An Entertainment File Filed Under Football: The Cost of a Mislabel

On Sunday, 4 October, the final of a reality television programme in Mexico ended. Karina Torres finished third. Paola Suárez — her close friend and a social-media influencer with a large following — appeared in a video, face flushed, saying she did not accept the result and would not take part in a future season. Within hours the clips were reshared, cut, captioned, and a wave of comment swallowed both names.

Ordinarily, that is an entertainment story. But this file was stored in a data drawer labelled football. No club appeared. No player, referee, league table or rule. The entire record consisted of two influencers, one television programme, and audience reaction to a vote. I read it three times to be sure I had not opened the wrong column.

The first rule I set for myself in 2026, when I edited the rules desk in Hai Phong, was this: before you analyse, confirm the sport. It sounds redundant. But in an automated intake system, where every record is labelled by an algorithm or by a hurried data-entry step, that confirmation is the only barrier between clean data and noise.

A file with no football in a football drawer is not the fault of Paola Suárez, not the fault of Karina Torres, and not the fault of the Mexican audience. It is a system fault. And a system fault, in the way I have been writing for seven years, is never an isolated incident.

What deserves analysis is not the anger but the mechanism that produced it — and that mechanism is real, it simply does not belong to football.

Three elements in this file transfer to football, at the level of media mechanism rather than sporting expertise. The expectation gap is the most visible: a fan believes the person they support deserved a higher position than the recorded outcome, and anger ignites inside that exact gap. Attached to it is a single-voice statement structure — one voice, no data, no second source, no reply from the party named. And a short life cycle, days to weeks, then it fades.

I have met all three intact on Vietnamese pitches, with different names attached.

In 2026, at Lạch Tray stadium, referee Nguyễn Văn Kiên sent off centre-back Nguyễn Hữu Phúc in the 68th minute for a challenge from behind. On replay, there was no dangerous play. I wrote that night, quoting IFAB Law 12, pointing out that a direct red card is correct only when the act is serious foul play or denies an obvious goal-scoring opportunity. The piece drew 45,000 reads, nine times the section average. The desk asked me to open a dedicated rules column.

What I remember is not the number. It is the hundreds of comments below insisting the referee had been bought. Not one offered a single piece of procedural evidence. Expectation gap plus single-voice statements plus a short life cycle — the same three ingredients, switched from a reality show to a match.

From then on I abandoned crowd-feeling analysis and moved to structure: quote the rule, set the context, analyse the consequence. Every later piece opened with a rules basis, leaving no room for vague interpretation.

The 2026 World Cup gave me my first quantitative tool. Based on my experience tracking matches in Russia, I logged every refereeing intervention across all 64 games. In the final between France and Croatia, referee Néstor Pitana awarded 31 fouls and showed only four yellow cards. The average match contained 3.2 minutes of dead ball caused by refereeing stoppages. With numbers, the argument changes character: people stop asking whether a referee is biased and start asking whether his intervention threshold is consistent. That question can be answered.

Euro 2026 was where I saw most clearly that technology does not erase controversy, it only relocates it. In the quarter-final between Spain and Switzerland, the opening goal was allowed even though forward Ferran Torres stood roughly 0.3 metres offside; the VAR technician never drew the offside line on the frame. I compared 48 matches and found the VAR error rate at the Euros was 1.8 times that of the 2026 World Cup. The VAR series proved controversial enough that UEFA had to issue an email explaining a new process to media outlets.

Then came the 2026 World Cup. I was among three Vietnamese journalists admitted to the VAR operations room in Qatar for the semi-final between Argentina and Croatia, watching semi-automated offside technology run in real time. After the tournament I wrote a three-part implementation guide for the VFF, with one number persuasive enough for anyone: offside-line error fell from 0.4 metres to 0.1 metres. The VFF referees committee used that document as the basis for a pilot project in V.League 2026.

I tell these three stories not to boast. I tell them to prove one point: every decent football argument ends with a question that can be verified. The entertainment file does not. It has no score, no goal, no rule, nothing to cross-check. It has only emotion, and emotion cannot be measured with any instrument I hold.

A record with no sport in it cannot be analysed with a sporting framework — mislabelling it manufactures false signal, and false signal spreads faster than the original error.

This is where I want to push back, including against my own side.

There is an easy argument to make: keep it, it might serve later as an example of fan reaction. I do not buy it. Football already has thousands of fan-reaction examples, each attached to match context, sequence and scoreline. A record with no match context cannot serve as an example; it only distorts the entity graph. A model trained on dirty data produces dirty conclusions, and dirty conclusions cannot be repaired by adding clean data afterwards.

The second easy argument: a label is a technical matter, fix one line and it is done. True, fixing one line takes ten minutes. But finding out why that line was mislabelled takes far longer, and admitting that an entire labelling process has a hole takes longer still. A pandemic did not create loopholes in the law — it knocked on every crack that already existed. This labelling error is the same: it does not create a new problem, it exposes an old one.

And here is the part I will not concede. If one file like this reaches the football drawer, it does not arrive alone. Where there is one mislabelled record, there is a mislabelling rule; where there is a faulty rule, there will be a next record. The right question is not whether to delete this file, but how many others sit in the same drawer.

A referee's mistake is never an isolated incident — it is the entire rulebook filing its own self-assessment. The same holds for data. One labelling error is the whole intake process filing its own self-assessment.

People see a viral video; I see a label field filled in too quickly.

So what should be done, concretely, so that readers do not have to trust a promise?

I propose a minimum checklist, short enough to be used in practice. Does the record contain at least one entity from the football industry — a club, player, referee, league or governing body? Does it contain at least one verifiable fact with a specific source and date? Is there a reply from the party named? And does the record belong to a competitive event that can be cross-checked with data?

Those four questions need no advanced technology. They need one person willing to read and sign. That is the kind of cost newsrooms cut first, and the kind that proves most expensive when missing.

As for the people in the original story: Karina Torres finished third, Paola Suárez was angry, the audience split in two. I am in no position to judge any of them, and I have no evidence that the result was interfered with. The source I read contained no reply from Karina Torres or the programme's producers. When there is no reply, no adverse inference may be drawn against anyone — that is a principle, not a courtesy.

My job is narrower: to keep the football drawer holding only football. It sounds trivial. But every decent sports-analysis system rests on a single assumption — that what goes in is what it claims to be. When that assumption breaks, every conclusion downstream is worthless, including the correct ones.

Changing one rule takes ten minutes; admitting the rule is wrong takes ten years. Changing one label also takes ten minutes. The open question for those running Vietnamese sports-data systems: when did anyone last check whether their football drawer was clean?