“TO THE ‘PEOPLE’...who made the monkey noises: Shame on you. Shame on you”, posted Mario Balotelli, then a striker for Brescia, on Instagram in late 2019. He was referring to fans of Hellas Verona, a rival football club, who hurled racist abuse at him that exceeded even the typical taunts from overheated “ultras”. At one point during the match, play was suspended as a distraught Mr Balotelli punted the ball at jeering hecklers and threatened to walk off the pitch.

Mr Balotelli did manage to score a goal that day, though Brescia still lost 2-1. However, recent research by Fabrizio Colella, an economics graduate student at the University of Lausanne, suggests that the striker’s strong performance in the face of racist abuse was more an exception than a rule.

By forcing sports teams to play games without fans, the covid-19 pandemic created a compelling natural experiment that has unleashed a flurry of academic research. Many studies, including one published last year in The Economist, have sought to measure how much of home teams’ advantage comes from the presence of supporters. Mr Colella, however, made use of the new data to examine a question with broader social implications.

All footballers get heckled, but not all of them suffer the same types of insults. At least in European stadiums, fans single out non-white players for racially tinged attacks. Do such barbs sting more than the standard battery of non-racial abuse suffered by white players?

The pandemic allowed Mr Colella to find out. For each player in Serie A, the top tier of Italian football, he compiled individual performance scores in every match during the past two years ranging from zero to ten, which were generated by an algorithm used for fantasy-sports competitions. (These numbers take into account contributions in all aspects of the game, not just offensive statistics like goals or assists.) Next, he classified over 500 players as either white or non-white using the Fitzpatrick scale for human skin color, which is commonly used in dermatology research. Finally, he compared how each player fared, on average, in matches played in front of fans against their performances in empty stadiums. Although fans may affect results in many ways, only racist chants could plausibly have a different impact on white players than on non-white ones.

The results were striking. On average, white players scored slightly worse without fans than they did in packed stadiums. In contrast, non-white footballers’ performances improved to a statistically significant degree when fans were absent, by an average of 1.2%. Mr Colella built a mathematical model that tried to account for these differences using other variables, such as players’ nationalities and teams’ overall quality. However, none of these controls eliminated the impact of skin colour. The effect was greater for the darkest-skinned players than for brown- or olive-skinned ones.

The quadrennial UEFA tournament for European national teams begins tomorrow, and will be the most prestigious international team sporting competition to occur since the start of the pandemic. Any hopes that the long hiatus of in-person events might have led to better behaviour from sport-starved fans have been dashed: just this week, Romelu Lukaku, a striker for the Belgian national team, said that “racism in football right now is at [an] all-time high.” Perhaps the only thing more dispiriting than the probable return of racist abuse to top-flight international football is the notion that it might actually work.