Classification Error: When an Azur Lane Cosplay Is Counted as Vietnam Esports Output

On a Saturday morning, I opened the esports content classification table of a research unit in Hanoi. In the esports column, after a series of articles about Liên Quân, PUBG and Valorant, there was a strange name: Shimakaze. I stopped, read it again, and opened the original article. The article did not talk about a match. There was no player, no meta, no tournament. It introduced a cosplay photo set of Shimakaze from Azur Lane. The cosplayer was known as Thiết Thủ Khiếu Thú. The main praise was about character resemblance, a mischievous expression, and the ability to transform through many outfits. A beautiful photo set. But it sat in the esports list. People remember goals. I remember the substitute who clapped for his teammates. That day, I remembered how a description turned a cosplay set into a sports data point. Azur Lane is a mobile gacha game developed by Manjuu and Yongshi. Players collect ship-girl characters, build fleets, and play PvE stages. The game has a large fandom, especially in Japan, China, and Southeast Asia. But its appeal comes from character design, skins, story and collection. It has no transnational professional league, no qualifiers, no regular competitive teams. Players do not climb a ranked ladder like in Liên Quân or Valorant. Azur Lane does not run on a competitive meta cycle. Its cycle is banner, skin and character events. The original article said Thiết Thủ Khiếu Thú transformed into Shimakaze, a Sakura Empire destroyer. The character is recognizable by rabbit ears and a naval uniform that mixes strength and cuteness. The article said the cosplay felt closer to the in-game version, and that the mischievous expression made the character lively. The whole piece stayed on an aesthetic level. There was no data on views, engagement, revenue or reach. No sports figure was mentioned. Not even the word competition appeared. This is a promotional fan-content post, not an esports report. Why was it classified as esports? I cannot access the survey unit's algorithm. But the structure of the data suggests three causes. First, the title and body contain the keyword game. Automatic classifiers often put anything game-related into esports. Second, the hosting page had a related-links block referring to PUBG Asia Stars, a real esports event involving Vietnamese player Himass, plus a copyright story. The algorithm sees those links and raises the esports weight. Third, many measurement units still use an open dictionary: the word game means esports. The result is that an Azur Lane cosplay article enters the esports growth list and competes with real tournament coverage. The damage is not just one wrong data line. It spreads through the whole ecosystem. Sponsors use reports to decide investment. Platforms use reports to build content packages. Media outlets use reports to shape editorial strategy. If a cosplay set is counted as esports output, then industry growth becomes artificially beautiful. I once analyzed Incheon United and noticed that a corner-kick drill repeated 47 times in three training sessions could generate 200,000 reads. The number was attractive, but it only told the story of a drill, not the team's strength. Distance covered and sprint counts are packaged as effort metrics, but useless running also produces good numbers. Esports data is the same. If we do not check the origin of a classification label, we use a pretty number to hide an empty story. Let me tell a small story. In 2026, at the World Cup, I mispronounced midfielder Jung Woo-young as Jung Young-woo three times on live radio. I could not sleep that night. I reviewed every match tape, recorded my voice, and practiced the names of all 23 squad members ten times a day. By the historic 2-0 win over Germany, I did not mispronounce a single name. Mispronouncing one syllable taught me how little I understood about that football culture. My way of fixing the error was to return to the smallest details. Now, when I see Shimakaze in an esports data table, I think our industry is making a similar mistake. The error is not in the cosplay set. The error is in how we name things. Calling a cosplay set esports is like calling a fashion show a match. Some readers may ask: if this is only a small article, why make a big deal? Because I have followed tournaments since 2026, from my days as an esports athlete and event organizer. I have seen bad data being used to sell ads, to price sponsorships, to persuade brands to spend money. When an industry uses an unclean data foundation, every decision built on top is fragile. My job is not to shout the loudest. My job is to keep the drumbeat so others can walk in step. Years ago, I held the story of young goalkeeper Park Seo-jun for six months and only published it when he officially debuted. I believe stories need time to ripen. Data also needs correct labels before it is allowed to speak. There is a contrarian angle. Separating cosplay from esports does not devalue the photo set. On the contrary, it gives the photo set the right home. A Shimakaze cosplay will shine in a gaming, pop culture, or entertainment section. There, it reaches the right audience: Azur Lane fans, the cosplay community, and lifestyle brands. Pushing it into esports just for traffic is a subtle insult to the people who made the set. They do not need a wrong label to be recognized. They need an audience that understands the value of transformation. In esports, an analyzable subject needs three attributes: competitive rules, a ranking system, and a tournament organizer. Azur Lane has none of those. Players may join community events, but those are side