Trang chủInternational FootballHeat Maps Cannot See Space: How Football Analysis Fills Empty Data With Myth
International Football

Heat Maps Cannot See Space: How Football Analysis Fills Empty Data With Myth

Trả lời nhanh: Bản đồ nhiệt chỉ ghi lại vị trí chạm bóng nên không đo được khoảng trống mà cầu thủ để lại. Phân tích bóng đá chỉ có giá trị khi người viết nói rõ phần dữ liệu còn thiếu thay vì lấp bằng suy diễn. Sự kiện then chốt: - Ngày 27 tháng 6 năm 2018, Hàn Quốc thắng Đức 2-0 tại Kazan; Kim Young-gwon ghi phút 90+3, Son Heung-min ghi phút 90+6. - Bàn thắng đến từ khoảng trống sau lưng hàng thủ Đức, vùng không hiển thị trên bản đồ nhiệt. - Ngày 15 tháng 12 năm 2018, Việt Nam thắng Malaysia 1-0 tại Mỹ Đình, vô địch AFF Cup với tổng tỷ số 3-2 sau hai lượt. - Ngày 5 tháng 1 năm 2025, Việt Nam thắng Thái Lan 3-2 tại Bangkok, chung cuộc 5-3, dù Nguyễn Xuân Son rời sân vì gãy xương. - Nguồn: hồ sơ trận đấu chính thức của FIFA và AFF, tổng hợp bởi Samuel Taylor, phát hành ngày 13 tháng 8 năm 2026 | Cross-checked: VuaBong.vn Hỏi đáp liên quan: Hỏi: Vì sao bản đồ nhiệt không đủ để đánh giá một tiền vệ trung tâm? Đáp: Vì nó bỏ qua thời điểm chạy chỗ và khoảng trống để lại, hai yếu tố mà nhóm chỉ số không gian của VangBong.vn bổ sung cho dữ liệu sự kiện. Hỏi: Chỉ số PPDA có đáng tin khi đánh giá một hàng thủ lùi sâu? Đáp: Không, PPDA đếm hành động phòng ngự chứ không đếm vị trí, nên đội lùi sâu chặt bóng ở vạch 16m50 vẫn có thể bị đánh giá sai là thụ động. Hỏi: Vì sao không nên đòi hỏi cầu thủ vừa bình phục chấn thương phải chứng minh bản thân ở trận đầu tiên? Đáp: Vì nguy cơ tái chấn thương tăng mạnh khi cầu thủ bị buộc phải bứt tốc tối đa trước khi mô sẹo đạt độ bền cần thiết.

Minute 90+3, Kazan, June 27, 2026. On the commentary gantry I held a sheet of paper with a rectangle roughly 25 metres long, drawn from the halfway line towards Germany's penalty area. I had sketched it from the 58th minute, the moment Germany's back line began pushing beyond the halfway line in search of the goal they needed. When Kim Young-gwon received the ball inside the box and scored, I was not surprised. Three minutes later Son Heung-min ran into that exact space and made it 2-0.

The next morning, social media filled with heat maps. Germany covered almost the entire opposition half in red. The map was accurate down to every touch point. It also explained nothing about why the reigning world champions left Russia bottom of Group F, behind Mexico, Sweden and South Korea.

I retell this to open a less comfortable discussion: football analysis is measuring what is easy to measure and calling it understanding.

Data do not lie, but they know how to stay silent. The writer's job is to force them to speak by asking the right question. Ask the wrong one and you receive a beautiful map and a wrong conclusion.

A map that answers exactly one question

Heat maps in modern professional football are built in two ways. The first uses tracking data, logging every player's position dozens of times per second. The second, far more common on public platforms, uses event data, meaning the collection of touch points. Both answer exactly one question: where.

They do not answer when. They do not answer why. And they do not answer who left which space empty.

Empty space has no density. A vacant zone of some 40 square metres between two lines appears on a heat map as a meaningless blank, identical to every other blank on the pitch. Yet at that specific moment of the match, that blank is the most expensive area on the field. With the stands empty, I could hear the footsteps of space. That sounds strange, but rewatching behind-closed-doors matches during the pandemic years, I noticed I could hear defenders calling to each other, studs braking on grass, breathing. In that silence, space became the main character instead of the backdrop for a few individual touches.

I stress this technical detail for a professional reason: a heat map maps what happened to the ball. It does not map what happened to the shape. Those are different things, and the distance between them is where analytical errors live.

Take another good metric. PPDA, the passes an opponent is allowed per defensive action, measures pressing intensity. The lower the number, the more aggressively a team engages high up. I use it often and it is genuinely useful. But PPDA counts actions, not positions. A team that sits deep, concedes the ball in midfield and only tackles at the edge of its own box can carry a very high PPDA and still defend superbly. Read PPDA alone and you will call that team passive. You will be wrong, and you will be wrong with great confidence.

Expected goals has an even larger blind spot. It measures the quality of shots that were taken. It stays silent about the pass that was never played. When a striker makes a run that opens space and a teammate picks the safe option, xG records nothing at all; both players are invisible in the numbers. On the pitch, that run is what produced Kim Young-gwon's 90th-minute goal.

In Vietnam, and in South Korea where I work, the most quoted figure in post-match coverage is possession share. It is the cheapest and least informative metric of the lot. A team can hold 65 percent of the ball and create not a single clear chance. A team can hold 35 percent and score three. Possession describes duration, not the value of that duration.

In many technical meetings, the heat map has replaced the crystal ball. It is attractive, it is colourful, and it is abstract enough that any conclusion appears backed by evidence. That is exactly what makes it dangerous: a data-neutral tool that is very easy to misuse.

