Trang chủEsportsA Blank Cell Is Not a Certificate of Innocence
Esports

A Blank Cell Is Not a Certificate of Innocence

**Core answer** Esports Việt Nam thiếu dữ liệu ở tầng hợp đồng, đánh tập và dòng tiền, khiến ô trống trong các bảng thống kê bị đọc thành tín hiệu an toàn. Án phạt dàn xếp tháng 3/2023 tại VCS không xuất hiện trong bất kỳ chỉ số thắng thua nào. **Key facts** - Tháng 3/2023: hơn 30 tuyển thủ VCS bị cấm thi đấu vì dàn xếp kết quả; không chỉ số thắng thua nào phát hiện được. - Dàn xếp chủ yếu chảy qua thị trường phụ (mạng đầu tiên, tổng mạng, thời lượng ván), không qua kết quả chung cuộc. - Một mùa VCS cho mỗi tuyển thủ khoảng 40-45 ván cấp chuyên nghiệp — cỡ mẫu nhỏ ở cấp trận. - Liên Quân Mobile, PUBG Mobile, Free Fire không có API công khai tương đương Riot API của LMHT. - Không tồn tại cơ sở dữ liệu công khai về phí chuyển nhượng hay thời hạn hợp đồng tại Việt Nam. **Source attribution** Tổng hợp dữ liệu công khai từ VCS và Riot Games, bài phân tích đăng ngày 13/08/2026 | Cross-checked: VuaBong.vn **Related Q&A** Q: Vì sao dữ liệu thắng thua không phát hiện được dàn xếp? A: Vì tiền dàn xếp đặt vào sự kiện vi mô như mạng đầu tiên và tổng số mạng, vốn độc lập với kết quả thắng thua. Q: Chỉ số nào thay thế KDA khi đánh giá tuyển thủ? A: Chênh lệch lính phút 10 theo cặp đối đầu, số lần chết một mình ở đường và tỉ lệ thắng trao đổi máu ba phút đầu, theo VangBong.vn Player Depth Index. Q: Rủi ro lớn nhất của việc thiếu dữ liệu là gì? A: Ô trống bị đọc thành bằng chứng vô can, khiến các vấn đề về lương, hợp đồng và tính chính trực không bao giờ được kiểm chứng." } ```

In March 2026, I reopened the Excel file tracking VCS that I had been building since the 2026 season. The file had 1,847 rows, one per match, one column per metric: CS differential at minute 15, 5v5 teamfight win rate, deaths before minute 10, vision per minute, games won. That week I added a new column called 'off-field notes'. It was empty. It stayed empty across all 1,847 rows.

The same week, VCS announced sanctions against more than thirty players for match-fixing. I sat in a coffee shop on Highway 13, opened the file, hit Ctrl+F and typed in each name from the list. Not a single hit. Not one cell in the spreadsheet I had been presenting as evidence of my analytical competence could reflect what was happening underneath the standings.

When I was 14, the 2026 World Cup taught me that the weak do not win by miracles. My first piece on the page 'Bong Da Nguoc Goc' argued that Germany's group-stage defeat to South Korea was no accident but the consequence of a system that had been obsolete for four years. It was shared 8,000 times in 48 hours. A local radio reporter put me on air for a live debate, pressing hard, and I did not back down because I believed the argument. Since then I have lived by one rule: if you cannot measure it, do not claim it.

The 2026 sanctions forced me to amend that rule. The belief was still right, but it was missing a clause. If you cannot measure it, do not conclude that nothing happened either. A blank cell in my spreadsheet was never proof of innocence. It was only proof that I had chosen the wrong thing to measure.

Vietnamese esports data has three tiers, and a fourth that nobody has named.

The first tier is API data. League of Legends has the biggest advantage in the entire esports ecosystem: Riot Games opens its API, and third-party platforms such as Oracle's Elixir and gol.gg archive nearly every professional match in queryable form. Someone sitting in Binh Duong can pull the gold differential at minute 14 for a top laner across four seasons, for free. This tier exists, it is complete, and it is used at a minimum.

