When the Frame Is Not Enough: The Craft of Saying "I Don't Know"

When the Frame Is Not Enough: The Craft of Saying "I Don't Know"

It was the 88th minute of a La Liga matchday-two fixture I was re-watching on tape when the referee pointed to the penalty spot. In the VAR room, four people rewound the same two frames: defender extends a leg, forward goes down. No camera angle showed the point of contact. After two minutes and fourteen seconds, the referee was called to the pitchside monitor, watched it back, and stood by his decision. The stands howled. In my notebook that night there was a single line: "D-13 — upheld, insufficient evidence to overturn."

The interesting part sits elsewhere. Across those two minutes and fourteen seconds, nobody in the VAR room said the words "we don't know." They said "there's nothing clear." The two phrasings sound similar, but on a professional scale they diverge sharply: one concludes something about behaviour, the other concludes something about the quality of the evidence. In modern football, the second is the harder conclusion to speak out loud.

In 2026, aged 23, I mispronounced striker Sardar Azmoun's name three times in a single live half, despite six pages of preparation. The lesson I took away wasn't about a surname. It was this: when people aren't sure, they tend to talk louder rather than slower. Referees are people too. That is why I track "not enough evidence" decisions as seriously as I track goals.

FOUR CATEGORIES, ONE THRESHOLD, THREE GENERATIONS OF TECHNOLOGY

IFAB's Laws allow VAR to intervene in exactly four categories: goals, penalties, direct red cards, and mistaken identity. A second yellow is not on the list. Outside those four categories the technology has no door to walk through, however loudly the stadium shouts.

What gets misunderstood most is the threshold. It is written as "clear and obvious error," or "serious missed incident." That wording concedes a legitimate grey zone, where an on-field decision stands because nothing has proven it wrong, rather than because it is right. This is the point most spectators reject, and I understand why: they came to watch justice executed, not to watch an evidence file appraised.

La Liga introduced VAR in the 2026-19 season. Seven seasons later the technology has moved through three generations. The first reconstructed incidents by hand from broadcast camera angles, entirely dependent on where the rights-holder chose to place cameras. The second brought in semi-automated offside cameras, cutting offside checks from minutes to seconds. The third builds a three-dimensional model of the whole incident, not just the offside line.

None of the three generations has touched the underlying problem: some incidents leave no clear physical trace in the frame. A hand on a shirt at 25 km/h, from a camera forty metres away, under floodlights, will always be an inference. Technology makes that inference faster, cheaper and more confident. It does not make it disappear.

In Vietnam, VAR arrived later and arrived unevenly. The national top flight runs VAR in some matches and some rounds, depending on pitch conditions and staffing. Lower divisions have almost none. I keep that in mind for the section on fairness below, because technology isn't allocated by need; it's allocated by cash flow.

THE 3F MODEL AND THE LIMITS OF EVIDENCE

I built the 3F model for myself after the 2026 World Cup, once I had watched dozens of incidents on repeat and realised I kept starting my analysis in the wrong place. 3F stands for Foul — Field — Frame, and the order of the letters matters more than the letters themselves.

Foul — the act: whether a law-breaking act occurred, judged in the sense of the Laws, not in the sense of feeling. It is the only question whose answer can be "yes" or "no."

Field — the field of observation: where on the pitch the act occurred, who has authority over it, and which of the four reviewable categories it falls into. Get this step wrong and everything downstream is meaningless, because an incident outside the box can be a red card but can never be a penalty.

Frame — the footage: whether the frame is sufficient to answer the question at step one. This is the step most television debates skip entirely. They argue about behaviour using footage that cannot establish behaviour.

The three steps must be taken in order, and step three holds a veto over step one. When Frame answers "not enough," Foul has no answer, however strongly the viewer feels one.

Since last season I have logged every VAR check in the La Liga matches I follow, using four data fields: check duration, outcome, number of frames used in the conclusion, and footage-quality grade. The quality grading is my own invention and my own responsibility, in three tiers: A — a decisive frame exists (point of contact, ball, foot position all visible); B — frames suggest but do not decide; C — no angle reaches the question.

| Tier | Footage characteristics | Typical outcome | |---|---|---| | A | Contact, ball and body shape all visible | Overturned, short duration | | B | Movement visible, contact not visible | Drawn out, mostly upheld | | C | No angle reaches the question | Upheld, argument continues after the whistle |

What I took from more than a hundred entries wasn't the overturn rate. It was that Tier C exists at a stable frequency, and that frequency has not fallen over time. Technology improves, Tier A expands, Tier B shrinks, and Tier C stays right where it is.

