Insufficient Information to Assess: The Four Words Vietnamese Football Has Not Learned to Say

On January 5, 2026, at Rajamangala Stadium in Bangkok, the final whistle blew and Vietnam won the ASEAN Cup. I was sitting in a small coffee shop in Shanghai, reopening the tournament data and counting. Three indicators I always put on the table before saying anything about a match: expected goals (xG), PPDA — the number of passes an opponent is allowed before each of your defensive actions — and average distance covered per player. That table was full. That night, data and emotion temporarily agreed with each other. That rarely happens. The next morning, I opened another analytical file. It was empty. No title, no source, no thesis, no entities, no timestamps. Every cell carried exactly one sentence: insufficient information to assess. A nine-part dossier, and all nine parts were silence. I could have deleted it and written something else. I kept it as it was. I was born in Vietnam, I am 38, I now live in Shanghai, and I work as a sports data analyst reporting on esports for the Chinese market. The root of my trade is football. On the night of the Shanghai derby, I chose the numbers over the entire city. In 2026, I was 29, a mid-level editor at a new football platform. After the derby between Shanghai Shenhua and Shanghai SIPG, SIPG lost 1-2 despite firing 20 shots and generating 2.8 xG against the opponent's 0.9. My boss asked me to write a piece praising Shenhua's fighting spirit. I refused, and used the data to show that the win was luck. Fans attacked me hard. The analytics community read it. Since then I have set myself a hard rule: never offer a judgment before at least three different indicators are on the table. Not because three indicators make me right. Because three indicators force me to stay silent when I do not know. In March 2026, I wrote a prophecy. The whole of Germany laughed. I analysed ten Germany qualifiers and found their average PPDA was 11.3, far above the 8.5 to 9.5 range of the top pressing teams. I wrote that Germany would be eliminated in the group stage because they could not press opponents. On June 27, 2026, Germany lost 0-2 to South Korea and finished bottom of Group F. The piece was shared more than 50,000 times after that night. I am not retelling this to praise myself. I am retelling it to say that I was taught a very expensive lesson about data being able to be right, and that I have still not finished learning the reverse lesson. The reverse lesson arrived in 2026, in the Euro semi-final. Denmark ran an average of 118.7 km per match; England only 112.3 km. Denmark took 18 shots per match, England 11. I said on a radio broadcast that the data said England would lose. England won 2-1 after extra time. What I overlooked was not an indicator. What I overlooked was squad depth and the mental lift of a substitute named Jack Grealish. Since then every piece I write carries a closing section: where could my assumptions be wrong. Now back to the empty file. The process I use has two steps. Step one extracts raw facts: what, whose, when, which number. Step two is the analysis. The file I received the next morning had completed step one and returned zero. There was no match to discuss, no tournament to rank, no player to assess, no money flow to check. A machine with weak discipline would fill that gap with prose that sounds extremely reasonable. A machine with discipline would write the same sentence nine times: insufficient information to assess. To me, that was the most honest output I had seen in months. And that is also why it is worth writing as news. Vietnamese football and Vietnamese esports produce empty dossiers like that every week. We just do not call them by that name. In March 2026, Riot Games announced sanctions against 32 individuals inside the VCS system, Vietnam's top-tier League of Legends competition. Thirty-two people, in a system where the total number of top-tier professionals is only a few dozen. That was the event I followed most closely that year, and it told me far more about structure than about individual morality. Esports betting erodes competitive integrity faster than traditional sport, and the reason is mechanical. In football, a fixed match needs at least two people, usually a goalkeeper and a defender, and the act shows up clearly on video. In League of Legends or Arena of Valor, a misplay in the 22nd minute looks identical to a misplay under pressure: same camera angle, same speed, same calm face. The margin between playing badly and throwing a match is far narrower in esports than in football. That is a structural feature of the game, not a moral failing of the player. At the same time, the in-play betting market in esports runs continuously and is sliced by objective, by kill, by dragon, by minion wave. Stake sizes are small and multiply very fast. Anyone with access to inside information, meaning five people in one practice room, holds an advantage no financial market would permit. Meanwhile, the income of young players on lower-tier VCS teams is not high, and careers are very short. That structure produces an equation with a bad solution. Here I must state my limits clearly: I have no data on total betting volume on the VCS, and I have no access to investigation files. Anyone who offers a specific figure for that volume is making it up. What I have, and what can be verified, is 32 individuals and one date. Enough to ask a question, not enough to conclude. On January 27, 2026, in Changzhou, Vietnam's U23 team lost 1-2 to Uzbekistan in extra time, in the snow. I watched that match in a late-night eatery and remember staying silent for a long time after the whistle. The whole country called it a miracle. I did not object. But when people call it a miracle, they are describing an outcome, not a method. Vietnam's U23 method that year can be described with data. A low defensive block, compressed distance between lines, and a large share of goals coming from set pieces and from Nguyễn Quang Hải's shots outside the box. That is a sustainable model in a short tournament, and it has a very clear ceiling over a long cycle. When you call a model with a ceiling a miracle, you are not praising it. You are setting yourself up for disappointment next time. That cycle repeats: win and it becomes legend, draw and it becomes crisis, lose and it becomes accusation. On my table, that is an emotional oscillation with very wide amplitude and very low frequency, the signature of a system with no data anchor. On February 1, 2026, Vietnam beat China 3-1 at Mỹ Đình, on the first day of Lunar New Year. I watched that match from Shanghai, and that night two stories ran in parallel: the story of a team that won, and the story of a team that had already been eliminated from the third qualifying round. Both were true. We choose to tell the