A prediction model anyone can try has given Brazil a 25 per cent chance of winning the men’s World Cup – but the results are sobering for many other nations
November 18, 2022
Brazil is the most likely winner of the 2022 World Cup, according to a prediction model from the Alan Turing Institute in London. The publicly available model gives Brazil a 1 in 4 chance, while England’s chances are estimated at less than 1 in 10.
Many people, from bookmakers to bankers, have developed models trying to predict the winner of the 2022 men’s soccer World Cup in Qatar, but most of these are done behind closed doors.
Nick Barlow of the Alan Turing Institute and his colleagues have developed a model that people can run on their computers at home, with 1000 tournament runs taking 15 minutes on an average laptop.
“For us, it’s very important for most things that we do that we open source,” says Barlow. “We encourage people to get involved, use and contribute to our Code.”
When Barlow and his team ran the tournament 100,000 times with their model, they found that Brazil won 25 percent of the time, with their closest rivals being Belgium at 19 percent and Argentina at 13 percent.
The researchers adapted a common method used for matches in national leagues that gives teams a defense and offense score to predict matches, but they optimized their model to eliminate home field advantage, common to all teams in Qatar except the home country will miss. and taking into account differences in the strength of teams playing against each other in international friendlies.
They also tuned it to give more weight to the results of certain games, like semifinals and finals, and newer games, and ran the model on previous tournaments to see how well its predictions matched real-world results, and optimize it based on its performance.
Barlow’s model is consistent with a model by Achim Zeileis of the University of Innsbruck in Austria and his colleagues, who ran their algorithm on a supercomputer and found that Brazil were also the likely winners, estimating their odds at 15 percent.
But other models predict other winners. Insurance company Lloyd’s used the collective insurance value of a team’s players to predict that England will win by beating Brazil in the final. The same model correctly predicted that Germany would win the Men’s World Cup in 2014 and France in 2018.
While Belgium was predicted as the most likely winner by a model designed by Oxford University’s Matthew Penn and colleagues, which correctly predicted the winner of the 2020 men’s EURO as Italy and six of the eight quarter-finalists. This model assumes that goals scored and conceded are evenly distributed around an average value.
More on these topics: