Thuy Linh, V.League and the Empty Report: When Vietnamese Sports Data Is Never Recorded

Hook

2:14 a.m. in Nagoya. I opened the deconstruction file I had been waiting three days for. The first data column returned zero. The second did the same. By the seventh column I understood: this file was empty, and it was empty not because someone typed it wrong, but because there had been nothing to type in.

Twelve hours earlier I had been re-watching a women's singles match at a BWF World Tour Super 300 event. It went three games. The third game had twenty-one points logged on the electronic board, every one of them, without missing a beat. But the shuttle-trajectory record in my hands captured four rallies. Four out of twenty-one. Not a software fault. The venue simply had no tracking camera system, and the tournament could not afford to rent one.

I am not writing this to complain about a file. I am writing because that empty file is the clearest diagnosis of the sport I follow every week.

Context

Sports data analytics runs on an unwritten rule few outsiders know: a blank cell does not mean zero. It means an unanswered question. But when software exports a report, it usually writes 0, and the next generation of analysts reads that 0 as an event that actually happened on court.

In V.League, most clubs still stop at event data: goals, cards, shots, pass completion. Only a handful have fixed cameras capable of producing positional data. In badminton the tiering is even sharper. Super 1000 and Super 750 events have shuttle-tracking systems that measure serve speed, shuttle trajectory and rally length. Super 300 and Super 100 events — the main battleground for Vietnamese players for years — mostly offer nothing beyond the scoreboard and a few broadcast camera angles.

That is why a player like Nguyen Thuy Linh can sit inside the world's top thirty in the BWF women's singles rankings while the number of her matches recorded at rally level can be counted on one hand. She is not under-tracked because she plays little. She is under-tracked because most of the events she plays fall outside the equipped tier.

Nguyen Tien Minh is the longer case. His career spanned nearly two decades, with a career-high world ranking of No. 5, published in the September 2026 BWF rankings. Yet if I wanted to reconstruct the rally-length distribution of a full season of his, I have no data to do it. I have memories. I have scattered recordings. I have result sheets. I have no data series. To an analyst, the distance between those two things is exactly the distance between a map and a photograph.

Core

Based on my experience tracking matches, I teach my students one rule before they are allowed to calculate anything: count how much of the data you are missing. If the missing share exceeds 15 percent, every conclusion that follows is a guess in makeup.

In badminton, the four metric groups I open first are: serve-error rate at decisive points — not total faults, but the fault rate from point 17 onward, when pressure is high enough to reveal who can carry whom; the distribution of points lost by court zone, meaning how many were lost at the net, how many at the rear court, how many by hitting outside the sideline; average rally length per game, because it reflects fitness and how energy is rationed; and win rate in rallies longer than 15 shots, the sharpest measure of nerve in a 21-point scoring system.

No Super 300 event a Vietnamese player has entered offers all four. Usually only the second exists, and only in coarse form: total points lost, with no zone breakdown. That means when a coach wants to know where a student is dropping points, the coach has to sit through the footage and count by hand. That takes six to eight hours for a three-game match. With three matches in a week, nobody can do it.

In football the gap takes a different shape. V.League has enough data to answer who scored, who assisted, who fouled. But to answer something that sounds simple — how high does this team press — there is almost nothing. The metric I still use is PPDA, the average number of passes an opponent is allowed before possession is recovered. To calculate it you must log ball position and the position of every player on each pass. Without fixed cameras and recognition software, you have no PPDA. You only have a feeling that a team pressed hard or pressed loose.

How many matches in a V.League season are recorded well enough to calculate PPDA by half? The honest answer is very few, and almost none outside the top group of clubs. The weekly debate over whether a team counter-attacks well or badly still happens on television, but it is driven by impression, not evidence.

Cost is the first reason cited, and it is real. A full tracking system for one badminton court needs four to eight cameras, a processing server and a technician per competition day. The minimum total prize money the Badminton World Federation sets for a Super 300 is 250,000 US dollars; at Super 1000 that figure is five times higher. A Super 300 organiser must choose between renting a tracking system and paying players. Nobody can blame them for choosing the second.

But the accumulated damage is not located in one tournament. It sits in the fact that an entire generation of players is judged by impression, and when judgement runs on impression, people always favour what is memorable. A beautiful rally is remembered. An effective but ugly rally is forgotten. That is systematic bias, not personal bias.

