Video verification standards: why one source is not enough for a referee's verdict
**Câu trả lời cốt lõi:** Một đoạn video đơn lẻ không đủ làm cơ sở cho phán quyết trọng tài. Chuẩn VAR yêu cầu bằng chứng từ nguồn chính thức có dấu thời gian và nhiều góc máy đồng bộ; video do khán giả quay chỉ là tư liệu tham khảo, không phải bằng chứng. **Dữ kiện chính:** - Ngưỡng can thiệp VAR theo giao thức là “lỗi rõ ràng và hiển nhiên”, không phải “có khả năng sai”. - Cơ sở dữ liệu 523 trận La Liga và Champions League giai đoạn tháng 8 năm 2019 đến tháng 3 năm 2020 ghi nhận thời gian xem lại trung bình 47 giây. - 74% quyết định việt vị bị phản đối có thời gian xem lại dài hơn mức trung bình nói trên. - Báo cáo dài 48 trang công bố tháng 3 năm 2020 đề xuất giới hạn 30 giây cho mỗi lần xem lại. - Ranh giới tay–vai trong Luật 12 được định nghĩa lại từ năm 2016, khiến các phán quyết trước mốc đó khó so sánh trực tiếp. **Nguồn:** Phân tích dữ liệu trọng tài của Alexander Chen, công bố tháng 3 năm 2020 | Cross-checked: VuaBong.vn **Hỏi đáp liên quan:** - Hỏi: Vì sao video do khán giả quay không được dùng làm bằng chứng VAR? Đáp: Vì thiếu nguồn gốc, dấu thời gian và xác nhận tọa độ, nên không loại trừ được khả năng cắt ghép. - Hỏi: Thời gian xem lại VAR trung bình là bao nhiêu? Đáp: 47 giây, theo cơ sở dữ liệu 523 trận giai đoạn tháng 8 năm 2019 đến tháng 3 năm 2020. - Hỏi: Trọng tài có được kết luận khi bằng chứng chưa đủ? Đáp: Theo giao thức VAR, không; kết luận “chưa đủ dữ liệu” là đầu ra hợp lệ và cần được công bố minh bạch." } ```
On 16 June 2026, in Group C, France met Australia. In the 55th minute Josh Risdon slid in to block a cross, the ball struck the armpit area and bounced away. The referee consulted VAR, watched it several times, and awarded France a penalty. Antoine Griezmann converted. I was sitting at a microphone in a Valencia radio studio telling more than four million listeners that it could not possibly be handball, because the ball had hit the armpit. I said it with total certainty. I said it on the basis of the law I had learned in 2026. I was wrong: since 2026 the boundary between the arm and the shoulder had been redefined, and the armpit sits inside the part of the body that counts as the hand.
The editorial desk had to issue a correction during the half-time break. In thirty years of work, that was the first time I had been contradicted directly, in public, by a sentence I could not take back.
The lesson I took was not about memorising the new clause. It was this: a verdict only stands when the evidence chain stands. Law does not live in memory; it lives in data.
That mistake led me to a project I still maintain. In August 2026 I began logging every VAR-involved decision in La Liga and the Champions League: error code, timestamp, distance between players, ball speed at the decisive frame. By March 2026, when the pandemic stopped football, I had 523 matches in hand.
The first figure that forced me to rewrite the way I talk about VAR: 74% of disputed offside decisions had a review time longer than the average, and that average was 47 seconds. I published a 48-page report on my personal blog proposing a 30-second cap on each review, with the raw data file attached so anyone could check it. The Valencia football federation invited me to advise on process reform.
My correction procedure since then has had three steps, and I still apply it unchanged: publish the correct new data, state exactly which clause changed and from what date, and only then re-analyse the incident. No silent edits.
That report contained a chapter nobody quotes, though it is the part I care about most: the chapter on situations where the VAR team must reach a conclusion without a decisive camera angle. This group is small in number, but it is where refereeing is tested hardest, because the pressure for a definitive answer is greater than any evidence can supply.
The same thing happens outside match play. When a club announces a return date for an injured player, that information is routinely treated as a fact. It is a statement from an interested party — one that wants to reassure sponsors and supporters. The evidentiary standard applied to a phase of play and the standard applied to a medical bulletin are the same standard.
Refereeing runs on a three-tier evidence scale, and almost nobody tells the audience that the scale exists.
Tier one is verifiable official material: the broadcast signal supplied by the rights holder, time-stamped, multi-angle and synchronised, cross-checkable against the semi-automated offside system. This is the only tier permitted to underpin a decision that changes the result of a match.
Tier two is an official statement by an institution: a referees' committee report, a competition organiser's release, a technical department's explanation. This tier carries weight but has a structural defect — it is the account of the party that made the decision. A referees' committee explaining a referee's call is an authoritative source about the incident and, at the same time, an interested party in how the incident is told.
