Lessons from 'La Universidad del Perreo': When Sports Data Gets Misclassified and the Consequences for Vietnamese Sports Journalism

I stared at the screen and asked myself: am I reading about tennis or about reggaeton music?

That's a question no sports journalist ever wants to ask. But this week, I received an analysis report from the Stage-1 system — an automated content processing pipeline designed to classify and analyze tennis articles. The result displayed: "Domain Label: tennis." But the content inside was entirely about Wisin — a Puerto Rican reggaeton artist — launching his album "La Universidad del Perreo" with 2 million fan registrations.

Not a single tennis player. Not a single tournament. Not a single serve or break-point statistic.

This is a story about a classification system failure. But deeper than that, it's a lesson about the value of domain expertise, the danger of blind trust in algorithms, and the responsibility of sports journalists — especially in Vietnam, where the sports data industry is still young.

---

The context of this issue begins with a perfectly normal article in Latin music media. Wisin, former member of the duo Wisin & Yandel, announced his project "La Universidad del Perreo" — an album combined with a "university" concept as a marketing campaign. The idea was simple: turn music into an "academy" where fans could register, participate, and learn about reggaeton culture. There were music professors (Ivy Queen was one of the first "teachers"), lectures (the album's songs), and students (2 million who registered on the website).

It was an excellent marketing campaign. But it was not sports.

So why did the system classify it as tennis? The answer lies in misleading keywords. "Tour" — Wisin wanted to embark on a tour with his album. "University" — the educational concept. "Teacher" — artists invited as instructors. These words, when fed into a simple keyword-based classifier, could falsely trigger the "sports" or "tennis" label. A system error, but with significant consequences.

---

I have followed tennis tournaments for nearly a decade. From Wimbledon to the Australian Open, from five-set epics lasting five hours to tearful press conferences after defeat. I've learned that data in sports is not just numbers — it's story, context, the connection between thousands of small details that only an insider's eye can see.

"Numbers don't lie. We just have to ask the right questions."

That phrase is the compass for every article I write. But if the numbers themselves — or rather, the way we classify them — are wrong from the start, then all subsequent analysis is meaningless.

Imagine a Vietnamese sports journalist receiving this report and not checking the original source. He might write an analysis about "brand-building strategies of tennis players" based on data from a reggaeton artist. The 2 million registration figure could be cited as an "engagement record" in tennis. Ivy Queen's participation could be mistaken for a new coach. And the entire article — however well-written — would be a scientific error.

This is not a hypothetical scenario. In the era of generative AI, where automated tools are gradually replacing the work of journalists, these types of errors are becoming more common than ever.

---

In Vietnam, sports journalism is at an interesting transition point. Local sports outlets are gradually adopting technology into their content production processes. But unlike developed journalism markets like the UK or Australia — where I've worked and witnessed strict editorial processes — Vietnam is still in the "learning to trust" phase with automated systems.

And that's when it's most dangerous.

"A new lineup, like a new watch, needs time to run accurately."

This saying was meant for sports teams, but it applies equally to AI systems. A content classification algorithm cannot be perfect from day one. It needs training, testing, error correction, and retraining. And most importantly — it needs human oversight.

In this case, an experienced tennis journalist would immediately recognize that the content was not sports-related. Wisin is not a tennis player. Ivy Queen is not a coach. 2 million registrations are not ATP points. But a junior journalist, or an editor who trusts the system too much, might miss these warning signs.

---

Let's talk about the number 2 million — the only quantitative data point in the entire original article. This is an impressive number for any marketing campaign. 2 million people registering for a "music university" after a single announcement. In a sports context, this number is equivalent to the fan following of a major tournament.

But the difference lies in the nature. In music, 2 million registrations could be the result of a well-funded viral campaign. In sports, a similar number could reflect the sustainable growth of a fan community over many years. Comparing them without context is a serious methodological error.

"Fans have the right to live in emotion; I have the duty to live in data."

But data also needs to be placed correctly. A registration number from the music industry cannot be used to analyze trends in tennis — unless there is a theoretical framework connecting them. And that theoretical framework, in this case, does not exist.

---

Interestingly, the original article about "La Universidad del Perreo" actually contains a story with a similar structure to many stories in sports. It's a journey from being dismissed to gaining mainstream recognition. Reggaeton, as the article describes, was once heavily criticized before becoming a million-dollar industry and being recognized at major awards like the Latin Grammy.

This journey — from "criticized and dismissed" to "institutional recognition" — is a familiar pattern in sports. Look at the history of professional tennis: from the amateur era to the Open Era, from modest prizes to millions in prize money. Or look at Vietnamese football: from struggling early days to the miracle at the 2026 U23 Asian Championship.

But structural parallels do not mean content similarity. A sports analyst can learn from how reggaeton built its brand, but cannot use reggaeton data to draw conclusions about tennis.

