When the Tracking System Returns Zero

In the 52nd minute of the 2026 World Cup semi-final between France and Belgium, the monitor in the operations room where I was sitting displayed what analysts fear more than any bad metric: an empty data cell. The sensor tracking Jan Vertonghen's distance covered dropped out for seven consecutive minutes. On the pitch, the Belgian centre-back was still running, still turning, still contesting — still a living variable. On my side of the screen, the system was silent. The match ended with a single goal from Samuel Umtiti, from a set piece, and Belgium went out. Most viewers called it an individual moment of brilliance. I kept a different label: a delayed report. Years later, writing this down, I realised I had spent most of my career talking about numbers that were recorded, while forgetting that half the job lies elsewhere: identifying the numbers that were never recorded. Professional football today runs on semi-automated tracking systems, multi-point cameras, GPS sensors in the back of shirts. A single league match can generate more than two million raw data points. But every data point can fail, and when it fails the system usually does not raise an error. It returns zero. The difference between a zero and a null is the entire problem. A zero means the player stood still. A null means we do not know what the player did. In most models the two are processed identically, and that is when analysis starts lying. I learned this lesson through a specific failure. In 2026, at 53, I took a data consultancy role at a club in Ho Chi Minh City. I built a system tracking 12 physical metrics per player: high-intensity running distance, pressing actions within the first 5 seconds after losing the ball, and the share of passes into the final third. On matchday 18, against Hanoi FC, the system reported a young midfielder named Nguyen Trong Huy at 8.2 km over 90 minutes — 15% below the team average. I recommended substituting him on 60 minutes. The coaching staff ignored it. The team lost 1-3. After the match I brought a 14-page analysis into the meeting room and did something I still consider correct: I audited my own data source before accusing the player. It turned out two of his twelve sensors had failed in the first half. The 8.2 km figure was a patched value assembled from missing data. The player was not lazy. My system was. From then on, every report I wrote carried a section nobody asked for: a map of missing data. Which positions had no signal, which time windows were noisy, which metrics were interpolated. As a result the head coach began to trust me more, not because I offered more numbers, but because I stated clearly what I did not know. The team finished fifth, four places better than the pre-season projection. A typical tracking system has three layers: collection, cleaning, modelling. If collection fails, the raw data is gone and nothing can save it. If cleaning fails, the raw data survives but is distorted. If modelling fails, the result is wrong in a perfectly plausible way. Of the three, the second is the most dangerous, because that is where someone must decide how to handle null values. No decision there is neutral. Interpolation is an assumption. Dropping the row is an assumption. Keeping the null is also an assumption. Data never lie, but the people reading them do. By Euro 2026, staged in 2026 after a one-year pandemic postponement, I applied the same principle at a larger scale. I studied the tournament's effect on the fitness of Southeast Asian players. The initial result showed Vietnam entered World Cup qualifying with six players who had played more than 2,800 club minutes the previous season. Nguyen Quang Hai was one of them. I filed a recommendation to manage his workload for the UAE fixture. It was not implemented. Quang Hai injured his ankle in the 23rd minute; the team lost 0-1. I then collected data on 40 Southeast Asian players who featured at Euro and the Tokyo Olympics. 57.5% of them declined by an average of 18% in performance over the two months after the tournament. A German researcher used that report in his work on post-tournament syndrome. The injuries of Euro 2026 were not a curse; they were a report that arrived late. But I am not writing this to claim credit. I am writing because a different trap is opening, and it is more dangerous than missing data. That trap is the habit of reading missing data as if it were evidence. When a player records no pressing actions in the second half, people conclude he is out of gas. When a team takes no shots for 20 minutes, people conclude they have given up. Both conclusions may be true, but they are drawn from a blank, and a blank is not an observation. It is the absence of an observation. In statistics this is the most basic and most common error: assigning a value to missing data. People fill the gap with an assumption, the assumption becomes a belief, the belief becomes a transfer decision. The transfer market is the only place where people pay for hope rather than output. A young player with three outstanding games in a season cut short by injury can be priced on precisely those three games, while the other 30 vanish from the file as if they never happened. Clubs buy a highlight reel, not a career. I once reviewed a transfer report in which a player's high-intensity running metric was calculated from six recorded matches, three of them low-intensity pre-season friendlies. Nobody in the meeting asked which matches the sample contained. The report looked professional. The charts looked beautiful. And the decision was wrong. I have sat through five World Cups and five times watched an emotional wave wipe out reasonable arguments within two weeks. The 2026 World Cup taught me that emotion is the hardest noise to filter. But the deeper lesson is elsewhere: emotion is not the enemy of data. Emotion is a variable, and a variable worth studying like any other. What is frightening is not a loud stadium. What is frightening is an analyst who follows the noise without ever checking the connection. If the majority is right this time, would I dare to write that I was wrong? That is the self-check I run before every report. Being 62 has not slowed me down; it has taught me which data are worth waiting for. I wait for a sufficient sample instead of a good match. I wait for a stable feed instead of a beautiful chart. I wait until null values are correctly labelled, instead of being stuffed into the zero column for convenience. Looking ahead, I believe the signal worth tracking this season is not in the league table. It is in the frequency with which tracking systems return null values across the last three matchdays. A team with many missing data points is usually a team rotating heavily, or a team with equipment problems, or a team hiding something about its fitness. All three possibilities deserve a question before anyone places faith in form. Just look at the numbers and you understand everything — provided you are willing to look at the blanks too. Every number is a confession, if we are patient enough to listen. A blank is an unfinished statement, and whoever leaves a statement unfinished usually knows the most.