The All-Seeing Eye Does Not Judge: The Power Struggle Behind Badminton's Instant Review System

At 19-19 in the deciding game, the line judge dropped the hand: shuttle in. The opponent immediately raised a hand to challenge. The big screen in the arena lit up. Ten infrared cameras began reconstructing the shuttle's trajectory within seconds, the stands held its breath, and the result appeared: the shuttle landed just a few millimetres beyond the sideline. A gap smaller than a strand of hair, yet enough to decide a ticket to the semifinals.

I have sat before countless screens like that. Having once handled data for sports programmes in Tokyo, what haunts me every time I rewind the footage is not whether the shuttle was in or out. It is the question behind all of it: when will people stop confusing "technology that judges" with "technology that proves"?

Context: from human eye to machine eye

Before the Instant Review System (IRS) became the norm, the authority to rule on court belonged entirely to humans: a referee, several line judges, and a service judge. Three layers of people, three fields of vision, but only one conclusion. The error lies in the fact that a human must see the shuttle land within a few thousandths of a second — a window shorter than a blink.

The Badminton World Federation (BWF) introduced the IRS into elite tournaments from the mid-2010s, initially at some World Superseries events and gradually expanding to larger ones. The mechanism closely resembles what viewers know from tennis: each player has a limited number of challenges per match, and if a challenge is correct, that challenge is retained. The system uses multiple calibrated cameras to reconstruct the shuttle's trajectory, then displays it on the big screen for the crowd and both sides to see.

What few notice: the IRS in badminton is not a "second referee". It is only a verification tool. The final decision still rests with the referee — the person who looks at the image, cross-checks the law, and announces. Technology does not replace people; it moves them from "judging by eye" to "reading the result". But that very shift creates a new paradox, as accountability is pushed toward an algorithm that cannot defend itself.

Analysis: three layers of a decision

Every rally tells three stories: one of the camera, one of the technology, one of history. The first is the story of the image — what the camera captures. The second is the story of the algorithm interpreting that image. The third is the story of the rulebook standing behind every ruling. These three layers rarely align perfectly, and the gap between them is where controversy is born.

The first layer, the camera problem. No camera sees the shuttle at the exact moment of landing perfectly. A badminton shuttle is not a geometric point; it is a small feathered sphere that deforms on contact with the court and moves at speeds that can exceed the sampling limit of a frame. The system must interpolate the trajectory between two frames — that is, it guesses the contact point. "Guesses" here is not a criticism; it is the nature of any model. But interpolation error can exceed the distance between shuttle and line in very tight rallies.

The second layer, the algorithm problem. The ideal parabolic trajectory we see on screen is a cleaned-up model. The real path is affected by air resistance, draughts inside the arena, spin, and even temperature. The system corrects for these factors, but correction always rests on assumptions. When the screen shows a smoothly curving shuttle path, viewers believe it is the truth. In reality, it is a version polished for easy viewing.

The third layer, the law problem. Badminton has grey zones that technology cannot fill. Take the service rule: the shuttle must be struck from below a defined height line (under current standards, roughly 1.15 metres from the court). But the "final moment the racket touches the shuttle" is extremely hard to determine by eye, let alone by camera. The service judge — a human — remains the primary arbiter. Trajectory review does nothing to assess the height of a serve. This is the loophole I believe will be the next major battleground for the sport.

I once spent hundreds of hours rewinding service faults across many major events, and what I took away was not about the players. It was about consistency. The same motion, the same height, but penalised by one official and ignored by another. That inconsistency is more destructive than a single wrong call, because it breaks faith in the system rather than merely breaking a scoreline.

Data never panics. People blind themselves when they rush on emotion. Across those hundreds of hours, I realised players do not react violently to a wrong result — they react violently to an inconsistent one. A player accepts losing to a close line call. They do not accept losing because the same call, a game earlier, was ruled the other way.

There is another dimension the press rarely mentions: the IRS shifts power from the referee to the system operators and to the teams themselves. Retaining a challenge when correct turns the challenge into a strategic resource. A clever player can save a challenge for the most psychologically important moment — not the clearest line call, but the one that could snap the opponent's momentum. This is a kind of reverse judgment: the player uses the tool of transparency to work on the opponent's mind.

Elite players such as Viktor Axelsen and An Se-young do not compete on technique alone; they compete on rhythm and on controlling an opponent's emotions. A well-timed challenge can slow the tempo, break a run of momentum, or simply extend the gap between rallies. So when I analyse the IRS, I never count only correct versus incorrect challenges. I count their timing.

Seen through the lens of professionalisation, a familiar pattern emerges, identical to what happened in football with VAR. At first, technology is celebrated as "fairer". A few years later, people realise technology does not erase controversy; it merely moves controversy from "in or out" to "is the system right". The debate shifts but never disappears. This is as true of badminton as of football.

Contrarian angle: when accuracy does not create fairness

Here I want to pose a hard question: what happens when technology rules correctly in technical terms but wrongly in match terms?

A shuttle three millimetres past the line is "out". Geometrically, that decision is absolutely correct. But in the context of a tense match, ruled by a gap thinner than a sheet of paper, does the player truly receive fairness? Or do they receive an accuracy that nobody on court can feel, verify, or accept by instinct?

I once wrote a long analysis on a calibration error in a ruling system, and I realised something chilling: the fault lay in the device configuration, not in a person. No one left any trace of emotion in that error. But the consequence landed on the shoulders of an athlete who had sweated for two hours. The system has no emotions, yet people bear the consequences of the system. That is structural injustice, and it is harder to fix than any individual mistake.

There is an uncomfortable truth: viewers tend to trust the screen more than the referee. When a shuttle path is replayed with elegant graphics, viewers nod as if they have witnessed truth. But they do not see the model; they see only the result. That naivety unwittingly grants the system a power the system never claimed for itself. And unchecked power always risks abuse — even unwitting abuse.

I am not proposing a return to the purely human eye. That would be a step backwards. But I reject the illusion that technology automatically brings fairness. Fairness comes from consistency of rules, from transparency of the model, and from people knowing the limits of the tools they use.

Takeaway

If badminton wants to go further, it needs three things: to publish the error margins of its calibration system; to extend technology to the service line, where the human eye still judges a millimetre-scale quantity; and to state clearly that every model has an error band — so that decisions falling exactly at that edge are handled by law, not by faith.

The question left behind: when a machine says "out", are we protecting the rule, or are we handing judgment to something incapable of regret?