisions in hard rallies. Vietnamese badminton faces exactly the same risk: a flattering ranking can price a player above their ability, and mispricing always charges interest later.

Third, and most important to me: when points are the only measure, people optimise for points. Scheduling is chosen to maximise points rather than maximise improvement. Every contract is a gamble, but the win rate lives in the spreadsheet — and here the spreadsheet is being built on the wrong criteria.

How Many Things in a Rally Are Worth Measuring

Now leave the ranking table and look at what actually happens on court.

A badminton court is 13.4 metres long and 6.1 metres wide at the outer sidelines. The net is 1.524 metres at the centre and 1.55 metres at the posts. A feathered shuttle has sixteen feathers and weighs between 4.74 and 5.50 grams, graded by speed from 76 to 79 depending on hall temperature and humidity. Those parameters have stood outside time for decades. The meta changes weekly, but the rules stand outside time.

In a standard rally at international level, stroke counts usually fall between six and twelve. Long rallies above twenty strokes are a small share of the total but consume most of the match's energy and heavily influence what follows. A top-level men's singles match lasts around forty-five to sixty minutes. Elite smash speeds exceed four hundred kilometres per hour; the record once verified stood at roughly four hundred and ninety-three kilometres per hour, and under modern measurement conditions such numbers are more spectacle than tactics.

So how many things in a rally are worth measuring?

I have built this list many times and always trimmed it. What I keep: rally-length distribution; point-win rate by rally-length band; net approaches and scoring efficiency when at the net; efficiency of the third shot after serve; unforced-error rate broken down by type — out of bounds, into the net, off the frame; performance from eighteen points onward; performance in the first five points after the eleven-point interval; and point-win rate when a rally exceeds fifteen strokes.

That list is not long. Yet in Vietnam, almost no tournament collects enough to calculate half of it.

The consequence is not a shortage of tables. The consequence is that human memory is heavily biased, and without data the bias becomes default truth.

First bias: spectacular rallies are remembered, quiet structure is forgotten. A smash into the corner gets retold many times. A twenty-stroke rally ending in an opponent's shuttle out of bounds is remembered by almost nobody, even though it is what produced the point. When coaching relies on memory, coaches teach the beautiful shot and ignore the accumulation.

Second bias: the final point is remembered, the sequence that produced it is not. In matches I have tracked, the eighteen-to-twenty zone is where everything turns. But because the scoreboard shows only the endpoint, spectators remember the error at twenty and forget that the player lost four straight points from eighteen. Fixing an error at twenty is a technical job. Fixing the cause of four lost points from eighteen is a data job.

Third bias: people trust winning, not how the winning happened. A 21-19, 21-19 win and a 21-11, 21-13 win go into the same result column. The physical cost, psychological stability and fatigue profile of the two are very different. Without data, both are simply a win.

Tactics do not live on the diagram; they live in how the data arranges itself. In Vietnamese badminton, the data has never been arranged, so tactics mostly live in a few heads.

Where the Money Goes in a Super 100 Event

To see what data costs, see where money goes.

Under BWF minimums for the most recent cycle, a Super 1000 event must offer a prize fund from roughly 1.25 million US dollars upward; Super 750 around 750,000; Super 500 around 420,000; Super 300 around 210,000; and Super 100 in the low six figures. Lower tiers such as International Challenge and International Series offer even less, sometimes only enough to cover minimum operating costs.

In a Super 100 event, most of the money does not reach players. It goes to venue, equipment, officials, medical staff, security, media and operations. Prize money for the deep rounds is a small share of total expenditure. And within that operating budget, the data line item is almost always absent.

That is a resource-allocation decision with its own logic. An organiser of a Super 100 must choose between three things: hiring an extra data recorder per court, increasing early-round prize money, or cutting ticket prices. For a decade, almost nobody has chosen the first. It is understandable: data is an investment you cannot see immediately, while prize money and ticket prices are felt instantly.

But the cost of not measuring does not disappear. It changes form.

It becomes coaching cost. A coach without data must review footage by eye, must remember, must guess. Every analysis session becomes a debate between competing memories, and the winner is whoever has the most authority in the room, not whoever has the best evidence.

It becomes the cost of training the next generation. A badminton nation with no data archive forces every new cohort to start from zero, inheriting nothing but oral description.

And it becomes opportunity cost, the largest amount that never appears in any ledger. Young players who are not analysed properly lose two to three years discovering weaknesses that data could have identified in two weeks.

I once spent a month learning to tell stories through people, after a failed broadcast night in which I presented a footballer's pressing numbers to an audience that did not understand them. The lesson applies to football and badminton alike: data does not speak for itself. That does not mean abandoning data. It means you need both a recorder and a translator.

When Provinces Raise Semi-Finished Products

This is the least discussed part of Vietnamese badminton, and in my view the most decisive over the long run.

In Vietnam, player development is tied to provincial units: Phu Tho, Bac Giang, Hanoi, Ho Chi Minh City, Dong Nai and a few other centres. A player begins in a provincial talent class, is funded by a local budget for seven to ten years, and then steps up to the national team. When that player achieves internationally, they become an asset of the whole sport.

The problem is structural: the province pays the development cost, while most of the value is captured upstream. This mirrors the loan-with-obligation-to-buy model in football — small clubs raise players, big clubs harvest them. Years ago I wrote that this model is destroying the finances of small clubs, and was understood to be talking only about football. I was not talking only about football. I was talking about any system where risk sits at the bottom and reward is drawn to the top.

In badminton, the concrete expression is this: a province invests ten years in a player, the player joins the national team and competes internationally on a federation schedule, and when the competitive cycle ends, the province still pays the salary for the entire post-career period. Meanwhile media value from the results is distributed differently.

Data can partially correct this imbalance — not by sharing money, but by sharing evidence. When every training session, match and recovery cycle is recorded to a common standard, a province's asset is no longer a name on a list but a quantifiable file. Such a file has value in budget negotiations, sponsorship applications, and proof of public investment efficiency.

Put another way: data is how the bottom layer keeps part of the value it created.

Correlation Is Not Causation

At this point I must argue against myself, because that is mandatory for anyone who reads numbers for a living.

There is an obvious correlation: every strong badminton nation has good data systems. China, Japan, South Korea, Indonesia, Denmark, Malaysia, Taiwan, Thailand — all fund analysis, all have data departments inside training centres.

From that correlation it is easy to conclude: to be strong, you must have data. I think that conclusion is true but useless in its bare form, and potentially harmful if used as a basis for budget allocation.

The reason lies in causal order. In strong badminton nations, data arrived after three things already existed: a playing population large enough to generate internal competition; a coaching corps thick enough to consume information; and a domestic tournament circuit dense enough to produce samples. Data is a multiplier, not a base. Multiplying a large number by an effective multiplier works well. Multiplying a small number by the same multiplier still yields a small number.

For Vietnam, this means buying an analytics system and expecting medals is a low-probability trade. The prior task is to increase competitive density and recording density at the same time, at moderate scale, over a long enough period for data to mean something.

One further point matters more than all of this: more data does not automatically mean more understanding. A bad table can generate false confidence faster than an empty one. An empty table makes people cautious. A bad table makes people act wrongly with an air of certainty.

There is one more entry in my self-rebuttal, and it makes me careful about my own position.

When global competition stopped in March 2026, every model I had built on historical data became useless within weeks. I tried to collect data from online training sessions of a club in Shanghai and got only a handful of data points per week — not enough to run anything. When I sent a report on post-lockdown fitness decline, the reply was that they needed solutions immediately, not long-term research. For the first time in my career, I admitted that data is not omnipotent.

That did not make me believe in data less. It made me distrust people who present data without stating its limits.

Limits of the Data

Every model, even a good one, has blind spots.

First, drift inside the hall. Badminton is a sport where a shuttle weighing under six grams is directly affected by air conditioning and airflow. The same stroke by the same player in two different arenas can produce two different outcomes. No current metric captures this fully.

Second, psychology in the decisive zone. Data can show that a player loses many points from eighteen onward. It cannot say why. It could be fatigue. It could be fear. It could be a small change in shuttle selection. Those three causes require three different interventions, and no table distinguishes them.

Third, injury and load management. A player entering a match at ninety percent condition produces data identical to a healthy player performing below capacity. No model separates the two without biomedical data attached.

Fourth, luck. A shuttle clips the net and falls on the other side, an instant review wins by a millimetre, an umpiring call goes against you. In a match decided by a few dozen points, a handful of random moments can change everything. Data describes probability; it does not describe the fate of one particular afternoon.

I have appended this section to every analysis since 2026. Not to reduce accountability, but to keep readers from being led by a number.

Statistics quantify a match, but they cannot quantify the heart of a fan. And in a country where every badminton medal is remembered for a long time, the unquantifiable part weighs more than the quantifiable one.

Signals for the 2028 Cycle

So what is worth tracking in Vietnamese badminton on the road to the 2028 Olympic Games?

The first signal is recording density. How to observe: count how many domestic tournaments publish detailed rally-level data files. Trigger condition: at least one national-circuit event publishes rally-level data openly. Expected impact: this is a hinge change, because it creates a domestic sample for every later analysis.

The second signal is scheduling density. How to observe: the number of entries at Super 300 level and above secured by Vietnamese players each year. Trigger condition: exceeding ten entries per year in both men's and women's singles. Expected impact: past that threshold, samples become thick enough to discuss trends rather than individual matches.

The third signal is stability in the closing zone. How to observe: point-win rate from eighteen onward among leading players. Trigger condition: sustained above fifty percent across three consecutive tournaments. Expected impact: this is the earliest indicator that a player is ready for knock-out rounds at a higher level.

The fourth signal is player movement between provincial units. How to observe: how many players transfer between training centres and how many stay long-term. Trigger condition: emergence of co-ownership or benefit-sharing mechanisms between units. Expected impact: without such mechanisms, the semi-finished-product model will keep eroding incentives to invest at the bottom.

None of these four signals requires a large budget to track. They require one decision: to record.

Who Will Record the Next Rally

I return to my forty-seven-column spreadsheet.

After that semifinal I thought for a long time about deleting the empty columns to tidy the sheet. In the end I kept them. I left them blank and wrote a small note beside each one: no source yet.

Because a sport needs to see its gaps before it fills them. Those forty-four empty cells are a map of the next ten years, if anyone is willing to read it.

The next rally will happen on some afternoon, in a hall with a draft and slightly harsh lighting. It will last eighteen strokes, end with a shuttle out of bounds, and nobody will remember it the following day.

Unless someone writes it down.

The question I leave behind is not for coaches, and not for the federation. It is for the person in the seventh row, with a laptop, with time, and with enough patience to fill in one empty cell.

When the whole world shouts, I read the spreadsheet again. Vietnamese badminton's problem is that nobody has written the spreadsheet yet.