The Empty Cells in Vietnam's Badminton Data

My spreadsheet has twenty-six columns. I can fill seven of them myself. The remaining nineteen sit blank, and they are blank through no laziness of mine. On a mid-September night in 2026, in a small apartment in Nha Trang, I rewound a main-draw match from the Vietnam International Badminton Championships. The match lasted 52 minutes. Rebuilding it into data took me nearly four hours. On the 41st stroke of a second-game rally, the home player reversed the shuttle from right to left with a cross-court whip. The shuttle landed tight to the sideline; the opponent was rooted. No line-camera system recorded that angle. No official statistics sheet preserved it. So I wrote it down by hand, into six small columns of a date-named file. What I understood afterwards had nothing to do with that player's technique. Vietnamese badminton analysis is short of raw material. Short enough that many people in the trade have quietly accepted it as a natural condition, like Nha Trang humidity or peak-season airfares. To see why, look at the data map of this sport. Professional badminton runs on the 21-point rally-scoring system, adopted by the Badminton World Federation (BWF) in 2026. Every rally ends in a point; there are no second serves. On paper this makes badminton ideal for data work: each exchange has an unambiguous outcome. But the tournament system is tiered, and the data tiers with it. At the top, the BWF World Tour Super 1000 and Super 750 events, including the All England, the China Open and the Indonesia Open, install Instant Review, high-speed cameras and publish stroke-level statistics. The middle tier, Super 500, Super 300 and Super 100, offers only basic numbers: points, a few aggregates, no stroke-level data. Below that, International Challenge, International Series, Future Series and almost the entire domestic calendar, offers nothing at all. The Vietnam International Championships sits in the second tier. Since 2026 it has been a BWF World Tour Super 100 event. That means a place on the world map, ranking points and international entries, but no data room. Beneath it lies the foundation where most Vietnamese players actually grow up: the national championships, the junior circuit, provincial opens, training-centre friendlies. There, points are counted on paper with a marker pen. Nguyen Tien Minh reached No. 5 in the world in 2026 and took bronze at the 2026 World Championships in Guangzhou, per BWF records. He competed at four consecutive Olympic Games. Yet if you want to look up how he distributed his smashes in a 2026 national semi-final, you will not find it. That data never existed, not because it was lost, but because nobody recorded it. Nguyen Thuy Linh broke into the world's top 25 in women's singles and played two consecutive Olympics, Tokyo 2026 and Paris 2026. The matches that lifted her there were mostly played in tiers with no data room. This is the starting point of every Vietnamese badminton analysis: a large void, not a dense table of numbers. Since 2026 I have hand-charted more than 200 badminton matches across every level, from provincial junior halls to packed Super 100 arenas. The method is nothing mystical. For each rally I record six columns: server, serve type (short, high deep, flick), rally length in strokes, rally winner, ending type (direct smash winner, opponent error, shuttle out, net contact), and the court zone where the shuttle landed. Those six columns are enough to rebuild a match's architecture. You can plot rally-length distribution, see who wins short rallies, who survives long ones, and see where errors come from. The other twenty columns vanish. Shuttle angle, shuttle speed, distance covered, jump count, spacing between movements. Everything that separates the world No. 40 from the world No. 4 lives outside every Vietnamese database. Based on my experience tracking these matches, this is domestic badminton's biggest bottleneck, bigger than the training-budget conversation. A badminton system without data coaches on faith. Faith is precious, but faith does not fix a faulty serve at 19-19. So I built my own measures. The Tempo-Switch Index (TSI). Picture a badminton rally. There are always two states: one side attacking, one side absorbing. Spectators sense it within three strokes. Nobody measures it. TSI works like this. For each rally, I identify which side holds the attacking initiative after the third stroke. Then I count whether the defending side regains the initiative within the next five strokes. The share of successful regains out of total rallies in which that player was passive is that player's TSI. Concretely: if a player is pushed onto the back foot in 60 rallies and flips the initiative in 21 of them, the TSI is 35 percent. My sample here is 412 rallies, drawn from four men's singles and five women's singles matches at national level and at BWF World Tour Super 100 level. I re-charted