Germany 0-2 South Korea, Kazan 2026: The Data Story the Scoreline Never Told

The Kazan Night, and a Notebook Opened Beside the Screen

On June 27, 2026, in Kazan, Germany walked onto the pitch as defending World Cup champions for their final group-stage match against South Korea. I sat in front of the screen, seventeen years old, a notebook open beside me, recording every shot from both teams in the most manual way possible: position, distance, angle, the situation that produced the attempt. I had never taken a course in data analysis at that point. I had only a naive belief that if I wrote enough down, the match would tell me something the scoreboard never would.

Germany held 74 percent of possession. Germany took fourteen shots. Germany strung together the familiar short passes, controlled the tempo, and pinned their opponent back for long stretches. If you looked only at those basic statistics, the story seemed predetermined: a great team imposing its game on a supposedly weaker one, with victory merely a matter of time.

But when I summed the expected-goal value of every shot I estimated by hand, the number that emerged looked entirely different. Germany reached roughly 1.2 xG from fourteen attempts. South Korea reached roughly 0.8 xG from just three attempts. The gap in possession was enormous, but the gap in chance quality had been squeezed alarmingly thin. A team holding the ball for nearly three-quarters of the match had failed to create meaningfully more dangerous chances than an opponent that had simply surrendered possession.

In the third minute of stoppage time, Kim Young-gwon put the ball in the net after a scramble in front of goal. VAR confirmed the goal. Minutes later, Son Heung-min ran half the pitch as Neuer had advanced to join the attack, and sealed the 2-0 win. That night, the defending champions left the tournament.

I sat for a long time after turning off the screen. What I realized was not "football is wonderfully unpredictable" or "the strong team lost through bad luck." What I realized was this: data does not lie, but it tells a different story from the one on the newspaper front page. The scoreboard said 0-2. The possession board said 74-26. But my notebook, with its clumsy self-calculated numbers, recorded a truth that Germany had not dominated as dangerously as their appearance suggested. I entered football on the night Germany collapsed against South Korea. From that night on, I began writing a data diary after every match, recording every metric by hand, and promising myself I would never make an emotional judgment without verifying it with numbers, however surprising the result might be.

Why I Trust Numbers, and How Far That Trust Goes

Before diving into the analysis, I want to make one thing clear: I am not someone who believes data can explain everything. I believe data is the only language through which a match will reveal its truth, but it is a language that must be translated, not a verdict that must be executed.

There is a vast distance between "reading data" and "understanding data." Someone can hand you a table with hundreds of metrics and you can still understand nothing, because data never stands alone. A number only becomes meaningful when placed in context: who the opponent is, where a player stands in his career, what structure the team operates within, what stage the match has reached. Detaching a number from its context is like reading a sentence with all the connecting words removed: you still recognize each word, but you do not understand the story.

Based on my experience following matches, I have noticed that the football public, and even part of the professional world, tends to make two symmetrical mistakes. The first mistake is ignoring data entirely and making judgments based on feeling, memory of a few highlight moments, or bias about a team's reputation. The second mistake is worshipping data blindly, turning numbers into absolute truth, and forgetting that behind every number is a human decision — what to measure, what to define as a dangerous chance, how to classify a phase of play.

Both mistakes are equally dangerous. Those who ignore data will be surprised by what they could have seen coming. Those who worship data will be deceived by what data cannot see. My position lies in the middle, and that is why I call myself a data storyteller rather than a data engineer. Engineers build systems. Storytellers use those systems to tell a story that means something to people.

So in this article, I do not want to hurl a barrage of numbers at you and treat that as a conclusion. I want to walk with you through each stage of my journey: from the clumsy Kazan night, to the pandemic season when stadiums stood empty, to Qatar when I learned to read pressing pressure, to the Euros and my work as a club consultant. At each stage, I will show what the data revealed to me, and how the data nearly deceived me.

A Match as a Reading Comprehension Test

Let us return to Kazan once more, but more slowly this time. When I added up every Germany shot, I noticed something that later became a working principle: the quantity of shots does not tell you the quality of the chances. Fourteen attempts can sound terrifying, but if most of them come from outside the box, from narrow angles, from situations where the defense has already reorganized, then their true value is far lower than the feeling the number suggests.

