Empty Stadiums and the Illusion of Possession: Two Football Experiments That Refuse to Learn From Each Other

On July 1, 2026, at Luzhniki in Moscow, Spain held 74% of the ball against Russia in the World Cup round of 16. They made over a thousand passes, controlled the tempo for almost the entire 120 minutes, and left the tournament after a penalty shootout. That defeat did not come from individual technique. It came from a belief: that holding the ball means controlling the match.

I sat in front of a screen in Tokyo that night and opened the expected goals (xG) table. Spain generated a mere 0.9 xG. Russia scored from a penalty, then goalkeeper Igor Akinfeev saved two shootout attempts to send his team through. 0.9 xG was all a side with 74% possession could produce. I posted online: "Possession is an illusion; pressing and transitions are the truth." Forty-seven traditional journalists criticised me. Six hours later, when Spain were eliminated, that post received 2,300 shares.

But that was only the first half of the story. The second half only appeared two years later, when football walked into a laboratory no one could have rebuilt under normal conditions.

In May 2026, the Bundesliga returned after the pandemic stoppage. The stands were empty. No singing, no jeering, no rhythmic clapping after each pass. I began collecting data from 87 matches in that period, and found something many coaches did not want to hear: the home win rate fell from 43% to 31%, while the draw rate rose to 29%.

This statistic is far from trivial. It is evidence for a hypothesis I had carried for years: fans do not "support" in the way the media keeps telling us. Fans create pressure. Referees award a few more fouls to the home side, assistant referees raise the flag a little faster for the home side, away defenders clear the ball a little more safely. When the stands are empty, all of those "little bits" disappear, and home advantage dissolves with them.

The reaction from the professional world was the telling part. Some called it temporary. Others said the sample was too small. But for the first time in modern history, we had a variable removed entirely from the equation, and the results changed immediately. Such an experiment does not need a large sample to force people to rethink.

At the same time, a second experiment unfolded on a completely different stage: Tokyo 2026, the Olympics postponed to 2026 and staged in silence.

I was living in Tokyo when the Olympics took place without spectators. At the same moment, the League of Legends World Championship final reached roughly 73 million peak online viewers. The planet's biggest traditional sporting event had empty stands; an esports event had tens of millions watching through screens. There is no coincidence here. Two systems for distributing attention are running on two different tracks.

I declared in 2026 that esports is the modern Olympics. The International Olympic Committee laughed. Now they are chasing us.

But before anyone brands me a hater of traditional football, let me state clearly what I actually pursue. I do not care whether esports wins or the Olympics loses. I care about a narrower question: tactics decide outcomes more than atmosphere, and we have just been handed the proof.

At 61, I no longer have time for the polite football played on paper.

Let us return to World Cup 2026, but this time from the other side. Spain held 74% possession and generated 0.9 xG. Russia held 26% and generated about 0.4 xG. Looking at the structure of chances, Russia gathered almost all of their xG from set pieces and a penalty; Spain spread 0.9 xG across dozens of harmless moves outside the box. Possession has become a measure of patience rather than a measure of danger.

Let us talk about pressing, which I trust far more than possession. The PPDA metric — the number of opponent passes before your side attempts to win the ball — shows that a high-pressing team does not need much of the ball to create danger. Russia did not need possession to win. They only needed to break Spain's rhythm at certain moments, then turn one moment into a goal.

That is why I say Tiki-taka did not die because it was defeated; it died because it was trusted for too long. It did not collapse under a specific opponent. It collapsed under the very people who believed in it so much that they stopped asking questions. I call that death by being too safe.

The same thing is happening to home advantage. No one beat it with a tactical plan. It dissolved on its own when the variable that produced it — the crowd — disappeared. The empty stadiums of 2026 were the laboratory; only now do we see the final product.

Let me build this case with numbers, the way I always do. Three pillars.

Pillar one: home advantage is a variable, not a constant. The naive assumption of traditional football is that home always delivers roughly three to four points a season. Strip away the crowd, and that figure shrinks almost to zero. Home advantage is composed largely of psychological pressure on referees and opponents, not of pitch conditions or travel distance.

Pillar two: possession is a leading indicator, not a result indicator. 74% possession does not create goals. Only moves that enter dangerous areas create goals. The difference between a possession-heavy team and a dangerous team lies in the quality of the last ten passes, not in the quantity of the first thousand.

Pillar three: attention has migrated off the pitch. 73 million people watching a League of Legends final while some Olympic events played to empty stands goes beyond the story of 2026. It is a long-term signal about where young audiences pour their focus, and football is not immune.

In Japan, where I have worked for more than two decades, the J.League taught me something opposite to my English instincts: certainty has value. Japanese teams rehearse structure obsessively, hold positions like keeping a promise, and stay patient to the point of sometimes slowing themselves down. That patience builds disciplined sides. It can also become a trap: when an entire league is certain, no one dares break the structure to create an unpredictable moment. That unpredictable moment is precisely what destroys data models.

During a transfer window, these numbers become even more dangerous. Models that price young players rely on output data — goals, assists, minutes played — and they routinely overrate youthful potential while underrating dressing-room chemistry. A 19-year-old with shiny metrics can be worth three times a 28-year-old who already knows how to keep a squad standing through winter. No algorithm measures the latter, and that is why clubs still fall apart after A-rated transfer deals.

Now comes the part where I may be wrong. And I will say it plainly, because I do not write to please anyone.

First, the 2026 empty-stadium data came attached to an unrepeatable situation: a compressed schedule, degraded fitness, players out of rhythm. Part of the drop in home win rate could stem from big clubs — which usually play at home before large crowds — suffering a double penalty through scheduling and the loss of their crowd. I have not fully separated these two variables, and an honest analyst must admit it.

Second, the limits of the xG model. xG counts chance quality based on position and shot type, but it does not measure the psychological pressure on the shooter, nor the quality of the opposing goalkeeper in any specific moment. Spain's 0.9 xG does not tell the whole story of how they repeatedly entered Russia's box without shooting because they lacked a genuine striker. Data is a witness, not the only judge.

Third, and this matters most: I may be wrong to place esports on par with football in tactical importance. Esports enjoys its online viewership advantage because it was born on the internet; football was born on grass. Comparing two different distribution models and concluding one is more modern than the other is a common mistake. Esports has the faults of youth; the Olympics has the crimes of old age. One needs to learn how to win, the other needs to learn how to let go. Both could learn from each other instead of despising each other.

Football's real blind spot is not a shortage of data. It is that too many people believe numbers speak the truth on their own, when numbers only speak to what someone chose to count.

So where do we go from here? I have no fixed formula to hand you, and if I did, I would not trust it. What I propose is a habit: every time you see a team dominate possession, ask how much real danger they created; every time you see a packed stadium, ask what share of the advantage comes from atmosphere and what share comes from squad quality.

My prediction — verifiable — is this: within three years, leading clubs will begin measuring home advantage as a dynamic, season-by-season index rather than treating it as a default constant. Whoever does it first will gain a transfer edge and a match-preparation edge that the rest will only notice once it is too late.

And as for you, if you still think fans just sit there to cheer, try explaining why 87 matches without crowds quietly rewrote the definition of home.