Home Advantage Before and After the Pandemic: Lessons from 64 Matches Without Fans

In mid-May 2026, when the Bundesliga became the first major football league to restart after the pandemic, I sat in my rented room in Nha Trang watching the Ruhr derby between Borussia Dortmund and Schalke 04. Signal Iduna Park was empty. No cheers, no sea of yellow, just the sound of the ball and players' shouts echoing through the silence. Dortmund won 4-0, but what caught my attention wasn't the scoreline. It was a number buried deep in the statistics: the average home win rate in the Bundesliga that season had dropped to 31.3 percent, compared to 42.7 percent the previous season.

I had spent years analyzing football data, and this was the first time I witnessed a natural experiment of such magnitude. Without fans, home advantage was separated from psychosocial factors. I decided to collect and analyze all 64 Bundesliga matches played without crowds, from May to June 2026. My objective was clear: to quantify exactly where home advantage comes from, and whether it truly exists without the noise of the stands.

Before the pandemic, home advantage in European football was one of the most consistently documented effects in statistical history. In the 2026-2026 Bundesliga season, home teams averaged 1.57 points per game, compared to 1.19 for away teams. Average home goals were 1.58, while away teams scored only 1.34. This wasn't unique to Germany. In the 2026-2026 Premier League, home teams won 45.3 percent of matches, scoring an average of 1.52 goals per game. In La Liga, the corresponding figures were 47.1 percent and 1.49 goals. This advantage was explained by multiple factors: familiarity with the pitch, travel fatigue for away teams, unconscious referee bias, and most importantly, pressure from the crowd.

When COVID-19 forced leagues to play without fans, I realized this was an opportunity to isolate each factor. If home advantage disappeared entirely without fans, it would mean crowd psychology played the decisive role. If it only partially declined, other factors like pitch familiarity and travel would still matter. And if it didn't change at all, we would need to reconsider our entire analytical framework.

The results from the 64 matches I collected painted a far clearer picture than what analysts typically assumed. The home win rate dropped from 42.7 percent to 31.3 percent, a decline of more than 11 percentage points. Average home points dropped from 1.57 to 1.28. More notably, average home xG fell by 0.19 goals per match, while away xG remained virtually unchanged. This meant home teams lost their ability to create chances, rather than away teams playing better.

Away teams' PPDA also improved markedly. Before the pandemic, away teams typically had PPDA averages about 1.2 units higher than home teams, indicating poorer pressing. Without fans, this gap narrowed to 0.4 units. Borussia Dortmund, the team with the best PPDA in the 2026-2026 Bundesliga, improved this metric by an additional 0.8 units in away matches without crowds.

I cross-checked with data from other leagues. In the Premier League, when the season resumed in June 2026, the home win rate across the remaining 92 matches dropped to 33.7 percent, compared to 45.3 percent the previous season. In La Liga, it fell from 47.1 percent to 35.2 percent. In Serie A, a similar decline, from 44.8 percent to 32.6 percent. This trend was consistent across Europe, reinforcing the hypothesis that crowd noise is the key factor creating home advantage.

But there is one point I need to be cautious about. Data from the no-fan period was not influenced solely by the absence of crowds. These matches took place in special circumstances: compressed schedules to complete the season, players competing after a long break, and psychological pressure from the pandemic. Some subsequent studies indicated that player physical output declined during this period, with average high-speed running distance falling 8 to 12 percent compared to pre-pandemic levels. These are confounding variables I cannot entirely eliminate.

However, even accounting for confounding factors, the decline in home advantage was too large to be explained by physical fitness alone. If fitness were the primary cause, both home and away teams would be equally affected, and home advantage would not change. Reality showed home teams lost their advantage, while away teams improved performance. This can only be explained by the disappearance of the psychological home factor.

There is another angle that I believe analysts have overlooked. Without fans, refereeing decisions also changed. A University of Reading study published in 2026 analyzed 1,267 matches across 17 European national leagues and found that yellow cards for away teams decreased by 0.3 cards per match without fans, while cards for home teams increased slightly. Penalties for home teams also decreased by 0.15 per match. This is evidence that unconscious referee bias from crowds is a measurable factor.

This brings me to a more important conclusion about the nature of home advantage. It is not a fixed attribute of a stadium, but a psychosocial effect that can be removed. Home ground does not automatically confer advantage; the crowd is what creates that advantage. For decades, football analysts treated home advantage as a constant in their predictive models. Pandemic data showed this was a serious error.

Looking back at history, there were similar cases but on a smaller scale. In 2026, when Italy banned fans from certain stadiums after the Catania riots, matches without crowds in Serie A showed home win rates dropping by nearly 10 percentage points. However, the sample was too small to draw firm conclusions. The COVID-19 pandemic provided a large-scale natural experiment that no one wanted, but it answered a question analysts had debated for decades.

This lesson has direct implications for how we analyze football today. When a team performs well at home, we need to distinguish between their actual ability and the support from the crowd. When a team performs poorly away, we need to consider whether it's a tactical issue or simply psychological pressure. And when predicting outcomes, we need to adjust our models based on the degree of crowd influence at each stadium.

I write my blog from a rented room in Nha Trang; now probability takes me everywhere, but the lesson from empty stadiums remains one I carry in every analysis. The match ends, but the data remains. And sometimes, what we learn from the absence of fans is more important than what we see when they are present.

The gradual return of fans to stadiums during 2026-2026 provided additional data for verification. When some English stadiums allowed 10,000 fans back in May 2026, the home win rate immediately rose to an average of 40.2 percent, approaching pre-pandemic levels. This reinforces the conclusion that the relationship between crowds and home advantage is causal, not merely correlational.

However, I want to leave room for skepticism. The relationship between crowd size and the magnitude of home advantage may not be linear. A packed 80,000-seat stadium may create a different effect than a 20,000-seat venue. We don't yet have enough data to determine the optimal threshold. And in modern football, as teams become increasingly professional in psychological preparation, home advantage may continue to decline even with fans back.

The data from the no-fan period also raises questions about the future of predictive models. If home advantage can vary by context, models need to be updated more flexibly. Factors like crowd density, head-to-head history at specific grounds, and team psychological state need to be incorporated with adjustable weights.

For those who follow football and are interested in data analysis, this is the moment to revisit your assumptions. An empty stadium doesn't need fans; it needs an analyst willing to look. What we learned from 64 matches in silence could change how we understand this sport for years to come.