A Tiger in the Football Feed: The Wrong Label and the Sports Data Pipeline

I was sitting at my desk in an apartment in Seoul, opening the newsroom's daily aggregation board. It was 6:40 in the morning, the coffee was still hot, and the twelfth line made me stop. The label was clear: football. The content behind it: a female Bengal tiger captured in La Barca, in the state of Jalisco, Mexico, after local residents reported missing cattle. No club. No player. No scoreline, no league table, no transfer contract. Just a tiger, a specialised trap, and a chain of agencies from municipal to federal level. I stared at that line for a few minutes. Seventeen years following football teams have taught me one thing: when the rhythm of the data drifts away from the rhythm of the story, that drift is always worth stopping for. Listen for the rhythm from the observation seat, where tactics first fall out of time. Today the drift was not on the pitch. It was in the very way we gather the news. Context: the pipeline no one sees Most readers picture a sports newsroom as a room full of people, where every article is the product of a reporter typing sentence by sentence. In many places, the reality changed long ago. A football news item today typically passes through at least three layers: raw data collection, tagging and classification, then human editing. The second layer is the least visible, and also the most fragile. Automated tagging systems work on keywords, entities and language patterns. They are very good when the text is clean and the topic is clear. They are very bad when a proper noun overlaps with a keyword from another field. A place name that carries an animal's name. A city that shares a nickname with a sports team. A word that is both a species and a club. Let one such piece slip in, and an entire wildlife report can be filed in the football drawer with nobody checking. This is the key point I want to stress: the error is not in the article selection. The error is at the tagging layer. And the tagging layer, unfortunately, is the layer for which most sports newsrooms have no named owner. When a mistopiced item enters the system, it does not echo like a stoppage-time goal. It sits quietly, waiting to be counted, aggregated, and eventually used as the basis for some later conclusion. What is actually in that item Setting the wrong label aside, I read the content carefully. The original story is fairly clear on the facts. A female tiger, reported to weigh around 100 kg, was captured in a coordinated operation by local agencies and then transferred to a wildlife rescue facility in Tlajomulco. The area involved spans several municipalities: La Barca, La Providencia, San José de las Moras, and neighbouring ones such as Zapotlán del Rey, Poncitlán, Jamay and Ocotlán. Participants included the Tlajomulco Wildlife Rescue Unit, the Jalisco State Civil Protection Unit, La Barca civil protection and firefighters, and a municipal animal collection and health unit. Reading that, a football reporter like me has to admit: this is a public-safety and wildlife report, written in the correct neutral, informative register. There is nothing wrong with it in itself. The error lies in dragging it into a list it does not belong to. One of the first principles I learned as a young reporter is this: never tell the story of a match by reading the scoreboard. Read the people. And here, reading the people and the event closely, I found a few things worth pausing over. First, there is a gap between headline and body. The headline says the tiger was captured after attacking livestock. The body says the tiger was reported for preying on cattle. A report of predation is not a confirmation of an attack. Between those two sentences lies a whole distance in certainty, and that distance is erased at the headline line. This is a kind of information compression that feels familiar in this trade: something merely reported becomes, after passing through a headline writer's hands, something that happened. Second, there is a contradiction inside the account itself. One expert is quoted saying the tiger is about a year and a half old, then the same person concludes from the teeth that it is an adult animal. A tiger of one and a half years is typically classed as near-adult, not fully mature. Two sentences side by side create a small crack, and the expert himself hedges, saying they cannot diagnose the age very well. Third, the figure of 100 kg for a one-and-a-half-year-old female Bengal tiger sits at the high end of the usual range. That further supports the possibility that the animal is older than stated, or not purebred. I have no data to conclude, but it is a question worth asking, not an answer to accept by default. Fourth, and this is what bothers me most as a professional, is the way the Bengal label is attached with no genetic or documentary basis. Precisely identifying a tiger subspecies normally requires genetic testing. An unregistered exotic big cat in western Mexico is more plausibly an illegally kept or escaped captive animal than a wild migrant. Yet the report simply calls it a Bengal tiger, and no one challenges it. Fifth, a professional point: of the 18 information points in the original, 12 carry no source at all. Where sourcing exists, it is mostly generic attribution to authorities, plus one named expert, the director of a municipal animal health unit. Every quantitative claim about the animal traces to a single source. There is no independent verification. For an event involving a big cat, one source is far too few. I recount these details not to nitpick a local report about a tiger. I recount them to show that even if we accept it as a wildlife story, it still has problems of sourcing and certainty. And yet it slipped into a sports data pipeline with nobody stopping it. If an item like this gets through, how many genuine football items are getting through with the same level of checking? That question haunts me more than the tiger. Nine measuring sticks with nothing to measure In my trade, when an article enters deep analysis, a standard toolkit is used: tactical systems, expected goals, pressing indicators, the transfer market, club financial structure, dressing-room ecology, public-opinion cycles, the rule framework and the industry's transmission channels. I tried applying that entire toolkit to this item. The result was the same in every cell: nothing to measure. No tactical system because there is no match. No performance data because