Germany was a team that controlled the ball in a way that forced opponents to shrink back. But possession, in the end, is only a means. The real question is: how efficiently was that possession converted into dangerous chances? When I calculated Germany's xG, I saw a team that held the ball a lot but created chances with unusually low efficiency. They circulated the ball into the final third very well, but from that final third into the danger zone, they hit a wall.

South Korea had built that wall in an organized way. Not through luck, not through spirit alone, but through a defensive structure that dropped deep, kept the distances between lines, and accepted surrendering the ball in areas where keeping it brought the opponent no benefit. They let Germany play sideways passes, safe circulation in midfield, and only pressed hard when the ball entered the danger zone.

This is what I want you to remember throughout this article: a defense can control a match precisely by surrendering possession. That was the first paradox I learned at Kazan, and it runs entirely against the common intuition that the team with more possession is the team in command.

When Kim Young-gwon scored in stoppage time, many called it a shock. But to my notebook, it was not a shock. It was the outcome of a match in which the data had distributed risk asymmetrically: Germany carried the risk of a relentless attack with low efficiency, South Korea carried the risk of a dense defense but owned those fleeting moments in the danger zone. When Neuer advanced in the closing minutes, South Korea had what they needed: the space behind a team whose shape had lost its balance.

That night in Kazan, I learned that reputation never appears in a dataset. My data table had no column reading "world champion." It had only columns for shot position, chance danger, and the structure of the move. And in those columns, Germany was no giant. Germany was merely a team struggling to turn control into chances, facing a team that knew exactly what it wanted.

The Pandemic Season: When Stadiums Emptied and Data Became the Only Echo

In 2026, the pandemic closed stadiums around the world. I was nineteen, a first-year student in Seoul, and like so many others I lost something I had never imagined could be lost: the noise of the stands. But instead of treating it as a tragedy, I saw a laboratory. For the first time in modern history, football was played in conditions where one enormous variable — crowd pressure — had been almost entirely removed.

I collected data from 380 K-League matches and 500 matches across five top European leagues. I compared home-win rates in the period with crowds and the period without. The result made me pause for a long time: the home-win rate fell from 45.2 percent to 37.8 percent without spectators.

That number, standing alone, was already interesting. But what pushed me to dig deeper was the next question: who suffered the decline most? When I broke the data down by team quality, an unexpected picture emerged. Weaker teams benefited more from the loss of crowd pressure. In other words, the noise in the stands was an asymmetric advantage, one that helped strong home teams more than weak home teams.

When I analyzed further by match phase, I found that most of the difference concentrated in the first fifteen minutes of each half. That is the phase when home teams typically harness crowd noise to create early psychological pressure, pushing opponents into a defensive shell, and sometimes scoring from that pressure itself. When the stands fell silent, this phase became more neutral, and weaker teams gained precious time to organize.

When the stadiums emptied, data became the only echo left behind. And that echo told me something ordinary football commentary does not say: that the atmosphere in the stands is not merely a cultural or emotional phenomenon, but a variable that can be measured, estimated, and shown to have a systematic effect on results. The pandemic taught me that the air in the stands is also a metric.

I wrote my first analytical piece during this period, published on a personal blog, combining statistical data with the social context of the pandemic. The piece was later widely shared by a Korean sports outlet. That was when I understood that my voice could not simply be the voice of someone reading numbers. It had to be the voice of someone who understands that behind every number is a person, an era, a context.

But the pandemic lesson had a deeper layer that took me several more years to fully grasp. The fall in home-win rates without crowds is a clear correlation. But correlation is not causation. How much of that decline came from the crowd itself, and how much came from accompanying factors the pandemic brought? A compressed schedule, changed substitution rules, physical shortages from interrupted preparation, or simply a different mentality among players who knew no one was watching?

I raise this not to deny what I found. I still believe the stands have real influence. I raise it to emphasize a working principle I will return to often: observational data shows you what goes with what, but it does not automatically tell you what causes what. The analyst's responsibility is to translate correlation into a hypothesis, then seek evidence to test it, rather than leaping straight from correlation to conclusion.

Qatar 2026 and a Metric I Trust Enough to Call "Burning Time"

By the 2026 World Cup in Qatar, I was twenty-one, and I had a far clearer method than on the Kazan night. South Korea's match against Uruguay ended 0-0, and if you looked only at the score, you might easily conclude it was a tense but dull affair, a draw that said little.

