Possession is not power: A data audit for Vietnamese football at the 2026 AFF Championship

At minute 90+4, the Viet Tri stadium fell silent. The ball moved from Nguyen Quang Hai to Vu Van Thanh and then back to defender Do Duy Manh. Vietnam had just finished a half in which they held 68% possession. Fourteen attempts, eight dribbles into the box, an xG of 2.1. But the scoreboard did not tell the same story. Thailand stood very deep, gave away the midfield, and waited for exactly one lost pass. When the ball was intercepted, they needed only three passes to reach the edge of Vietnam's penalty area. I do not blame emotion. I do not rush to praise dominance. I open a spreadsheet and trace every number. My question is not how much better Vietnam played than Thailand. My question is: what story are these numbers telling, and what are they hiding? That is how I have worked for seven years. A single number can be a polite lie. Sixty-eight percent possession sounds safe, but if it does not create shots from dangerous areas, it is only expensive jewellery. That match was not an exception. In several matches at the 2026 AFF Championship, Vietnam controlled possession comfortably but still had to wait for set pieces or an individual moment from Nguyen Xuan Son to break through deep defensive blocks. This reminds me of an old mistake. In 2026, I published a self-made xG model for a match between Becamex Binh Duong and Hanoi FC. The model looked at possession, looked at pass numbers, and concluded that my team was dominating. The score was 0-3. Hanoi FC held only 38% possession but produced eleven shots from inside the box. They did not need to keep the ball. They only needed to keep their shape and wait for Binh Duong to make a mistake. The data is not wrong; the reader is wrong — and I used to be that reader. After 2026, I rewrote my algorithm. I added PPDA, the receiving positions of defensive midfielders, and the speed of central attacks. I also made a rule: a number only means something when it is placed next to at least three other variables. Possession must be paired with the zones of control. Passes must be paired with direction. xG must be paired with chance quality, not shot volume. My reading of data starts with context. In the first leg of the 2026 AFF final, Thailand deliberately gave the ball to Vietnam in front of their own penalty area. They did not press in their own third. They retreated into a low block. When the ball was at the feet of a Vietnamese centre-back, Thailand's pressure was almost zero. But as soon as the ball crossed the midfield line, two Thai wingers moved inside. They did not chase the ball. They cut passing lanes into the striker. As a result, Vietnam passed the ball around the block, while Thailand waited for one misplaced pass. If you look only at possession, Vietnam looks stronger. But if you look at where Thailand won the ball, the picture changes. They recovered the ball about 45 metres from Vietnam's goal. From there, they needed two or three passes to reach the space behind Vietnam's full-backs. In my spreadsheet, this is called the starting point of a dangerous attack. A team with less possession but with dangerous starting positions creates more chances than a team with more possession but attacks always starting from their own half. That story repeats across Southeast Asia. Against teams that surrender possession, Vietnam often holds 58% to 65% of the ball. But the conversion rate is not proportional. The reason lies in the tempo of ball circulation. The ball moves sideways more than forward. Full-backs receive the ball too far from goal and are forced to cross into crowded areas. Nguyen Xuan Son can win aerial duels, but he cannot appear in three dangerous spots at the same time. In modern football, a cross from the byline has a scoring probability of around 3-5%. That number gets even lower when the defence has already placed five players in front of goal. PPDA is the metric I use to measure pressing intensity. In many matches, Vietnam's PPDA was between 8.5 and 10. That means opponents were allowed to make around nine passes before a Vietnamese player tried to challenge. That number is not bad, but it is not consistent between zones. Vietnam pressed hard near the touchline, but almost did not press when the ball sat in the centre in front of the two centre-backs. Opponents noticed this quickly. They passed back to the centre-back, waited for Vietnam to step up, and then played one forward pass into the space behind the full-back. This is why high possession did not come with a high number of clear chances. I have a principle: every time I analyse a match, I also record the data that goes against my view. With Vietnam, that part is the ability to create chances from open play. If you remove set pieces, the team's average xG drops significantly. That does not mean Vietnam played badly. It means the team relied on dead-ball situations more than desirable. Goals from corners and free kicks still count, but they are harder to repeat. A team that wants to go far in a major tournament needs chances from both states of play. A 30% probability is not an excuse; it is a reminder that I am right only seven times out of ten. Every time I make a prediction, I leave an exit door for new data. After the semi-final against Singapore, I received praise for my prediction model. But I know the model can fail in the next match. Football does not live in a spreadsheet, but the spreadsheet helps me see football more clearly. I need that humility because every model of mine is built on mistakes that were once mocked. One of my biggest mistakes happened at the 2026 World Cup. I wrote a preview before the final between France and Croatia. Based on xG, I concluded that France could not beat Croatia. The article received more than 200,000 reads and a lot of criticism. France won 4-2. When I reviewed it, I realised I had not adjusted the data for the quality of knockout-stage opponents. Croatia faced weaker teams in the group stage, which inflated their xG. France faced Argentina, Uruguay and Belgium. Every match against a strong opponent reduces the value of raw xG. I rewrote the whole analysis and published it openly. Since then, I never offer a single conclusion. I write in this direction: if you do not consider factor A, the model gives result B; but when you add factor C, the result reverses. For Vietnamese football, factor C is usually physical condition and squad depth. At the 2026 AFF Championship, the fixture list was tight, leaving players only about three days between matches. When fitness dropped, the team's press could not be sustained beyond the 70th minute. The defence had to drop deep, and the midfielders could not move inside as quickly as in the first half. Opponents knew