The Blank Report: Reading Badminton Through What Nobody Records

In March 2026, at a small arena, I sat in the fourth row and looked up at the statistics screen. No shuttle speed. No net-point win rate. No rally length. Just two players' names and a score ticking over.

I opened my laptop. A twelve-page analysis file I had prepared the night before sat there, neatly tabled, every section with a row waiting for data. Almost every cell came back with the same word: N/A.

Fifteen years in this trade, I have written hundreds of analyses for football and badminton. For the first time, I sat in front of an analysis with nothing to analyse. And I realised that was a match too — a match between the analyst and the silence.

Numbers do not lie; they only stay silent until you know how to listen.

A sport narrated by 5% of its matches

Professional badminton runs on a clear hierarchy. Since 2026, the Badminton World Federation (BWF) has staged the BWF World Tour across Super 1000, 750, 500, 300 and 100 tiers, plus the BWF World Tour Finals. Above the individual circuit sit the team events: the Sudirman Cup for mixed teams, the Thomas Cup for men, the Uber Cup for women.

At the top tier, data is rarely missing. The All England — first staged in 1899 and the oldest Super 1000 — has Hawk-Eye cameras, a live statistics feed, shuttle speed, landing-point maps. So do the Indonesia Open, China Open and Malaysia Open. A single Super 1000 semi-final can generate thousands of individual data points.

Lower down, the picture changes. A Super 100 qualifying match usually offers an electronic scoreboard and a fixed-angle YouTube stream. No Hawk-Eye. Nobody records how many centimetres shorter a player's second serve is than the first.

The gap between those two worlds is where I work. And it shapes almost everything spectators believe they know about this sport.

For Vietnamese players such as Nguyễn Thùy Linh or Lê Đức Phát, having to compete across every tier means living in both data worlds: one week analysed through thousands of data points, another week reduced to a single scoreline.

Because the media does not stratify the same way. An article about the modern attacking trend in badminton is usually built from a handful of Super 1000 events. That conclusion is then applied to a system of more than two hundred tournaments a year. We narrate badminton through the 5% of matches that are fully recorded, and call it the truth.

I have made exactly that mistake. In 2026, when the World Cup was played in Russia, I was so immersed in a detailed event dataset that I believed every football match could be measured the same way. Germany touched the ball 735 times against South Korea, three times their opponent, and I used PPDA — the number of passes an opponent is allowed before each pressing action — to show that the team's possession was futile. The metric was beautiful. It was persuasive. And it made me forget that most matches in the world do not have a tenth of that data.

Measure the report, not the match

In that small arena, I decided to do something unusual. I stopped trying to analyse the match and started analysing my own report.

I built a measure called the Information Density Index (IDI) — the number of independent data points recorded per minute of play. The calculation is simple: count the distinct data fields the system or the human recorder actually fills in, divided by total minutes of live play. A field counts as independent if it cannot be derived directly from another.

I ran it on my own log — 40 matches scattered across the 2026–2026 season, from Super 100 qualifying to Super 1000 semi-finals. It is a small sample, and I will state its limits below.

In the Super 1000 group, IDI hovered between 60 and 80 data points per minute of live play. In the Super 300 group it fell to roughly 8 to 12. In the Super 100 group it hit bottom: 2 to 3, essentially just the score and rally duration.

The distance between 80 and 2 does not measure badminton quality. It measures visibility.

Then I added a second measure, the Blank Coefficient (B) — the share of empty cells in a match report. A complete report has B near 0. My twelve-page report that night had B of roughly 0.71.

Here is what I learned: once B passes about 0.5, the report stops describing the match and starts describing the writer's ignorance. It becomes a confession formatted as a spreadsheet.

That sounds like a failure. To me, it was data.

In 2026, when the Bundesliga returned in empty stadiums, I met a similar kind of void. Average metrics jumped erratically: genuine attacking sequences rose, yet goals from set pieces fell about 22% year on year. Nobody could explain it, because the familiar data did not contain the answer. I had to build a new measure — the average distance between lines when a team lost the ball — and it showed that empty stands pushed teams higher, exposing space behind the defence. The emptiness of the terraces became a variable.