there are no players. No contract, no transfer fee, no wages, no sell-on clause. No league table, no relegation race. No owner, no coach, no captain. No fan sentiment, no betting odds. Places like La Barca or Poncitlán are municipal jurisdictions in Jalisco, not clubs or football markets. A nine-part toolkit applied to a subject with no matching part. And this is where I have to state plainly something this trade sometimes forgets: when there is no data, the correct answer is insufficient information to assess, not a fabricated comparison that sounds plausible. I know some people will try anyway. They will look at the hunt for the tiger and call it a high press. They will look at the animal being cornered into an area and call it a low block. That is metaphor, not analysis. And metaphor, pushed into the place of data, becomes false information presented as true information. In my trade, that is the gravest offence. This is the point I want to keep. For years I have sat in the observation seat, counting the rhythm of a team. That rhythm comes from very small things: a player making a run half a second early, a back line a step out of sync, a call that no one answers. I learned to trust those small details more than neatly presented numbers. But I also learned that a small detail only has value when it is real. A fabricated detail, however small, is still fabricated. There is another aspect of timeliness worth noting. The item records the event as taking place on Monday, September 28, but gives no year. September 28 fell on a Monday in 2026, 2026 and 2026. This detail alone is enough to say that the item's news value cannot be verified from the item itself. For an industry that lives on the freshness of news, a vague time stamp is a problem, not a triviality. What is frightening is not the tiger If we stopped at the fact that a tiger got the wrong label, the story would end in a chuckle. Anyone could nod: a system error, an everyday thing. But I do not think that is the frightening part. The frightening part is that this error is almost certainly not alone. Keyword-based tagging errors tend to appear in clusters, because they come from the same pattern, the same misreading. If one wildlife item slipped into the football drawer, then very likely neighbouring items in the same collection batch carry similarly wrong labels. Once inside the analytical layer, they blend into statistics about entities, sentiment and trends. From there, they quietly bend the picture the reader sees. I have witnessed a smaller version of this story. In 2026, when FC Seoul switched from a 4-4-2 to a 3-5-2, I did not redraw the formation. I scrolled through 127 comments on the supporters' forum. 68 percent in favour, 23 percent worried, 9 percent furious. From the fans' fear, I wrote a piece about the position of the number 10. It was shared thousands of times. But what I learned was not a formula. What I learned was this: 127 opposing voices, one truth: the pitch always answers for itself. Raw data, however noisy, must return to the pitch to be verified. The same holds for text data. A label does not become true simply because it sits in a spreadsheet. It must return to the content to be verified. And in this case, the content answered long ago: this is not football. There is a paradox I want to let stand on both sides, without resolving it too quickly. On one hand, automation lets the sports industry process an enormous volume of news every day, something humans cannot do by hand. On the other hand, automation itself creates an intermediary layer with no accountable owner, where errors can live a long time before being found. I do not want to deny automation. I only want to say that speed without a checkpoint is just being wrong faster. And here is where I think my trade needs to examine itself. We demand honesty from players in interviews. We demand accountability from coaches for their statements. We scrutinise every word of an agent for a whiff of something shady. But behind the scenes, we accept labels no one checks. An empty stadium still holds a rhythm; a small mistake only knocks one beat out, it does not kill the song. But a wrong label, repeated often enough, can knock the whole song out. Seen from two homelands I was born in France and work in South Korea, so I often look at an event through two lenses. In both football cultures, people share a habit of trusting the system. In France, it is faith in process and institution. In South Korea, it is faith in speed and the perfection of data. Both habits are beautiful, until they make us stop asking questions. The item about the tiger in Jalisco is a gentle reminder. It reminds us that a system, however sophisticated, can still mislabel an animal half a world away, and that the error is only found when someone sits down, reads slowly, and notices the drift. In my trade, that person is usually not the fastest. That person is the most patient. I am not writing this to tell the story of a tiger. The tiger has been taken to a rescue centre, and that is a good ending for it. I am writing this to tell the story of the label. Because the label is what remains, repeats, and spreads to other items. What does not appear in an item sometimes tells a more important story than what does. Here, what does not appear is football, and that very absence is the thing to read. What to do next If it were up to me, I would not just fix one line. I would place a gate before data enters the system: check entities, not just keywords. An item that wants to be filed under football must contain at least one club, one player, one competition, or one football governing body. If none is present, it does not belong there. It is that simple, and yet it is what many pipelines skip because it slows things down. And I would assign one person to own the tagging layer. Not to punish, but so that a name stands behind every classification decision. When no one is accountable, an error stops being anyone's error, and that is exactly how an error becomes permanent. I am still in my observation seat. The pitch is still there, and its rhythm is steady. But tonight I will spend time re-reading every line of the news list, slower than usual. Because if a tiger can slip into the football drawer unnoticed, then the real question is not what else might slip in, but what we are reading without checking at all. I leave that answer open. Because in football, as in data, what we cannot see often tells a more important story than what we can.