first, then call it analysis. At the 2026 ASEAN Cup, the national team under Kim Sang-sik gained a new variable: a naturalised striker, Nguyễn Xuân Son, who changed how this team creates chances. A naturalised striker is not a tactical solution; it is a change in squad structure, dragging with it changes in how the whole team defends and transitions. Very few commentaries in Vietnam during that period measured that shift. Most simply counted goals. I have to say this before others do: my profession has a problem. Over the past ten years, data analysis departments have moved from the stands into the dressing room itself. A V.League club today may have one or two people working full time with data, and every week they send the coaching staff reports thicker than what a coach has time to read. The conclusions in those reports are often statistically correct and rhythmically wrong. A model says your team should pass more to the right flank. It does not know that your right-back has been sleepless for three nights because his child is ill, and that the tempo of the match was slowed by two beats because of rain. The spreadsheet is an altar, and I offer myself to each number. But I have learned that the altar is not truth. It is only the place where I am most honest. Based on my experience following the V.League across many seasons, I see that the PPDA of most teams sits in a middle band and does not fluctuate much. Matches tend to be decided by physicality and set situations more than by pressing structure. Gegenpressing has been decoded in Europe, and mid-table teams there turned to physicality to turn football into athletics. The Vietnamese version of that trend is a style built on intensity and second balls, where the prettiest indicator is not xG but the number of ball recoveries within thirty metres of the opponent's goal. The problem is that nobody measures it. And what is not measured gets told, in words. With no crowd, football changes shape. I found that out and was rejected for it. In 2026, I collected 250 Bundesliga matches after the restart and found the home win rate fell from 43 percent to 31 percent, with average goals per match down 0.4. My editor asked me to add an optimistic message about recovery. I refused. That research was later cited by some Bundesliga coaches, and I lost a freelance contract because of my rigidity. What I took from it was not a figure but a habit: state the environmental context of every dataset before using it. Empty stands or full stands, fixture density, weather. Vietnamese football does not have that habit, and so comparisons between one V.League season and another are usually technically meaningless. Transfers are a fertile gamble, but I count cards before placing a bet. In the domestic V.League market, a player can be valued by the number of goals scored in a season, regardless of position, regardless of minutes, regardless of chance quality. A striker who scores 12 goals from 30 shots inside the box and a striker who scores 12 from 90 shots from all distances receive the same figure on a contract. In probabilistic terms, those two are not in the same profession. The following season, one of them will cost a club dearly, and nobody will trace back the contract to ask why. I will say something many colleagues in Vietnam do not want to hear. An empty dossier is not a failure of the analytics trade. It is the only honest output when the input contains nothing. The real failure is the number of decisions in Vietnamese football made on the basis of an empty dossier, then filled in with adjectives. A contract signed because of fighting spirit. A coach sacked because of cultural mismatch. A tactic kept because it had won three in a row. All three sentences share the same structure: a conclusion, and behind it, a void. Every crowd is wrong. The only thing that is not wrong is probability. I do not say that to look down on crowds. I say it because I was once part of a wrong crowd, in 2026, and I still keep the recording of that broadcast. There is a misunderstanding about my work I want to correct. People think a data analyst is someone who delivers answers. More accurately, we are the ones who set limits on answers. When I said Germany would be eliminated, I did not say Germany would certainly be eliminated. I said their elimination probability was higher than the market was pricing. If Germany had advanced, my model would not have been wrong; it would simply have been beaten by an outcome sitting in the tail of the distribution. The distance between those two things is my entire profession, and it is also what most debates about Vietnamese football do not have. From the Bundesliga to Worlds, I look for the same thing: a truth that can repeat. Champions do not repeat. Models do. What I am waiting for in the next cycle is not a trophy. I am waiting for a V.League club to publish its match data in raw form, for everyone, every matchday. I am waiting for a Vietnamese esports competition to publish its full competitive integrity monitoring process, including both resolved and open cases. I am waiting for a coach to say, at a press conference, that his team lost and he does not yet know why. None of those three things has happened. But in the empty dossier I received on the morning of January 6, 2026, there were nine lines stating that there was insufficient information to assess. I read them as a reminder: the hardest part of this trade is not finding an answer, but recognising that you do not have any answer at all. Data context The indicators in this piece come from three different sources with differing reliability. The PPDA figures for Germany in the 2026 World Cup qualifiers are numbers I calculated myself from public event data, with error possibly reaching half a unit depending on how a defensive action is defined. The Denmark and England figures at Euro 2026 come from official broadcast data, more reliable, but measured under conditions where crowd levels differed between matches. As for the V.League observations, I must be blunt: they rest on multi-season observation rather than a complete dataset I have access to, because that dataset does not exist publicly. Where could my assumptions be wrong? My largest assumption is that indicator-based assessment is always better than assessment based on direct observation. That is wrong in at least two cases: when the sample is too small, and when the most important variable is not measured. The Euro 2026 semi-final is an example of the second case, where I failed to measure bench depth. My second assumption is that competitive integrity problems in Vietnamese esports are structural rather than individual in origin. I have no data to prove that, only inference from market structure. If someone hands me a dataset on VCS player incomes and average career length, I am ready to revise that entire section.