The Camera Angle at the Edge of the Frame: How Geometry Decides a VAR Call in the V.League

The referee runs to the edge of the touchline in front of Stand B. His hands draw a rectangle in the air, a gesture anyone who has watched football for more than three years recognises instantly. The stand goes silent for two seconds, then the noise returns, louder than before. On the big screen hanging in the corner of the ground, the word VAR appears alongside a spinning circle. Fifteen thousand people sitting there see nothing else.

In the technical room a few dozen metres away, someone is scrubbing the footage back at slow speed. They lay a line on the grass, parallel to the last horizontal line of the pitch. Then another line. Then another. Three lines, three measurements, and three answers that are not entirely identical.

I have sat in something close to that position. Not in an official VAR room, but in an editing suite a few kilometres from a stadium, with two monitors, a spreadsheet, and an odd habit: every time I was unsure, I wrote down a number. Frame count. Camera count. Centimetres of variance between measurements.

Years later I reached a conclusion few people want to hear. Most errors do not sit at the centre of the pitch. They sit at the edge of the frame. Referees rarely err in the choice they make. They err because they have been placed in an observation position that geometry does not allow them to get right.

---

VAR arrived in Vietnam later than most of the region, and that delay left marks

FIFA wrote video assistance into the Laws of the Game in 2026, after a long trial period. The Thai League and several regional competitions began using it in the 2026-2026 window. Vietnam's V.League 1 implemented VAR later, during the 2026 season, operated by the Vietnam Professional Football Joint Stock Company (VPF) together with the Vietnam Football Federation (VFF), with FIFA technical approval.

That gap was not merely a scheduling matter. It produced a specific consequence: by the time VAR started operating in Vietnam, audience expectations had already been shaped by the Premier League, La Liga and World Cups. Viewers were used to semi-automated offside lines, 3D reconstructions, and overhead frames looking straight down at the turf. Then they went to a Vietnamese stadium and saw something else.

The difference was not in the software. The offside-line software is an off-the-shelf product, licensed by FIFA and installed to a fixed procedure. The difference was in where the software's input data came from: which cameras were positioned where, at what height, and whether the frame was obstructed.

Consider how offside measurement works. The software needs to fix two moments precisely: the instant the ball leaves the passer's foot, and the body position of the relevant players at that exact instant. To determine depth along the pitch — whether a player is ahead of or behind the last defender — the system needs either a view almost perpendicular to the horizontal plane, or a pair of cameras far enough apart to reconstruct three dimensions. With only a high but laterally offset camera, the software is forced to infer depth from perspective. And inferring depth from perspective is where error is born.

At many stadiums in Vietnam, camera placement is constrained by the architecture of the stands themselves. Roofs sit low, stands are not tall enough, or the technical area lacks a safe anchoring point for equipment. The main broadcast camera — the one that determines most of what a referee sees in the VAR room — is usually set at a moderate height, sufficient to cover the match, insufficient to create an ideal horizontal plane for offside measurement.

This is the crux that almost nobody raises when arguing about VAR. The system does not fail because the software is poor. It fails because it is forced to answer a geometric question with incomplete geometric data.

---

The price of a system, and the price of saving on one

A match with VAR does not merely need a computer. It needs a control room, a dedicated transmission system that does not share a signal with the broadcast feed, a roster of on-site technicians, and above all a dedicated camera cluster at the positions FIFA requires.

FIFA distinguishes two kinds of image. The first is broadcast imagery, serving television audiences, produced by the broadcaster with the goal of beautiful angles, readable play, and visible player emotion. The second is officiating imagery, with a completely different goal: measurement accuracy, positional stability, and independence from a director's editorial choices.

That divergence of purpose creates structural tension. In well-funded competitions the two camera systems are entirely separate. In budget-constrained competitions they must share a signal. And when they share, the first thing sacrificed is always the geometric quality of the officiating data.

Running VAR for a single match in Southeast Asia typically costs from tens to hundreds of millions of dong, depending on camera count, transmission system and staffing. In a league where many clubs still depend mainly on owner funding and short-term sponsorship deals, paying for a fully compliant VAR system across every round is a problem without an easy solution.