I once lived inside that situation. In 2026, while working in the analytics department at Nagoya Grampus, I submitted a fourteen-page report on a young striker named Ryo Kato. His expected-goals figure was 0.82 per match, the highest in the squad. But he had scored only four goals in nine hundred minutes. My conclusion was clean: Kato was being pulled away from the box, where he hunted the ball best. The head coach dismissed it, on the grounds that Kato was too small against J-League centre-backs.

At the end of that season Kato moved to KV Kortrijk for 1.2 million euros and scored twelve goals in the Belgian top flight. My data was not wrong. I failed to communicate it. Fourteen pages persuaded nobody, because I handed people a table when they needed a story. Since then, every analysis of mine opens with a concrete on-pitch situation, and only afterwards pulls out the number to prove it.

In 2026 I was invited onto television for the World Cup in Russia as a data commentator. Before Japan played Colombia I said exactly one thing: across three qualifiers Japan had allowed opponents only 6.8 passes on average before recovering the ball, and if they held that level, Colombia would break early. Japan really did win 2–1. But the switchboard took dozens of calls complaining that I spoke in strange jargon. PPDA 6.8 is a number, and I am only the man who copies reality down — but viewers do not need a copyist. They need a translator.

In 2026, when the pandemic froze every league, my contract was cut by forty percent. I did not go looking for new data, because there was no new data to find. I re-watched 547 J-League matches from 2026 to 2026 and asked one question: what happens when a team leads at minute 70 and starts dropping deep? The result: once PPDA rises above 12, the probability of conceding an equaliser is 38 percent. 547 nights of matches taught me this: football froze, but the numbers did not.

I tried to run the same result backwards onto V.League. The question is unanswerable — not because Vietnamese football is different, but because I do not have 547 matches recorded at that level. Data is never in a hurry. It waits for me to be patient enough to understand it — but it can only wait if somebody recorded it in the first place.

There is one more subject I want to raise here, because it sits inside the area I track. Since V.League adopted VAR, the public data on reviewed incidents, intervention counts and decision-reversal rates remains thin. What stands out is that VAR's central clause — clear and obvious error — is itself a vague phrase. There is no quantitative definition of clear. Which means that even at the refereeing system's highest technological tier, the room for subjective judgement is wider than audiences assume, and we argue about it without even a unified dataset to compare.

Contrarian

There is an assumption I want to break: that missing data is missing evenly. It is not. Missing data has an order, and that order mirrors the power hierarchy of a sport.

Look at women's badminton. Women's singles draws get fewer cameras, fewer sponsors, fewer analytical pieces. Female players from countries without a strong badminton tradition are recorded even less. The result is that anyone trying to evaluate a female player has less evidence to work with and is forced back on impression. A sport that runs on impression will always reproduce the old hierarchy, and it does so unconsciously, so nobody is held responsible.

The same argument applies to esports. A women's competition run as a closed ecosystem, where teams only play each other internally, will never produce a genuine star — because there is no open competitive data to prove who is better than whom. The absence of open competition and the absence of data are two sides of one sheet of paper.

Here I have to warn myself. Precedent is a compass, but a compass also points wrong if you read it as prophecy. When I found that a PPDA above 12 leads to a 38 percent equaliser probability, that was a correlation inside one specific J-League dataset from 2026 to 2026. It is not a law. A team dropping deep can concede because the defence is weak, because the schedule is congested, because the referee adds four minutes, because the opponent brings on a better substitute. If I called that causation, I would have turned data into religion. Before concluding, I always ask myself: what structural factor sits behind this number, and if I remove it, does the number still stand?

In 2026, when Saudi Arabia beat Argentina 2–1, the world called it a miracle. I sat up one night, went back through the footage and counted five successful offside traps in the first half alone. The average distance between their two lines was 18 metres, exactly the method I had built in 2026. I wrote that it was a plan executed almost perfectly, and I was called cold, accused of stripping the match of its wonder.

I understand that reaction, and I think it is partly right. Belief, passion and the crowd's anger are things xG cannot measure. Since that episode I always add a small section called the limits of data at the end of every piece, spelling out what the numbers cannot see. And I no longer write certainly. I write most likely.

Takeaway

The signal I am waiting for next season is not a trophy. It is the first time a V.League club publishes complete positional data for an entire season to the public, or the first time a Vietnamese badminton player has a whole competitive season recorded at rally level.

When that day comes, someone will say it changes nothing. Nagoya does not read my reports, but data does not need a reader — it only needs someone willing to ask the right question. Every pass is an answer. I am only the man asking the right question. What remains is whether we record the question at all.