The Empty Report: Lessons on Silent Data in Vietnamese Youth Football

Hai Phong, a November afternoon. In the technical meeting room of a V-League club, a scouting file on a seventeen-year-old midfielder lands on the table. It has a name, a date of birth, a parent club. Every other field is empty: minutes played, duel success rate, distance covered per ninety, injury history, growth-rate data. The presenter closes the file and says the sentence I have heard no fewer than twenty times in eight years in this job: "No red flags, we can sign him."

Nobody cheated in that room. Nobody was lazy. A blank report had simply been read as a clean report, and a decision worth several hundred million dong had just been made on the silence of the data. That moment always reminds me of what my technology colleagues call a silent failure: the system does not crash, does not raise an error, it simply returns nothing, and every link downstream keeps running as if everything were normal.

A blank cell and a zero are two entirely different facts. A blank cell means nobody measured. A zero means someone measured and found nothing. In scouting files, the two states get blended together every week, and the price of blending them wrongly usually shows up late — after the contract is signed, after the injury has happened, after the player is twenty-two and nobody remembers why he was brought in.

Vietnamese football has data. The VPF publishes match statistics every round; the big academies such as PVF, HAGL-JMG, Viettel, Song Lam Nghe An and Nutifood all have people breaking down video and taking notes. The difficulty sits elsewhere: most of our data stops at the outcome layer — goals, assists, cards, pass completion — while the decision to sign a young player or not depends on layers far beneath that.

I was born in France, I work as a player development consultant, and I have lived in Hai Phong long enough to understand that importing European benchmarks wholesale is a very polite way of fooling yourself. The BMI threshold of a French academy does not account for a fifteen-year-old boy in Nghe An cycling seven kilometres to training, eating three meals, and playing on a flooded pitch for four months of the rainy season. But local calibration does not mean lowering standards. It means measuring the right thing.

Data is topsoil, and I always dig three layers deeper.

The first layer is biomedical context: what growth stage the player is in, whether there is a history of ligament or growth-plate injury, bone density, whether sleep and nutrition are disrupted by schoolwork. The second layer is competitive environment: which league he plays in, what shape his team operates in, how many touches he gets per match in that role. The third layer is opponent quality and match context: does the opponent press high, is the defender opposite him twenty-eight or seventeen, was it an away match in bad weather.

Skip those three layers and a spreadsheet can push you to the wrong conclusion in both directions: praising a harmless player and dismissing one who is recovering.

There is another trap sitting inside the metrics that look most objective. Distance covered and sprint counts get packaged as effort indices, but ineffective running also produces beautiful numbers. A midfielder who covers eleven kilometres because he keeps chasing the ball from the wrong position looks more industrious than one who covers nine but is always in the right place to cut out a pass. I have sat in meetings where distance was used to defend a player, while the video showed he had been pulled out of the defensive structure fourteen times in a single match.

The same holds for spectacular moments. A flashy piece of skill is more memorable than a pass that opens space, but the thing that decides matches is usually the second one. Elite football, like every structured contest, is settled at the layer of control and vision more than at the layer of display. Our modern statistics measure the layer of display far better.

In 2026 I was a senior specialist at the Viettel youth football training centre. I underrated the midfielder Nguyen Duc Nam, then sixteen, because his BMI and speed tested below the national U17 benchmark. I concluded he lacked the physical foundation. I overlooked a detail sitting right in his medical file: Nam had just returned from a ligament injury and was in a catch-up growth phase — his body was adding height faster than his musculature could adapt, which made his measured speed in those three months lower than his real capacity. Three months later Nam debuted for the first team in the V-League and produced four assists in five matches.

That mistake made me add a column to my data table: "biomedical context". From then on, any file missing that column was not allowed to support a physical conclusion. A player is not a metric, but the metric is where I start digging.

In 2026, when global football paused for COVID-19, Song Lam Nghe An invited me to review their academy. The old data showed striker Tran Van Cong, eighteen, producing 0.8 goals per ninety minutes — the best in the academy — but he cramped frequently and rarely played. The conventional reading was: good output, insufficient fitness, wait longer. I did not read it that way.