The result is a compromise. The league has VAR. But VAR under infrastructure that is not yet synchronised. Fans receive reassurance in principle, while still confronting decisions that VAR itself turns into fresh controversy.

I once kept a tracking sheet across many rounds of a Southeast Asian league, classifying each VAR incident by three criteria: whether the incident was reviewed at all, how many camera angles the referee was given, and how many of those angles were genuinely suited to answering the question under dispute. The result made me stop. In most contentious offside incidents, exactly one camera angle was geometrically suitable — and in many cases it was the last angle the referee was shown.

---

The geometry of a single instant

Error in video offside measurement has three main sources, and all three can be quantified.

The first is spatial resolution. A frame captured from several dozen metres away, at a given focal length, has a physical limit on how far apart two points it can distinguish may be. At that distance, each pixel in the frame corresponds to a real distance on the pitch. That distance usually runs from a few centimetres to more than ten. Even if everything else were perfect, the system still could not separate two positions closer together than that threshold.

The second is frame rate. A standard broadcast camera records at 25 frames per second, roughly 40 milliseconds between consecutive frames. A sprinting player can cover about half a metre in that window. A player moving diagonally can change position along the depth of the pitch substantially in the same window. So the question "which frame did the ball leave the foot" carries an error that translates directly into tens of centimetres along the length of the pitch.

The third is perspective. This is the largest source of error and the least discussed. When a camera does not sit on the horizontal plane of the last defender, every pixel in the frame is distorted according to a determinate perspective rule. Reconstructing a player's true position from that frame requires knowing the camera's exact position in three-dimensional space, its viewing direction, focal length, and lens distortion. Error in any one of those parameters propagates straight into the final result.

In leagues with fixed, calibrated camera systems, these parameters are measured once and held constant through the season. Where cameras are re-positioned before each match, or adjusted mid-match to serve the broadcast, calibration becomes continuous work — and any delay in that work opens a gap of error.

I spent a great deal of time asking whether the human eye can compensate for these errors. It cannot. The eye is very good at estimating relative distance at short range, and very poor at judging depth at long range, especially without fixed reference points. It took 37 re-watchings of the same clip at different speeds before I understood that the human eye is not a measuring instrument — and that anyone claiming otherwise is confusing confidence with accuracy.

---

Broadcast cameras and officiating cameras do not serve the same purpose

This is the point I consider most important in the entire VAR debate, and the most widely misunderstood.

Fans watch football through broadcast imagery. They see a high, clear, attractive angle, and they form a conclusion. When the referee decides otherwise, they conclude the referee is wrong, or VAR is wrong, or both.

But a beautiful angle and a correct angle are two different things. A beautiful angle shows the viewer the flow, the emotion, the tempo of the match. A correct angle for VAR has its optical axis close to the horizontal plane of the incident, sufficient height to minimise occlusion, and precise calibration.

In many cases those two requirements conflict directly. The angle that shows viewers most clearly that a player appears offside is often the angle with the largest perspective error. The angle that yields the most accurate measurement is often hard to watch, partially obstructed, or so oblique that viewers cannot judge it themselves.

This is why seemingly obvious situations on television become deadlocks in the VAR room. Not because the referee cannot see what the audience sees, but because the referee is asked a different question from the one the audience is asking.

The audience asks: was that player offside? The referee must ask: do I have enough data to assert that player was offside beyond reasonable doubt? The two questions sound alike but are logically different structures. The first seeks truth. The second seeks evidence. And truth does not always come with evidence attached.

---

"Clear and obvious error" — a clause that cannot be defined

The global VAR protocol is built around one phrase: clear and obvious error. Video officials may intervene only when the on-field decision is wrong at that threshold. Situations in the grey zone — where a decision could be right or wrong but not clearly enough to overturn — must stand.

In principle this is sound. It protects the flow of the game, limits excessive review, and concedes that some incidents genuinely cannot be settled absolutely.

In practice the phrase is a problem with no fixed answer. The "clear and obvious" threshold depends on the standard of each competition, each referee group, and even each phase of a season. Referees are trained to recognise the threshold, but the threshold is not a number. It is a judgement.

