Human Eyes, Machine Eyes, and the Boundary of Justice on the Badminton Court

In the video analysis room of a television station in Tokyo, I replayed the footage of a shuttlecock landing right on the sideline. The frame rate was dropped to its lowest setting. Each millisecond passed like a silent film. The shuttle rotated, wobbled, then touched down. The line judge raised a hand. The stands erupted. And on the other side of the net, a player collapsed, both hands covering his face. I have watched that moment no fewer than forty times, and each time, I saw something different. The first time, I saw a great rally. The tenth time, I saw a controversial decision. The thirtieth time, I saw a system struggling to hold on to the audience's trust. Every shot tells three stories: one of the camera, one of the technology, one of history. And in badminton, those three stories rarely align. That is why I sit here, writing these lines, rather than simply filing a results report. Because behind every score lies a judging apparatus that runs on human eyes, machine eyes, and rules written nearly a century ago. Badminton is the fastest racket sport in the world. A smash from a professional player can send the shuttle flying at speeds exceeding four hundred kilometers per hour in the first moments after it leaves the racket face. At that speed, the human eye can barely follow the trajectory, let alone judge the point of contact to the nearest millimeter. It is precisely this gap between human capability and the speed of the game that gave rise to an entire industry of judging technology. The Badminton World Federation, an organization founded in 2026 that now governs the entire global professional tournament system, introduced the Instant Review System to major events around 2026. The core technology behind this system is Hawk-Eye, a trajectory-tracking system using multiple cameras placed around the court, reconstructing the shuttle's flight path on a large screen for referees and spectators to view together. The principle sounds perfect: machines do not play favorites, machines do not fear crowds, machines do not get tired. But the story is far more complicated than that. Before technology arrived, badminton matches were governed by a purely human judging system. A head referee sat on a high chair, observing the whole court. Line judges stood at the four corners, eyes fixed on the lines. A service judge checked the posture and the point of contact between racket and shuttle. Together they formed a distributed judging machine, where each person was responsible for only a small fragment of the truth. In 2026, badminton switched to the 21-point rally scoring system, meaning every rally scores a point, regardless of who serves. This change shortened match times and increased drama, but it also made every referee decision weigh heavier. Under the old system, a misjudged rally could be offset by the right to serve. Under the new system, a single wrong point can be the deciding point of the whole match. This is the context I want you to grasp before we enter the hardest part: does technology actually solve the problem of justice, or does it merely shift it to another layer? I have spent many years working as a video analyst for television stations, and my experience watching matches has given me a rather peculiar belief: technology does not make referees fairer, it only makes their mistakes more visible. When a human eye misjudges, spectators can argue. When a machine eye misjudges, spectators do not know whom to argue with. Let us begin with a technical paradox few people notice. The shuttlecock is not a hard object. It has a distinctive shape with a cork base and a flared feather skirt. As it flies, it rotates around its axis, generating enormous drag and decelerating extremely quickly. This means its trajectory is not a simple parabola like a football or a tennis ball. It is a complex three-dimensional curve, influenced by airflow, humidity, temperature, and even indoor arena winds. Hawk-Eye reconstructs the trajectory by sampling at multiple points and interpolating the flight path between them. In tennis, where the ball is hard and the trajectory more stable, the system's margin of error is published at a few millimeters. In badminton, the variables are greater, and the margin of error is larger too. When a shuttle lands near a line, the distance between the real contact point and the point the system displays can fall within a gray zone — a zone where neither human nor machine can be one hundred percent certain. This is exactly where I want to use the word "decode". We tend to look at the big screen and believe it is absolute truth. But the big screen is only a reconstructed image, not the truth itself. The gap between that image and the truth is precisely the space where all controversy is born. I remember a match at a top-tier event that I followed as an analyst. A leading player requested a review of a rally at a deciding score. The system displayed the shuttle touching the line within a millisecond. The referee upheld the decision. But when I reconstructed the footage from the broadcaster's independent camera angle, at a higher frame rate, I saw that the shuttle appeared to have landed just outside the line by a very small margin. I did not publish that. Not because I was afraid, but because I knew both images carried margins of error, and releasing an opposing conclusion based on a different camera angle alone would have been irresponsible. The lesson here is important: in referee analysis, what is frightening is not error, but false certainty. A referee who admits uncertainty is an honest referee. A system advertised as perfect is a system hiding its limits. Now, let us talk about the least-analyzed aspect: the service rule. In 2026, the Badminton World Federation implemented a rule requiring the shuttle to be struck from a height not exceeding one point one five meters at the moment of contact. Previously, the rule was based on the server's waistline, a relative standard dependent on each athlete's body. This change was meant to create a more objective, measurable, verifiable standard. But it created a new problem. How do you measure one point one five meters precisely within a millisecond, when the racket moves at high speed and the human eye cannot read the number? Major tournaments began using electronic measuring devices and specialized cameras to assist service judges. But at many smaller events, where budgets are limited, judgment still rests on the human eye. This is a textbook example of what I call the technology gap. The same rule, enforced to two different standards in two different places. Athletes at major events are protected by machines; athletes at smaller events are judged by feeling. Can we call that fair? I do not think so. And I think this is a far more structural problem than individual controversies over a single rally. Over many years of watching, I have realized that referee controversies in badminton usually revolve around three groups of issues. First, the lines, where technology has intervened most strongly. Second, the serve, where technology has intervened unevenly. Third, collision situations, racket touches, net touches, and subtle technical faults that cameras struggle to capture. In the third group, the problem becomes especially complex. When two players' rackets touch during a high-speed rally, determining who touched first, where, and whether it was a fault requires a near-perfect viewing angle. At many tournaments, there are not enough cameras to cover every angle. Referees are forced to decide based on what they see, and what they see may differ from what