This AI Simulated the 2018 World Cup 100,000 Times to Predict a Winner
A group of researchers used AI and machine learning to predict that Spain and Germany are the most likely winners of the 2018 World Cup in Russia.
During the 2010 World Cup, a psychic octopus named Paul picked the right team in 12 out of 14 games where he was asked to predict the winner, including the final. Eight years later, for the 2018 World Cup in Russia, we have a more scientific method to foresee the future: an AI that simulated the tournament 100,000 times.
According to to these simulations, Spain is the most likely winner, followed by Germany and Brazil.
A group of researchers from the German Technische Universitat of Dortmund and the Technical University in Munich, as well as the Ghent University in Belgium, came to this conclusion using machine learning based on several factors, such as the FIFA rankings, each country’s population and their gross domestic product (GDP), bookmakers’ odds, how many of the national team players play together in a club, the player’s average age, and how many Champions Leagues they’ve won.
Taking these factors into account, the researcher’s AI simulated the soccer tournament 100,000 times, calculating each team’s probabilities to pass to advance past the group stage, and then pass each following round all the way up to the final. The table below shows each team’s odds in the group stage.
And the next table shows the teams’ odds for the whole tournament, including the most likely winner. Spain is the most likely winner, slightly ahead of Germany, according to the researchers’ calculations.
Interestingly, however, the most likely course of the knockout stage has Germany taking home its second World Cup in a row.
The researchers explain that this is because Germany is most likely to beat Spain in the semifinals, but has a much harder road to get there, and thus more chances of getting knocked out.
Of course, take these predictions with a grain of salt. The researchers themselves say that given “the myriad of possible constellations this exact tournament course is still extremely unlikely.”
Plus, their machine learning model could not have predicted that Spain would fire and replace its coach two days before the team’s debut, causing an uproar in the country and within the team.
This article originally appeared on Motherboard.