And any judgement has a distribution. Give the same incident to ten VAR officials and there is a high probability of more than one conclusion. Most leagues acknowledge this privately in technical meetings, and rarely say it in public.

I think this is where fans deserve fuller information — not to lower expectations, but to understand the tool correctly. VAR is not a court of law. VAR is a tool for re-checking a human decision, operated by humans, against a standard defined by humans in natural language — and natural language always contains grey areas.

---

A lesson from a World Cup 2026 incident

I have a personal reason to care about camera angles, and it does not come from Vietnamese football.

In the 2026 World Cup final between France and Croatia, referee Nestor Pitana used VAR to identify a handball in the penalty area and awarded France a penalty. I spent days afterwards measuring the incident from every available broadcast angle. The result drew my attention to a detail unrelated to football: only one angle showed the arm in a position that could be considered unnaturally extended, and that was the angle the referee was shown in the VAR room.

I did not conclude the decision was wrong. I concluded it was reached on a narrower set of data than the set of data that existed. That is a problem solvable by process, not by argument.

From then on I adopted a structure I call the decision chain. Instead of judging the final outcome, I record each link: when the incident occurred, how many camera angles existed, how many were given to the referee, how long the review actually took, and how long would have been needed to reach a certain conclusion with sufficient data. Presented this way, most controversies become far more concrete and far less emotional.

I brought that method to Southeast Asian football, and it produced similar findings. The problem does not lie in referees' characters. It lies in the data infrastructure they are given.

---

What VAR exposes

There was a period when stadiums around the world stood empty. Football continued, the ball still sounded, but no crowds filled the stands. During that time I noticed something about the sound of a match: most of what creates the feeling of football does not come from the ball. It comes from the crowd.

VAR works in a similar way. It does not create justice. It creates a process, and that process exposes what already existed but was hidden.

Before VAR, a wrong decision was treated as human error, accepted as part of the game. With VAR, a wrong decision becomes a system failure. The emotional difference is enormous. Previously, fans were angry at a referee. Now they are angry at a machine — and a machine cannot apologise.

This is something I think we have not fully processed culturally. Empty stadiums in 2026 showed me one thing: VAR does not save football, it exposes football. It exposes that refereeing quality depends on infrastructure quality. It exposes that standards are written not in numbers but in tacit agreements between parties. It exposes that fairness in football has always been a target rather than a state.

And in leagues where infrastructure is not yet synchronised, VAR exposes that more clearly than anywhere else.

---

The contrarian view: are we seeking justice, or a prettier angle?

When a contentious VAR decision lands, the standard public response is to demand the release of the audio between referee and VAR. The proposal sounds eminently reasonable, and I once supported it. Several major leagues have implemented it, and it genuinely improved transparency.

But I have partly changed my view. Releasing audio solves a transparency problem, not an accuracy problem. It tells fans what the referee was thinking, not whether the referee had enough data to think correctly.

In many cases, releasing audio makes things more complicated. Fans hear a short, apparently superficial exchange and conclude the process was sloppy. In reality, most of the checking happened earlier, at stages the audio does not capture: camera selection, line calibration, identifying the source frame.

There is a temptation both fans and media find hard to resist: the belief that one more camera angle would reveal the truth. In fact another angle only helps when it is geometrically suitable. Another wrong angle does not reduce doubt; it increases the number of possible conclusions.

We think we are seeking justice. In most controversies, it turns out we are only seeking a prettier angle. And a prettier angle is usually not a more correct one.

---

Placing the V.League in a longer cycle

There is a trend in how Vietnamese football is discussed that I think needs adjustment. When a problem appears — a contentious refereeing decision, a national team defeat, a club in financial trouble — the reaction is usually framed within one match or one season. That framing turns every problem into a crisis, and every solution into a substitution.

But refereeing infrastructure problems do not operate on a season cycle. They operate on an investment cycle. A compliant camera system has a lifespan of years. A VAR training programme needs years to produce a sufficiently deep roster. A calibration procedure only stabilises after hundreds of repetitions.

This means judging VAR in the V.League through a handful of controversies is measuring in the wrong unit. We are measuring in matches while the problem is measured in years.

