Nine Empty Cells in Manchester: When the Only Correct Verdict Is Insufficient Information

2:14 a.m., Manchester. On the screen is a spreadsheet with nine tabs, one for each analytical dimension, and all nine are empty. I have been sitting there for four hours with an empty file, coffee long gone cold, and one very specific temptation: fill in a cell. Just one. A name, a metric, a sentence that sounds reasonable would be enough to make the spreadsheet look like a report, and enough to stop the editor from asking a single follow-up question. I did that once. In 2026, aged 19, I filled the "player cautioned" cell with a name that was wrong. Six weeks later I was still logging the 189 card incidents of the 2026 World Cup to rebuild the only backup I had left: reference data. Tonight I am filling in nothing. The report will be published with nine identical lines: insufficient information. My job is not to report who won. I reconstruct the umpire's decision-making process: who saw what, at what moment, under which rule framework, and in what language that decision entered the official record. A professional tennis match leaves four layers of documentation — multi-angle footage, sensor data from the electronic line-calling system, the chair umpire's official report, and the statistics the tournament publishes afterwards. Those four layers rarely match perfectly, and the gap between them is where I work. This file reached me as the output of a first extraction pass. That pass is meant to pull out the headline, the information points, the named entities, the core viewpoints, time sensitivity and source quality. The result came back empty. No headline, no entities, no player, no tournament, no timestamp. In my workflow this case has a name of its own: null-value handling. The label attached to every cell in that situation is a phrase I use constantly: insufficient information. It does not mean "nothing happened." It means "the available sources are not enough to reach a conclusion." Those two things sit very far apart, and in my trade, confusing them is the origin of almost every serious error. Every empty cell is a moment when I refuse to put pen to paper. In a newsroom, refusing to write is the most expensive act there is, because the cost is not silence — the cost is having to own a product that looks unfinished. A nine-dimension analysis with every cell blank reads like helplessness. But if I fill it with a player who does not exist in the source, what I have produced is not analysis. It is fiction formatted as a record. The uncomfortable part is that I have everything I need to fabricate convincingly. My nine dimensions already have their scaffolding: playing style, form data, tournament system, the landscape of the tour, rules and governance compliance, team and player management, risk, media narrative, and the industry transmission chain. Each dimension has tables, rating scales, and a one-to-five-star field. All it takes is a single name and the whole machine runs smoothly. There is no name. So I start somewhere else: the provenance of the data. Electronic line calling now replaces line judges at many major events, and the operator publishes an average system error of a few millimetres, commonly cited as roughly 3.6 mm. That threshold matters more than any wording in a news story. A ball landing on the line within a margin smaller than that threshold sits in a zone the tool cannot adjudicate. Which means when I see a ball rendered right on the edge of the line, the first thing I check is not the decision. It is the configuration and calibration of the system that day. When data contradicts the eye, trust the data — but never forget to check where it came from. That second clause is the one I paid to learn. In 2026, as a first-year sports science student in Manchester, I volunteered as a data analysis assistant for a local amateur club. In a regional league match, I found that the official record did not register two fouls inside the penalty area. I spent three days reviewing the full video, counting every collision and building a minute-by-minute comparison against the record. The two fouls stayed where they were — absent from every official document. The lesson was not that the referee missed them. The lesson was that the official record does not coincide with what happened on the pitch; it is a data layer with its own error margin, written by someone with a limited view under limited time. From then on, every piece I write carries a cross-check section I never skip: two independent sources per event, and a note on what motive each source might have. Back to the empty spreadsheet. The strongest temptation does not come from the "player name" cell. It comes from the metrics. Without a named player I cannot assign numbers, but if I allowed myself to build a template profile from professional memory, I could write several very smooth pages instantly. I know exactly what I would write, because that is the trap I have met most often. Distance covered and sprint counts are packaged and sold to audiences as effort metrics. They measure movement, not effectiveness. In tennis, a player dragged along the baseline through a fifth set will post a very handsome distance figure. That number does not say he is fighting. It says he is being led from corner to corner. Running without purpose still produces beautiful numbers, and beautiful numbers always find someone willing to quote them. The way I check this is to place effort metrics beside positioning metrics: contact point location, the rate of balls returned in play from the fifth stroke onward, the share of points won in rallies that force a player out of the central zone. When effort rises while the win rate in long rallies falls, the effort metric is measuring the tail of a tactical problem, not the quality of the performance. A metric only means something when it has a baseline. In 2026, I found an anomaly in the disciplinary data: the Portugal national team received cards at a rate roughly 41 percent higher in matches officiated by French referees. I rebuilt 23 matches from 2026 to 2026, set them against head-to-head historical data, and wrote a 3,500-word investigation. The piece was later used by a refereeing researcher at UEFA as reference material when assessing the consistency of officiating teams. I stated plainly in that article that 23 matches is a thin sample. The confidence interval is wide, and part of the gap could come from opponents, from the round, from the surface, rather than from the man with the whistle. The fact that a researcher used the piece does not make it more correct. It says one thing about the trade: what carries value in an analysis room is not the conclusion, but the way the evidence is built. Whenever I conclude early, I force myself back through three layers of checking: player name, minute of the incident, type of card. My 2026 error lived in the first layer. I wrote that a University