Milan and the Lesson of the 0.2-Second Sensor Lag: When Data Lies

In modern football, no one still debates the importance of data. But there is a bare truth few are willing to admit: most tactical decisions are built on numbers that have never been verified. I saw this with my own eyes, on an autumn afternoon in 2026 at the Milanello training centre, when a 0.2-second sensor lag nearly buried an entire season. I was 48 then, working as a member of AC Milan's coaching staff. The board assigned me to validate the movement dataset from 20 Serie A matches in the 2026-17 season. This seemed like tedious work: cross-checking expected goals (xG) against actual results, looking for anomalies. But the deeper I dug, the stranger it got. Milan's xG at home at San Siro was 1.85, far higher than 1.02 away. The curious part: actual goals scored in both venues were nearly equal. Looking at the raw numbers alone, you would conclude Milan attacked better at San Siro but were unlucky in finishing. It was a reasonable conclusion, logical, and completely wrong. When I cross-checked video footage with sensor data, I found the problem. The sensor at the south-west corner of San Siro was lagging 0.2 seconds. In football, 0.2 seconds is enough time for the ball to travel from the goalkeeper's hands to the halfway line. The result: every build-up from the goalkeeper was recorded with the wrong position, wrong timing, and wrong tactical context. Automated analysis models read this data and painted a completely distorted picture of how Milan constructed attacks. I wrote a 14-page internal report recommending recalibration. Coach Vincenzo Montella read it and decided to increase ball circulation to the right wing based on the corrected data. Milan won 5 of their last 8 matches and secured a Europa League spot. This story is not to boast. It is a reminder that data only tells part of the story — the rest lies in knowing how to listen. What I want to highlight is not just a broken sensor. The problem is systemic. In elite football today, every club collects millions of data points per match. They have analytics teams, algorithms, predictive models. But very few ask the most basic question: Under what conditions was this data measured? A tracking metric without measurement context is a meaningless number. An xG model without input verification is an expensive arithmetic exercise. I have spent 41 years in sports, from F1 broadcast booths to football coaching staff. In both fields, I see the same pattern: people trust data as truth, forgetting that data is created, maintained, and interpreted by humans. A sensor lagging 0.2 seconds in Serie A is like a G-force sensor mounted in the wrong position on an F1 car. Both produce convincing but false conclusions. Look at how modern clubs evaluate players. They use expected goals, expected assists, progressive passes, pressures. These numbers help, but they don't speak truth on their own. A player with high xG is not necessarily a great striker — he may simply play in a system that creates chances. A defender with a low successful pressing rate is not necessarily poor — he may be playing in a deep defensive block. Every tracking number needs to go to the operating table, not the altar. Back to the Milan story. What I learned was not which sensor broke, but the importance of the verification process. When I noticed the abnormal home xG, I did not jump to conclusions. I reviewed video. I checked sensor specs. I calculated error margins. I tested hypotheses by comparing data from different sources. This is basic scientific method, yet rarely applied in football analytics. The problem becomes more serious when data drives transfer decisions. Clubs spend tens of millions of euros based on tables provided by third-party companies. They don't verify how data was collected, don't cross-check with video, don't consider tactical context. The result is failed blockbuster signings — not because the player is bad, but because he doesn't fit the system, and no one spotted that from the data. I once attended a transfer meeting at a Serie A club. The analytics team presented a shortlist of 10 potential strikers, ranked by expected goals per 90. No one asked in which system those goals were scored, against which opponents, at what stage of the season. Such basic questions were deemed unnecessary because the computer had already calculated it. This is the most dangerous blind spot of data-driven football. I am not dismissing data. On the contrary, I believe data is the most powerful tool football has gained in two decades. But powerful tools must be used correctly. An F1 car can hit 350 km/h, but if you don't know the track, you'll hit the wall. Data is the same. It cannot replace the eye of a football person. It only expands vision — provided you know where it is looking. At AC Milan in 2026, we were lucky to have a verification process. The coaching staff trusted the internal report, verified with their own eyes, and adjusted tactics based on reality. But not every club can do this. Many still operate on blind faith in numbers, paying the price with inexplicable defeats, disappointing signings, seasons veering off course. Every collapse has a precondition — few bother to look ahead. The 0.2-second sensor lag at San Siro was a small precondition, but it represents a larger problem. When you trust data without verification, you build on sand. When you use numbers to decide without understanding context, you play an expensive gamble. And when you believe the number is the truth, you lose your own judgment. From the Milan training ground to esports screens, the law of space remains the same. Data does not generate meaning by itself. Humans create meaning from data. And to do that, humans need to verify, cross-check, and sometimes dare to doubt the very numbers they trust most. The question I want to pose to modern football people is not how much data you have. It is how much of it you have verified.