The Silence Between Two Numbers

7:12 a.m. on a Tuesday, in a small apartment on Nguyen Thien Thuat Street, Nha Trang. I open an analysis file a partner sent over during the night. The file holds thirteen standard fields: original title, source, article type, one-sentence summary, author stance, article purpose, list of information points, related entities, time sensitivity, source quality. All thirteen carry one identical line: insufficient information, cannot assess.

I sit still for about three minutes. My right hand, out of habit after thirty-one years in this trade, has already settled on the keyboard. My index finger finds the number keys, where I usually type xG, PPDA, shot conversion rate. The body wants to write before the brain can think. That is the most dangerous reflex of a person who makes a living from data: the urge to fill an empty space with anything at all, as long as the space disappears.

I close the file. And I decide to write this piece, about that very moment.

Speaking of the "Vietnamese sports data analysis trade" sounds grand. The reality is far smaller. It is a quiet group of people: a few dozen scattered across Hanoi, Ho Chi Minh City, Da Nang, Nha Trang and a few smaller provinces. Some work for clubs, others for news sites, and the rest freelance like me. We have no laboratories, no research centres, no department vouching for every figure we publish. We have a computer, a few paid data accounts, and a span of time that is always cut short by a familiar message: needed in two hours.

That two-hour window is what creates most of the problem I want to address today.

In 2026, I began my career as a fact-checker for an international sports magazine. The job was unglamorous: reread other people's drafts, underline every unverifiable figure, send it back with notes. I learned something that has not changed in the nearly thirty years since: a number without a source is not data, it is a rumour written in the shape of a digit. Readers cannot tell the two apart. Writers must. That is the only professional line I do not allow myself to cross.

In April 2026, I opened my first data blog. In 2026, I bet on xG. The V-League answered with a shock.

The specific match was Hanoi FC beating Thanh Hoa 3-2 at Hang Day Stadium. The InStat data showed Hanoi generating only 0.9 xG while Thanh Hoa produced 1.7 xG. The next day the media praised the home coach as a tactical genius. I wrote a short, dry piece saying the result came from an unsustainably high conversion rate. Hanoi's subsequent drop of points confirmed what I had said. The professionals called back. From then on, I opened every analysis with a raw data table instead of a feeling about the match.

But the 2026 shock taught me a second lesson, larger than the first: data can betray you if you read it at a single slice and believe you have read the whole truth of the match.

A year later, the 2026 World Cup taught me: data is never a single layer.

Before that tournament, a major football site asked me to predict the champion with my own model. I fed total xG and PPDA from the whole group stage into the machine and put Brazil on the throne. Brazil were eliminated by Belgium in the quarter-finals. France won. After the tournament I reviewed every match and found my own error: I used data pooled across the whole tournament, while France improved their PPDA from 11.2 in the group stage to 8.7 in the knockout rounds. Every champion changes how it plays in each phase. I applied a single fixed number to a seven-match journey.

Since then, each of my analyses separates group-stage data, knockout data, and match-context notes. Layering data is how I stay calm amid the madness of a transfer window.

Speaking of transfer windows, I must tell another story, perhaps the one closest to this morning's empty file.

One year, in the period when stadiums had no spectators, a V-League club brought me in to advise on mid-season transfer strategy. They handed me a list of targets, each name with a price. I asked one question that silenced the meeting: where is the transfer precedent for these players, and who set the first price. No one could answer. Those prices had been repeated from one source to another, with no one tracing them to the origin. When the stands emptied, I found the law of the transfer market. That law is: most transfer numbers in Vietnam are not market value, they are expectations legitimised by a semicolon. Market administrators do not manage money flows. They manage expectations.

I tell those three stories not to show off the trade. I tell them to arrive at a question this morning's empty file threw straight at my face: when there is no data, what should an analyst do?

The technically correct answer is simple. The analyst should stop.

But that answer, in the actual professional reality in Vietnam, is one almost no one wants to hear.

