The Misapplied Esports Label: A Cosplay Photo Set and the Content Industry's Taxonomy Gap

04:47 in the morning, Shanghai. The window of my room looks out onto Wukang Road; the streetlights are still amber, and the vegetable trucks have already set off for the market. I lie on the couch, the phone screen glowing, my thumb scrolling. A piece appears, and in its top corner there is a small tag: esports. The photo inside shows a young woman in a sailor outfit, a pair of rabbit ears on her head, a model cannon in her hands, standing against a blue studio backdrop. No stage. No stands. No scoreboard. Just a cosplay photo set of the character Shimakaze from the game Azur Lane, with a few lines describing the design, the expression, and how close it comes to the original. The photo set is not bad. The tag is what made me sit up. I remember an evening in 2026 in the LPL Summer studio in Shanghai, when I mispronounced Clearlove's name as Clear-lake three times in a row during the first game between Team WE and EDward Gaming. The chat flooded with hashtags, my face burned, my commentary rhythm broke. That night I went home and understood something I have written about again and again since: the wrong name on the screen, the right lesson for a lifetime. Calling a person by the wrong name strips them of part of their identity. Mislabeling a piece of work does the same, except the unit being mislabeled here is not a person. It is an entire reading system. What I found in that article, in terms of substance, is fairly simple. Azur Lane is a mobile gacha game developed by Manjuu and Yongshi. It has no professional circuit, no franchised league, no competitive ladder structure comparable to League of Legends, DOTA 2, CS2 or Valorant. Its life cycle runs on character banners and skins, on revenue from character design, not on balance patches. So when a piece about cosplay of a character from that game is filed under esports, the error sits at the level of classification, not at the level of the article's quality. The character in question is Shimakaze, a destroyer belonging to the Sakura Empire faction. Her design carries rabbit ears, a sailor outfit, and a warrior-but-cute spirit — a formula engineered to spread through a fandom. One detail stands out: this character can transform through a great many outfits. In the language of the gacha market, that is an asset. In the language of the arena, it is nothing at all, because there is no arena in which to measure it. The article is signed by the author Tuan Hung, on a Vietnamese-language outlet. At the end there is a block of related links, and that is where I stopped longest. It contains news about PUBG Asia Stars, about the player Himass facing a possible competitive suspension, about KRAFTON having to issue an apology, about a dispute between Soopi and Mr. Pha, and a line about the CEO of Box Vietnam being sought for copyright infringement. The outlet's entire real esports output sits in that related-links block, not in the main article. The main article is a photo set. The related links are a newsroom. This picture has become deeply familiar to me after more than fifteen years standing between the Vietnamese and Chinese markets. In Vietnam, the word esports has turned into a basket. Anything with young people in it, with a screen in it, with an online community behind it, can be thrown into the basket. Cosplay goes in. Gacha goes in. Magazine shoots go in. Even a livestream selling T-shirts printed with anime characters gets called esports content. The people tagging this are not lying on purpose. They tag by habit, by keyword, by topic cluster suggested by a system. The problem is that such habits accumulate into a standard, and a wrong standard produces wrong data. Let me reconstruct the real value chain behind that article. At the top sit the publishers Manjuu and Yongshi, owners of the Azur Lane intellectual property. They design Shimakaze, give her rabbit ears, a sailor outfit and an open wardrobe. In the middle is the cosplayer — the person who rebuilds the design out of fabric, wigs, makeup and a blue screen. At the bottom are the fans, who tap the photo, leave a comment, and may return to the game to open loot boxes for the newest skin. Not one link in that chain is a competitive link. There is no coach, no player contract, no standings table, no pick-and-ban phase. It is a marketing flywheel for a gacha title, powered by the emotional loyalty of its players. That flywheel is real, and it is large. It is simply not esports. The difference lies in where the revenue comes from. An esports team lives on prize money, sponsorship deals, league slots, and a payroll paid out every month. A gacha title lives on the share of players who buy skins and open boxes. Those two money flows run on different rhythms: the rhythm of a season, and the rhythm of a banner. Standing in the middle, content people like me are tempted to merge them, because merging makes the chart look better. But a good-looking chart and an accurate chart are two different things. I made that merging mistake for years. Back when I was mid-level on the content side, I once proposed building a virtual watch room for the LPL, opening a voice chat