I collected PPDA data from open sources. PPDA measures the intensity of pressing, calculated as the number of opponent passes allowed within the pressing zone divided by the number of defensive actions in that zone. The lower the figure, the more aggressively and consistently a team presses. South Korea recorded a PPDA of 7.2, a figure on par with the top European national teams at that tournament.

That figure, alone, does not tell the whole story. But combined with the team's shape and the match situation, one thing became clear: South Korea had not played passive defense against a physically tough South American side. They actively pressed high, forcing Uruguay to pass under constant pressure, and controlled space in a way the 0-0 completely concealed.

I wrote a roughly 2,000-word analysis arguing that the draw did not reflect South Korea's superiority in controlling space against an opponent rated higher in physicality and experience. The piece caught the attention of an editor at a major sports channel in Seoul, and from there I gained regular collaboration.

But what I want to dwell on is not a personal achievement. What I want to dwell on is an understanding of pressing I built from that very data. South Korea's pressing in Qatar was not burning energy — it was burning the opponent's time.

This is an idea that needs careful explanation, because the ordinary way of talking about pressing is often emotional: people speak of "fighting spirit," of "determination," of "tireless energy." All of these are partly true, but they conceal the real mechanism of pressing.

When a team presses high in an organized way, what it actually does is not exhaust the opponent's body but exhaust the opponent's processing time. Every second cut from the time a player has to receive, scan, and decide is a second taken from the quality of the next pass. A good pressing team does not necessarily make the opponent more tired; it makes the opponent decide faster, and therefore less accurately.

Viewed through this frame, pressing is not a purely energy-consuming action. It is an investment. The pressing team spends energy and position to buy time and space. The right question is not "does this team press," but "what is this team buying with the energy it spends, and at what price."

I began analyzing matches this way as a cost-balance sheet. Every run a player makes is an expense. Every position taken is an investment. Every press is a transaction whose return is uncertain. When I look at football this way, I see things the statistics table alone cannot show: that some teams run a great deal to no purpose, and some teams run less but place the opponent in difficulty with every step.

Euro 2026 and the Problem of Tactical Transfer

In 2026, after graduating, I joined a professional club in Seoul as a data consultant. This was when I moved from observing and writing about data to using it to change other people's decisions. That shift was bigger than I had imagined.

Observing data and recommending action are two different jobs. When I wrote blogs, I could present a complex picture, acknowledge uncertainties, and let readers reflect for themselves. When I worked with a head coach, I had only a few minutes to present, and everything I said had to lead to a concrete decision. No one has time for long-winded analysis before a training session.

During Euro 2026, I studied how Spain used Lamine Yamal to exploit the right flank. This very young player created an average of 2.1 key passes and 4.3 touches inside the box per match. Those numbers, to me, were not only about individual talent. They were about a structure: how the team created space for a wide player, how they moved to open passing lanes into the box for him, and how they used him as an anchor to stretch the opponent's back line.

From there, I proposed the coaching staff trial a similar tactic with an eighteen-year-old talent at the club playing as a right-sided attacking midfielder. My proposal was not "let us copy Spain." Copying is the fastest route to failure. My proposal was to extract the structural principle behind how they create chances for a wide player, then test whether that principle fits our current squad, fitness, and style.

This is an important lesson about tactical transfer. When you see a big team succeed with an idea, what you should take is not the outward form of that idea but the mechanism that produces its effect. You must ask: what in that structure is actually generating the advantage? Is it the player's position, the teammates' support, the tempo of circulation, or the threat from another player on the opposite side that forces the back line to spread thin? If you cannot separate the mechanism from the form, you will copy an empty shell.

And this was also when I realized something I had long suspected: data in a professional environment is not meant to prove who is right or wrong. It is meant to give the decision-maker a second view, another lens through which to test their own intuition. A good coach usually already has correct intuition about the match. My job is not to replace that intuition with numbers, but to provide evidence that strengthens or challenges it, and to point out the blind spots intuition cannot see.

A Contrarian Angle: When Data Becomes Fortune-Telling

At this point I must turn to critique something I consider the greatest danger of modern football data analysis: the heat map.

