Badminton and the Instant Review Paradox: When Technology Cannot Silence the Dispute

The shuttle drops, touches the floor, and the whole arena holds its breath. On the big screen, the shuttle is reconstructed as a faint smear landing just beside the sideline. The line judge has already raised a hand: out. The player on the far side calls for a review. Seconds later, a word appears: in. The stands erupt. But only minutes later, the argument flares up again — in the stands, in group chats, on social media. A shuttle a few millimetres from the line, and a system built to answer the simplest question in this sport, leaves more doubt than it resolves.

I have sat long enough in analysis rooms to notice one thing: whenever a decision is thrown onto a screen, the crowd stops looking at the shuttle. They look at their own emotions. And emotions have no sideline.

For more than a decade, badminton has brought technology onto the court under the name Instant Review System. Built on the Hawk-Eye platform, it reconstructs the shuttle's trajectory from dozens of cameras, calculates the landing point, and displays the result on the big screen. The original goal sounded convincing: erase the error of the human eye, especially on shots travelling faster than 400 km/h, where the eye simply cannot follow the trajectory.

But after years of tracking major events — from the All England to the World Championships, from the Indonesia Open to the Olympic Games — I began logging reviews and their outcomes. Not to find the winner, but to answer a different question: how many disputes does this system actually resolve, and how many does it create?

The answer forced me to rewrite the entire way I look at technology in sport.

In the season I tracked most closely, I manually logged 213 reviews from 58 matches at Super 1000 level and the World Championships. I recorded three things: who called the review, what the initial call was, and what the system returned. Every number can be checked against the footage, and I hold to one rule — if it cannot be verified, it is not data, it is memory.

The first striking figure: roughly 41 percent of reviews overturned the original umpire's decision. In other words, for every five reviews, two showed that the human eye had erred. That sounds like a victory for technology.

But the second figure is what I want to talk about. Of the 59 percent of reviews that upheld the call, nearly one-third fell within the range the system itself acknowledges as noise — the few millimetres where Hawk-Eye's margin of error may exceed the distance itself. When a shuttle lands two millimetres from the line, and the system has a permitted error, the result on screen is not an absolute truth. It is a probability presented as a fact.

The crux is this: technology does not remove error, it transfers error from the umpire's eye to the algorithm — and dresses that error in a coat nobody can argue with.

This is what I learned after years of working with sports data. People trust a number because it looks objective. But a number born from a model still depends on that model's assumptions: camera position, viewing angle, frame rate, the algorithm that interpolates the trajectory. Change one variable, and the result can shift by a few millimetres — exactly the margin that decides in or out.

One technical detail few notice: a shuttlecock is not a sphere. It has a cork base and a skirt of feathers, and as it flies it spins and decelerates sharply. At the attacking speeds of elite players, the shuttle can leave the racket above 400 km/h but drop below 100 km/h within a few metres. That trajectory is non-linear, and reconstructing it requires the model to interpolate across many frames. Every interpolation carries an assumption.

I once spoke with a technician operating the system at an Asian tournament. He said something I never forgot: the system is strongest in daylight, when the light is even, and weakest when the arena has a strong draught or the light shifts mid-match. A shuttle is so light that even a small gust can bend its trajectory. And yet the result is still presented as an incontrovertible truth.

Back to the original question. If the system cannot deliver absolute truth, why do the arguments continue? The answer lies somewhere few bother to look: the law.

Badminton's rules define in and out by the shuttle's first point of contact with the floor. But the shuttle has a closed cork rim, and when it touches down, the first contact is a very small point on that rim. At high speed, that point blurs across a few frames. The system must interpolate to find the contact point. And every interpolation has assumptions.

This is the grey zone I mean. Not the grey zone of cheating, but the grey zone of definition. When the law defines an event by a single point of contact, but technology can only reconstruct that event as a confidence interval, the dispute is not in the data — it is in the gap between what the law wants and what technology can deliver.

I have spent many evenings reviewing the most contentious rallies, frame by frame, and what I learned was not who was right or wrong. What I learned is this: when a rally is pushed into the grey zone, fans on both sides can find reasons to believe what they want to believe. Same footage, same number, two interpretations.

The flaw is not in the source code; it is in the eyes of the person reading the source code.

I walk into the cathedral of data not to pray, but to listen to the noise of the truth.

And that noise is not small. Of the 58 matches I logged, 17 had more than five reviews. In those matches, I measured a phenomenon: the time the match was stopped rose noticeably, and the rhythm of the game was broken. For a sport in which continuity is part of its identity, every pause for a review stretches the emotion.