Badminton works the same way, only at a different scale. In lower-tier events, what decides a match is usually the first three beats: the serve, the return, and the third shot. That is where most errors and most tactical intent live. Yet no system records those three beats at Super 100 events, because fixed cameras lack the resolution and nobody pays someone to sit and count.

So we count what is easy to count. Smash winners. Unforced errors. They appear on screen because they are easy, not because they matter.

People remember the winning shot; I remember the twelve passes before it. In badminton, those twelve passes are twelve rallies nobody bothers to record.

I once watched a young Vietnamese player win a qualifying match in which the official statistics recorded exactly four lines. Four lines for seventy-two minutes of play. He won by constantly varying his serving rhythm, something no metric captured. A few months later, at a Super 500 event with full cameras, the same approach was read by his opponent and he lost quickly. Lower-tier data is not merely thin — it is thin in a way that makes us misjudge a player in both directions.

At the same time, I think of Morocco at the 2026 World Cup. Before their match against Spain, almost every prediction leaned toward the team averaging 68% possession. But when I dug into the defensive data, Morocco allowed opponents just 2.3 touches in their own box per match, the best in the tournament, despite holding only about 30% of the ball. That metric was not in the mainstream tables. I had to build it myself, and it was right.

A gap is not evidence

This is where temptation appears, and I have to warn myself.

When data is thin, an analyst falls easily into two symmetrical traps. The first is filling empty cells with inference and calling it a discovery. The second is subtler: worshipping the gap itself, treating everything unmeasured as the most profound thing there is.

Both are wrong. The absence of data proves neither quality nor mediocrity. It proves only that we have not looked closely enough. The correlation between having no numbers and mattering little is created by the observer, not by the match.

When the stands fall silent, every team strips off its mask. But an empty stand does not create truth — it only removes the noise that covered it. The listener still has to know how to listen.

This is also why I increasingly distrust the heat map. It is a visualisation so beautiful that people forget which data drew it, at which tournament tier, from how many observation points. A Super 1000 heat map and a Super 100 heat map look alike, yet their reliability differs by an order of magnitude. The heat map has become a new form of divination: the vaguer it is, the more readily it is believed.

And I have to admit my own limits. IDI and the Blank Coefficient are measures I built myself, run on a sample of 40 matches I logged myself. They are useful only for describing how available data is; they do not measure badminton quality. A match with a high IDI is not necessarily better than one with a low IDI. Anyone using these two indices to compare players has picked up the wrong tool.

A crowd's emotion is data too, and it is no less valuable. When a stand leans into a rally, that is information about a match's pull that no metric replaces. Critiquing emotional conclusions does not mean denying emotion. It means wanting that emotion to stand on firmer ground.

A signal for the next cycle

There is one signal I am tracking, and it will change how we read badminton.

Computer-vision systems are getting cheap very fast. What once required a control room with Hawk-Eye can now be approximated by a few commercial cameras plus a recognition model. If that reaches the Super 100 and Super 300 tiers, we will for the first time have data on the first three beats of thousands of matches that previously existed only in spectators' memory.

When that happens, what collapses is not the rankings but the stories we tell about them. Players dismissed as unremarkable because they only compete at lower tiers may reveal a skill set that top-tier data never touched. And conversely, some legends built on fully recorded matches may look thinner than we assumed.

Football is not short of miracles — but even miracles have a probability distribution. The same holds for badminton. The problem is that we have never had enough sample to estimate that distribution.

Every season is a lifetime of practice; every error term is a meditation. That blank report taught me nothing about the two players on court. It taught me that a sport narrated through a fraction of what happens on court will always leave behind champions nobody remembers. And the analyst's job, sometimes, is to sit still in front of an empty cell until you are certain what is missing — before rushing to fill it with a story.