No Report Written in Advance: Why Referee Data Must Pass Through Two Sources

In September 2026, at Groupama Stadium, I recorded in the Lyon–Marseille match report that Dimitri Payet had committed three fouls in the first half. Four days later, the Ligue 1 organisers returned the file with a single line: the figures do not match the video. I rewound the footage at minutes 27, 34 and 41 — still three incidents. Then came minute 44, when Payet leaned into an opponent and pulled his shirt in an off-the-ball tangle, and the referee blew his whistle. I missed it because my eyes were following the ball rolling towards the left flank. Four fouls, not three. An error like that was enough to put me in front of the footage of all 12 Marseille matches that season for four weeks. The job of a league disciplinary reporter is record-keeping, separate from commentary. We log punishable behaviour, with the minute, the position on the pitch and the type of incident. Every entry in the report goes straight into the file of the Ligue 1 Disciplinary Committee, which decides suspensions and fines. A foul counted short can keep a player who has already reached a card threshold on the pitch for another round. A foul counted twice can turn an ordinary duel into a bad mark in the personal record a player carries through his career. That is why every organisation with a reporting system requires two independent scribes, reconciled only at full time. Nobody is allowed to close the book before the final whistle, because while the ball is still moving, behaviour is still being generated. My mistake in the 2026 World Cup qualifying campaign taught me: the report is never written in advance. After the Groupama incident, I built a personal error-code table of 47 codes, grouped by type: tactical fouls to stop counter-attacks, fouls inside the penalty area, off-the-ball fouls, dissent towards referees, time-wasting, and the behaviours that only appear from the camera behind the goal. Each code must carry at least two independent sources: the primary video, and a second, separate data source — a broadcast feed, an analyst's log, or the referee's own match sheet. If the two sources disagree, the incident does not enter the system. I call it the two-key rule. In the summer of 2026, when Russia hosted the World Cup, my editor assigned me a feature on yellow cards. I decided not to chase the heavily televised fixtures. My method for covering the 2026 World Cup was a net: small mesh, no fish missed. I selected the 14 lowest-scoring group-stage matches — the ones the media treated as dull — and logged every counter-attack-stopping foul. The result sat with Carlos Queiroz's Iran. Across three group games, Iran committed 23 counter-attack-stopping fouls, the highest figure in the tournament. Not the highest in the matches I watched; the highest of all 32 teams. The European referees' council later cited the piece, and I accepted an invitation to collaborate with So Foot magazine. What matters is that those 23 incidents barely appeared in mainstream analytical output. Automated tracking classifies a foul by where it happened, not by what the offender intended. A body-check in midfield in the 78th minute, with team-mates pushed up, is tagged as a midfield foul exactly like a mistimed tackle while your own team is in possession. Separating tactical fouls from the rest requires rewatching the incident, counting players from both sides in the attacking half, and noting the timing. A similar case sits in the group of metrics packaged as measures of effort. Distance covered and sprint counts are usually presented as proof of commitment. Running without purpose still produces good numbers. A centre-back can cover 11 kilometres in a match without once interfering with the opponent's attacking structure; a midfielder can cover 13 kilometres merely by being dragged towards the ball. In my disciplinary log, distance covered has no code at all. What a player does without the ball is what I number. Since VAR arrived, the evidentiary standard for a disciplinary report has shifted. A foul once described through the feel of a man standing on the touchline is now re-measured from multiple camera angles, and the referee can review his own decision. That ought to make my job easier. In practice the pressure has risen, because every note I write can now be checked against footage the whole committee can see. The counter-intuitive part runs against popular expectation. More data gives error more places to hide, because a wrong figure still looks tidy and still fits the format. When I wrote three instead of four, the tables stayed balanced, the charts stayed neat — only the video objected. A system with a single source will never catch that. The second source exists to contradict, not to confirm. Another blind spot is rarely discussed: the referee's threshold for punishment is not constant between matches. The same shirt-pull is whistled by referee A in the 10th minute and waved away in the 85th, once the game is settled. VAR's clear and obvious standard must differ from the standard for a disciplinary record, because one serves match management and the other serves the file. Feeding both standards into the same data table is a methodological error. The sports data industry is expanding faster than its rulebooks are updated. I have written about esports betting and found that the pace there is quicker still, because match cycles are shorter and integrity records are thinner. The mechanism is identical: weak records, slow regulation, powerful data. When those three drift apart, the integrity of the result is the first thing to pay. If I had to pick one task for the season now running, it would be checking the second source before publishing any table of fouls. My mistake in the 2026 World Cup qualifying campaign taught me: the report is never written in advance, and it is never written by one person alone. And which Ligue 1 side this season is committing the most counter-attack-stopping fouls that nobody has noticed? I am still counting.