FIFA is an organization that is no stranger to alleged bias and corrupted processes (see Sepp Blatter, the Russian and Qatari World Cup selection controversies). But to what extent does bias permeate The Best FIFA Men’s Player award, the annual award is given to the most outstanding player in men’s football? To find out, myself and Michael Johnson produced the study: "Cultural similarity and impartiality on voting bias: The case of FIFA’s World’s Best Male Football Player Award".
This post was adapted from a piece produced by Steven Gilstead about our work.
The winner of this annual award is selected via a ranked-ballot vote by an international jury of representatives and media from each national men’s team. It should come as no surprise that a significant amount of bias permeates this process – but precisely how much, and along what lines is what myself and Michael Johnson set out to uncover in this study.
I noticed some years ago that the English captain, Steven Gerrard, included Luis Suarez, at the top of his 2013 ballot. This revelation was the catalyst for us to investigate potential bias in award voting. At the time, Suarez was arguably a top 10 player, but he certainly wasn’t the best player. So why did Gerrard vote this way?
Luckily, FIFA publishes the voting results, making it easy to study. Each member country has three voting members, the coach and captain from the men’s national team, and a member of the media who all vote on a short-list of 23 candidates. Each voting member submits a ranked ballot with three votes - worth five, three and one point(s) respectively – the player with the highest point total wins the award. We analyzed the voting records against the best players’ statistics for each season from 2010 to 2016, comparing the number of points a voter awarded to a player against the average points that same player received. We discovered a significant amount of voting bias in terms of three factors: cultural similarity, in-group bias, and impartiality of the voters’ countries.
The cultural similarity factor was measured using cultural distance, cultural clusters, and collectivism, while the in-group factors were measured using nationality, club, league, geography, ethnicity, religion and language. The impartiality factor for the voter’s country refers to whether or not that country’s government treats all individuals in an impartial manner regardless of race, ethnicity, family/political ties, or social standing.
We found that some of the strongest determinants of voting bias were voter-player similarity in cultural distance, cultural clusters, and collectivism, and that voters and players from neighbouring countries, and who’s capital cities were close, geographically speaking were more likely to be biased in their award voting. Interestingly, we also found that voters from more diverse countries were less likely to demonstrate voting bias.
Captains demonstrated the highest amount of voting bias, followed by coaches, then media. The captain’s bias can be explained by their close proximity to players and the strength of in-group bias. Back to the Gerrard/Suarez example, we found that that particular example can be explained by the in-group bias factors. when voters and players share nationality, club, league or are country neighbours, then voting bias is positively correlated. In the Gerrard/Suarez example, at the time, Gerrard and Suarez played for the same club, Liverpool F.C, which is the second-best team in the city of Liverpool (I'm biased).
Interestingly, if only media representatives voted during the seven years of our study, it would have resulted in different winners in 2010 (Sneijder instead of Messi) and 2013 (Ribery instead of Ronaldo).
In sporting awards, from soccer to basketball, to cricket, there is a whole range of data where you can empirically say, ‘this is the best player’” But what we're showing is that beyond the top three superstar outliers, there is massive levels of cultural bias in terms of where you actually rank and appear.
It seems like these biases are part of human nature, that are only really suppressed by strong institutions. On the other hand, homophily, or the tendency of individuals to associate similar others, is common in cultures across the world – after all, birds of a feather flock together.