so fans could comment on every game together. In 2026, the pandemic forced the league to play in empty stadiums. I organized thirty matches, and the Summer Final on 27 August between JDG and TES, ending 3-2, remains an unforgettable memory. When JDG came back in game five, five thousand people were on the voice chat, thousands of voices breaking at once. I sang along in a stadium with not a single person in it. That night I learned that physical emptiness cannot kill emotional resonance, but I also learned something more bitter: a crowd of feeling does not automatically become usable data. To analyze, you have to count. To count, you have to classify correctly. The Azur Lane article shows me that gap clearly. Not a single number appears in it. No views, no shares, no follower count for the cosplayer, no revenue, no contract. Only qualitative assertions: the design is rather impressive, the spirit is mischievous, it is close to the in-game original, it was carefully invested in. If this were a match report, I would have gold difference, damage dealt, vision control time, kill counts, Herald timing. I can sit with a game like the MSI 2026 final between RNG and DK, a 3-2, for hours and read its structure out loud. In the deciding game that day, my eye went to Ming, a twenty-seven-year-old support with three barren years on the international stage. He picked Nautilus, not to deal damage but to absorb roughly forty-five thousand points of enemy damage onto himself and buy his teammates room to live. Forty-five thousand. That is a number you can weigh, put on a table, argue about, refute. That is what analysis needs. With a cosplay photo set there is nothing to weigh. And here is the point I want to push one step further. The absence of numbers is not merely missing evidence; it changes the nature of the text. A piece about a product with no metrics is forced to lean on praise, and praise always has an owner. In the content industry, a strong cosplay set usually reaches a newsroom one of two ways: the outlet finds it, or a content-side buyer sends it. The second path is far more common than audiences assume. When that happens, lines like impressive, close to the original, carefully invested are in fact placement copy rewritten in an editorial voice. Nobody lies, but nobody verifies either. I stress this because it breaks a naive assumption in Vietnamese esports data circles: that if a piece sits in the esports section, its content has already passed esports standards. It has not. On a single day, one section can hold three kinds of text with entirely different natures: a match report with statistics, a transfer story with contracts and fees, and promotional design content with nothing measurable in it. Merging those three into a single metric called esports content volume is an act of systemic self-deception. I have sat in meetings where people excitedly presented double-digit growth for the esports vertical, while a meaningful share of that growth came from cosplay galleries and gacha news. Numbers like that are not technically wrong in the counting. They are wrong in the definition. And a wrong definition replicates itself through every later report. Back to the related-links block at the end of the article, where I found the real esports. This is where the industry needs to look. The story of the player Himass at PUBG Asia Stars, the possibility of a competitive suspension, KRAFTON's response as publisher having to apologize, and the dispute between Soopi and Mr. Pha — all of it sits in territory I call the territory with rules. There are tournament regulations, violation procedures, appeal rights, suspension terms. There is also real money: withheld prize payments, frozen contracts, a family that does not receive its transfer at the start of the month. Based on my experience following matches and disciplinary cases, a regional suspension is usually reported with far less weight than it carries. Fans see a notification line. The player sees three months of salary and a future under a question mark. Beside it sits the line about the CEO of Box Vietnam being sought for copyright infringement. On its own, that is just a legal item. Set next to a cosplay photo set, it becomes a striking analytical pairing: within the same content ecosystem, one side is the publisher's deliberate tolerance of community work, the other is a legal line crossed. Gacha publishers tend to be generous with cosplay, because cosplay keeps their characters alive for free. But that generosity is not always repaid with respect for the boundary. In other words, fan content and copyright infringement do not sit at opposite ends of a straight line. They sit right next to each other, and intent is the only thing separating them — and intent cannot be measured. At this point I have to talk about what I believe is the deepest cause of a cosplay set slipping into the esports basket: the economics of attention in the Vietnamese-language content market. Outlets have to survive. A match report demands that a writer be present, have time, understand the patch, be able to read statistics, and have the relationships to conduct interviews. A cosplay set demands an image source and two hundred