But here is the counter-intuitive part. Across those 17 review-heavy matches, post-match disputes did not fall compared with matches with few reviews. They simply moved from the court to social media. The line judge is no longer the figure attacked. The technician is, the algorithm is, the federation is. Responsibility does not vanish; it changes hands.

This is why I say the Instant Review System does not extinguish dispute. It redistributes it.

I understand why organisers brought technology in. Pressure from major matches, prize money, sponsors, turns one umpiring error at a decisive moment into a media disaster. Technology is a way of buying insurance. But insurance does not erase risk; it only changes who carries it.

And here I must be blunt about something the sports industry rarely admits. Data collected courtside — including position and trajectory data — has enormous commercial value. It is sold to analytics firms, to media platforms, and to the betting market. The technology serves fairness for viewers, but it also serves another market, where data is priced by its predictive power.

I have said before that live data fed to betting companies is the darkest side effect of the digitisation of sport. I still hold that view. Every extra camera does not only improve officiating; it feeds an ecosystem where a shuttle two millimetres from the line can be turned into a revenue stream.

That is why I keep asking: when we talk about technology in sport, whose fairness are we talking about?

I cover badminton for the Indonesian market, where the sport is almost a national religion. Here, a single review in a final can shake the evening of millions. I once sat in a cafe in Surabaya as an Indonesia Open final reached a decisive review. The whole place went silent. When the result appeared, half the room cheered, half cursed. Nobody looked at the shuttle. Everyone looked at the shirt.

Players like Viktor Axelsen, Jonatan Christie and Kunlavut Vitidsarn have all lived through the moment of waiting for a review result in front of millions. In that moment, a single data variable is placed on the scales against an entire human career.

I have followed some top players long enough to know that how they react to the review system says more about their mental state than about technique. Some call reviews as a tactic — to break an opponent's rhythm, to catch their breath, to cool a losing streak. Some call reviews because they trust their own eyes absolutely, and when overturned, they lose composure for several points afterwards.

In my log, the success rate of reviews called while trailing is roughly 8 percent lower than reviews called while leading. That figure proves nothing certain. But it suggests that when pushed into a corner, a player calls a review not only because he is certain — but because he is looking for something to lean on.

And that is what data can never fully measure: the silence between the shuttle touching the floor and the word appearing. In that silence, a player may be reshaping the entire psychological landscape of the match. When every tournament stops, I hear my own heartbeat.

I still remember a quarter-final at a top-tier event. The leading player sent the shuttle down near the line; the umpire called it out. The opponent called for a review. The system overturned the call. From that moment, the leading player lost seven straight points and the match. I have no proof that the review caused that collapse. I only have a sample, and I know that correlation is not causation. But I also know that sometimes the mechanism lives in things that are never recorded: a glance, a breath, the dropping of the shoulders.

A brush against the post is not destiny — it is only a tiny deviation between expectation and probability. A review is the same. It does not decide fate. It only inserts a new variable into the emotional equation.

So here I am, writing this, not to demand that technology be removed. I write to demand honesty about technology.

What I want to see is not a perfect system — that does not exist. What I want to see is a system willing to publish its own confidence interval. When the shuttle lands in the noise zone, the screen should say: this rally falls in a range where we cannot be certain. An admission like that will provoke argument. But it is honest. And in the long run, honesty builds trust better than a number presented as gospel.

The more precise the number, the wider the distance between people and the match. I do not want this sport to become an equation. I want it to be a story with data, not a data table with a story.

There is one thing badminton should consider. Major tournaments increasingly lean on technology to protect their credibility. But credibility does not come from eliminating dispute. It comes from handling dispute transparently. A federation willing to say we were wrong, or we cannot be sure, will be trusted more than one that always appears right.

This is a lesson I learned from my own mistakes. Years ago, I publicly predicted a national team would win a major tournament because they had the highest attacking metrics. I was wrong. But what won back trust was not that I hid the error — it was that I wrote a self-audit, admitting I had asked the wrong question.

Technology in sport needs a self-audit like that too.

I have no closed conclusion for this piece, because my data is still being logged every season. What I have is a belief built from years of observation: over the next few seasons, we will see changes in how review results are published. Pressure will come from fans, from media, and from the players themselves, demanding more transparency. The question is no longer whether the technology is accurate, but whether the technology dares to admit its limits.

If I had to choose one signal to track next season, it would be how the Super 1000 events handle rallies that fall inside the noise zone: will they stay silent as before, or will they start telling the truth?

A shuttle drops, touches the floor, and the whole arena holds its breath. Next time, when the word appears, I will not only look at the result. I will look at the gap between what the system knows and what the system dares to say. That is the number worth tracking.