The training ground was closed, so I interviewed Cong's family online and re-analysed archived GPS data from earlier sessions. Two things surfaced. First, his cramping clustered in the final fifteen minutes of the second half in hot conditions, rather than spreading across the match — a hydration and warm-up signal, not a weak physical base. Second, his high-intensity distance across the first sixty minutes sat in the academy's leading group. I recommended signing him professionally before the league resumed. In the 2026 V-League season, Cong scored six goals.

What I learned was not "always trust young players". What I learned was that when data is interrupted by an external event, people tend to abandon the data rather than find a substitute layer. Old data still has value if you know the conditions in which it was measured.

In 2026, following Hai Phong's winter transfer window, I reviewed the loan contract of defender Le Van Son from Ho Chi Minh City. Son's total successful tackles in the AFC Cup looked excellent: twelve. But broken down match by match, his three direct errors leading to goals all came in away games, in the first twenty minutes of the second half, with his team pushed high. I advised the club against a long-term deal. Two weeks later Son was injured and the contract was cancelled.

There is nothing mystical there. It is a repeating pattern: a player performs well in a controlled game state, but when pushed into fast decisions in tight space under crowd pressure, his decision speed drops. Total tackles cannot show that. Only situational breakdown can.

In 2026, at the Euros and the Paris Olympics, I was invited to mentor a group of young journalists. I found that Spain's midfielder Pedri dropped eighteen percent in distance covered after the seventy-fifth minute, and predicted he would decline if pushed into extra time. The coaching staff did not rotate, and Pedri left the tournament injured. I was right, and I still felt uncomfortable, because my method then reacted too slowly to the actual tempo of the competition. I started learning machine-learning methods to supplement the work, not to replace the eye, but to shorten the time between observation and warning.

Injury does not delete a talent, it simply moves that talent down into the sediment.

From 2026 to now, the way I handle a report has changed in a fairly monotonous direction. Before anyone is allowed to draw a conclusion, a file has to pass a gate: a minimum number of observed minutes, a minimum number of matches with complete data in both phases, a biomedical context note, and one line on opponent quality in the matches used for evaluation. If any item is blank, the file goes back to collection and does not proceed to judgement.

Based on my experience watching matches in the V-League and the national youth competitions, I would argue that most bad scouting decisions in Vietnam do not come from misreading the data, but from reading a blank cell as a confirmation. The opposite risk exists too: a complete file that applies European thresholds directly to Vietnamese U17s will wrongly kill late developers.

The contrarian angle I want to put on the table is this: the urgent need in Vietnamese youth football is not to buy another data system, but to build the habit of refusing to conclude when the data is silent. A club can pay for analytics software within a week, but getting a technical committee to accept that "we do not have enough information, we will not sign" takes several seasons. The bottleneck is organisational culture, not technology.

There is a second paradox worth naming. Promotion-and-relegation pressure in the V-League pushes clubs toward short-term safety: signing a player with numbers in a familiar league rather than being patient with a youngster who needs two more seasons to reveal his real value. A data map can point you in the wrong direction if you cannot read the terrain. And our terrain has very specific features: short off-seasons, pitches that change with the monsoon, school schedules overlapping training schedules, and a domestic transfer market where contract information is usually foggier than performance information.

It took me three years to understand that data also needs catch-up growth. When a club upgrades its training conditions, changes its physical curriculum, or changes head coach, all the old data on that player becomes data from a different environment. Reading it without adjustment is like measuring a player's speed immediately after he has grown seven centimetres.

What I want to leave behind is not a conclusion, but a testable experiment. If over the next two seasons Vietnamese academies and clubs apply exactly one rule — no signing decision on a file with blank cells in mandatory fields — then the share of youth contracts written off within eighteen months should fall. If it does not fall, my hypothesis is wrong, and I will be the first to rewrite this report from the first layer of soil.