Yan's Knee, Taekwondo's Sensors: When Combat Sports Hand the Verdict to the Screen

YAN'S KNEE, TAEKWONDO'S SENSORS: WHEN COMBAT SPORTS HAND THE VERDICT TO THE SCREEN

I don't watch the goal; I watch the camera angle that filmed the goal. I first wrote that line at seventeen, after the night in Kazan, and at 2 a.m. on March 7, 2026, Seoul time, it was still the only sentence I could use to describe the work I was doing.

On screen was UFC 259, held at the UFC Apex in Las Vegas, Nevada, on March 6, 2026 local time. Round four. Petr Yan, the reigning bantamweight champion, was controlling the fight against Aljamain Sterling. Yan stepped in, raised his right knee, and drove it into Sterling's head while Sterling was on the canvas.

The referee was Mark Smith. He stopped the fight, called the doctor in, and walked to the judges' table. I did not rewatch the knee. I started rewinding the camera angle behind Sterling's back, the one the official highlight package almost never uses, because it is not pretty.

It took exactly seven viewings, at the slowest speed my browser allowed, to confirm a detail neither camp mentioned at the press conference: Yan's right knee made contact for less than a quarter of a second, and Sterling reacted roughly four-tenths of a second later. That number does not prove who was right. It only told me the argument about to unfold would not be about the strike. It would be about the rule.

Context: a rule that measures geometry, not force

The provision applied at UFC 259 sits inside the Unified Rules of Mixed Martial Arts, overseen by the Nevada State Athletic Commission. It is short: a fighter is considered "downed" if any part of the body other than the soles of the feet is touching the canvas. When an opponent is in that position, a knee or kick to the head is a foul.

What matters is not the provision itself but the way it classifies conduct by body position rather than by force. A knee that barely grazes the head of a kneeling fighter is a foul. A knee three times harder to the head of a standing fighter is legal. The rule does not measure force. It measures geometry.

This rulebook did not appear from nowhere. It was codified in 2026 by the New Jersey State Athletic Control Board, then copied across states and countries over the following two decades. Every revision added a definition to plug a hole created by a previous fight. That is how combat sports law works: it always writes about the past and is always judged in the present.

For Yan's knee, Mark Smith chose the harshest sanction available: disqualification. Sterling was declared the winner and became bantamweight champion. It was the first time in UFC history that a title changed hands by in-fight disqualification.

Combat sports history is full of verdicts that were questioned. On October 1, 2026, at the Jamsil Gymnasium in Seoul, Roy Jones Jr. met Park Si-hun in the Olympic light middleweight final. According to statistics published afterwards, Jones landed 86 punches; Park landed 32. The three judges still scored it 3-2 for Park. The fallout led to the suspension of all three judges and became one of the forces pushing Olympic boxing to overhaul its scoring system repeatedly over the next two decades.

I live in Seoul. I pass through Jamsil every week. And I still wonder: if 2026 had eight camera angles like 2026, would the story have changed, or would it simply have moved somewhere else to be argued about?

In taekwondo, that question has been tested with real money. World Taekwondo introduced the electronic Protector and Scoring System, known as PSS, at the 2026 World Championships in Copenhagen and made it mandatory from the London 2026 Olympics. The principle is simple: sensors inside the protector register a kick that lands with sufficient force. A kick to the trunk no longer needs a human eye.

But only the trunk. The head wears no sensor. Head-kick points are still awarded manually by three judges pressing buttons. In other words, within a single round, two body zones of the same athlete are scored by two different epistemologies: one binary, one probabilistic.

Analysis: when four camera angles are all correct

This is where I want to spend most of the space, because it is the shared architecture of every modern combat sport, not just MMA or taekwondo.

The core idea: technology does not erase controversy in combat sports; it moves controversy from the referee to the rule, and then from the rule to the architecture of the scoring system.

At UFC 259, the controversy never claimed the knee did not happen. There were enough angles, and every angle confirmed contact. The controversy claimed three other things: whether the knee hit the shoulder or the head; whether Sterling was genuinely hurt or exploiting the situation; and whether intent matters if Yan never registered that Sterling had gone down.

Legally, all three questions are meaningless. The Unified Rules do not require the referee to establish intent. They require him to establish position. The knee was aimed at the head. Sterling's head was exposed. Sterling's knee was on the canvas. Those three conditions were enough for Mark Smith to do exactly what he did.

So the object worth analysing is not the referee's decision. The object worth analysing is the gap inside the rule.

The same structural gap takes a different shape in taekwondo. World Taekwondo publishes its scoring ladder: one point for a punch to the trunk, two for a kick to the trunk, three for a kick to the head, four for a spinning kick to the trunk. The value of a head kick has been changed at least once in the past decade, raised to four points under the 2026 rules and then lowered back to three. Every time the number changes, the federation says it is protecting the sport's appeal. That is a statement about television, not about martial arts.

Alongside it sits the gam-jeom mechanism: each penalty awards a point to the opponent, and ten gam-jeom in a single round results in disqualification. The format is three two-minute rounds, best of three. It is a system designed never to produce a draw. A system that forbids draws must manufacture a winner at any cost, and every mechanism that must manufacture a winner tends to push controversy toward the person who carries final responsibility.

Meanwhile the trunk is scored by a machine and the head by humans. A fighter can receive points from a system that sees nothing but force, and fail to receive points from a system of three people who can missee, press late, or be distracted by the crowd. That asymmetry is the unsolved problem of modern taekwondo.

I spent the summer of 2026 coding 1,247 refereeing decisions from the 2026 World Cup and three K League 1 seasons. When I cross-referenced how combat sports operate, one common thread appeared: every review system assumes the problem lies in the capacity to observe. No review system assumes the problem lies in the rule being observed.

In football, that is semi-automated offside and the drawn lines. In MMA, it is the definition of "downed." In taekwondo, it is leaving the head outside the sensor system from Copenhagen 2026 to today.

At the Paris 2026 Olympics, taekwondo ran from August 7 to August 10, 2026, at the Grand Palais Éphémère, with 128 athletes across eight weight categories. It was the Games at which the two-tier architecture operated in full. South Korea won two golds: Park Tae-joon in the men's under-58kg and Kim Yu-jin in the women's under-57kg. Thailand's Panipak Wongpattanakit successfully defended her women's under-49kg gold. Althéa Laurin delivered France's first Olympic taekwondo gold in the women's over-67kg. Viviana Márton won Hungary's first gold in the sport. Ulugbek Rashitov defended his men's under-68kg title.

Those numbers are outcomes. Beneath them runs something else: the video appeal system. Coaches hold a limited quota to request a review, usually one per round, returned only if the appeal succeeds. The mechanism is close to a replica of football's VAR, with one important difference: in taekwondo, a coach must decide to appeal within seconds, before the round moves to its next rhythm.

We can look at the problem from five angles, just as I rewatched that UFC 259 sequence seven times.

Angle one, the spectator's. The spectator sees a knee and a fighter going down. Immediate conclusion: the attacker wins. But the attacker was disqualified.

Angle two, the commissioner's. They see conduct violating a clearly written provision. Conclusion: the decision was correct.

Angle three, Yan's corner. They see a situation in which their fighter had no time to register his opponent's position within half a second. Conclusion: the rule ignores human reflex limits.

Angle four, Sterling's. He felt genuine contact to the head and had an opportunity taken from him. Conclusion: the sanction was justified.

Angle five, the broadcaster's. We see a finish where nobody is happy. Nobody won the way people want to win. Conclusion: this is the type of finish that generates the most engagement, which is why it gets replayed the most.

The first four angles are all correct. That is precisely the problem. When four different viewpoints are each correct in their own terms, the error is not in the viewers. The error is in the thing they are all looking at.

The rematch between Sterling and Yan took place on April 9, 2026, at the VyStar Veterans Memorial Arena in Jacksonville, Florida, at UFC 273. Sterling won a split decision, with two scorecards of 48-47 for him and one of 47-48 for Yan, according to the published scorecards. This time there was no illegal knee, no disqualification, no doctor. And nobody remembers it the way they remember the first fight.

That points to a law I recorded in my 47-page notebook at seventeen: controversy is not a by-product of sport; it is one of the main products. A decision that is lawful but unpopular outlives a brilliant fight nobody argues about.

There is a venue detail I cannot skip. The same VyStar Veterans Memorial Arena in Jacksonville hosted UFC 249 on May 9, 2026, the first major American sports event to return during the pandemic, inside an arena with no spectators. Two years later, in the same building, Sterling and Yan rematched in front of a full house. The 2026 season had no spectators, but it had a very large pair of ears. During that stretch I recorded a small but consistent shift: with no crowd noise, referees reacted to the sound of impact and the sound of breathing rather than to crowd pressure. Many decisions became slower, and also more accurate. When crowds returned, that lag vanished within a single season.

That is why I always question the decision-making environment before questioning the decision-maker.