actually happened. This is where I want to speak of the concept I call "adjudication". Referees do not merely apply the law; they interpret it. And in interpretation, the human element is always present. An experienced referee knows when to be firm and when to be flexible. An inexperienced referee may apply the law mechanically and produce decisions that are correct on paper but wrong in spirit. In badminton law, there are clauses granting referees final authority based on personal judgment. For example, a referee may call a fault when a player deliberately slows the match, when a player behaves disrespectfully, or when a situation falls outside the law's foresight. These clauses exist because the law cannot cover every situation. But they also create room for subjectivity. I have witnessed matches where the same behavior, at two different moments, received two different decisions from the same referee. That is not bias. It is inconsistency, and inconsistency in adjudication damages trust more than a single error. This leads me to an observation about data. In other sports, one can measure a referee's consistency through statistical indicators. Number of cards issued, number of faults called, number of decisions overturned after review. In badminton, collecting this kind of data remains limited. We know a great deal about smash speed, points scored, and win rates, but very little about the behavior of the judging apparatus. This is a major gap. Because if we cannot measure it, we cannot improve it. And if we cannot improve it, every controversy will keep repeating along the same pattern: a controversial rally, a wave of outrage on social media, a few explanations from organizers, then everything fades into oblivion until next time. Data never panics. People blind themselves when they rush into emotion. I have seen this too many times. A player who loses because of a controversial decision will speak of injustice. A spectator who rewatches the footage will speak of bias. A journalist who needs a story will speak of the collapse of justice. But if we sit down, measure, and analyze calmly, we often see a different picture: a system operating within its limits, and those limits can be expanded if there is enough will. Now, to the most counterintuitive part of the story. People often assume that more technology leads to more justice. I do not believe that. I believe more technology, if not managed properly, can lead to more dependence and less responsibility. Imagine a future where every on-court decision is made by machines. Line judges disappear, service judges disappear, leaving only an automated system and a supervisor. It sounds efficient. But when a mistake occurs, who is responsible? No one. The mistake becomes a technical glitch, a bug to be patched. And in a world where no one is responsible, trust cannot be built. Justice in sport lies not only in the correct result. It lies in the process by which that result is produced. A correct decision made by a system people do not understand will not create trust. A correct decision made by a referee people trust will create trust, even when that decision is controversial. This is the biggest blind spot in how we discuss sports technology. We focus on accuracy and ignore transparency. We want machines to be right, but we do not demand that machines explain. And a system that cannot explain itself is a system that cannot be questioned, cannot be improved, and ultimately cannot be trusted. I once witnessed a situation where the review system produced a result different from what the referee had seen with the naked eye. The referee upheld his original decision, and was heavily criticized afterward. But I understood his reasoning. He was not certain of the system's accuracy in that specific situation. He chose to trust his own eyes over an image he could not verify. That was not conservatism. It was the caution of a man who understands that technology also has limits. The problem here is that we have not yet built a clear hierarchy among sources of information. When human eyes and machine eyes conflict, which takes priority? Under current law, the answer is not always clear. And this ambiguity creates space for controversy, for emotion, and for inconsistent decisions. Another rarely mentioned aspect is the psychological pressure on the referees themselves. They know that every decision they make can be reviewed, analyzed, dissected on social media. In that context, many referees tend to choose the safe option: uphold the original decision to avoid being criticized for inconsistency. But that safety is sometimes a form of injustice, when a wrong decision is maintained simply out of fear of change. I have observed referees for many years, and I have realized they are the most poorly understood group under pressure in sport. They have no fans. They have no sponsorship deals. They have only a difficult job, judged by millions of people within less than a second. And when they err, they become the target of collective anger. This does not mean we should ignore their mistakes. It means we should analyze their mistakes systematically, rather than merely reacting with emotion. So where should we go from here? I believe the future of badminton adjudication lies not in replacing humans with machines, but in designing a system where humans and machines complement each other. Machines provide data and accuracy. Humans provide context and judgment. And most importantly, humans take responsibility. To achieve that, three specific changes are needed. First, there must be a transparent standard for margin of error. When the review system displays a result, it should also publish its margin of error. Spectators have the right to know how certain a decision was. This would reduce controversies built on false certainty. Second, data collection on adjudication behavior must be expanded. Not to monitor referees in a threatening way, but to build a database that lets us measure consistency and improve training. A referee who knows he tends to call faults in a specific situation can adjust. A system that knows it tends to err at a specific angle can add a camera. Third, there must be a mechanism for explanation. Whenever a controversial decision is made, there should be an official channel to explain the reasoning, based on law and data. Not to justify, but to be transparent. Spectators may disagree with a decision, but they should be able to understand why it was made. These three changes do not require new technology. They require will. And will is harder to create than any device. I return to the footage in the analysis room in Tokyo. The shuttle still lands near the sideline. The line judge still raises a hand. The stands still erupt. But this time, I am not seeking a definitive answer about whether the decision was right or wrong. I am seeking something else: did the system that produced that decision have enough transparency for people to trust it, even when they disagree? The answer, in most cases, is still no. But it can become yes, if we choose to build trust rather than merely pursue accuracy. Badminton has come a long way from the days when only human eyes watched the court. It will go further still. The question is not whether machines will replace humans. The question is whether we are wise enough to preserve the human part in what machines cannot grasp: caution, transparency, and responsibility. Because justice on the badminton court, in the end, is not measured by the number of points judged correctly. It is measured by the trust of those who sit watching, that the game they love is still being run fairly. And that trust, no camera can reconstruct.