Badminton's Data Void: What Story Is the World's Fastest Racket Sport Actually Telling?

I once stood in the press room of a Super 1000 event, holding the statistics sheet handed out by the organisers. It had the score of every game, the number of service faults, the total match duration. I needed one small fact: how long the longest rally of the match lasted, and where it fell in the deciding game. That box was blank. The media officer shrugged and said the system does not record at that level.

Years later, hosting a match day at the Sudirman Cup, I still carried the same question in my jacket pocket. At the same time, at a second-tier European football match, the expected-goals figure for every player appeared on my phone less than ten seconds after the final whistle.

The distance between those two worlds is the subject of this piece.

A tour that runs without a measurement system

The BWF World Tour currently comprises four Super 1000 events, six Super 750 events, nine Super 500 events and a larger group of Super 300 tournaments, plus a scattering of Super 100 events across Asia and Europe. The season stretches almost the whole year, from the Malaysia Open in January to the World Tour Finals in December, with the Thomas Cup, Uber Cup, Sudirman Cup and World Championships slotted in between.

The world ranking is calculated from a player's ten best results over 52 weeks. It is a fairly sophisticated points system: it rewards consistency, punishes players who only explode at a few big events, and it creates the points-defence pressure that anyone following an Olympic qualification race understands well. To earn a seeding at a Super 1000, to hold an Olympic place, a player must compete often and go deep in events that are anything but glamorous.

But that entire system measures outcomes, not matches.

If you want to know how Viktor Axelsen turned a third game around, at which point An Se-young changed the tempo, how many seconds Kunlavut Vitidsarn needed to recover between rallies in the final game — you have to rewatch and count yourself. The federation publishes draws, results, head-to-head records and ranking points. Data on rally length, shot distribution, court coverage and shuttle speed per rally barely exists in any public, queryable form.

There are exceptions, and they make the gap clearer. In heavily televised matches — the Olympics, World Championship finals, a few Super 1000 matches on the show court — the broadcast graphics occasionally flash a smash speed. But that is the image of a moment, not the data of a career. It looks good on air and vanishes the moment the programme ends.

Four reasons behind the void

The first reason is money. The total prize purse of the All England Open, badminton's oldest and most revered tournament, has hovered around 1.3 million USD in recent seasons. Divide that across a multi-camera motion-tracking system, plus operating, storage and data-interface costs, and it eats a significant share of a tournament's budget. European football spends billions of pounds on broadcast rights per cycle, which is why it can drop a few million on a data system as loose change. Badminton has no loose change.

The second reason lies in the nature of the movement. The shuttlecock has strange aerodynamics: it travels fastest immediately after leaving the racket — it has been recorded above 490 km/h under test conditions — then decelerates violently within a few metres. Such a trajectory demands very high frame-rate cameras and a tracking algorithm far more complex than following a round ball on grass. The line-call system has achieved this at a minimal level, but for a narrow purpose: determining the point of landing for a ruling. Expanding it into tactical data is a different problem, and many times more expensive.

The third reason is governance structure. The Badminton World Federation is an organisation whose resources are several orders smaller than FIFA's or UEFA's. National federations run their own tournaments and decide their own investment in analysis and filming. The result is an uneven ecosystem: a match in Denmark may be filmed carefully from several angles, while a qualifying-round match at a Super 300 in Asia may have a single fixed camera and one person operating the scoreboard.

The fourth reason belongs to storytelling culture. Badminton fans were raised on feel. We talk about the feel of the shuttle, about two players understanding each other from a single glance, about the silence in an arena before a serve at 20-19. Those things do not need numbers. They even resist numbers, because anyone who has sat in a badminton arena knows that what makes a crowd hold its breath is not an index.

I met Mbappé not to talk about football, but to learn how to count the silences before the ball rolls. That lesson came from a sport with far too many numbers. And here, in badminton, I have met a sport with far too few.

What I learned from counting on my own

Over two seasons, I logged the rhythm of more than three hundred matches I watched, mostly on television and a few times sitting in the arena itself. The method was crude: a notebook, a stopwatch, recording the length of every rally and the rest interval between rallies in the third game of long matches.

A few things emerged, and they made me rethink how I read a badminton match.

First, in elite matches, the rest interval between rallies is a bigger variable than I assumed. Players who win third games are often not the ones with the longest rallies, but the ones who best control the rest rhythm — stretching it when they need air, shortening it when they need momentum. That is data that a stopwatch can capture completely, and it exists in no public statistics sheet whatsoever.