The contrarian angle: technology relocates responsibility, not just error

Here I want to push against a common assumption.

The assumption is that technology exists to make sport fairer. I argue that technology in combat sports exists to make the power to judge harder to trace.

When Mark Smith disqualified Petr Yan, he had a name, a face, and a decision criticised by millions for weeks. He carried responsibility. But when a sensor inside a taekwondo protector fails to register a sufficiently forceful trunk kick, nobody is criticised. There is nobody to criticise. The sensor manufacturer does not hold a press conference. The federation does not suspend a circuit board.

That is a relocation of responsibility few people name correctly. For more than a decade, combat sports federations have been shifting judging authority from people to systems, and each time they announce they are reducing error. They do not announce they are reducing accountability. But that is what has happened.

Kazan erased a goal, but opened an eye. I wrote that line about how VAR taught me to see differently. There is another version I rarely use: technology can open an eye and cover a face at the same time.

The second blind spot, and perhaps the more important one: we are confusing two kinds of error.

The first kind is error in execution. The referee did not see, saw late, or saw wrong. This kind can be fixed with technology, and it partly has been.

The second kind is error in rule design. A provision is written to solve a problem of the past, and when circumstances change, it stands unchanged. The downed-opponent rule was written to protect fighters already in a position where they cannot defend themselves. In practice it also protects fighters actively lowering their stance to evade strikes, and in the specific case of UFC 259 it created a half-second window in which a referee had to treat two completely different situations identically.

Technology cannot fix the second kind. It only makes the second kind more visible, and because it is more visible, it generates more controversy. This is the central paradox of every modern combat sport: more cameras, more arguments. Not because people got worse, but because cameras finally showed us the gaps we previously lacked the data to see.

In taekwondo, that moment arrived when the PSS began awarding points for kicks the human eye could not see, and withholding points for kicks the human eye could see clearly. The modern taekwondo athlete fights for two audiences at once: three judges at the table and a circuit board inside the protector. Technique must be calibrated for both. It is a skill no coach can teach, because it has no name yet.

There is a coincidence I have always found interesting. In 2026, I wrote a 2,100-word piece about Raheem Sterling's penalty in the Euro 2026 semi-final between England and Denmark at Wembley. Sterling fell inside the box; I stood up inside the lecture hall. That same year, in a different sport, a man sharing the surname Sterling also went down in a contested moment. Two sports, two rulebooks, two governing bodies, and the same question: when a man goes down, who decides what the fall means?

Way forward: separate the incident from the rule

If I had to propose a direction, I would separate the review process into two independent tiers, the way financial regulators separate inspection from adjudication.

The first tier handles the incident: did the strike land, was the position correct, was the force sufficient. This tier has public criteria and can end in a number.

The second tier handles the rule: is the applied provision still appropriate for the sport being played. This tier does not handle individual fights; it operates per season, and its minutes must be published.

Without the second tier, federations force a single referee to do two jobs at once: apply the current rule and apologise for the current rule. That is an impossible assignment, and we are handing it to the most exposed people in the system.

The rule is the only thing that never enters stoppage time. At UFC 259, the fight ended in round four. In taekwondo, the bout ends after three two-minute rounds. But the rule is always still there, word for word, waiting to be rewritten.

And one question to close, not to answer immediately: if a decision is proven lawful yet wrong against the sense of fairness of nearly everyone watching, is the thing to fix the rule, or our sense of fairness?

I do not know. But I know I will still get up at 2 a.m. to rewatch it, because that is my part of the work.

Source data for this article

• Event: UFC 259, March 6, 2026 local time, UFC Apex, Las Vegas, Nevada.
• Bout: Petr Yan vs. Aljamain Sterling, bantamweight.
• Referee: Mark Smith.
• Result: Yan disqualified in round four; Sterling won and took the title.
• Rematch: UFC 273, April 9, 2026, VyStar Veterans Memorial Arena, Jacksonville, Florida; Sterling won by split decision.
• World Taekwondo's Protector and Scoring System (PSS): first used at the 2026 World Championships in Copenhagen, mandatory from the London 2026 Olympics.
• Paris 2026 Olympics: taekwondo ran from August 7 to August 10, 2026, at the Grand Palais Éphémère, 128 athletes, eight weight categories.
• Seoul 2026 Olympics: Roy Jones Jr. lost to Park Si-hun 3-2 after landing 86 punches to 32, according to statistics published afterwards.
• The Unified Rules of Mixed Martial Arts were codified in 2026 by the New Jersey State Athletic Control Board.