activities. A few PvP modes do not turn a game into esports. League of Legends and Dota 2 have long tournament circuits, rules, and professional players. Dota 2 has The International with tens of millions of dollars in prize money. Liên Quân has domestic and international leagues. PUBG has qualifiers. Azur Lane has seasonal campaigns, new character banners and gacha rates, but there is no competitive season in which a team becomes champion. This seems obvious, but Vietnamese media is blurring the line. I observe a pattern: outlets attach the word esports to anything with game characters to attract traffic. This is especially common in cosplay. A Shimakaze cosplay article can easily get tens of thousands of views if labeled as a hot game topic. But that blurs the boundary between entertainment and competition. A consumer goods sponsor may look at a report and believe Vietnamese esports is growing fast. They do not know that most growth comes from fan content, cosplay, and entertainment streamers. They put money on the wrong billboard. In 2026, I followed the South Korean Olympic team. Lee Kang-in was placed in a free role behind the striker, unlike his usual wing position at Valencia. In the quarter-final against Mexico, despite losing 3-6, he made 12 chance-creating passes, the most in the tournament. When I wrote that Lee Kang-in does not need to play on the wing, many people misunderstood it as criticism. In fact, I was reclassifying his role. Putting a player in the right position, putting an article in the right section, putting a cosplay set in the right space: these are all acts of respecting truth. Based on my experience tracking matches, I know that a small classification mistake can lead to a completely wrong conclusion. When I work with brands, they often ask: how fast is Vietnam esports growing? I answer: it depends on your definition. If you include cosplay videos, reaction videos, and fan art, the number is much larger. If you only count tournaments, teams, players, ticket revenue, and media rights, the number is more modest. Sponsors should not pick the prettiest number. They should pick the number that matches their commercial goal. A contract is a signed farewell. When a sponsor signs onto a false report, they separate money from real value. Some people say a mislabeled cosplay article is only a small stone dropped into a lake. The water ripples and then calms. I disagree. In the age of algorithms, a small stone can be multiplied millions of times. One wrong label can be copied by other measurement units. In a few quarters, it becomes an esports growth trend that nobody checks. I am old enough to know that pretty numbers often hide uncomfortable questions. I have buried many stories in my notebook, waiting for the right moment. I buried a story for six months because no one was ready to hear it. This time, I am not burying it. I am writing it out, because Vietnamese esports data is growing so fast that everyone wants to talk about it, but few people stop to check what they are actually talking about. The Vietnamese market has its own characteristics. Mobile games are more popular than PC games. The cosplay community is closely tied to the anime community. Many esports media outlets rely on entertainment content for traffic. Therefore, a classification standard should be built by media professionals, not just by algorithms. We need to separate game, esports, cosplay, entertainment streamer, and industry news. Each type has a different unit of measurement. Only then can reports be meaningful. I do not have a perfect formula. I only have the habit of checking names before publication. Before every article, I ask a Korean colleague to listen to how I pronounce player names. Before every data table, I should ask: what is the real name of this item? Classification units need to distinguish three layers. The first layer is competitive esports, where players, tournaments, rules and results exist. The second layer is gaming in general, where players experience a product. The third layer is fan content, where cosplay, fan art, creative videos and fan culture live. These layers can overlap, but they must not be merged in one report. A Shimakaze cosplay set may be made by someone who loves Azur Lane. That person may also watch Liên Quân tournaments. But that does not turn the photo set into a match. Audience overlap does not change the nature of the product. I do not expect all industry reports to change overnight. I only want to remind everyone that before we talk about growth, we must talk about categories. An Azur Lane cosplay set can be a wonderful work of art. It can attract gamers and anime communities. But it is not a match. When we confuse the two, we lose the ability to measure accurately. I stood on the Incheon training ground many times to watch how players tied their shoelaces before kickoff. Spectators watch the score. I watch how they tie their shoelaces before the ball rolls. Those small details give me the story. In data, the smallest detail is the classification label. If the label is wrong, every story behind it is wrong. I wrote this article slowly, weighing each sentence. Because I know that a rushed article will create a wrong answer, and a wrong answer can be copied hundreds of times. The Incheon training ground still remembers every step I took while waiting. Data is the same. It remembers every label we attach. If we attach the wrong label, it will repeat that mistake for years. Let cosplay live in the world of cosplay. Let esports be measured by numbers that truly come from competition. As for me, I will keep the drumbeat, quiet and precise, until the listeners can walk in step with a single rhythm.