The heat map has become such a common presentation tool that it is almost indispensable in any analysis. People overlay a pitch with patches of blazing red and cold blue, point at them, and declare something about how a team operates. But I would argue the heat map has become a new form of fortune-telling. It creates a sense of science, a sense that we are seeing objective truth. But in reality, it often conceals a player's true role within the tactical system.

The reason is simple: a heat map tells you where a player was, not what the player did there. A player can appear in a blazing red zone without generating any value; another can appear in a pale blue zone yet be the decisive link in a specific move. The heat map collapses all actions onto a single scale — position — and erases all differences in quality, timing, and context.

Worse, the heat map collapses an entire match into one static picture. It does not distinguish minute ten from minute ninety. It does not distinguish when the team is leading from when it is trailing. It does not distinguish live play from dead-ball situations. It turns a match, which is a sequence of decisions ordered in time, into a cloud of positions without time.

This does not mean the heat map is useless. It can be a starting point for asking questions. But it should never be the endpoint of analysis. When I see a heat map, I do not ask "where did this player operate." I ask "why was this player in those positions, at those times, and did the tactical system place them there or did they choose it themselves." The answer to the second question is what is worth discussing.

I include this critique not to appear skeptical of every data tool. I include it because I believe a data storyteller has a duty to distinguish between tools that aid understanding and tools that create the illusion of understanding. In an era when anyone can access an enormous volume of data, the greatest risk is not a lack of data but too much data presented without being understood.

The Blind Spots of a Data Person

If I only criticized others, I would be a hypocrite. Professional data people have their own blind spots, and I want to confess a few of them, because the very collisions with them have shaped how I write today.

The first blind spot is the temptation to use data to win an argument. When you have all the metrics in hand, it is easy to select the numbers that support the view you already held and present them as objective evidence. But that betrays the very nature of the work. A tactical translator should not use data to win debates; they should use it to illuminate the truth, even when that truth forces them to change their initial view. So I always ask myself one question before publishing any conclusion: "What does this number change?" If the answer is "nothing, it just confirms what I already believed," then I know I have not analyzed deeply enough.

The second blind spot is the temptation to over-contextualize. I have said that context is essential, and I believe it. But there is a fragile line between contextualizing to illuminate and contextualizing to avoid having to draw a conclusion. If every time I offer a judgment I add ten caveats about "but it depends on," then in the end I have said nothing. Readers need something to lean on, a claim to hold onto, even if they may later dispute it. My self-correction is to write a single thesis sentence before starting, then ask myself whether every paragraph serves that thesis.

The third blind spot is conflict avoidance. As someone with a tendency to want to reconcile, to see the right side of every party, I once wrote analyses that ended with "both are right." But football does not run on fence-sitting conclusions, and readers sense that hesitation immediately. I learned that caution does not mean vagueness. I can acknowledge that data does not give me the whole truth and at the same time still deliver a decisive judgment about what the data really indicates. So I try to end with a calm, decisive sentence rather than a hovering question.

These blind spots do not disappear. They simply become more visible over time, and I learn to work with them rather than pretend they do not exist.

The Dual Lens of China and Korea, and the Question of Neutrality

There is an aspect of my work I have never spoken about publicly, and I think this is the right moment. I am a Chinese person living and working in South Korea. I carry two frames of reference for football, and this is both an advantage and a trap.

Korean football, in my observation, operates on a logic of organization, time, and discipline. Tempo is distributed, roles are assigned, and strength comes from everyone knowing what to do in each moment. Chinese football, at its deepest level, operates on a logic of individual impulse. Moments of individual breakthrough carry great value, and sometimes collective strength is mobilized behind an outstanding individual.

Both logics have strengths and weaknesses. Organized football can sacrifice creativity for stability. Impulse football can sacrifice stability for explosive moments. The most interesting question, to me, is not which logic is superior, but whether a hybrid solution can be built that exploits the strengths of both: a structure disciplined enough to withstand pressure, yet flexible enough to let individual impulses erupt in decisive moments.

But I must confess there is a subtle trap in the position of carrying two cultures. That trap is hierarchical language. When you stand between two football nations, it is easy to slip into comparisons that rank, saying "this football nation is clearly superior to that one," followed by many other judgmental ways of speaking. I once wrote a few such sentences, and I regret them.