In the period I have been tracking recently, there is one genuinely positive signal: Vietnamese football has made strides at national team level and at club level, with concrete results in regional competitions. Those results generate resources and attention. The question is whether those resources flow into infrastructure — cameras, training, data systems — or into things that can be seen immediately.

The history of leagues suggests the answer usually leans towards the latter. But history also shows that the leagues which changed that had one thing in common: they began measuring things that cannot be seen.

---

One concrete proposal

If I could propose a single change to how VAR operates in the V.League, I would not propose changing the laws. I would not propose releasing audio. I would propose something far smaller and measurable.

That is: publish, after every round, the number of camera angles provided to the VAR in each review, together with the relative geometric position of that angle against the horizontal plane of the incident. This is technical data, not sensitive data. It reveals no identities and no dialogue. It answers one question only: did the referee have enough data to answer the question they were given?

Once that data is published over a few seasons, two things happen. First, debates about referees shift from judging character to judging infrastructure. Second, investment pressure acquires a specific target, because once you know exactly what you lack, filling the gap becomes a budget problem rather than an argument about opinions.

This is the kind of transparency that is achievable even before budgets change. And in most systems, transparency about weakness is the first step towards fixing it.

---

What I still cannot answer

At 45, after many years in this work, I realise I have softened on one specific point: I no longer believe every controversy can be settled with better data.

Part of football runs on emotion, and that part cannot be measured in centimetres. When a player celebrates a goal only to have it taken away by a line thinner than a hair, the sense of loss is real, whatever the geometry says. When a supporter drives two hundred kilometres to watch their team and returns feeling robbed, no technical report soothes that.

I think the right answer is not to choose a side. It is to build a system good enough that the arguments that remain are the arguments that genuinely need to be had — not arguments produced because we did not have enough cameras in the right places.

In most of the incidents I have analysed, the problem was never that a referee did not know the law. The problem was that the referee was asked to apply the law under observation conditions that did not allow the law to function. That is a technical problem, and it has a technical solution.

The remaining question is whether that technical solution is treated as a priority, in a system where what happens on the pitch always attracts more attention than the cameras placed at the edge of the frame.

---

Some notes on method

What I have set out above rests on personal observation, publicly available league data, and FIFA technical documents on the VAR protocol. I have no access to internal VPF or VFF data, and I do not present my inferences as official conclusions.

My method is simple and repeatable by anyone. For each incident I record four parameters: the number of camera angles that existed, the number used, the actual review time, and the minimum review time needed to reach a certain conclusion with sufficient data. The gap between the last two is a useful indicator of how adequate the data was.

Applied across a large number of incidents, patterns emerge. And the clearest pattern I found is the one I set out at the beginning: most errors cluster at the edge of the frame, where perspective begins to distort geometry, and where nobody thinks to place one more camera.

---

What to watch next

Against the noise of the transfer window and league restructuring, transfer talk usually drowns out the more meaningful signals. For Vietnamese football, I think three signals matter more than most.

First, the structure of refereeing infrastructure. If stadiums are upgraded, the question should not only be capacity but camera position. A stand twenty metres higher can change VAR data quality more than any software update.

Second, the VAR training pipeline. The number of licensed VAR officials is a long-term capacity indicator for a league. As that number rises, decision quality tends to stabilise, even if the number of controversies does not fall immediately.

Third, the degree of technical data disclosure. A league willing to publish process data about its refereeing sends a different signal from one that publishes only results. That signal matters to investors, to broadcast partners, and to fans themselves.

---

Closing

The referee in the VAR room looks at the screen. They have one question, a few angles, and a limited amount of time. They do not have the ability to see what geometry does not allow them to see.

At the edge of the frame, where the camera stands a few metres off the horizontal plane, there is a margin of error nobody can measure exactly. That margin exists through every round, every season, and becomes part of what we call football.

The task is not to abandon VAR. The task is to move those cameras into the right places, and then patiently measure again. Because a league can only begin to answer the question of fairness once it has enough data to see the truth — and the truth, in modern football, tends to sit where few people think to look: at the edge of the frame.