Empty Data Cells and the Hunger for Signal: How Esports Is Learning to Read the Unwritten

Minute 42, Game 5 of the 2026 League of Legends World Championship semifinal in Iceland. Faker stepped into the enemy jungle alone. His staff swung through darkness where no ward had been placed. No teammate followed. T1 lost to DWG KIA 2-3. On the sheet of paper in front of me that night there was exactly one number: T1's vision score in the river area, Game 5 — four units. Four units of vision for a decisive game. I was sitting in front of the livestream as a commentator, pen still in hand, and the only question looping in my head was: what happened in those final seven minutes that nobody saw? I retell this story not to reopen a fan's sorrow, but because it taught me something I only absorbed after years in the profession: an esports writer must trust the numbers and distrust the silence of the numbers. When data disappears, that is not a full stop. That is a new question. Over the years, I have watched Vietnamese esports and the wider regional scene grow accustomed to two extremes: either inflating a single metric into a legend, or shrugging off an empty stat sheet entirely. Both are tragedies for the writer. An empty cell in a data table does not mean nothing happened. It only means nobody has bothered to dig deep enough. The world of esports rests on a paradox. Never before have viewers had so much data. Platforms such as Leaguepedia, Games of Legends, Gol.gg and Oracle's Elixir publish statistics so detailed that every team fight, every ward placement, every Summoner Spell cooldown is queryable. Every major match at events like Worlds, MSI, the LPL, the LCK and the VCS comes with hundreds of sub-metrics, and organizers release the raw data files after each game. But never before have readers been fed such hollow articles. Sensational headlines, content that is mostly a compilation of tweets and community comments, with genuine tactical analysis occupying only a small fraction. This imbalance comes from three sources. First is time pressure. Esports is a sport that runs on the clock. A match ends at nine in the evening; the analysis piece must be up by ten. The writer has no time to dig through detailed datasets, so they use what is on hand: emotion, community commentary, and the highlights that already went viral. I fell into exactly this trap during my first two years in the industry. Second is the language gap. Esports is a multilingual field but not equally multicultural. A statistic published in English on an international analysis site takes nearly a day to reach Vietnamese readers, by which point the match ended the previous night. Writers know this, and sometimes choose the simplest route: silence. Third, and this is the sorest point, is the culture of reading "empty" as "nothing." In professional data analysis there is a convention serious analysts must always remember: N/A does not mean no risk. N/A means insufficient information to assess. The two are worlds apart. Yet in esports I have seen no small number of internal reports fall into exactly this trap. When data on a team was missing, instead of writing "insufficient information, further investigation required," the analysis department wrote "no issues." Transfer decisions were then made on that confusion. The consequences were often concrete: signing a player without knowing his wrist had an injury problem, or contracting a team without knowing its internal salary dispute. To see the problem clearly, look at something very specific: a team's first-15-minute stat sheet. That sheet contains roughly twelve different metric rows. General-interest media usually extract only two: kills and gold. Those two rows are enough to write an entertainment piece, but not enough to tell a tactical story. The other ten rows — vision score, jungle control rate, CS, kill participation, damage taken, Summoner Spell cooldowns, wards killed, wards taken from opponents, snowball rate, and team composition temperature differential — are where the real story lives. Yet those ten rows are usually left behind in the Excel file. This is why I always tell young people entering the profession: learn to read a data table before learning to write a sentence. Prose can wait. Data cannot. I will draw out four personal moments to illustrate how to read data — and how to read the absence of data. The first moment is EDG versus RNG, group stage of the 2026 LPL Summer Split. I was sixteen, having just taken a contributor role at an esports outlet in Guangzhou. After EDG won 2-1, I was standing in the interview area when I met head coach Clearlove. He asked me a question I still consider the most serious test I have ever received: little sister, do you know what jungling is? I held up my tablet. On it was the stat sheet I had just pulled from the match data. I answered: sir, EDG held 62.4% jungle control in the first 15 minutes, but RNG had a river-area vision score 1.7 times higher. That is why EDG's first two deaths were both lost from brush. Winning Game 3 came from switching to a lane-pressure strategy. The coach was silent. Then he nodded. My article "The Girl in the Dragon's Den" reached ten thousand reads overnight. What I learned was not that I was smarter than anyone. It was this: if I had not opened the dataset, I would forever have been just a girl standing outside the interview area, writing anodyne pieces about a sport she did not truly understand. The river-area vision score, that dry number nobody noticed, was the evidence that RNG had read the map better than EDG in the first two games. That is not a subjective opinion. That is a conclusion with citations. The second moment is Pun, the Saigon stray cat, 2026. When global sport was paralysed by the pandemic and League of Legends competitions moved online, I happened to follow a small tournament between amateur teams in southern Vietnam. There was a young player nicknamed Pun, playing Pyke in the support role, holding a twelve-game winning streak with a kill participation rate of 87%. No sponsor. No coach. He played from an internet cafe in Saigon. My piece "The Saigon Stray Cat" went viral to the point that professional teams began asking to buy Pun in the next transfer window. But if you read that article without reading the dataset, you would think it was a touching story. It was not merely a touching story. It was a behavioural profile: a