The Sports News Pipeline and a Story That Does Not Belong to It

The Sports News Pipeline and a Story That Does Not Belong to It On the evening of Sunday, 13 September, at El Dorado International Airport in Bogotá, Colombia, a man named Samuel Adrián was detained the moment he stepped off his flight. He works as a content creator and had just returned from Panama. Before departing, he had pre-recorded a video to be released in the event of his arrest, and that video was published right on cue. The civil sanction he is now serving consists of ten days of detention and a fine equivalent to fifteen legal monthly minimum wages in force in Colombia. I read this story in an item labelled football. What made me stop was not the content of the case. That is a legal and cultural matter, belonging to another section. What made me stop was the label. In twelve years of watching sports news pipelines from both the Vietnamese and the Japanese side, I have grown used to mislabelled items. This time was different. Of the eighteen information points in the source, not one mentioned a club, a player, a coach, a competition, a contract, or even a match. Not a single line was about football. A story with no football in it, sitting inside a football section. For me, that contradiction is itself a passage worth dissecting. The mechanics of an error Most large sports outlets run on an automated layer. Bots collect items from many feeds, a keyword classifier assigns categories, and editors only touch a handful of prominent pieces. When a source item sits inside a general aggregator feed shared across sections, the label follows the feed rather than the content. A civil case in Cali ends up in a sports section not because someone read it wrong, but because nobody read it at all. Here is the point I want to anchor before going further: a label is not a description, a label is a decision. And once that decision is automated, it stops being a decision. It becomes a habit of the system. In my own analytical work, I always place every judgement inside a separate part I call the context variable, before drawing the diagram. The same formation, the same shape, but home or away, with or without a crowd, a dense or a sparse fixture list, all of it changes the meaning of each pass. For a news item, the section label is the context variable. Reading a piece under the wrong label is like watching a match without knowing which side is at home. You still see the passes, but you no longer read their meaning correctly. What the source actually contains Stripped of its label, the source contains a fairly compact legal structure. The matter began with an acción de tutela, a constitutional mechanism in Colombia, filed by the doctor Mario Angulo, on the grounds that his materials had been used in a documentary without consent. The civil court of the municipality of Cali ordered the removal of specific recordings, and did not order a blanket ban on the documentary for its opinions. When the order went unenforced, the incidente de desacato mechanism was triggered, leading to ten days of detention and the fine described above. There is one legal detail I consider central to the whole story. The scope of the order is far narrower than the way the matter is usually recounted. The order targeted recordings made without consent, not the thesis of the film. The gap between the order's real scope and the broader public framing is where the truth gets bent, not because anyone lies deliberately, but because a frame is chosen in advance. There is one more detail, and it concerns conduct. The order already existed. Samuel Adrián knew it was waiting for him. He boarded the flight anyway, having pre-recorded a video from Panama to be released if he were detained. Enforcement occurred only once he set foot on Colombian territory. That is a very clear illustration of the territorial limits of civil-contempt enforcement, a point whose relevance reaches well beyond a single individual's case. The blind spot of reading by label This is where I want to be slightly counter-intuitive. The first reaction of most people to a case like this is to pick a side. One side reads it as viewpoint persecution. The other reads it as defiance of a court order that deserved sanction. Both readings skip something in the middle, and the thing in the middle is the thing worth analysing. The court order was narrow. The storytelling was broad. The space between those two is where the magnifying glass belongs. As an observer, I see a familiar behavioural pattern here: a legal event converted into a media event, with the conversion planned before the event occurred. The pre-recorded Panama video is not a reaction. It is a distribution asset. The same moment is both news and content at once. And when an event is simultaneously news and content, its commercial value can rise thanks to the very legal cost it incurs. Data has its limits too: I can read a behavioural sequence, but I cannot measure intent, so I stop here at the level of structure. This is not true only of one individual. It is true of the way the sports industry now operates. Sports news outlets today do not merely report on football. They operate as attention-harvesting machines, and anything that generates attention can be drawn into the pipeline. A case in Colombia, a cultural dispute, a political controversy, all of it can travel through that pipeline, as long as it produces reads. The problem is not that they appear. The problem is that they appear under a label that does not belong to them, and the reader is not told. There is a trap the sports industry rarely looks at directly. If a sports section can contain anything, then a sports section is no longer a sports section. It has become an aggregation funnel. And once a section becomes a funnel, the value of being labelled correctly falls with it. An empty stadium is football's coldest laboratory, and an automated news pipeline is just as cold: it hates no one, loves no one, it merely classifies. But precisely because it is cold, it never corrects itself. Readers trust the label There is a detail in the source I find notable and very easy to miss. Almost every information point carries no named source, apart from two items citing case-file material reported by Colombian media and a court resolution. That means the durability of the story does not rest on its origin, but on the follow-up reporting. If the follow-ups travel through the same pipeline that applied the wrong label, the error multiplies instead of being corrected. Readers trust the label. That is the only implicit contract between a newsroom and its audience. When that contract is broken systematically, the loss is not confined to one article but spread across the entire classification system. For a football reader, the cost is even more concrete. When you search for information on a match, you rely on the label to filter. If a search about football returns a case in Cali, you lose both time and trust. Neither loss is entered into any column of a newsroom's dashboard. Based on my experience following matches and news cycles, I have observed a fairly stable pattern: mislabelling never occurs as isolated incidents. It comes in waves, usually coinciding with periods when a newsroom accelerates output, and it disappears when someone sits down to check. Which means the fault is not in the technology. The fault is that nobody has been assigned to check the label. A small thing, a large system I do not want to exaggerate this story. One mislabelled article does not bring down a newsroom. But I do not want to downplay it either. In my work I always remind myself that a judgement must start from the context variable before it moves into the diagram. In journalism, the context variable is the section. If that variable is wrong, everything downstream stands on a skewed foundation, no matter how accurate each individual sentence may be. Every match is a maze, and I merely redraw the map. In this story, the maze is not in Colombia. It is in how accustomed we have become to reading by label, and how easily we forget to check the source. Japan taught me that leading by two goals is still not a match. The same holds for a news item with a traceable origin. Having a source does not mean having the right section, and having the right section does not mean having the truth. Those three layers must be verified separately, and none can substitute for another. Takeaway The accuracy of a news item does not come from the honesty of the writer. It comes from the discipline of the process. When the process automates classification, we trade accuracy for speed, and we rarely know we have made that trade until a story that does not belong to our section appears right on the homepage. What I will check next time is simple: whether the follow-up reporting on this case remains under the wrong label or is moved to the right section. If it stays there, then it is no longer a classification error. It is a choice. And as I always say when discussing defence: a system never confesses its own weakness. It only reveals it when it is attacked in exactly the right place.