the whole set twice, in October and November 2026, to reduce coder error. The discrepancy between the two passes was 4.7 percent of rallies. And here is the part I must state before anything else: with 412 rallies charted by one person, the error margin remains wide. Everything below is a hypothesis awaiting verification, not a law. The index is useful only for comparing players within the same tournament, same court, same shuttle conditions. Carrying it across continents is self-deception. The second measure is simpler, which I call the Dead-Zone Index. I divide each player's half-court into six zones: two front-court, two mid-court, two rear-court. Whenever a rally ends, I mark the zone where the shuttle landed. Across 412 rallies, one pattern held fairly steady in the women's singles group I charted: the left mid-court was almost empty. Fewer than 6 percent of rallies ended there, while the right rear-court accounted for nearly 31 percent. There are two readings. First, that is where the cross-court smash naturally travels, since most players are right-handed. Second, that is a zone very few Vietnamese players deliberately target. This is where data touches emotion. When you watch a nineteen-year-old waiting in the left mid-court for a shuttle that, by the numbers, will almost never arrive, you understand why she loses. She is not slow. Nobody ever drew her that map. At junior level it gets harder. A fifteen-year-old may jump higher than peers simply because puberty arrived six months earlier. Build indices on that group without normalising for biological age and you will pick the wrong athletes. It is a classic scouting error, and it thrives where longitudinal data, tracking the same player across years, does not exist. TSI and the Dead-Zone Index share one blind spot. Both measure actions taken, not intentions. A cross-court smash may be the best option the player saw, or the by-product of feet that failed to move. From outside, the two look identical. Telling them apart requires movement data, available only at the top tier and only for the world's top thirty players. That is why I tell young analysts one thing: do not learn badminton analysis by rewatching Super 1000 finals. You will learn how a superstar handles a shuttle, but not how a world No. 60 survives qualifying. Two different problems, two different datasets. If a coach asked me four questions, I would answer none of them. At which stroke of the rally does this player lose most points? Nobody counts. After dropping the first game, how does he change his serve type? No record. At 18-18, what share of his choices are attacking? It does not exist. And from which minute does his physical output decline, expressed through which indicator? Unmeasurable. Those four questions are four empty columns in my file, and I suspect every national-team coach has wanted the answers. People remember the winning smash; I remember the twelve strokes before it. Those twelve strokes decide the match, and they are exactly where Vietnamese data is emptiest. A belief is spreading fast through the region's sports-analysis circles: more data means more accurate analysis. It sounds reasonable and fails at one crucial point. Instant Review does not measure truth. It measures what it was designed to measure, with its own error, and the rest of the match lies beyond its reach. When we pour all our attention into the fully instrumented tiers, we quietly create survivorship bias: conclusions about world badminton are drawn from thirty elite players and then applied to thousands of others. Another trap is familiar to anyone who has done statistics: correlation is not causation. A player with a high smash-winner count is not necessarily a better smasher. He may simply have faced three weak opponents in a row. Or the match may have been played in an arena with the air conditioning running too hard, making the shuttle fly faster. Shuttle speed depends on temperature and humidity, and in a hot, humid city like Ho Chi Minh City that factor is large enough to distort any cross-tournament comparison. Finally, an empty cell is data too. It says nothing about the player. It says something about us: which tournaments get instrumented, who gets recorded and who gets left out. Numbers do not lie; they stay silent until you learn how to listen. But silence sustained too long becomes a habit, and that habit is shaping how an entire generation of Vietnamese players sees itself. I entered this trade through a spreadsheet, but I stayed for the stories inside it. Vietnamese badminton's story is unwritten for a simple reason: nobody has built enough columns to write it. Every season is a lifetime of practice; every error term is a session of meditation. The next chapter will not be decided by any new index of mine. It will be decided by someone at a provincial training centre staying behind after practice, opening a blank file and recording the first stroke.