solitary predator in the support role with a kill participation rate far above the role's average. Without the number 87%, the story of a boy playing from an internet cafe is only emotion. With that number, the story becomes a discovery usable for scouting. That is the difference between a storyteller and an analyst. The third moment is Faker, minute 42 of the 2026 World Championship. I have already told the story of the catch at the start of this piece. But there is a deeper layer I have never written in full: that night, the frightening thing was not Faker's death. The frightening thing was that in the post-match dataset, T1's river-area vision score from minute 35 to minute 42 was essentially zero. Not a single T1 ward survived in the enemy jungle across the final seven minutes. A legendary team, in the decisive game of a World Championship, walked into darkness without a single lamp. The seven silent minutes of the dataset are the single greatest explanation for the catch at minute 42. Yet no outlet led with that number. They led with the name Faker. Because Faker sells more clicks than an empty stat sheet. This is the hardest lesson in the profession: data does not tell stories on its own. It only speaks the truth when someone bothers to translate it. Without a translator, seven silent minutes in a dataset get read as "nothing happened." But those seven minutes contained exactly what readers needed to know: why T1 lost. The fourth moment is the trap of the empty cells labelled N/A. I once worked with an internal esports analysis unit that published weekly reports on regional teams. In one report, the section assessing a team's financial risk was left entirely blank, with a short note: N/A. Nobody in the group questioned it. Three months later, that team owed its players two consecutive salary periods and dissolved. Nobody in the analysis group saw a single warning sign, because they had grown used to treating N/A as safety. In esports, N/A is an enemy dressed as a friend. It wears a neutral coat, speaks in a calm voice, and lulls the reader with reassurance. But underneath that coat is a pit. A writer without data must write from their own not-knowing, which means stating clearly that they do not know. Only a liar says "nothing" when in fact they know nothing. Here I must argue against myself, because I know the demanding reader will say: it is all very well to talk, but where do you get the data to write every day? Esports has hundreds of matches per week, and not every match has a detailed stat sheet. True. This is the greatest weakness in my argument, and I concede it. There are three categories of match for which the data table is genuinely empty. One is tier-three and tier-four tournaments, where no official data provider exists. Two is friendly matches or leaked scrims, where data is incomplete. Three is matches in smaller esports disciplines, where the data infrastructure has not yet developed. In all three cases the writer must work from the raw: direct observation, interviews, and frame-by-frame video analysis. But even then, one principle must not be violated. The writer must state clearly what they are missing, rather than disguising that lack behind ornate prose. If I do not have vision score, I must write plainly: this match has no publicly available vision score data; I can only analyse based on video. That is honesty. It is also the minimum courtesy owed to readers who spend their time on my work. Conversely, I must also push back against the esports fan community, including my younger self. We love stories that go from the mud to the spotlight. We love telling of a player who rose from poverty to the top. But that passion, unchecked, turns the writer into a romanticizer of facts. A player who wins three matches in a row gets turned into a phenomenon, while the data shows all his opponents were bottom-of-the-table teams. I call this the minute-88 syndrome. In football, minute 88 is the boundary between a legend and a forgotten story. In esports, every play has its own minute 88. And a serious writer is one who knows that not every minute-88 play deserves to be told as a legend. Sometimes that play is just a mistake disguised by the halo of victory. One reader once messaged me after an analysis of a second-tier team. She wrote: after reading your piece, I finally understood that for all this time I loved a team because of what I imagined, not because of what the team actually did. That was the greatest compliment I have ever received. Because it meant my data accomplished what my emotion could not: restoring the truth to a subject the community had misread. So where should an esports writer begin, if they want to escape the trap of empty cells? Based on my experience covering and writing about the scene, there are three steps that cannot be skipped. Step one is always to open the raw dataset before opening the comment section. Step two is to cross-check at least three signals before asserting a tactical judgement, because a single signal, however beautiful, is still only a single signal. Step three is to state the limits of your own knowledge when data is insufficient. These three steps do not guarantee a good article, but they do guarantee an honest one. And in an industry where attention is currency, honesty is the most durable asset of all. I still keep the note from the Iceland night of 2026. It has yellowed now, but the line "vision score: 4" is still clear. I do not keep it to remember T1's sadness. I keep it to remind myself that whenever the dataset is empty, I am not permitted to write "nothing to say." I must write "there is something that has not yet been seen." Vietnamese esports is in a transformative phase. Tournaments grow more professional, data grows richer, yet the pool of analysts and writers with real depth remains thin. That is an opportunity for the next generation: those who choose to read the dataset down to its final row, rather than write before they understand. In the years ahead, as the region's scouting academies begin building serious data systems, the best writers will no longer be the best storytellers, but those who place the right question into the right empty cell. Vision score never lies, but it does not know how to tell a story either. The storyteller is us. And we should only tell it when we have read everything the numbers want to say — even when what they want to say is silence.