this. They moved the ball wide in the first twenty minutes, forcing Vietnam to run constantly, then accelerated at the end of the match. This is not the fault of any single player. It is the result of relying on too few substitutes with similar quality. Squad depth is the biggest advantage in short tournaments. I still remember 2026, when leagues played behind closed doors during the pandemic. I analysed 400 matches in the Bundesliga and K League 1 to measure home advantage. The results showed home teams won only 31% instead of the usual 44%. That data taught me that fans affect referees, players' psychology and the speed of decision-making. Football without fans proved one thing: data without context is only half the truth. The same applies to 68% possession. It means nothing unless I know that the opponent willingly gave the ball away, where they gave it away, and what they planned to do after winning it back. If I had to summarise Vietnam's AFF Cup 2026 campaign with three numbers, I would choose the number of passes into the final third, the number of recoveries in midfield, and the number of shots from less than 18 metres. These three metrics reflect tactical intent better than possession. In the matches where Vietnam played best, these three numbers were high. In the difficult matches, possession stayed high but those three numbers dropped. The problem is not having the ball. The problem is what the team does with it. Nguyen Quang Hai is an interesting example. He is not the type of player who runs the most, but his receptions between the opponent's lines are always high. When Hai receives the ball 15 metres before the box, he can create the decisive pass. But if the ball reaches him too late, when the opponent's defence is already set, his value drops. The team's issue is that ball circulation in the final third is not quick enough. Safe sideways passes slow down the attack and allow the opponent to recover. Thailand, in contrast, moved the ball with one-touch or two-touch passes, going forward more often. They did not need to keep the ball for a long time to be dangerous. This leads to a big question: What should coach Kim Sang-sik prioritise after the tournament? In my view, the answer lies in reducing the number of harmless passes in the defensive half. Instead of always building from the back, the team could use long diagonal passes into the space behind the opponent's full-backs. Nguyen Tien Linh and Nguyen Xuan Son are strong enough to hold the ball and wait for support. When you have a striker who can win duels, you do not need 65% possession. You need around 45% possession but more direct attacks. Southeast Asian football is moving in that direction. Indonesia and Thailand are using fast wingers who skip the possession stage to finish quickly. I am not saying possession is meaningless. I am saying it must be placed next to other metrics and, more importantly, the context of the match. If you are losing and the opponent drops deep, high possession is normal. If you are winning, low possession is also normal. The only honest way to use data is to record the intentions of both teams. I often ask: what is the chance this is just background noise? If the answer is above 30%, I stop and write about the noise. With the AFF Cup 2026, the noise is that many teams did not have their strongest squads because of the tight schedule. Several key players missed matches through injury or suspension, which distorts the statistics. However, there is one signal I believe is real. Nguyen Xuan Son ended the tournament as top scorer. The remarkable thing is not the number of goals, but the variety of his finishes. He scored with his left foot, right foot, headers and through tight-space finishes. A striker like that frees the team from depending on one attacking pattern. But to make the most of Son, the team must deliver the ball into the box faster. Every extra second in midfield is a chance for the opponent to reorganise. I want to see the team accept more risk, accept losing the ball in dangerous areas, in exchange for clearer shots. That change must start at the clubs. When I watch V.League, I see too many teams choosing to defend in numbers and wait for opponent mistakes. That creates a generation of players used to slow tempo, afraid to receive the ball under pressure and lacking skills in tight spaces. The national team cannot sustain high pressing for 90 minutes because of that. This is not Kim Sang-sik's fault alone. It is a problem for the whole youth development system. If we want to improve the data, we have to begin on training pitches, not in international matches. I am often asked: who are my articles for? The answer is: for people who want to verify things themselves. I publish sources, calculation steps, and even numbers that go against my view. In this article, the opposing data is about the team's defensive performance. Despite pressure, Vietnam had solid statistics in interceptions and duels. Do Duy Manh and Bui Hoang Viet Anh are centre-backs who read the game well. The problem is not the centre-backs. It is the space behind the full-backs when they push forward. That weakness can be fixed with a deeper holding midfielder or with only one full-back advancing at a time. The story of Vietnam's possession is not new. It appeared under Park Hang-seo, continued under Philippe Troussier and now under Kim Sang-sik. But each period has a different shade. Under Park, the team accepted lower possession to counter-attack. Under Troussier, the team tried to control more but lacked sharpness in the final third. Under Kim, the team is looking for balance. The 2026 AFF Cup showed that Vietnam can control the ball when needed, but can also play directly with fast wingers. That balance is the foundation for the future. In the 2026 World Cup qualifiers, Vietnam's opponents will no longer be Southeast Asian teams that surrender the pitch. They will be teams that can press high and keep intensity for 90 minutes. If Vietnam only knows how to keep 65% possession without clear chances, they will be punished. On the other hand, if they learn to transition quickly, from defence to attack in three passes, they can surprise stronger opponents. I want to end with a question, not a conclusion. In the next five years, do we want Vietnam to be a team that keeps the ball more but wins less, or a team that keeps the ball just enough and wins the important matches? Data will not answer for us. It only reflects the choices we have made. Football does not live in a spreadsheet, but the spreadsheet helps me see football more clearly. And what I see most clearly after the 2026 AFF Cup is that Vietnam is standing at a crossroads. They can continue to enjoy the safety of sideways passes, or accept risk to reach a new level.