The reason is not merely a matter of politeness or sensitivity. The reason is that hierarchical language destroys the role of a neutral tactical translator. The moment you declare one football nation inherently superior, you stop analyzing. You start ranking. And when you rank, you no longer see what both sides do well and do poorly, because the conclusion has been decided in advance.

My position, after many years, is not to rank football nations. I only analyze models. Korean football and Chinese football are two different operating models, each with its own problems and its own solutions. My job is to understand both deeply enough to move ideas from one side to the other when that is useful, and to remain honest with both.

Why the Scoreline Always Arrives After the Data

Let me return to the central question of this article: why does a data storyteller believe data can see what the score has not yet said?

My answer has three layers.

The first layer is time. A match unfolds over ninety minutes, but the score is only a final moment. For most of that time, no goals are scored at all, yet countless decisions are made: decisions to hold position, to choose a pass, to press or drop back. Data records those decisions, and therefore it can record a trend before that trend crystallizes into a goal. The score is the consequence. The trend is the process. And the process always comes before the consequence.

The second layer is asymmetry. Football is a game in which results do not always correspond proportionally to chance quality. A team can create better chances and still lose. This makes the score an unreliable measure in a single match. Data, by measuring chance quality rather than only outcomes, can show that a team played better than the score suggests, or the opposite. This is not a way of saying "the losing team deserved to win." It is an acknowledgment that football has a large random element, and data helps us distinguish quality from luck.

The third layer is prediction. Because data records trends, it can be used to predict what is likely to happen next. If a team has created many high-quality chances over several consecutive matches without scoring, the likelihood they will score in the next match is high, unless something structural changes. This is what I always remember when writing: do not only look at what happened, look at what is forming.

I do not believe in beautiful goals. I do not believe in beautiful goals. I believe in correct goals. A correct goal is one that comes from a process, from a trend, from a structure that created the chance. A beautiful goal can be luck. A correct goal is the consequence of something repeatable.

The Regular Season: Reading Signals Beneath the Table

We are in a regular season, and this is the moment to adjust how we read data. A regular season is a long stretch of time, and that means short-term fluctuations should not be read as long-term trends. One defeat tells no story about a team. Neither does a three-win streak.

Based on my experience following matches, I have found that the regular season demands unusual patience when reading data. The real story does not lie in the table but in the currents beneath it: changes in pressing intensity, physical wear across consecutive matches, systematic refereeing controversies, and the tactical adjustments coaches make only after accumulating enough data about their own teams.

When opening a season analysis, I usually start with a tactical signal. For example, I will point out that over the last three matches a team's PPDA has fallen notably, meaning they are pressing more aggressively. Or I will point out that the running gap between a team's two halves is widening, suggesting a fitness problem that may become larger in the coming rounds.

What I avoid is turning these signals into absolute claims. My caution about absolute claims is not innate temperament. It is the result of having witnessed a great team collapse, and of understanding that football changes faster than we think. A new tactical trend that appears after a single successful match or season can be an illusion of a small sample. Before calling something a trend, I want to see it repeat across different contexts, against different types of opponents, and with different outcomes.

Correlation Is Not Causation, and the Times Data Nearly Fooled Me

This is the part I consider most important for anyone reading data analysis, and also the part I have struggled with most in my career.

I once fell into the trap of believing that because two variables move together, one causes the other. I once saw that high-pressing teams tend to have better results, and nearly concluded that high pressing was the cause of success. But then I realized something simple: strong teams often have the conditions to press high. They have faster players, better fitness, and higher organizational ability. So is pressing the cause of success, or merely a symptom of owning a better squad?

The answer is not simply which comes first. The truth is often both, in a causal loop with feedback. But recognizing that complexity, rather than choosing a clean and wrong conclusion, is what separates the serious analyst from the one merely seeking numbers to support a predetermined view.

I have also nearly been fooled by data when analyzing a player's performance. Goals are an attractive measure, but they depend on many factors beyond a player's ability: the quality of teammates, how the team plays, the position in which the player is used. A good striker in a weak team may score few goals; an average striker in a strong team may score many. If I looked only at goals, I would misjudge both. This is why I always seek to measure what a player controls, not only what the player benefits from.

These lessons taught me to write differently. When I offer a conclusion, I try to indicate what is evidence and what is inference. I try to distinguish between what data tells me, what data merely suggests, and what I believe from experience but lack sufficient evidence to prove. Readers deserve to know which level of certainty I am standing on.

This may seem to conflict with readers' desire for decisive conclusions. But I believe transparency about degrees of certainty does not weaken analysis. It strengthens its credibility. Someone who tells you "I am certain of this" in every sentence is less trustworthy than someone who tells you "this I am sure of, that I only suggest."

Data, Discipline, and the Silence of the Observer

There is a question I often receive: why do this backstage work? Why not write flashy commentary, why not stand in front of the camera, why spend hours facing data tables almost no one sees?

My answer will probably disappoint many, because it is not exciting. I do this work because I believe what happens under the spotlight is only the tip of the iceberg. The match you watch on television was shaped long before the ball rolled: in training sessions, in tactical meetings, in the coaching staff's decisions based on data the audience never accesses. I want to be near those decisions. I want to understand the mechanism behind what everyone sees.

There is a stillness in this work that I love. When the match ends, when the stands go dark, the data remains. Every number I read is a confession the match never spoke aloud. A match never tells you on its own where it was decided, which team lost its balance at which moment, which chance was missed because of a wrong positional decision in the thirtieth minute. Data says those things, but only to those who sit still and listen.

I realize this resembles how a quiet observer works. They do not try to make noise. They do not try to impress. They only try to see correctly. And in a world where tactical noise, the noise of emotional commentary and unsupported claims, grows ever louder, that silence is a form of resistance.

Looking Back Four Years at a Single Number

When I look back at the four years from Kazan to the writing of this article, I see a clear shift in how I think about data.

At seventeen, I believed data was a discoverable truth. That if I calculated correctly, I would know how the match really unfolded. At nineteen, during the pandemic season, I began to believe data was a measurable truth, but one that must be placed in context. At twenty-one, in Qatar, I began to believe data was a story that could be told in different ways, and my job was to choose the telling that best served the truth. At twenty-three, as a consultant, I began to believe data was a tool to change decisions, and its value is measured by whether it produces better decisions.

And now, writing these lines, I believe data is a language. Not the truth, not ultimate reality, but a language for approaching the truth. A language with its own grammar, its own limits, with things it can express and things it cannot. My job, and perhaps the job of anyone in this field, is to learn that language well enough to express the nuances ordinary language misses, and to be humble enough to acknowledge what it cannot express.

I still keep the notebook from the Kazan night. It is no longer used for calculations, but I keep it as a reminder. It reminds me that I entered football through clumsiness, through a naive belief that if I wrote enough, everything would become clear. Years later, I know everything never becomes entirely clear. That is the nature of football, and perhaps of everything worth understanding.

Signals for the Next Round

As we enter the coming rounds of the season, there are a few signals I will be tracking, and perhaps you should too.

I will track changes in teams' pressing intensity in the middle of the season. This is the stage when fitness begins to wear down, and teams that rely too heavily on high pressing often start to show signs of decline. Based on my experience following matches, PPDA usually rises at this stage for teams without sufficient squad depth, and that is an early sign of a problem that can affect the title race.

I will track the gap between xG and actual goals for teams. Teams that score less than their xG over a long period tend to regress to the mean, meaning they will score more in the coming stretch. Conversely, teams that clearly outscore their xG may be benefiting from luck or from an outstanding opposing goalkeeper, and that advantage is unlikely to last.

I will track the tactical changes coaches make only after accumulating enough data about their own teams. These are often the most important changes in a season, because they are made based on evidence rather than a knee-jerk reaction to public pressure.

And I will track the growth of young players. I believe the best way to judge a young player is to observe what they control, not only what they are lucky to receive. A young player who creates chances his teammates fail to convert may be better than a young player who scores many goals from favorable situations.

But above all, what I will track is what the table does not say. Because I have learned, from the Kazan night and many nights since, that the most important things in football are often the things you must go looking for, not the things that appear before your eyes.

A team's pressing, in the end, is not the impulsive action of strong legs. It is a decision about how to allocate resources, about accepting energy expenditure today in exchange for time and space tomorrow. And if I had to choose one sentence for you to carry after finishing this article, it would be this: in modern football, a team does not win with its legs, but with how it chooses to spend them.