Data does not blow the whistle, but it illuminates the angles the naked eye skips. Here it illuminated an uncomfortable one: the limits of the very technology we have spent heavily to trust. Some questions cannot be answered with pictures, and never will be, because the answer isn't in the input data.

Put plainly, Tier C is the only tier where the correct conclusion doesn't sound like a conclusion: "not enough evidence." Those four words never appear on a scoreboard, never appear in a bulletin, and never appeared in any analysis I wrote before 2026, because I also once believed an analyst must always have an answer. I was wrong.

WHY "NOT ENOUGH EVIDENCE" IS A PROFESSIONAL CONCLUSION

In statistics there is a concept I learned in my second year at university and needed years to fully grasp: the confidence interval. An estimate without a confidence interval is a meaningless number. In refereeing, "not enough evidence" is a confidence interval spoken out loud.

A referee saying "not enough evidence" is doing three things at once. First, admitting the limits of the data available. Second, returning the decision to where the Laws place it — the field. Third, refusing to manufacture a false certainty.

False certainty is the most dangerous thing in this profession. A referee who overturns a decision on a blurry frame is no worse than a referee who ignores a clear one. He is worse in a different way: he teaches everyone that evidence doesn't matter, only that the conclusion sounds decisive.

In 2026, when the Euros were played in empty stadiums, I joined the referee-analysis desk of a large sports outlet. My job was to build an index I later named the Referee Sensitivity Index: for each referee, take the last 40 matches and measure card frequency, foul-whistle tendency, penalties per match, and willingness to follow VAR. Colleagues called me mechanical at first. When a piece on a referee carrying an 0.8 VAR errors-per-match figure was shared several thousand times, they started believing.

That index taught me something I had never read anywhere: Tier C is not distributed randomly across referees. Some referees land in Tier C twice as often as their peers, and the reason is not eyesight. It is positioning, the moment they start moving, and whether they choose to trust an assistant or decide alone. Those things are measurable, fixable, and teachable. This is the kind of data a refereeing ranking should use, instead of counting how often a referee was criticised on television.

CASE STUDY: MATEU LAHOZ AND ARGENTINA — NETHERLANDS

At the 2026 World Cup, the quarter-final between Argentina and the Netherlands, refereed by Mateu Lahoz, entered history in the way no referee wants. He issued 18 yellow cards, the highest number ever recorded in a World Cup finals match, and added a red card for Denzel Dumfries after the penalty shootout. The second half ran more than ten minutes of added time.

My post-match analysis, built on the clearest camera angles I could assemble, identified four incidents I believed should have been handled differently. A former FIFA referee called me a "fantasy academic." My content director defended me, not because I was right, but because every conclusion in the piece was tied to a specific frame and a specific timestamp. People can dispute a conclusion; they cannot dispute the data.

What I did not write in that piece, and write now: Lahoz did not fail at seeing. He failed at match management. A referee can read every incident correctly and still lose control, the way a doctor can diagnose every symptom correctly and still lose the patient. Argentina — Netherlands is the cleanest example of how being right on each incident does not mean running the match right.

This is also where I have to check myself. If I use frames to dissect every Lahoz decision, I must accept that the same set of frames can be used to dissect every sentence I write. I chose that approach, because the alternative is writing by feel, and feel cannot be verified.

THE ECONOMICS OF EVIDENCE

This is the part I weighed longest before writing.

VAR runs on money. A VAR room needs four to six people, a fibre transmission system, at least twelve camera angles, and a redundant feed. That cost only makes sense where broadcasting contracts are large enough to carry it. Where they aren't, football carries on with one referee, two assistants and a pair of eyes.

I have followed Spain's lower divisions long enough to know what happens there. A wrong decision in the third tier has no VAR room to review it, no footage to analyse, no article to demand justice. It happens, it ends, it disappears. Players there absorb the punishment for someone else's mistake and have no channel to appeal through.

Every season, the media tells a fairy tale about a small club knocking out a top-flight side in the Copa del Rey, and every season that tale is consumed within two weeks and thrown away. The debate about resource allocation — broadcasting money, VAR money, professional refereeing money — never follows the story. Fairness, at its deepest layer, is a budget question, and football has never wanted to say so.