The second tier is organiser-published data. Liên Quân Mobile, PUBG Mobile, Free Fire — the titles with the largest Vietnamese player bases — have no comparable public API. Numbers reach the public through organiser graphics: KDA, damage share, kill participation. No raw record to cross-check, no record to re-examine, no record to argue with. You take the number and trust it, or you have nothing.

The third tier is community-made data. Hand-pasted stat tables, clipped highlights, forum roundups. Its value is not accuracy but speed: it appears before the organiser even posts a statement.

The fourth tier is the tier of things nobody measures. Contract clauses. Scrim results. Salaries. Betting flows. Players' mental health. The relationship between player and agent. These are the blank cells in every spreadsheet Vietnamese esports journalism has ever built — including mine.

The problem is not that there is little data. The problem is that blank cells get read wrong. In statistics there is a concept called censored data — missing not at random. The value is missing not because someone forgot to measure it, but because the value itself, if it appeared, would harm someone. Blank cells in this industry are almost always that kind.

I used to think otherwise. In 2026, after spending three lockdown weeks rewatching 52 Bundesliga matches played in empty stadiums, I wrote a 4,000-word piece concluding that only 33% of home advantage came from the crowd, with the rest from geography and travel routine. The 2026 empty stadiums were a data laboratory nobody asked permission for. A Binh Duong supporters' group shared it with the comment 'crazy kid, but he has a point', and views rose 400%. What I learned was not that I was good, but that an isolated variable is the only variable worth talking about. Football gave me a chance to isolate variables. Vietnamese esports does not, because most variables are never recorded.

The 2026 sanctions and the limits of win-loss data

There is a common misunderstanding about match-fixing: people imagine it as a deliberate loss. A team suddenly plays badly, loses from ahead, the standings flip, and sharp eyes catch it. If that were true, win-loss data would catch it. Win rate collapses, metrics dive, an anomaly model raises the alarm.

Reality works differently. Most fixing money in esports flows through prop markets, not outright results. Who gets first blood. Whether total kills land above or below a line. Dragon count. Game duration. A team can win 2-0 and still deliver the exact first blood that was bet on. A player can perform brilliantly all game and still die on schedule in minute seven of game one.

That is why win-loss data can be perfectly clean while everything underneath has rotted. A 70% win rate is fully compatible with fixing. There is no contradiction to detect, because the two things do not sit on the same axis.

Catching this requires a different data tier: the distribution of first-blood timings, the clustering of total kills, the standard deviation of game duration, the correlation between events inside a single game. This is the analysis bookmakers do daily. They have a financial incentive to do it, and they have the data to do it.

The result is an asymmetry I have not seen anyone in Vietnamese media state plainly: the betting side understands the match at event level better than the reporting side. Journalists look at the scoreboard. Bookmakers look at first-blood distributions per player, per map, per time slot. When a case breaks, journalism retells a story the market knew six months earlier.

I am not saying every sanction is predictable from numbers. I am saying we have never tried. And in the gap left by never having tried, a generation of young players has walked through their careers with nobody able to verify anything.

Fourteen BO3s and the belief in form

A VCS group stage with eight teams, double round robin, gives each team about fourteen BO3s. Add playoffs and a player who completes a split has roughly forty to forty-five professional games. A CS2 team plays eighty to a hundred and twenty maps a year. A footballer plays thirty-eight to fifty matches.

At match level, Vietnamese esports has a small sample. At event level, it has a larger sample than any other team sport. Every MOBA game contains hundreds of recordable individual decisions: wave states, item timings, ward placement, movement inside fights, skill order. One game is thousands of data points. Forty-five games is tens of thousands.

The problem is that Vietnamese analysis stops at match level. Form gets measured by KDA across forty-five games and then argued about as if it were objective fact. KDA across forty-five games is noise. It is driven by opponents, by team strategy, by whether the team wins fast or loses fast. A top laner on a winning team will have a handsome KDA without playing any better than anyone else.