words. The cost differs by dozens of times, while the traffic differs far less — and the photo set may even spread more easily, because it asks readers to know nothing about the game. This is a perfect economic equation, and it solves itself. The result is that the esports section, after a few years, takes the shape of a general market: the professional sports stall sits jumbled together with the photo stall, the anime stall, the lifestyle stall. Readers get used to it, and then nobody sees anything wrong anymore. I do not object to mixture. I object to undeclared mixture. Community content has its own value, and that value is real. That Shimakaze photo set is a product of labor: construction, makeup, shooting, post-production, posting, waiting. Whoever made it deserves a place. That place is called the entertainment section, or the game culture section, or the community section — any name that is true. The wrong name hurts both sides. The esports side gets diluted, the cosplay side gets judged by a standard not meant for it, and the photo set gets read with the wrong question: what does this prove about the competitive level of the national esports scene. It proves nothing, because it was never made to prove anything. Now comes the part where I must be most careful, because this is where I am most likely to romanticize. There is a very appealing way to tell this story: that esports is growing up, spilling off the stage, blending into mass culture, becoming a common language for young people. It sounds lovely. But it is a narrative that has been verified as false. The boundary between esports and fan content does not blur because esports got stronger. It blurs because the label is too cheap. Tagging a photo set as esports costs nothing and pays off in search ranking and advertising. If that label were as expensive today as a medical or financial label — where one wrong word can cost a license — pieces like this would sit in their proper drawer. Label blur is not a sign of maturity; it is a sign of a system with no quality control department. And here is the other side, the side I want the reflexive critics to hear. The easiest thing is to say cosplay has nothing to do with esports, that it is junk content, that the people producing it only want clicks. That position is as lazy as the thing it criticizes. Because in the structure of the industry, the fan layer is precisely where every esports revenue stream originates. Without fans paying to watch, paying for in-game items, paying to show up at meetups, no sponsor signs a contract. Fan culture is not the decoration of esports; it is the foundation. That cosplay set, functionally speaking, is an expression of that foundation. The problem is not that it exists. The problem is that it is named wrong and counted wrong. The third face of romanticization sits on the opposite side, and I have made this mistake myself. It is when we look at something like a player's suspension and turn it into literary tragedy. I once wrote that way. On 20 October 2026, in Busan, I sat in the press row and watched RNG lose 2-3 to G2 in the World Championship quarterfinals, after a year in which they had won nearly everything. I watched Uzi drop his head onto the keyboard, both hands covering his face. Colleagues around me rushed to write pieces blaming the protect-the-ADCarry style and patch 8.19. I could not write like that. I wrote three thousand words under the title The Weight of a Dream, and I still remember the last line: the tears do not belong to RNG, they belong to those who believed. It was widely shared, and hundreds of people wrote to thank me for crying with them. That night taught me two things, and only one of them is beautiful. The beautiful one: empathy is a legitimate way to read, and sometimes the only way that preserves the dignity of the loser. The bitter one: a piece that touches a crowd's emotions is very easily mistaken for analysis, when it is only a good piece of writing. Since then I have set myself a checkpoint. When writing about failure, I must include a concrete cost: a contract cancelled, a prize payment withheld, a tournament slot lost, a season gone and unrecoverable. If I cannot name the price, I am making poetry, not news. The same principle applies to mislabeling: if I claim that wrong classification causes harm, I have to show where the harm is. As someone who has worked between the Vietnamese and Chinese esports scenes, I see that harm in three places. The first belongs to the reader. A fan enters the esports section looking for news about a match, a transfer, a disciplinary ruling, and encounters a photo set. They tap in, find nothing, leave, and gradually form a reflex: this section cannot be trusted. When a genuinely significant esports item appears — like the PUBG Asia Stars case with the player Himass, KRAFTON's response, the dispute between Soopi and Mr. Pha — it gets buried under light content and read with half attention. The second belongs to the content maker. A young writer entering the trade, looking at the structure of the newsroom, learns a false lesson: that writing about data is hard labor with few readers, while tagging is fast and gets traffic. They optimize for