To understand why, we need to look at how a sports data analysis is produced. A proper analysis passes through two stages. Stage one is extraction: from the source article, the analyst pulls the title, source, type of content, information points, a list of related people and organisations, time sensitivity, source quality. Stage two is the real analysis: technique, tactics, player data, event systems, competitive landscape, rules, coaching staff and development pipelines, risk surfaces, public narratives, and the industry transmission chain.

If stage one returns zero, stage two has nothing to stand on. Any conclusion drawn afterwards is not analysis but fiction.

In developed sports markets this rule is written into workflows and checked automatically. In Vietnam, the rule exists as a saying passed between practitioners: do not invent. It sounds simple. But production pressure turns that simple saying into the hardest thing to keep.

I have seen how that pressure works. An editor receives a piece about a table tennis match at the national championship. The writer has no data on the spin of the rallies, no figures on serve efficiency, no footage clear enough to analyse ball placement. Instead of writing that the data is insufficient, the writer fills the piece with adjectives: explosive, soaring, resilient. It goes to press. No one objects. Readers are satisfied because they were confirmed in what they already believed before reading.

This is the point I want to dwell on longest, because it is the centre of this entire piece.

In table tennis, the sport I come from, reading spin is a foundational skill. A ball can be topspin, backspin, sidespin, or no-spin, and the same hand motion can produce different spins depending on the contact point. Beginners watch the hand to guess the spin. Good players watch the contact point, the ball's height, and the rhythm. Professionals read the spin before the opponent touches the ball. The difference between those three levels is not in the eyes, it is in which data layer they choose to read first.

When I moved from table tennis to football analysis and then to esports, I carried that habit with me. In football, a shot is not just a shot. It is a combination of position, timing, pressure, strong foot, shooting angle, goalkeeper condition and even the scoreline. In esports, I learned that tempo is also a data layer. Spectators mistake a flashy teamfight for a high-level match. But close viewers see that vision control and map control decide the outcome, while the teamfight is only the tip of the iceberg. A team that wins the fight but loses vision control is a lucky team, not a strong one.

And this is where the heat map enters, in my view, as one of the most abused tools today. The heat map has become the new divination. It hides a player's real role in the tactical system. A player with a heat map smeared across the pitch may not have played well. Perhaps he played well because the system demanded he roam. Or perhaps he roamed because the system was leaking and he had to run to patch it. The heat map does not distinguish those two cases. A reader who reads the heat map without reading the system is reading a beautiful but meaningless picture.

If I had to say the single most important sentence of this piece, I would say: correlation is not causation, and in Vietnamese sports analysis most fatal errors are correlations presented as causation.

A player who scores many goals is not necessarily the most important player. A team that wins many matches does not necessarily have the best system. A praised coach is not necessarily the one who made the right decision, he may simply be the one lucky at the right time.

Here I must return to the initial question, the one this morning's empty file raised. When there is no data, what should an analyst do?

I believe the answer is: say there is no data, and point out exactly what is missing.

It sounds like surrender. I think the opposite. Saying "insufficient information, cannot assess" is one of the most valuable statements an analyst can make, because it protects readers from their own beliefs. An honest empty analysis is better than a full one built on nothing.

There is an uncomfortable truth about this trade in Vietnam that I have observed for years: we reward certainty and punish hesitation. A piece asserting firmly, even wrongly, spreads better than a piece modestly saying the data does not yet permit a conclusion. Readers like people who say "certainly." So do social media algorithms. And gradually, practitioners learn that certainty is the shortest path to recognition. Epistemic humility becomes a competitive disadvantage.

That is the economic reason empty data files keep getting filled with speculation. No one wants to hand an editor a blank page. No one wants to tell a coach they need three more days. No one wants to admit to an audience that they do not know.

I understand that pressure. Thirty-one years in, I too have typed numbers I knew I was inflating, just to get the piece to press on time.

But I also know the cost. After seven years, I believe in the silence between two numbers.

What is that silence? It is the span between the moment we read a metric and the moment we understand it. It is the compulsory hesitation before assigning meaning. In table tennis, that silence is the slight pause before a strange serve. In football, it is the three seconds before we release a number. In esports, it is the instant a player realises a fight is happening without yet understanding why it began.