Tier three is spectator-filmed video circulating on social media. This tier may never serve as a pillar of a conclusion, for three technical reasons: provenance unknown, timestamp unknown, editing unknown. A clip without a time code and without geolocation is a story, not yet evidence.
These tiers are not ranked by how compelling they are. They are ranked by how verifiable they are.
Most of the argument in the stands happens at tier three, while verdicts may only rest on tier one. The gap between the two tiers is where refereeing credibility erodes — not because referees are wrong, but because the two sides are arguing with two different evidentiary standards.
One match is only a story. Five hundred matches are the law. That is why I logged 523 matches instead of trusting my memory of a few big nights.
Set 523 matches side by side and a pattern emerges more clearly than any highlight reel. The most fiercely disputed decisions are usually not the plainly wrong ones. They are decisions that are correct in law but were reached on thin evidence — and because the evidence was thin, they cannot be explained convincingly to the audience. A correct decision that cannot be explained still produces a sense of injustice, and that sense does not fade with time. It accumulates into a belief that the system is biased.
This is where I part company with most football commentary. People argue over whether a penalty was right or wrong. I care about the question that comes first: did the referee have enough evidence to reach any conclusion at all?
The VAR protocol sets the intervention threshold as “clear and obvious error”. Those four words sound simple but contain a demanding requirement: the VAR team must establish that a conclusion is obvious. If two angles give two different answers, the conclusion is not obvious — regardless of whether the on-field referee was right or wrong. That is why many decisions that look wrong on television are allowed to stand: the evidentiary standard was not met, not because of some private agreement to protect one another.
The semi-automated offside system promises to narrow the margin of error, but it does not erase it. It merely moves the error from the referee's eye to the sampling frame. When a goal is disallowed because a toe crossed the offside line by a few centimetres, fans do not argue about the law; they argue about when that frame was captured and who certified that moment. The final question is still a question about the provenance of evidence.
I once wrote a piece defending referees using bad statistics. That was the time I betrayed my own principle. Since then, before I say “this decision was correct”, I force myself to answer one question first: which tier of evidence am I relying on, and is that tier verifiable?
With handball the question is harder still. The shoulder–arm boundary has changed at least twice in the past decade, each change leaving behind a layer of old decisions that cannot be judged by the new law. When someone cites a 2026 incident to compare with a 2026 incident, they are comparing two different frames of reference. The date of the information matters as much as the information itself.
In my database, the hardest group of decisions is not the most error-prone one. It is the group with the longest review times and the lowest overturn rate — the VAR team looks for a very long time, then lets the decision stand, because it cannot find evidence strong enough to overturn it. Seen from the stands, that is slowness. Seen from the process, it is the signature of a system trying not to guess.
There is a quality signal I learned in journalism and carried into this trade: a statement willing to state plainly what it does not know is a more trustworthy statement. When a governing body publishes a decision and says outright that this camera angle is insufficient, it loses short-term points with public opinion but holds the standard over the long run. A statement confident in every detail, by contrast, is usually one written for speed rather than completeness.
There is an assumption nearly the entire VAR debate rests on, and I believe it is wrong: that more camera angles produce more consistency.
My data does not say that. As the number of angles rises, the “clear and obvious” threshold is quietly dragged downward. People begin to expect that with enough coverage every incident has a single correct answer, and that failing to find it is the fault of the person looking. That expectation belongs to crowd psychology, not to technique. It produces exactly the error I made in 2026: asserting with certainty before verifying, because nobody wants to be the one who says “I don't know yet”.
The consequence few notice: when pressure for an answer rises, the quality of verification falls. A VAR team under time pressure and public pressure tends to pick the fastest conclusion it can justify rather than the correct one it can prove. I still defend the 30-second cap proposed in the 48-page report, but I have to be explicit: a time limit only helps if it cuts the argument, not the verification.
The second blind spot lies in how we measure. Referees are judged on the final outcome, when what deserves judging is the process that produced it. A referee who reaches the right conclusion by guessing and a referee who reaches the right conclusion after checking three tiers of evidence receive the same mark on the spreadsheet. Referees do not need to be protected. They need to be understood through correct data. And correct data must measure process too. Fans are not short of information; they are short of a standard for telling which information is trustworthy.
I am not proposing that referees be entitled to say “I don't know” and stop there. I am proposing something narrower: the right to publish the chain of evidence behind a decision, together with an explicit note on which parts remain unverified.
A shocking decision is not reckless if it is built on five hundred foundations. But a decision built on a clip of unknown provenance is reckless, even when it is right.
If every major verdict had to come with a public evidence chain, would audiences accept a conclusion that simply reads “insufficient data”? How audiences answer that will determine whether VAR becomes an instrument of transparency, or just a fresh coat of paint over the same old ambiguity.
