The Empty Analysis Sheet: A Lesson on Data Integrity in Esports

A young colleague sent me a nine-dimension analysis of an esports match. I read it from beginning to end, slowly, the way I read every report. Nine sections. Each section had a heading, tables, an assessment framework, a "conclusion" and an "evidence" field. It looked highly professional, exactly the standard of a document meant to be presented to an editorial board.

But all the content inside repeated a single line: "insufficient information, cannot assess". No tournament name. No game title. No team. No player. No coach. No time anchor. No source. A report perfect in form and empty in substance, to the point that it labeled itself a "framework placeholder" — a mold waiting for content to be poured in later.

I sat still for a moment, the paper still in my hand. Then I realized: what I was holding was a mirror. It reflected the very disease of the esports analysis industry in recent years — where the skeleton is built first, and the truth is stuffed in afterward, or never stuffed in at all.

Data never lies; it is simply that we have not yet asked the right question. But when we have no data at all to ask, the most correct question is: what are we missing?

Why an analysis can be empty yet still look complete

Unlike traditional football, esports is far more fragmented when it comes to data. A League of Legends, Dota 2, CS2 or Valorant match does not share a single yardstick. League of Legends KDA does not speak the same language as CS2 Rating. Pick and ban rates exist only in certain titles. Gold-per-minute, vision, lane pressure — each title defines them differently. And every patch rewrites the rules of the game, turning last season's data into history rather than present.

That means the number-one prerequisite in esports analysis is identifying the correct game title. No game title, no frame of reference. No frame of reference, and every number is meaningless. An analysis sheet that does not name the game is like a weather report that does not name the city: you can read it, but you cannot act on it.

I have watched this industry for 18 years, from my role as an athlete in 2026, to a tournament organizer, and then into esports media. I learned one painful thing: most mistakes in analysis do not come from a wrong model, but from wrong input data — or worse, no input data at all.

When an analysis begins with "insufficient information", that is not necessarily the analyst's failure. It is usually the failure of the upstream data-collection process: missing match records, missing roster information, missing time anchors, missing sources.

Nine dimensions, and the cost of each gap

Imagine a standard analysis with nine dimensions. First, patch and meta analysis: which update is changing how the game is played, who benefits, who suffers. Second, tournament format analysis: BO1, BO3 or BO5, how the bracket runs, how dense the schedule is. Third, team and player analysis: paper strength, chemistry, bench depth. Fourth, the regional picture: where Korea, China, Europe or wildcard regions stand. Fifth, club finance and business. Sixth, rules and governance. Seventh, the risk profile. Eighth, public narrative and expectations. Ninth, the esports industry's transmission.

Take the first dimension — patch analysis. If we do not know which version the match was played on, we cannot know which champions were strong, which tactics were dominant. A team that won on an old version can collapse on a new one if the patch cuts exactly into their strengths. But if the analysis does not record the version number, we are comparing two different things while thinking they are one. The second dimension — format. A BO1 and a BO5 are two different games in terms of probability: BO1 rewards explosion and luck, BO5 rewards depth and consistency. The third dimension — roster. Without knowing who starts, who is benched, who is just back from injury, we cannot assess chemistry.

Those nine dimensions, with sufficient data, produce a picture. But when the data is empty, they produce only nine empty frames, equally beautiful and equally useless.

I once fell into the opposite trap: too much data, but not knowing how to ask the right question. In 2026, at 25, I was a reporter for a new football site in Binh Duong. I recorded data by hand from 182 V-League matches off video. I found that Long An had the lowest PPDA in the league — 7.8 — meaning they let opponents hold the ball comfortably, yet conceded only 0.7 goals per match thanks to lightning counterattacks. I wrote "Low pressing is not cowardice", and a veteran coach dismissed it as soulless statistics.

But the young assistant of Binh Duong club invited me to build a pressing map for the team. That debate shattered my traditional way of reading a match. V-League is a mess, but every mess has its own rules.

What I learned was not "having data means winning". It was: data only has value when it answers a specific question. Those 182 matches were useful only because I asked one right question: is low pressing really bad defending?

In 2026, I was sent as an analysis reporter to the World Cup in Russia. After the quarter-finals, I predicted Croatia would beat England, because Croatia's average xG was 2.3 against England's 1.1, even though Croatia had to play many extra times. In 2026, I staked my entire career on a probability model named Croatia. Colleagues laughed and said football is not mathematics. Croatia won 2-1 after extra time. My piece "Goals from probability" was shared more than 10,000 times, and my boss gave me my own column titled "Seeing by numbers".