Most of my analytical errors live inside the silence I skipped.

That does not mean I advocate paralysis. An analyst cannot forever say I do not know. Epistemic humility is different from abdication. The line is this: when there is data, we must analyse it to the end; when there is none, we must state clearly that there is none, and describe what is needed to have it. True humility is knowing exactly what you know and do not know, then acting accordingly.

I return to this morning's file.

After reading thirteen lines of "insufficient information," I had three options. Option one: invent an analysis, based on my background knowledge of this phase of the annual season, about relegation pressure and the title race, about how refereeing affects a few recent results. That is the most attractive option, because I have enough material to make it sound very convincing. Option two: refuse and stay silent. Option three: write about the very gap.

I chose option three, not because it is easy, but because it is the only option I could take without betraying myself.

And I think that gap is, in fact, a far larger topic than a single personal analysis.

Looking at the broader context of Vietnamese sport, we are at a stage where data is talked about a great deal but understood very little. In football, advanced metrics such as xG, PPDA and progressive passes have become keywords in every argument, yet are rarely defined correctly. In table tennis, the notion of data still means little more than the scoreline, while most information lies in what is not recorded: spin, placement, tempo, the rate of serves that win points outright. In esports, data is far richer thanks to the digital nature of the game, yet it is read carelessly.

In all three environments, we are missing the same thing: a standard for when it is permissible to conclude.

That standard, if I may propose one, should have three layers. Layer one, data sources must be traceable to their origin. A number without a source should take part in no analysis. Layer two, every conclusion must state its degree of certainty and sample size. A conclusion drawn from two structurally different matches is not the same as one drawn from twenty. Layer three, when data is missing, an analyst must be allowed, and encouraged, to say there is insufficient information.

The third layer is the hardest, because it runs against the human instinct to fill gaps and against the economic incentive of the trade to have a product to sell.

I have spent nearly my whole career standing between those two forces.

Here, perhaps I should answer a question some regular readers always ask me, really only a variant of the initial question in more popular form: if there is no data, should one then say nothing at all?

My answer is: one should say what is missing, and why it matters. An analysis of the absence of data is still analysis. An analysis of what we would expect to see if data existed is also analysis, provided we name it correctly and do not present expectation as fact.

The mistake is not in writing when data is missing. The mistake is in writing as though you have data.

I think about my three options this morning and realise that the third, writing about the gap, is in fact the highest-value option I could give readers. Readers do not need another prediction about the annual season. They need a reason to trust that what they read is trustworthy.

And that trust is not built with firm assertions. It is built with transparency about limits.

There is an irony I have learned over the years. The more transparent I am about my limits, the more I am trusted. Not because I am right more often, but because when I am wrong, readers know I warned them beforehand. Trust in this trade is like trust in table tennis: it does not come from winning every point, but from never losing honesty about the point you are losing.

Recently I had a conversation with a young analyst in Ho Chi Minh City. She asked how to write faster. I replied that the right question must be how to know when to write slower. She was quiet a while, then said no one had taught her that. That is another gap, and it is larger than any data gap.

If I had to describe my trade in one sentence, I would say: my trade is spending most of my time telling apart what I know from what I want to know, and refusing to write about the second.

There is another signature line I have carried for years, and it grows truer as I age: some seasons can only be read through xG, not through the eyes.

But after this morning's empty file, I must add one more line for myself: there are also seasons, matches and analyses that cannot be read through anything at all, simply because we do not yet have enough data to begin.

And in those seasons, matches and analyses, the most honest thing a practitioner can do is shut the laptop and say: here, I do not yet know.

It sounds like little. But in an environment where everyone rushes to assert, a person who dares to say they do not know is a rare thing. And rare things, in sport as in data, are usually the most valuable.

I do not intend to end this piece with a conclusion. I have only a question to leave for the next cycle of the annual season, a question I think will follow me and others in this trade for many more years: if we rewarded honesty about the unknown the way we reward certainty about what we believe, how many empty data files would stop being filled with speculation?

I have no answer to that. And I think, this time, having no answer is more honest than inventing one to meet a quota.