The Empty Analysis File: Why I Refuse to Write Before the Data Arrives

At 2:40 a.m. in Shenzhen I opened a nine-section analysis file handed over by the text-extraction stage. Every field contained words. There were section headers, a tactical comparison table, a six-row risk matrix, even a source-attribution block. I read all four pages and realised I had just read a product containing no club name, no player, no scoreline and no date. It was a flawless file about a match that never took place. Sitting still in that small room, I thought of another night. On 10 July 2026, the World Cup semi-final between France and Belgium in Saint Petersburg. I was seventeen, in the eleventh grade, filming and editing my own analysis videos for social media. I went live and stated flatly that Didier Deschamps would send France into a high press, winning the ball in Belgium's half. He did the opposite: France conceded the ball by choice, dropped their block and waited for the exact moment to attack in transition. France won 1-0 through Samuel Umtiti's header in the 51st minute, from an Antoine Griezmann corner. The comments under that video were not gentle. What I did next shaped my career. I did not delete the video. For seven straight days I reopened the full 90 minutes and logged every touch by France's centre-back pair, every off-ball movement in midfield, every sideways pass Belgium were forced into once the space had closed. Belgium had more of the ball, yet the genuinely dangerous chances leaned towards the team that chose to sit deep. I understood what I had been missing. It was not reading of the game. It was evidence. I had told a tactical story built on guesswork, then dressed it in the tone of a man who knew. A match without a roar still tells you more than a whole loud season. From then on I set myself a hard rule: never write analysis before verifying at least three independent sources and reviewing the full match footage. Every tactical claim had to carry concrete figures, from touches and distance covered to sprint counts, rather than a feel for the rhythm of the game. The rule sounds simple. It has saved me many times from publishing sentences I could not defend. Modern football analysis runs in two stages. The extraction stage reads the source text and pulls out entities, events, timestamps, quotations and time sensitivity. The deep-analysis stage takes that data package and opens several dimensions: tactics, club finance and the transfer market, results and the opinion cycle, league landscape and team positioning, rules and governance, management and the dressing room, risk, media narrative, and the industry transmission chain. If the input package is empty, every dimension is a house without foundations. A writer can present as many sections as he likes, but none of them survives one simple question: where did you get this. The file I opened that night was exactly in that condition. Extraction returned a single label: football. No original headline, no outlet name, no article type, no one-sentence summary, no information points, no identified entities. The only field with content was the domain label, which is the one thing you can guess without reading the article at all. Every other field was either blank or contained an instruction for whoever ran the next step. The difference between usable analysis and plausible-sounding analysis is this: usable analysis always shows what it rests on. The biggest files in European football I have followed all rest on dated documents. In February 2026 the Premier League announced 115 financial charges against Manchester City. In November 2026 Everton were deducted 10 points for breaching loss limits; by February 2026 that deduction was reduced to 6 points on appeal. In March 2026 Nottingham Forest were deducted 4 points. In January 2026 Juventus were docked 15 points in Serie A, and by May 2026 the deduction was revised to 10 points. Four files, hundreds of pages, and every conclusion tied to a specific date, a specific body, a specific clause. That is the standard of the trade. An accusation only carries weight when the person making it accepts being checked. Analysis is the same. If I write that a team's midfield lost control of the centre, I must be able to say who I compared it against, in which match, using which metric, and on which date. Without those four things, the sentence is just an opinion presented beautifully. In the 2026-2026 season I followed Shandong Taishan through a compressed fixture period, when competitions were played under restricted attendance. I had access to the training ground and the dressing-room area. The team went five matches without a win and dropped places in the table. In a league without fans, I hear boots striking grass more clearly than the referee's whistle. When that run stretched on, the media's first reflex was to blame the defence. I requested GPS data on distance covered and sprint counts for the whole squad across the last five matches. The results located the problem in midfield: ball recoveries had fallen, the distances between lines had stretched, and the back four were facing far more one-on-one situations than earlier in the season. The defence suffered the consequence; it was not the cause. Had I written from feeling, I would have pointed at the right pain but the wrong root. Collapse does not come from a single conceded goal. It comes from hundreds of small details ignored. During that period I also logged two internal signals: a young midfielder losing focus after an internal disciplinary action, and the first-choice goalkeeper showing signs of shoulder pain while still playing. I did not publish those signals as conclusions. I put them into an internal report with a request for a medical check and a cross-reference against match data, because the dressing room is where truth outlives any contract. An unverified observation remains an observation; it becomes a conclusion only when a second