Second, the quality of the lift decides matches more than the attacking shots do. The statistics sheets handed to the press at some events count only winners and errors. But a winner appears at the end of a sequence, while the lift raises the quality of the whole sequence behind it. Counting the last shot without counting the first is miscounting, and this miscounting is shaping how fans evaluate a player.

And the last thing kept me thinking longest: when a player loses their legs in the third game, the signs appear three to five points before the scoreline changes. If an index measured recovery time between rallies, it would predict the match outcome more clearly than any attacking statistic.

I checked this against matches with recorded footage. In most of the cases I watched, the player who dropped a run of points had begun stretching the preceding rest intervals a few points earlier. Not in every case. But often enough that I believe this is an index worth building seriously. And the surprising part is how cheap it is: a timekeeper and a spreadsheet, something any Super 300 tournament could manage.

When the pandemic emptied stadiums, the K League placed cardboard cut-outs in the stands and piped in crowd noise through speakers. I surveyed 120 matches that season and found home win rates fell by roughly six percentage points against the previous season. Fake spectators, fake noise, but the emptiness was real — I recorded it with my ears. Badminton is in a comparable position along a different axis: the feeling is real, but it is not recorded, and because it is not recorded it is easily replaced by cheap stories.

Lessons from sports that went first

Tennis was once in a position close to badminton's. Until the early 2000s, fans had only scores and a few basic serve statistics. Then line-tracking systems were introduced, landing data was digitised, and broadcast analysis became part of the viewing experience. Table tennis followed a similar road when its new tour structure forced events to standardise data for television.

What those two sports share is that they did not begin by hunting for a magic metric. They began by recording what already existed: where the ball went, where it landed, what the tempo was. The clever indices came later, once the data layer was thick enough.

Badminton is doing the opposite. We want a metric good enough to tell stories right now, while nobody will record the rally lengths of a second round at a Super 500.

The trap of borrowing someone else's toolkit

The instinctive response to a void is to demand that badminton get its own expected-goals index. I think that is a mistake.

Expected goals was invented to answer a specific football question: how well does a result reflect the quality of chances created. Football needs that question because its scorelines carry enormous noise — a match can finish 1-0 with fifteen clear chances for one side. Badminton has a different structure: every point is a closed contest, the point winner is the one who ends the rally, and the noise in the result is far lower.

Borrowing a tool designed for a different problem and bolting it onto a sport that does not have that problem produces an index that looks good on air and means nothing in an analysis room. Football has been through this: data did not make arguments smarter, it made arguments cheaper. People cite an index to end a conversation rather than to open one.

But there is a counter-argument worth weighing, and it is stronger than I first assumed. Without data, badminton cannot defend itself against false stories. When a player withdraws from a Super 300, the default story is lack of motivation. When a player performs badly three events running, the default story is a decline. When a women's singles player complains about the schedule, the public's first reaction is to doubt her attitude.

With no data on actual match load, on the real gaps between events, on how much physical capacity declines across games, fans can only speculate. And speculation always tilts negative, because that is how humans handle missing information.

From Eriksen collapsing to Su Bingtian standing still, I see a long track called loneliness. Badminton has that track too, but it is longer because nobody draws it out. A 27-year-old men's singles player competes in twenty events a year, flies across four continents, and when his body breaks down people call it a loss of form. An index on rest intervals and match load would make that sentence impossible to utter.

What will come first

I do not think badminton's first metric will be an attacking one.

It will be a metric about recovery. About rest. About rhythm. Because that is what this sport is actually decided by, and it is also the cheapest thing to measure. A Super 300 with a modest budget can still hire a timekeeper, still add rhythm data to the press information sheet, still archive it and open it publicly after each season. If such a system were standardised at federation level, it would change how fans read a third game more than any smash-speed graphic.

And it would change how we tell stories. Because then, a player losing their legs at 14-15 would no longer be a commentator's subjective observation. It would be a fact, and facts can be argued with, contested and corrected. That is the real value of data in sport: not to prove who is right, but to make unfounded claims more expensive.

The generation playing now — An Se-young, Kunlavut Vitidsarn, Kodai Naraoka, the post-2026 cohort — plays a different sport from the one the current data system was built to serve. Their rallies are longer, their games are heavier on the body, and the number of events they must play to hold ranking is greater than the previous generation's. When a sport changes faster than its measurement system, what gets left behind is the measurement system, not the sport.

Years from now, when someone sits in an arena at eleven at night, watching the third game of a final with no speed graphics, no analysis panel, only a notebook and a stopwatch — that person may be doing work more important than a whole room full of machines. Because the right question usually arrives before the data. And in badminton, the right question is still waiting to be asked.