the structure, not for the truth. Ten years later we have a generation of commentators who write excellent headlines and cannot read a damage chart on screen. The third, and the one I care about most, belongs to the data. The content industry lives on data, and data is born from classification. A mislabeled piece does not stay put; it enters tables measuring content volume, interest levels, reader demographics. Fifty such pieces a year and the trend has shifted. A thousand and an entire reporting segment becomes meaningless. We will have documents saying national esports is growing at double digits, while most of that growth is rabbit ears. The price of a cheap label does not show up immediately. It shows up when someone outside the industry asks a simple question: according to your numbers, which title is rising. And we have no answer. I still keep a habit from the night I got a name wrong. Before every match I write myself a character file: competitive history, nicknames, peak moments, and the depths as well. I never open the microphone before I know how to pronounce a name correctly. Later I extended that habit to editorial work: never tag a piece before knowing what it was born to do. A match report is born to record a match. A transfer story is born to count money and read intent. A product introduction is born to sell something. Those three need three drawers, three counting methods, and three sets of reader expectations. Merging them is an act of carelessness, not of simplification. I use the word carelessness in its literal sense, not as an accusation, but to point out that the absence of malice can still cause loss. On the night I mispronounced Clearlove's name, I bore no malice. I simply was not prepared. So I propose something very concrete, and I write it down not to please anyone. Pull the Azur Lane piece out of the esports data pool and return it to the games and community content drawer. Do not attach any inferiority to it; it is simply in the wrong place. At the same time, review the tagging rules, because a piece like that cannot slip into the basket if the rules are tight. The test question is simple: does this article contain at least one competitive entity — a team, a professional player, a tournament, a patch affecting balance, a transfer transaction. If none of those are present, it is not esports. Done. No further debate necessary. And I propose something harder, something I know will run into the inertia of my own industry. Rewrite the related-links block at the bottom of pieces like this. If an outlet is capable of reporting that a PUBG player faces a competitive suspension, that a publisher had to apologize, that two names in the same community are in dispute, then that outlet is capable of serious reporting. Yet that news sits at the margin, as secondary links, while the center of the page is a photo set. That inversion of priorities reads like a self-declaration of where the newsroom stands: light content as the main goods, rule-bound news as an accessory. A newsroom that ranks itself that way will never have a voice in debates about regulations, player rights, or prize money. And those debates are what will shape the industry over the next decade. I still keep one thing from my days in Busan. When the whole press room runs in one direction, the person who stays behind alone is often right — but not because they are smarter. They are right because they are one beat slower, and that slower beat gives them time to see the detail the crowd skipped over. Sometimes that detail is just a name mispronounced, a tag attached to the wrong thing, a link pushed to the bottom of the page. I learned to bow to the match after a night of calling a person by the wrong name. Now I am learning another bow, to a photo set that did nothing wrong, but was filed where it does not belong. Someone will say this is a small matter. One tag, one photo set, gone in a day. True, it will sink. But the tag will not sink, because it stays in the database, in the quarterly report, in the record of an industry trying to prove it has grown up. We spent twenty years convincing the world that esports is a serious sport, with rules, discipline, measurement, and both physical and mental demands. We have the numbers to defend that argument: audience scale, contract values, tournament counts, prize pools. Then we throw a set of rabbit ears into the basket with our own hands, and call it esports with our own mouth. A fan opens their phone at nearly five in the morning, sees that tag, and believes it. That is the only part of this story that genuinely hurts me. The tears do not belong to RNG, they belong to those who believed. Reader trust works the same way: it belongs to the readers, not to the person attaching the tag. And once readers have been led wrong a few times, they do not come back. By then, correct tagging will not matter, because no one is reading. The question I leave with the taggers, the editors, the people sitting in growth-report meetings: if tomorrow you had to read through your entire content archive and call every item by its true name, would you dare? And if you would not dare, then what is your basket holding, for whom, and to what end?