person confirms it independently. Around the same time I had one unforgettable shift. On 3 July 2026, the Euro quarter-final between Ukraine and England in Rome. I was twenty, working as a data contributor for a football site. The first half had just ended when I was taken to hospital with appendicitis. The match kept running. England won 4-0: Harry Kane opened the scoring in the 4th minute, Harry Maguire headed in on 46, Kane completed his double on 50, and Jordan Henderson sealed it on 63. I sat on the hospital bed with a drip in my arm, using a laptop and a phone to cover the remaining 45 minutes. I split the work with two colleagues remotely: one handled the data, one reviewed the flow of play, while I set the structure and edited. The process had four steps: identify the core information, classify the data, assign the work, cross-check. The piece was finished twelve minutes after the final whistle. Writing from a hospital bed, I learned that the pulse of a match never waits for anyone. Looking back, that night in Saint Petersburg, that season in Shandong and that hospital shift all taught me the same thing: what I must protect in every article is not a viewpoint but a path back to the evidence. So when I opened that nine-section file and found it hollow, I knew exactly what had broken and where. The first gap was the source field. The file named no original article, no publication date, no credibility tier for the outlet. In football data, the source is part of the fact: a transfer line from a journalist with direct access to an agent carries very different weight from a line recycled on social media. Remove the source field and everything else automatically loses its checking value, however professionally it is presented. The next gap was the entity list. No club, no player, no competition, no coach. This is odd, because football journalism almost always names people. Even a purely opinionated piece has to mention someone, since football is a game played by specific humans. That emptiness indicates the extraction step never ran, rather than the original text having no entities to extract. The following gap was structural. The template defined two fields as derived: the entity list and source quality were both to be taken from the information points. When the information points are empty, those two fields are guaranteed to be empty as well, with no further error required. One broken link collapses the whole chain, and the reader at the end of the chain sees only the fallen column, not the break. Another gap was time sensitivity. The file stated plainly that timestamps had not been assessed. In sports reporting, time is half the content: a note on form from three weeks ago is entirely different from a note on this morning's injury situation. Without timestamps, a writer cannot separate hot news from stale news, and a reader has no way to check. The remaining gap is the most dangerous because it is invisible: formal completeness mistaken for substantive completeness. The file had tables, a matrix and headers, so it felt checked. A hurried reader will sign it off. An automated system will pass it on. I once sat in a meeting and watched a report get approved purely because it ran to forty pages. Collapse does not come from a single conceded goal. It comes from hundreds of small details ignored. In this case, one large detail was ignored: there was nothing to analyse. The common misunderstanding about analysis from the outside is the belief that the more complete the product looks, the more trustworthy it is. In football, belief always flows towards the pre-packaged story. Media loves the underdog because the upset generates traffic, and an analysis with every section filled will be shared far more widely than one that dares to say there is not enough data to conclude. Yet the underdog-miracle narrative is precisely where truth bends most. Only by following a weak team for a full season does a writer see what miracles cost: weeks short of players, night flights, contracts not renewed, footballers playing out of position just so the team can field eleven. There is a second, opposite misunderstanding: that data analysts have walked fully into the dressing room and understand the team better than the people standing in it every day. Data does not lie, but it does not place itself correctly either. Distance covered only means something when you know the player's assignment. Sprint counts only mean something when you know which flank he plays and what he is carrying in his body. A spreadsheet is always correct at its own level, and always incomplete at the human level. Finally, process is not a shield. I have publicly championed process discipline when everything collapses, but process only stops me panicking; it does not absolve me of the words I publish. The four steps of identifying core information, classifying data, assigning work and cross-checking become meaningless if I use them to legitimise a page I know is hollow. That is a line I do not intend to cross again. The internal signals I will watch next week are very specific. The entity list must contain at least one club name. The source block must contain at least one news organisation with an absolute publication date. The time-sensitivity field must no longer be blank. And every tactical claim in the final draft must trace back to a metric or a piece of footage. Those four signals are enough to decide whether a file gets opened or sent back, and enough to separate a writer at work from a writer performing. That season, I learned that football's most fragile form is also its most durable: it can vanish in an afternoon, and it can be rebuilt from the smallest details. If tomorrow I open another file full of words with not a single name in it, I will return it with one note: tell me who we are talking about, which match, and on what date. Would you be willing to sign your name to an analysis you cannot cite a single source for?