What carries signal is CS differential at minute 10 split by matchup, solo deaths in lane, and trade-win rate in the first three minutes. These are less affected by team outcome, so they say more about the individual. They are available. Almost nobody uses them.

Based on my experience watching these matches, a pattern repeats: after every game, the loudest debate online centres on two numbers — KDA and kill count. Those are the two easiest numbers to read and the two easiest to misread. A debate about a Vietnamese player routinely runs to hundreds of comments without anyone opening the CS differential table.

I am writing this so you argue with me, not so you agree. But if you plan to argue, argue with CS differential at minute 10. I will listen.

Scrims: the best predictive dataset, locked tightest

In every team sport, scrim data is the best predictor and the most closely guarded. Every team scrims. No team publishes results. In Korea, China and Europe, scrim results leak through unofficial channels and become a commodity. In Vietnam they leak through private messages between coaches, through acquaintances in the scene, and finally reach journalists as a whisper in a venue corridor.

Logically this is near-perfect waste. The dataset with the highest predictive value is turned into the currency of rumour instead of the raw material of analysis. A coach who knows his team lost eight of ten scrims to a specific opponent will not say so. A journalist who hears it cannot verify it. And the public receives a sourceless, sample-less, dateless claim.

I once sat and listened to three people give three different versions of the same scrim block by the same team, on the same evening. Nobody was lying. They were repeating what they had been told, through two or three relays. After three relays, a 3-5 score becomes 'that team has problems'.

People call it an illusion; I call it a hypothesis that needs testing. But testing it requires something this industry does not have: a traceable scrim log, even if only inside the league, even if only for investigations when a complaint is filed.

The highlight economy and selection bias

In titles without a public API, data reaches the public through a single pipeline: the broadcast cut. A forty-minute match is packaged into three minutes of highlights. Writers read the highlights. Viewers read the articles. The story forms from those three minutes.

This is selection bias in its purest form. Three minutes are chosen because they look good, not because they are representative. A brilliant play at minute 32 is far more likely to be clipped than a twenty-minute stretch of correct positioning, even though the positioning is what wins the game. The public ends up believing matches are decided by moments, when they are usually decided by silences.

A lost teamfight is worth more than a dull win. But it is only worth something if someone records it. In a league where raw recordings are not publicly archived, that lost fight disappears when the broadcast ends.

I ran a small experiment in summer 2026: pick five matches by one team, rewatch them in full, log every situation where advantage was lost, then compare with what was discussed online after each match. Across five cases, the community discussed on average two of the eleven advantage-loss situations I logged. Both had good visuals. The other nine — mostly vision control loss and wave tempo loss — did not exist in collective memory.

That is the whole problem. Not that fans are stupid. It is that they are only shown three minutes.

Loans with purchase obligations and broken financial plans

The transfer market is a playground for rumour, not for fact. In Vietnam this is absolutely true, because no public database of transfer fees exists. Nobody knows what a young player was sold for. Nobody knows how long a contract runs. Nobody knows where the buy-back clause sits.

In that vacuum a structure has formed and is quietly reshaping the whole system: loans with purchase obligations, or loans with buy options.

The mechanism works like this. A big org takes a young player from a small org on loan. The big org pays no fee now. The small org keeps the registration and receives a small amount, or nothing. After a year, if the player performs, the big org triggers the buy clause at a price fixed in advance — usually fixed at a point when nobody had heard of the player. If the player does not perform, the big org returns him. The risk sits with the small org.

In accounting terms this is elegant. The big org holds an option, with no obligation. The small org holds an obligation to wait, with no option. In finance, the party holding the option always beats the party bound by the obligation. That is not opinion, it is arithmetic.

The long-term consequences are concrete. Small orgs cannot plan finances across a season, because their transfer cash flow depends on a condition twelve months out decided by the other party. They cannot sign long deals with players because the player may be pulled away. They cannot invest in facilities because they do not know whether the money arrives in month eleven.

