When Football Analysis Tools Encounter Data Gaps: Lessons on the Boundary Between AI and Sports Journalism

In the editorial office of a sports magazine in Marseille, I once witnessed a scene that made me think deeply for many nights. A young colleague, proficient in xG and PPDA algorithms, held in his hands an eight-dimensional analysis of a Champions League match. He looked at me with a bewildered expression: 'Teacher, the system is reporting an error. All data fields are empty. No information points, no team names, no results. Only a domain label saying "football" remains.' This scene reflects a concerning reality in modern sports journalism: we are building increasingly sophisticated analytical tools to the point where we forget that the foundation of any analysis is real data, and real data comes from investigators, reporters, and editors who sit in the stands or in the locker rooms. Thirty years of following matches from small district pitches in Central China to the massive stadiums of Ligue 1, I have learned one thing: data does not score goals, but it knows where the ball is going. That is why I always carry a notebook to record moments that no statistical metric can capture: how a midfielder changes his running rhythm to create space for a teammate, how a goalkeeper takes a small step forward before facing a penalty kick, how the emptiness of an empty stadium transforms players into ghosts running across the grass. These details are not in any database, but they are the core of football storytelling. Recently, I was introduced to an eight-dimensional analysis framework designed to comprehensively evaluate all aspects of football: from tactics and technique, to club finance and the transfer market, from sporting results and public opinion cycles, to league context and team positioning, from rules compliance and governance, to locker room and coaching staff analysis, from risk profiles to media expectations analysis, and finally, football industry dissemination analysis. This is a fairly comprehensive system designed to cover every angle of the beautiful game. However, when applied to an article with no specific content, this system immediately exposes an obvious but often overlooked truth: without input data, there is no output analysis. This may seem obvious, but in reality, many modern analytical tools operate in the opposite way. They are designed to fill gaps with guesses, to generate analyses even with minimal information. Some platforms even pride themselves on this capability, claiming that with just a tweet or a short summary, the system can produce a thousand-word analysis. This is where the boundary between sports journalism and artificial intelligence products becomes blurred and concerning. The eight-dimensional framework I mentioned will struggle with this situation. In the Tactical and Technical Analysis dimension, it needs data on formations, playing styles, pressing schemes, and player usage. In the Club Finance and Transfer Market Analysis dimension, it needs information on salary structures, broadcasting revenue, commercial revenue, and net debt. In the Sporting Results and Public-Opinion Cycle Analysis dimension, it needs standings, form sequences, fixture lists, and fan emotion cycles. Each dimension requires a certain amount of data to make meaningful assessments. And this is the most important lesson: in sports journalism, nothing can replace being present, observing, and recording. What concerns me most is the trend of using AI tools to generate football content without verifying reality. I have seen articles created entirely by algorithms, with tactical analyses based on assumed data, transfer predictions made without any source, and even locker room psychological analyses where no one actually entered that locker room. This is not sports journalism; this is the product of a system trying to fill knowledge gaps with guesses presented as facts. Returning to the original situation: an eight-dimensional analysis framework applied to a completely empty input. What is noteworthy is that even with complete data, such an analytical tool can only provide part of the picture. It can tell you that a team has low PPDA, meaning they press aggressively. It can tell you that a player has high xG but low conversion rate, meaning they create many chances but don't score. It can tell you that a club is running a deficit and at risk of Financial Fair Play violations. But it cannot tell you why a player suddenly changes their playing style, why a coach makes a strange tactical decision in the 70th minute, or why a team decides to sell their star at the most crucial point of the season. In thirty years of work, I have witnessed countless matches decided by factors that no analytical tool could predict. This is why I believe that in the age of increasingly sophisticated AI tools, the most important skill of a sports journalist is not the ability to operate complex algorithms, but the ability to go out and gather information. To visit training grounds, sit in bars where players gather after matches, talk to the most passionate fans and the most skeptical ones. Only from those experiences does real data emerge. And only with real data can analytical tools fulfill their purpose.