Home Is Not an Advantage: What 28 Matches Without Crowds Revealed

Home Is Not an Advantage: What 28 Matches Without Crowds Revealed

On May 16, 2026, the Bundesliga returned after nearly two months of suspension and became the first major league in the world to reopen after the COVID-19 lockdown. That night I sat in front of a screen in Nagoya, not to watch for entertainment but to count. Signal Iduna Park stood empty, and the shouts of the coaching staff rang out so clearly that I could hear the referee's footsteps through the broadcast microphone. I opened a new spreadsheet, named it empty_stands_28, and began logging every phase of play.

By the end of those first 28 matches, one number jumped off the sheet: home teams had won only 7 games, roughly 25 percent. In the same league, measured the same way, that figure had been 41 percent the previous season. A gap of 16 percentage points cannot be assigned to form or scheduling, because the quality of teams does not change in a matter of months. The equation had one variable removed: the crowd. Home was never a gift from fortune; it is a psychological and physical structure you can measure, and one that can vanish. Home has never been an advantage, only noise encoded into goals.

A natural experiment no one designed

In sports analysis, a clean natural experiment is as rare as a match with no margin of error. You cannot force teams to play at home without a crowd just to measure the advantage, because that would break the very nature of a sport played before stands. The pandemic of 2026 accidentally created exactly those conditions, across Europe, for months. I treated that period as a rare observational sample rather than a marketing disaster.

My method was painstaking. For each match I recorded the result, the goals, the xG of both sides, the PPDA — the number of passes an opponent completes per defensive action, a measure of pressing intensity — fouls, cards, added time, and the number of fouls the referee called on the home side versus the away side. I split the data into two groups: matches without crowds and matches with crowds earlier in the same season. Comparing the two is like comparing two worlds running in parallel.

What kept me patient was a narrow question: of all the components of home advantage, how much belongs to the crowd and how much to geography? The crowd can leave the stadium. Travel distance, familiarity with the pitch, the habit of sleeping at home — those cannot. If home advantage collapses when the stands are empty, I know most of it lives in the crowd, not on the map.

I do not remember matches; I remember their heat maps. And the heat map of an empty stadium always has a cold patch exactly where the crowd used to roar.

The evidence chain from 28 matches

The first number is the home win rate. In the first 28 matches without crowds, home teams won 7, drew 12, and lost 9. In the 2026/19 season, the Bundesliga home win rate was 41 percent. This is a fact you can cite and verify from German league history. A drop of 16 percentage points is larger than any random swing usually seen between consecutive seasons.

The second number is the distribution of goals over time. Matches without crowds tended to produce fewer first-half goals but more late goals. The most plausible explanation is tempo. Without the stands pressing them on, teams start more slowly, hold the ball more carefully, and wait for the opponent to err. By the end, as stamina drains, gaps open and goals arrive later.

The third number is the home team's PPDA. In matches with crowds, home teams usually press harder in the opening 15 minutes of the first half, with PPDA roughly 1 to 2 units lower than their own away figure. In matches without crowds, that gap nearly disappears. This fits a simple hypothesis: the stands are a power source, and when that source runs dry, early-match pressing intensity drops.

A concrete example is the Ruhr derby on May 16, 2026, when Dortmund hosted Schalke in an empty stadium and won 4-0. Erling Haaland opened the scoring, and what stood out was not the scoreline but the fact that Dortmund could not sustain its usual second-half pressing intensity. Without a crowd, the home side lost the drive to accelerate after taking the lead, and the match drifted toward a settled outcome instead of a chase.

I arrived at PPDA not through the Bundesliga but through a more painful night. In 2026, at 17, I sat in front of a TV in Nagoya watching the World Cup round of 16 between Japan and Belgium. Japan led 2-0 through goals from Genki Haraguchi and Takashi Inui. Then they lost 2-3, with Nacer Chadli settling it in the 90+4th minute. No one cared about the data; they blamed spirit.

I scraped every phase and took notes. In the first half, Japan posted a PPDA of 6.7 — an extremely high pressing level. In the second half, PPDA fell to 14.8, the midfield stopped closing down, and Belgium's xG rose from 0.7 to 1.9. The team had not lost its quality. It had lost its intensity, and lost it while leading. When Japan pushed up, I did not see a miracle; I saw the formula of collapse. Every pressing move is a testimony, every number a confession.

From then on I understood that a pressing midfield does not rely on lungs alone. It relies on the feeling of being backed. The stands do not run for the players, but they change a player's pain threshold. When sixty thousand people rise together after a challenge, a player runs for the next phase without stopping to calculate. In an empty stadium, the calculation returns, and the legs grow heavier.

Noise encoded: referees, tempo and psychology

The most interesting part of the dataset is not the goals but the whistle. In matches with crowds, referees tend to call more fouls against the away side than the home side, and added time tends to run longer when the home team is trailing. In matches without crowds, that gap narrows sharply. Referees did not become more honest. They simply stopped hearing the pressure.

The pandemic did not kill football; it stripped off the makeup. When the makeup vanished, you could see what was pure skill and what was a reward for context.

Match tempo reflected the same thing. Home teams in empty matches often increased possession time while reducing the number of decisive passes into the box. They controlled more but threatened less — a form of sterile control I always warn readers against. Check xG per pass into the final third, not just possession percentage. Beautiful control numbers can be the result of passing the ball around while no one stops you.

Psychological pressure vanished in a measurable way. At a home ground with a crowd, a player who wastes a corner hears a distinct sigh. In an empty stadium, that feedback does not exist, and players perform as if in a serious training session. That is why I like watching the closing minutes of crowdless matches: with no crowd to blame, only decisions and consequences remain.

Parallel to the Bundesliga, I followed a similar experiment on the grandest scale. The Tokyo Olympics, held in 2026, took place almost entirely without spectators at most venues. Sports that depend on the emotional rhythm of a crowd lost part of their identity, and the metrics that capture arousal dropped. This did not give me a complete number, but it reinforced a belief: the crowd is a technical variable, not just an emotional one.

What the data does not prove

Here I have to stop myself. A strong correlation does not equal absolute causation. A sample of 28 matches is small, and it sits in a peculiar context — disease, distancing, a compressed schedule, players performing without the psychological habits of adaptation. Any conclusion drawn from it must carry a margin of error.

Some parts of home advantage remain intact. Travel distance remains. Familiarity with the pitch and the weather remains. Sleeping at home instead of a hotel remains. Crowdless matches removed only the portion of the advantage that comes from people in the stands, not the whole concept of home.

I also do not want to turn a pandemic into a perfect experiment about sport. Players in 2026 performed in an unusual mental state, with a crowded calendar and eroded motivation. Part of the drop in the home win rate may come from those factors rather than from the stands alone. I write out the condition for it to hold: if home advantage returns close to its old level when crowds return, the crowd-as-cause hypothesis is reinforced. If it does not return, I must look for another variable.

This is the boundary between signal and noise. People need belief to place a bet; I need data to be certain. I do not want to build a conspiracy out of a shock, nor to praise a conclusion simply because it sounds sharp.

A signal for the next round

What I carry away from this dataset is not a slogan but a habit: whenever a team wins at home, I split the result into three parts — quality, geography, and crowd. In a regular season, when the calendar is dense and stamina is a life-or-death variable, the crowd is usually the most underrated part.

In the next round, watch the home team's PPDA in the first 15 minutes. If that number drops below its season average, the team is living on its stands. If it holds steady whether home or away, the team is living on its system. The data, so far, still says that part of home advantage dies the moment the stands empty — and that part is worth exactly the noise we traded to keep inside our heads.