THE FINAL TWENTY MINUTES AND THE FATE OF TIER C

Since the Laws allowed five substitutions, the physical architecture of a match has changed. A team can swap half its attack on the hour and still keep a substitution in reserve. The early second half has become a tactical testing ground, while the final twenty minutes have become genuine attrition warfare: higher speed, wider gaps between lines, more collisions.

For anyone tracking referees, the consequence is concrete. More incidents require judgement, the referee's reaction window shrinks, and footage quality drops because everything happens faster. Tier C — the tier with no decisive frame — clusters heavily in exactly that window.

Which means the hardest, most contested and least verifiable decisions land precisely when the referee is most tired, the teams are most tired, and the crowd is hottest. That is a structural problem, not a problem of individual competence. No training course fixes it, because it does not live in people.

A contingency plan isn't for avoiding a crisis; it's for standing firm inside one like a referee in a storm. During the relegation season I followed, I built a four-part series to answer one question: if everything collapses, what percentage chance does this club still have? My conclusion then was not pretty, but it was verifiable. The lesson repeats: a bad plan that is clear beats a beautiful belief that is vague.

A LESSON FROM AN EMPTY REPORT

There is a professional story I have never told. In my workflow, every analysis passes through two stages: the raw material is decomposed into information points, and only then is it analysed. Stage one is the fragile one. If the source document is unreadable, or fails during extraction, stage two receives an empty dataset.

The worst outcome isn't the empty dataset. The worst outcome is an empty dataset that looks like a normal one. When every field is blank, they don't raise an error, they simply go quiet. And a rushed analyst fills the gap with guesswork, because in this trade a gap is always more uncomfortable than a mistake.

My system now has an integrity gate: if there is no title, no source and not a single information point, the item is flagged as failed, not as weak. That distinction matters in football as much as in a data room. "No problem found" and "no problem assessable" are different sentences, and blending them is the fastest route to a decision nobody can audit.

Inside VAR, Tier C is the pitchside version of the empty report. The correct response to an empty report is not to invent content; it is to say clearly that it is empty, then start again from the original document. That is why I believe a referee brave enough to say "I have no basis" protects the match better than one who guesses to get it over with.

THE CONTRARIAN ANGLE

I know what I have written above will be read as an excuse.

Supporters hate the phrase "not enough evidence," and they have a legitimate reason: they paid for tickets, they paid for broadcast subscriptions, and they were promised that technology would settle the arguments the human eye could not. When technology answers with "not enough evidence," that promise breaks. The feeling of being cheated is a valid feeling, and no technical argument erases it.

But there is a paradox worth putting on the table. The same stands, when a referee overturns a decision on a blurry frame — and referees sometimes do so to avoid looking hesitant — howl even louder. Viewers want decisiveness, and viewers want correctness. In Tier C, those two cannot coexist. Every VAR argument, at its deepest level, is an argument about who is entitled to be disappointed.

The second contrarian point I apply to myself: if I demand that referees be brave enough to say "not enough evidence," I must accept that readers have the same right to say it to me. Across more than fifteen years writing about football, the number of conclusions I published that my data could not carry exceeds the number I would like to admit. A mistake on live broadcast is like a mistake on the pitch: look straight at it, learn, and blow the whistle for the next match. That principle only has value if it applies to the writer too.

And there is one more thing I must say, even though it is not easy listening for my own colleagues. A great deal of television refereeing debate is led by people who have never opened the Laws of the Game, which is fine as entertainment but wrong as information. A pundit can say "that's a red card" a hundred times without once asking which frame proves it. That habit has produced a generation of viewers who know exactly what they want and very little about whether what they want has any basis.

TAKEAWAY

Accuracy is a technical goal; fairness is a moral one, and the two do not always point the same way. A decision that breaks no law can still be wrong in substance; what people need is fairness, not merely precision.

If I could propose one reform for this season, it would not be more cameras. It would be releasing VAR room audio to the public, in full, uncut. When three million people hear a referee say "I can't see contact, I have no basis to overturn," those words stop being an excuse and become part of the match. Football has learned to live with goals nobody saw coming. It will have to learn to live with answers nobody wants to hear.

Some situations have no absolutely correct answer, only a person with enough courage to own the decision. People remember the goals; I remember the whistles that protected them.