---

This misclassification incident raises three important questions for Vietnamese sports journalism:

First: How to build an accurate sports content classification system? The answer lies not in having more keywords, but in understanding the true nature of each field. A sports classification system needs to be trained on pure sports data, with specific keywords (tournament names, player names, technical terms) and cross-checking mechanisms.

Second: Who is responsible when automated systems make errors? In a traditional newsroom, the editor bears ultimate responsibility. But when content is created or classified by AI, the line of responsibility becomes blurred. Vietnamese newsrooms need clear regulations on this issue.

Third: How to train sports journalists with enough expertise to detect system errors? This is perhaps the most important question. A good journalist not only knows how to write — he must deeply understand the field he covers.

---

"I don't remember what I wrote. I remember what I counted."

When I was a beat reporter following the Australian national team at the 2026 World Cup, I learned a valuable lesson about being cautious with new trends. My colleagues were excited about Graham Arnold's high-pressing tactics. But I spent two days reviewing Australia's last three matches, calculating the conceded rate when pressing high (1.8 goals per game) versus when sitting deep (0.9). My conclusion: the tactic was unsustainable. Argentina scored two goals from space behind the defense.

The same lesson applies to AI systems: don't trust a classification result just because it comes from an automated system. Check, verify, and be ready to ask questions.

"The beat keeper doesn't compose the music, but without him, everything falls out of rhythm."

In sports journalism, humans remain the irreplaceable "beat keepers." AI can help classify, summarize, and even draft content. But only a journalist with expertise can detect that an article about reggaeton music is not tennis content.

---

This incident also raises a larger question about the future of sports journalism in the AI era. Can automated systems completely replace human journalists? The answer, based on evidence from this very incident, is no.

AI can process thousands of articles per second. It can extract data, create summaries, and even write articles. But it lacks the ability for nuanced contextual judgment — something that only experience and expertise can provide. An AI might know that "tour" is a word in tennis, but it cannot understand that "tour" in the context of a reggaeton artist means a music concert tour.

This is not a mere technical error. It's an error of essence: a lack of real-world understanding that only humans possess.

---

"Transfer rumors are a math problem: too few facts, too many variables, all hypothetical."

Similarly, wrong data analysis is an unsolvable equation. When the input data is wrong, all conclusions are worthless. This is the GIGO (Garbage In, Garbage Out) principle in computer science — and it applies equally to journalism.

For Vietnamese sports journalists, the lesson is clear: always verify your data source. Never blindly trust any system, whether it's AI, algorithms, or even colleagues. Ask questions. Verify. Be ready to say "no" when things aren't right.

---

Looking at it from a positive angle, this incident is a valuable learning opportunity. It reveals weaknesses in automated content processing workflows and what needs improvement. Specifically:

1. Cross-domain detection mechanisms are needed: When an article contains keywords from multiple different fields (e.g., "tour" in both sports and music), the system should flag a warning rather than auto-classify.

1. Human verification step is needed: For articles with low classification confidence, journalist intervention is required before analysis proceeds.

1. Model retraining is needed: This incident provides valuable training data to improve classification accuracy.

---

"There are things that only reveal themselves when you sit still longer than a set."

In this case, simply sitting still and carefully reading the original article — instead of rushing to follow the automated classification result — was enough to detect the problem. You didn't need to be a tennis expert to know that Wisin is not a tennis player. You just needed to read.

This is the simplest but most important lesson: in the age of automated information, the skills of reading and verifying remain paramount.

---

My conclusion from this incident is not a criticism of AI systems, but a reminder of the value of human expertise. Automated systems can make mistakes — that's the nature of technology. But humans have the responsibility to detect and correct those mistakes.

For Vietnamese sports journalism, this is a golden opportunity to build quality editorial processes, train journalists with deep expertise, and develop AI systems suited to the specific characteristics of Vietnamese sports.

"I don't write to shock. I write to answer the question no one dares to ask."

The question I want to raise today is: Are we blindly trusting technology? And if so, what will we lose?

The answer, as this incident has shown, could be a lot. A misleading article, worthless analysis, and — worst of all — a false belief that technology can completely replace human expertise.

---

I will end this article with an observation from the incident itself. The original article about "La Universidad del Perreo" was actually an interesting story about brand building and community connection. If analyzed in the right domain — music and marketing — it could provide significant value. But when forced into a tennis framework, it became an error.

Similarly, AI can be a wonderful tool if used correctly. But if we expect it to do things beyond its capability — like completely replacing human judgment — we will be disappointed.

"A big win is not yet a revolution."

And a data classification error is not yet a disaster. It's just a reminder that we — journalists, analysts, sports media professionals — still have an important role. The role of the beat keeper, the checker, the one who asks the right questions.

And that, perhaps, is the biggest lesson from the story of "La Universidad del Perreo" — a story about music that taught us more about sports than anyone could have imagined.