Data Gaps Are More Dangerous Than Bad Data

There is a moment in this trade that taught me something scarier than reading a wrong number: the moment a data sheet comes back empty. Late one evening, I sat in front of a screen in an apartment in Surabaya, waiting for the dataset of an esports match I had been assigned to analyse. The file opened: empty title, empty source, and the most important column — the list of information points — had not a single line. The only sounds in the room were the fan and a group chat urging the analysis along. In the middle of that gap, a very human pressure appeared: fill the blank with something that sounds plausible.

I used to think bad data was the analyst's greatest enemy. I was wrong. The greater enemy is the silence of data, because silence does not incriminate itself — it invites us to fill it with imagination. A wrong number can be caught by tracing its source. A gap cannot, unless the writer actively refuses to fill it. And in sports analysis, where speed is rewarded and emptiness is treated as failure, the pressure to fill is greater than anywhere else.

A few years ago, my process had only one layer: gather the numbers, then write. In 2026, working as a data coordinator for Surabaya United in Liga 1, I reported that my team held 63% possession against Persib Bandung and recommended pushing the line higher. The result was a 0-3 defeat, with the space behind both full-backs exploited all night. I sat with it for three nights, reviewing every phase, and found what I had missed: the opponent's PPDA. They had not been passive. They had deliberately conceded the ball to counter. I wrote a ten-page self-critique and proposed a cross-checking process before every match. The mistake in Surabaya taught me to question data, not to trust it.

From then on I built a habit that looks rigid: every claim must rest on at least three independent data sources, and every number must answer three questions — where it comes from, under what conditions it was collected, and where it is absent. The third question is the one I use most. Because in sports analysis, what decides an outcome often is not the number printed on the page, but the number left off it.

Many people imagine sports analysis as a machine: data in, conclusions out. In reality, most of my time is not spent computing, but checking whether the data actually exists to compute. A professional analytical framework is useful, but only when there is an input. When the input is empty, the framework becomes a mould waiting to be filled — and the human instinct is to fill it.

Imagine a report on a match with no tournament name, no team name, no game version. There are nine familiar analytical layers to work through: patch and meta, tournament format, roster and player form, regional landscape, club finance, rules and governance, risk profile, media narrative, and the industry transmission chain. It sounds thorough. But not one of those layers runs without a concrete entity: a game, a team, a player, a tournament. The more detailed the framework, the greater the temptation to fabricate, because an empty mould always pressures the writer to fill its shape.

Take the first layer. To speak about a patch, I need a version number, I need to know what was buffed, what was nerfed, and I need at least one affected team or player. Without those, every statement about the meta is a guess dressed in terminology. The format layer is the same: without knowing single-elimination or round-robin, best-of-one or best-of-three, nothing can be said about upset probability. A single game and a five-game series are two different worlds, and merging them is a professional error.

The roster and form layer is even stricter, because position semantics change entirely between titles. Comparing one player's metrics with another's at a different position is meaningless, even if both numbers are correct. The regional layer depends on the title: a region can be strong in one game and weak in another, so any claim that a region is strong without naming the game is unverifiable.

Then comes club finance and governance. These are the two most sensitive layers, because they touch money and reputation. When data is missing, I would rather say the risk cannot be assessed than say there is no risk. A gap in a financial record is not evidence of health — it is only evidence of a missing record. The same holds for allegations of cheating or rule-breaking: with no allegation, no accused party, and no named governing body, any conclusion is speculation, and speculation here can do real harm.

I have seen this happen in both football and esports. An analyst receives a transfer report with no figures, and instead of saying there is not enough data, he writes a story about a reasonable fee, a wage bill, a release clause — all of it fluent and groundless. An article about an esports patch with no data can turn into a report with version notes, pick rates, win rates — numbers born from nothing but wearing the shape of truth. The danger is that they do not incriminate themselves, because they are written in the exact grammar of truth.

The paradox is this: a wrong data sheet can still be caught, because it leaves a trail. An empty data sheet leaves nothing. When I trace a number, I usually find its root. When I trace a gap, I only find explanations for why there is nothing — blocked collection, content filtering, or simply a source that never existed. These three causes lead to three different actions, and merging them into a single phrase, missing data, loses the most important information of all.

So in daily work, I treat gaps more strictly than errors. A gap is not no risk. It is risk cannot be assessed. Those two sentences sound almost identical in a report, but they lead to two completely different actions: one is reassurance, the other is pausing and going to find data.

There is a lesson here from the defensive-data room. In 2026, working as a data editor for a football site, I analysed the France versus Argentina match. Public opinion criticised France's defence. But when I counted tactical fouls in the middle third, they registered 14 per match — the highest in the tournament. The 2026 World Cup lifted the trophy with tackles nobody remembers. Those fouls never appear on a scoreboard, nobody remembers the names, but they were what kept the match on the right rhythm. If I had only looked at what was recorded — goals, assists — I would have missed the very thing that mattered most.

In esports this is even clearer. A transfer story can recite a deal with not a single figure: no fee, no contract length, no confirmation from anyone. The writer feels safe because the story sounds plausible. But plausible is not the same as true. The transfer window is when noise drowns out signal, and the only way to separate them is to follow the money, follow the contracts, follow the agent's moves — not to follow how far a rumour spreads.

At this point people often ask me: so should we distrust everything we see in data? Not exactly. Debating for the sake of difference is another game, and it is no less dangerous than blind trust. What I mean is narrower: empty data and weak data are two different things, and merging them is the most common error in quick analyses. A match with little data can still be analysed, if we are honest about the limits. A match with no data cannot be analysed, no matter how many sections our framework has.

Even with complete data, correlation is not causation. I have seen a team with a high possession share lose repeatedly, and someone conclude that possession is useless. Wrong. The problem was not possessing the ball; it was possessing it in zones that cannot hurt anyone. The number was right, the reading was wrong. If even a correct number can be misread, a fabricated number will be misread far worse. And when a fabricated number enters a transfer report, it is not merely wrong — it pushes up prices, pushes up expectations, and pushes a whole chain of decisions behind it.

My process has one step many find strange: after finishing, I list on paper everything the article asserts without a source. If that list is longer than three lines, I rewrite. The step is unglamorous, but it is what keeps an analysis from becoming an advertisement for the writer's own imagination.

At a broader layer, an entire industry is shaped by how we handle gaps. When game publishers, teams, streaming platforms and sponsors all make decisions based on analyses, a fabricated number at the lower layer can flow upward and become a bad investment decision. That transmission chain moves faster than people think, and it does not distinguish between real data and data that merely sounds reasonable.

For readers, I suggest a simple filter. Every analysis should name at least one concrete entity — a team, a player, a tournament, a version. Every number should come with its collection context. And when an article can provide neither, read it as an opinion, not as evidence.

I did not write this to tell the story of an empty spreadsheet. I wrote it because, in a season where everything is measured, the scarcest thing is not data but honesty about the data that does not exist. The signal for the next cycle is simple: when a sports analysis is perfectly confident yet has not a single entity to anchor to, read it as a warning, not a conclusion. And when I myself am placed before a gap, the right answer is not to fill it, but to leave it empty and go find real data.