of Manchester defender received a yellow card in the 23rd minute of the derby against the University of Liverpool, when the card belonged to a teammate. My editor reprimanded me, I had to write an apology, and I spent the following six weeks memorising the disciplinary ladder in the rules of the game alongside the 189 card incidents of the 2026 World Cup as a baseline dataset. Since then I separate two kinds of sentence in every piece: event verification and interpretive commentary. An event-verification sentence needs at least two sources and one timestamp. An interpretive sentence must be flagged as opinion, with a stated level of confidence I am willing to own. That separation sounds like bureaucracy, but it is the only thing that keeps a 3,000-word piece from collapsing on itself. The history of tennis law gives me another layer of protection. The serve clock — the 25-second mark competitions adopted from the 2026 period — arrived after years of argument about match pace. Electronic line calling, which replaces line judges, was expanded gradually and then applied fully at many major events from the 2026 season, Wimbledon among them. Off-court coaching, once a prohibited act, was legalised at major events in the same 2026 season, with limits on timing and method. Those three changes share one property: each was decided under incomplete data. Nobody had a perfect study proving that a 25-second clock would improve match quality, or that removing line judges would reduce controversy. Governing bodies issued verdicts under insufficient information, then published review criteria to adjust later. That is the practical definition of sports governance. A card placed in the wrong slot can change the flow of an entire season. I have been the one who wrote that wrongly. But there is a subtler kind of error: the card goes into the correct slot while the communication lands in the wrong place. The 2026 US Open women's singles final is the example I use when I speak to students. The disciplinary ladder in the rules operates in steps: warning, then point penalty, then game penalty. A player committing three successive violations travels the whole ladder, and the final result is a game awarded to the opponent. The first violation concerned off-court coaching, the second racket abuse, the third language directed at the official. Technically, the ladder was applied in the exact order in which it is written. The controversy did not sit in the ladder. It sat in the fact that the player and the crowd were not clearly told which rung they were on, and in the fact that the same coaching violation had been handled differently on different courts in the same tournament week. Consistency in applying the rules matters more than the individual decision. Seven years later, off-court coaching became a permitted act under the rules. That argument turned into a line of amendment. The way I handle this kind of file is to reconstruct the process rather than judge the outcome. For each decision I need to know what the umpire said, when they said it, which rung the player was told they were on, and whether a comparable precedent existed earlier in the same event. When those four questions have answers, the controversy narrows on its own. When one of the four comes back blank, I enter the familiar label into the file. In 2026 I was assigned to follow an African national team after they reached the semi-finals of a World Cup. I spent four weeks analysing their 12 matches and counted 87 tactical fouls. The notable figure sat elsewhere: their average card rate was about 32 percent lower than European teams at the same stage, while their number of clearances was higher. What I needed to explain was not how well they defended. It was why their card rate fell below the baseline while they were clearing the ball more than anyone. The answer was structural: their defensive system rested on cutting off the man without the ball and blocking passing lanes, not on direct duels. Tactical fouls happened at the preparation layer, in positions where a referee has little reason to reach for a card. The operational lesson is concrete: a low metric does not mean a lack of aggression, and a high metric does not mean dominance. A low card rate can signal tactical discipline, or it can signal fouls committed where nobody is watching. To tell those apart, I have to count where on the court the fouls happen, not how many cards were shown. Back to tonight's empty spreadsheet. The blank is data too. It tells me something no fully populated analysis table can tell me: the input source does not exist, and every conclusion downstream will be a product of the writer's imagination. A piece like that can read smoothly, can be shared widely, and is worth absolutely nothing. This industry rewards decisiveness. A headline with a clear verdict always travels further than one that says the evidence is not there yet. That incentive structure flows backwards into the analysis room and produces a very hard pressure to resist: the pressure to reach a conclusion, any conclusion, to prove the work was done. I have seen data tables filled in purely so they would look like data tables. I have done it. The asymmetry of time is what makes me gentler with other people. An umpire on court has a few seconds, one angle, and a loud crowd. I have three days, nine camera angles, the original record and a spreadsheet. That asymmetry obliges me to be generous towards someone else's decision. But generosity must not become laziness, because laziness is precisely what produces wrong reports and lets them survive for years. My first mistake was not the red card given to the wrong player. It was believing I would never give one to the wrong player. So I propose a small procedural change for disciplinary reports: every conclusion carries a published confidence level, and every file preserves the items that could not be resolved. A report made only of confident conclusions is a report that has hidden the hardest part of the job. A report brave enough to write "insufficient information" exactly where it belongs is a report that can still be checked ten years from now. VAR is not wrong. The person operating VAR is wrong. And that is precisely where my work begins. Next season I will track three things. First, how much detail electronic line-calling data is published at, because the level of detail determines whether I can audit calibration at all. Second, off-court coaching figures, a dataset that barely exists even though the rule has changed. Third, card rates broken down by officiating team, something I still have to build by hand because nobody publishes it. A tournament is a system. Every refereeing decision is a variable. My job is simply the act of verification. Tonight the verification returned a null result, and I will publish that null result, because an honest verification of an empty input still beats a full verdict assembled out of thin air.