The Empty Board and the Discipline of Silence in Data Journalism

That night my screen held a table of twelve cells. All empty. No tournament name, no player name, no date, no move, no opening system, not even a single game to hold on to. Only short notes lined up neatly like overturned chess pieces on a board. I sat looking at it for a long time. Not because I was confused, but because I already knew the pressure that would follow: the newsroom waiting for copy, readers waiting for copy, and at least three very elegant opening lines about some "generational handover" in chess already sitting in my head. I could type a headline, assemble a few numbers from memory, and have a persuasive analysis in minutes. Readers would not know. Editors would not know. Only I would know, and that knowing is exactly why I did not write. People call it a shock; I call it data that has not been read yet. But this time it was different: there was no data to read at all. When the source will not speak My job is to retell the truth of a chessboard through numbers. Not through the feeling after watching a beautiful game, but through things that can be verified: Elo ratings, win rates, move-quality indices, age curves, schedules, tiebreak formats. Chess is one of the few sports where data is almost perfectly transparent — every high-level game is recorded, every rating is public, every ranking change has a traceable origin. That is precisely why, when a source returns an empty result, I do not treat it as a small matter. An empty analysis table does not say the tournament was quiet. It says the pipeline that feeds information into my hands broke somewhere. Perhaps the original page sat behind a paywall. Perhaps the content was rendered by dynamic scripts the machine could not read. Perhaps the link was simply wrong. Or, more simply, the input record was a blank someone created to test the system. What I learned over the years is to distinguish two entirely different kinds of risk. The first is missing a real story that is happening and still hot. The second is inventing a story that sounds real. These two risks demand opposite responses, and the only way not to confuse them is to accept standing still until you know which branch you are on. That empty table did not tell me which branch. So I stood still. The three-source discipline and the scar of 2026 I am known in the newsroom as the slowest writer. It is a fame I never wanted, but I keep it like a good habit. Before writing, I check three independent sources. Not three reports citing the same number, but three sources born from three different origins. If all three trace back to a single data table, I count that as one source, not three. The scar formed in July 2026. I was forty-five, one of seven female data journalists accredited at a World Cup. While colleagues wrote emotionally charged pieces, I stayed behind and dissected the numbers of a famous match. I remember showing something that ran against the crowd's feeling. A veteran editor brushed my piece aside with a sentence I still recall verbatim: women looking at football data only know how to pick the number that suits them. I published that piece in a regional digital sports daily. It drew more than two hundred thousand reads. But the loss was not the rejection; it was the lesson. When there is no data, people default to believing the smoothest story. And the smoothest story is usually the easiest to write, not the most correct. Since then I set a fence for myself. Never write about a match without a basic data frame. Never conclude without a traceable source. Never turn memory into evidence, because memory always favours the storyteller. Numbers are asceticism: you must give up convenience before you can see the truth. In March 2026, when Asian leagues paused, I applied that same discipline to a club with the highest average age in its league. I collected sprint data from the previous three seasons and compared it with recovery data after six months of shutdown. The results showed a clear drop in pace, with the older cohort recovering far more slowly. I predicted that club would collapse in the closing stretch. The editors laughed. By year's end, that club lost a final on fitness, running more than six kilometres less than its opponent. Age is the only variable that never lies. Then came a European Championship, when the whole football world was intoxicated by a feverish pressing school, and I went to check the numbers of the team considered slow. One midfielder completed the most passes in the tournament, with accuracy above ninety percent, but his share of long line-breaking passes was very low. That team's pressure index was also below the tournament average. My conclusion then: this team does not press madly; it controls through positioning and short passes. The piece foretold their title, while mainstream coverage was still worshipping two other teams. xG does not replace emotion; it explains why our hearts race. Those three moments taught me the same thing: the power of data lies not in confirming that I was right, but in forcing me to choose between an honest answer and a pleasant one. The board and the invisible sinews Chess demands that discipline even more strictly than football. A football match can drift on emotion, but a chess game forgives no one. Every mistake leaves a trace. Every rating can be cross-checked. A player's rise can be seen in a multi-month rating curve, not in one beautiful win. A new opening system can be assessed by its engine match rate, not by a commentator's feeling. I often ask myself what would happen if I applied my own analytical frame to chess. The answer always leads to the same point: a conclusion has value only when it is anchored to a named event, a named player, a specific date. Without those anchors, every analysis is merely literature. The competitive landscape of elite chess also demands double caution. There is a champion tier, a challenger tier above twenty-seven hundred, a rising-star tier, and a reserve pipeline behind them. From outside, that picture always looks stable — but the stability of a picture is never proof of the stability of reality. Nothing extracted does not mean nothing happened. Silence at the data layer says nothing about the event layer. And here is what I want to state clearly, because it is the centre of everything: an analytical dimension left blank may not be labelled clean. If I cannot verify some aspect, it is in an unexamined state, not a transparent one. That is the difference between a room no one has entered and a room that has been swept clean. Outsiders see the same thing. Professionals must tell them apart. What an empty table actually reveals Here I want to pause on the part I consider the most valuable of this story, because it is counter-intuitive. When an analysis table returns all-empty cells, the writer's natural reflex is to reach for a familiar narrative frame to fill it. In chess, that frame is usually a generational handover, or a wave from an ascending nation, or a dispute over the rules. These frames exist because they always seem right, and because readers are used to them. That is why they are dangerous: they are unconditionally right, so they verify nothing. A journalist with backbone must accept that when the input is empty, the only defensible finding is a procedural one, not a specialist one. An empty result is not a judgement about chess. It is a quality signal about the pipeline that produced it. And that signal, read correctly, is more useful than a full analysis, because it points exactly to what needs fixing. The irony is that our media rewards confidence, not honesty. A writer who dares to say "I do not have enough data" looks weaker than one who dares to assert something grand without backing. But here is a truth few admit: in sport, the loud talker is usually noisier than the correct one, and noise is easier to remember. I am not opposed to prediction. On the contrary, I believe in it so much that I publicly stake my reputation on it before every major event. But a prediction without a falsification condition is not a prediction; it is a prayer written in the form of a conclusion. If data X appears, my conclusion collapses — a serious writer must be able to say that before saying the conclusion. The transfer market is the only place where people pay for unverified numbers. But a news article is different. An article should not be a market. Signals to track in the next round There is one thing I always check at the end of every process, and I suggest anyone in data work check it too: are my sources genuinely separate, or are all three merely shadows of the same number? In chess this matters even more, because an entire news network often leads back to one rating feed. Three citations are not three sources. With that empty table, what needed doing was clearly not writing. What needed doing was re-running the extraction step on the raw text, recovering the publication date, identifying the outlet, and only then allowing anyone to interpret. The date is the first field to fix, because every judgement about information lag is anchored to it. No date, no season, no championship cycle, nothing at all. I still keep my principle after all those years. No source, no article. No date, no conclusion. No data, I choose silence. That silence is not surrender. It is a professional decision, made on a clear understanding of the cost of a fabricated article: it destroys trust faster than any other error. A missed story can be written tomorrow. A fabricated story can never be recalled. We live at a moment when machines can produce fluent text on any topic, including a chessboard they have never seen. The only thing that separates a data journalist from a text-generating engine is not speed, not style, but the ability to say "I do not know" when one genuinely does not know. That empty board remained empty that night. And that, in the end, is the only correct thing I could write.