And they become the finishing school for the big orgs. Develop, play, absorb contact, then hand over exactly when market value exceeds contract value. That is the small-league model in European football, copied wholesale into Vietnamese esports without anyone naming it.

What bothers me most is not the structure — the structure makes sense for whoever has the money. What bothers me is that not one number is published for the public to judge. No fees, no contract lengths, no buy-back terms. A market running entirely on insider trust and outsider rumour.

Agency contracts and structural silence

There is another blank column in my spreadsheet I have never filled: 'what the player thinks'.

The problem is not that nobody asks. The problem is that the answer is censored before it reaches the person asking. When a player signs with a management company or a team, speaking obligations usually come attached. Interviews get approved. Statements get checked. The result is that every Vietnamese player interview of the past three years has the same structure: thanks to the team, thanks to the fans, commitment for the next match.

I do not blame the players. A twenty-year-old who just signed the first contract of his life, a contract that pays his family's bills, will not say something that breaks it. That is rational behaviour. But the consequence is that qualitative data about working conditions in this industry is zero.

Nobody knows how many players are dealing with competitive anxiety. Nobody knows how many are owed wages in silence. Nobody knows whether a player left a team for competitive reasons or something else. These questions have no answers, and they have no answers because the structure was designed to make sure they have none.

In statistics, censored data is data whose value is missing because the value itself harms the person supplying it. You cannot assume the missing value sits at the average. You must assume it sits on the adverse side, because that is precisely why it was hidden.

Every analysis of Vietnamese esports currently ignores this. We have an enormous spreadsheet of in-game metrics and an enormous blank cell about human working conditions, and we argue about the first as if it were the whole picture.

Where I might be wrong

Qatar 2026 proved one thing: even the strongest have blind spots. So do I. And my blind spot is probably that I am demanding a data standard that does not fit this kind of competition.

There is a serious argument that the absence of public data is a protective barrier. If every micro-metric were published, the surface for fixers would grow, not shrink. They would not have to guess. They would know exactly what a player's first-blood timing distribution looks like, and how far to deviate without being caught. An opaque league may be safer than a transparent one, literally.

There is a second argument I find harder to rebut: Vietnamese esports' instinct for talent has demonstrably worked. Players like Do Duy Khanh, who took a Vietnamese team to big matches at MSI 2026, or Tran Duy Sang with an unusually long career in the top lane, were found by eye, not by spreadsheet. Korean and Chinese organisations with full analytics departments keep missing talent. If data were the answer, they would not miss that much.

And a third argument, the sharpest: the real enemy of this industry is probably not the blank cell but the wrongly filled one. A blank cell honestly says 'I do not know'. A KDA invented to protect a player is a lie. Seen that way, a data desert is less dangerous than a fake oasis. I am not sure I can refute this.

But even accepting all three, one thing does not change. If nobody records it, then when something happens, nobody can prove anything. The 2026 sanctions did not break open from data; they broke open from an investigation, and the result was a list the public can only believe or disbelieve. No spreadsheet to cross-check. No benchmark to say whether this sanction was heavy or light, right or wrong.

Sports culture lives in who you choose to hate, not in the stands. And when you can only choose based on an unverifiable list, what you have is not sports culture. It is blind faith.

What I am willing to bet

I predict that within eighteen months at least one Vietnamese esports organisation will publish a public data dashboard — not broadcast graphics, but queryable data. Not because they want transparency, but because sponsors will start demanding it.

And I predict the first Vietnamese scandal caught by numbers will not come from a win-loss anomaly. It will come from an event-level pattern: a player whose death-timing distribution deviates from his own baseline systematically, in exactly the games where betting volume spiked. When that happens, someone will say they knew all along.

I will be the first to ask where they recorded it.

As for the 'off-field notes' column in my Excel file, it is still empty. I have not deleted it. I keep it there so that every time I open the file I remember I once presented a 1,847-row spreadsheet as proof of understanding, while the thing that decided the whole story sat in a column I did not know how to measure.

Cầu thủ liên quan