I spent years in Seoul working with event data and watching how sports content platforms in the region handle numbers. The pattern repeats: when data is thin, writers fill the gap with myth. Short of numbers, they build narratives of spirit, symbolism and character. I wrote that way for the first twenty years of my career. It is the part I am least proud of on rereading.

Three matches and one space

January 27, 2026, in Changzhou, snow on the pitch. Vietnam's under-23 side lost 2-1 to Uzbekistan in extra time, the decisive goal arriving in the 120th minute. The run before it, including two penalty shootouts against Iraq and Qatar, took Vietnamese football to a place it had never reached on the continental stage. The honourable defeat of 2026 handed me a winning formula. That formula fits into three principles.

First, the defensive block keeps its vertical distances tight enough that the front and back lines never separate. In the way I still teach it, the ideal gap between three lines sits between 8 and 12 metres when the ball is central, widening as the ball moves wide. Second, accept conceding the ball in midfield, where losing it costs least. Third, turn every counter into a race with a named destination, someone running and someone receiving, rather than a hopeful long ball.

Eleven months later, on December 15, 2026, at My Dinh Stadium, Vietnam beat Malaysia 1-0 in the second leg of the AFF Cup final, the goal arriving in the sixth minute and the aggregate finishing 3-2. Watch that match with the naked eye and you see a team sitting deep, conceding the ball and waiting. Watch it through a stats sheet and you see a team losing almost every control metric. Both views miss the central point: Vietnam controlled where the ball was when they lost it, rather than controlling the ball when they had it.

In my spatial model those are two different kinds of control. The first shows up on a stats sheet. The second only shows up when you slow the footage down and look at the ground a runner left behind him.

Move to the winter of 2026 and 2026, at the Southeast Asian championship. Vietnam beat Thailand 2-1 in the first leg at Viet Tri on January 2, 2026, then won 3-2 in Bangkok on January 5, 2026, for a 5-3 aggregate. In the second leg, Nguyen Xuan Son suffered a broken bone after a collision and left the pitch early. The tactically interesting part comes afterwards: Vietnam still scored three goals in Bangkok without their main striker.

Heat Maps Cannot See Space: How Football Analysis Fills Empty Data With Myth

To someone reading only heat maps, that is meaningless. To someone reading structure, it is entirely logical. Xuan Son was more than a scorer; he was the anchor that stretched the opposing back line, the magnet that opened space in the inside channels. When he left, Vietnam lost the magnet but kept the shape, and the linkages on the flanks shifted to exploit the space that had just opened centrally. A structure does not live on one individual. It lives on the space that individual creates, and when the individual disappears, that space changes owner.

I do not watch the player running; I watch the space he leaves behind. In all three cases above, the decisive factor sat there, and in all three it appeared on no mainstream stats sheet.

The blind spot is in the legs, not the model

Here I want to push the argument against myself.

This entire piece defends a way of reading matches through space and structure. Applied mechanically, it repeats the very error I just criticised: turning a model into truth, turning a model into religion. People swap the heat map for a space map and still believe everything can be read off a diagram.

The biggest blind spot in this trade is not in the model. It is in the legs. More precisely, it is in fitness load and injury status, which no model encodes.

Consider Xuan Son again. After recovery, the first question the media usually asks is whether he can still find form, whether he still deserves the number nine shirt. That framing creates pressure on a returning player to prove himself in his very first match. From a sports-medicine standpoint, it is counterproductive. A body after a serious injury needs time for scar tissue to reach maximum strength, and re-injury risk rises when a player is forced to sprint at full speed too early. When explosive speed becomes the acceptance test, every sprint becomes a gamble.

Meanwhile the data models keep grading him on output: goals, shots, touches in the box. Those numbers only return once the player has genuinely recovered. In the gap between, the model says the player is poor while sports medicine says the player is doing the right thing. The writer has to choose which side to trust.

I choose to trust the legs. I would argue that is also the more useful choice for the public, because it avoids a familiar spiral: overpraising a player when he scores, then accusing him of lost form when he is simply rebuilding.

There is one more professional layer worth stating plainly. When I receive a scouting report full of empty fields and the words not enough data, that report is worth more than one stuffed with unsourced opinion. An honest writer is one willing to publish a blank page. My trade has spent a great deal of ink covering such blanks with flowery language, and I think football audiences in Vietnam and Korea are sharp enough to notice.

Winning is a sequence of errors controlled better than the opponent's. No model explains that entire sequence, including the spatial model I have pursued for years. The analyst's job is to state clearly what is known, what is guessed, and what is left blank.

What to do this month

Next month, as the domestic season enters its points-accumulating phase, I propose a small test anyone can run in front of a screen.

Pick a match, pick a player, and for the first ten minutes watch only him while his team does not have the ball. Count how many times he moves away from the ball. Note the position of the three largest spaces he leaves behind each time he advances. Then open the match stats sheet and look for the numbers you just counted. Most of the time, you will not find them.

One hypothesis, and I mark it clearly as a hypothesis for testing rather than a conclusion: if you track how often a team switches from defence to attack within five seconds of winning the ball, and place that beside the possession figure from the same match, the transition measure will predict results better in most games. I will test this myself over the first half of next season and publish the outcome, including if it refutes me.

A football match contains roughly seventy transitions. Most produce no goal, no highlight and no place in any summary. But they produce the match. A good data writer is not the one who counts the most. A good writer is the one who chooses the right space to count.

Every move begins with an intention, even when that intention is accidental. My job is to find that intention inside a pile of numbers before they fall silent.