Croatia was not a miracle, but a well-managed variance. What we call a "miracle" is just the name we give to what we have not yet measured.

Then in 2026, when the pandemic paralyzed tournaments, I spent my time analyzing 252 Bundesliga matches from May to June, the matches with no spectators. The result: the home-win rate fell from 43% to 29%, and away teams ran 6% more. I tweeted the comparison table, and The Analyst shared it, treating it as scientific evidence about home advantage.

The applause on the empty stands recorded a truth no one wanted to hear. Home advantage lies mostly with the crowd, not with the pitch.

Then at EURO 2026, I published a study of 342 penalties in 5 European leagues, showing Donnarumma dived to his right 72% of the time against right-footed players. I predicted Italy would beat Spain on penalties. The piece was mocked as fortune-telling. The semi-final came, Italy won 4-2, and Donnarumma saved two shots to his right. The piece reached 1.2 million views, and an international sports channel invited me as a data expert for the 2026 World Cup.

Those four stories share a common denominator: they began from concrete data, with sources, time anchors, and subjects. They did not begin from a skeleton. And they all admitted their limits — a probability model, not a prophecy.

The Korea–Vietnam lens: when two speeds of data meet

Standing between the mature esports scene of Korea and the booming market of Vietnam, I see a clear mismatch. In Korea, match data has long been standardized: the same match, the same set of metrics, stored across many seasons, enabling time-series analysis. In Vietnam, data is often collected by hand, unsystematic, and easily lost when a tournament ends.

That mismatch is both a disadvantage and an opportunity. A disadvantage, because we lack a basis for long-term comparison. An opportunity, because those who arrive later can learn from the mistakes of those who came before without paying the price. But only if we admit we are short on data — rather than pretending we have it.

When emptiness is a wake-up call

Here, I must argue against myself, because a counter-intuitive angle has value only when it explains something.

The esports analysis industry is making a subtler mistake than merely lacking data: filling the gaps with assumptions without labeling them as assumptions. A beautiful heatmap can hide a player's real role in the tactical system. The heatmap has become the "new astrology" — it gives us the feeling of science without giving us evidence.

Similarly, in the field of injury and return, medical confidentiality leaves fans and media blind. Clubs only announce the injuries that benefit their share price. The result is that we analyze a roster without knowing who is in pain, who is recovering, who is being hidden. That is analysis on a deliberately distorted data foundation — more dangerous than empty data, because it looks complete.

We think we understand the game, until the data sheet opens our eyes. And sometimes, the empty data sheet is the most honest wake-up call of all.

Here, correlation does not mean causation. A team winning after a coaching change does not prove the new coach is better. A player with high KDA does not prove he is the most important. And an analysis that looks complete does not prove it has value.

When I received the nine-dimension report full of "insufficient information", my first reaction was irritation. But my second reaction was relief. At least it did not fabricate. An analysis that dares to say "I don't know" is more honest than a confident analysis that is wrong.

The biggest trap for a data person is bending the numbers to win an argument. I nearly did that after the Croatia bet — when you are right once, you easily believe you are always right. But a model that is right once can still be wrong ten times. So, before writing, I force myself to look for counter-evidence. If I find none, only then do I trust my conclusion.

And I must remind myself that a counter-intuitive angle has value only when it explains something common sense overlooked, not when it is merely there to shock. A piece that only 5% of readers understand is a failed piece. So every hard term must come with a concrete match example. When I write about PPDA, I do not explain it with a formula, but with the image of Long An dropping deep and counterattacking. When I write about probability, I do not draw a distribution curve, but tell the story of Croatia. Data only lives when it touches a specific memory in the reader.

The signal for the next round

So what is the signal for the next round?

First, treat data integrity as part of the analytical skill, not as an administrative procedure. An analysis with no game title, no time anchor, no source is not analysis — it is decoration.

Second, learn to say "insufficient data" without shame. In an industry where everyone wants to appear knowledgeable, honesty about your own limits is itself a competitive edge.

Third, ask the right question before reading the numbers. When the data sheet is empty, the right question is not "who wins", but "what am I missing in order to answer".

Vietnam's esports industry is booming. But a boom without data discipline will produce a great many "miracles" — and very few verifiable truths.

I still keep that empty nine-dimension report in my drawer. Not to criticize my colleague, but to remind myself: before building a nine-storey building, you must check whether the foundation exists.