It’s Belgium to win the European Championships football tournament. But only after extra time against Italy in the final.
That, at least, is the curiously precise prediction from analysts at Goldman Sachs, based on a dataset of 6,000 games since 1980 and factors including current form and home advantage.
The analysis was led by Christian Schnittker and Sven Jari Stehn. “Speaking as Germans, we checked this prediction a number of times,” they said in a note released on Tuesday.
The Euro 2020 competition kicks off on Friday after a one-year delay due to Covid-19, with the winners from the 24 competing countries due to be crowned on July 11. Unusually, there is no host country – games are spread around 11 European countries, with the final stages due to be played at Wembley Stadium in London.
Goldman analysts built a statistical model to simulate the tournament. It is based on expected goals scored by each team, calculated by the strength of the squad (using the World Football Elo Rating); goals scored and conceded in recent games; home advantage; and “a tournament effect”.
The latter shows that some countries tend to outperform at tournaments compared to their rating – notably Germany, Croatia and the Netherlands. Meanwhile, the model calculates home advantage is worth 0.4 goals per game.
“Our framework predicts that Belgium will win the Euros for the first time in history, narrowly beating Italy on July 11,” the Goldman analysts predict.
The model gives Belgium the edge primarily due to its high Elo score, where it is ranked first amongst European nations, and only behind Brazil globally.
The Goldman analysts predict a close race between Belgium, Italy, Portugal and Spain, who are expected to be the semi-finalists.
Other predictions include Germany making it out of its difficult group ahead of France, but then losing to England at Wembley during the round of 16 (this was presumably the point that the two checked their spreadsheet was working). England then lose to Spain in the quarter-finals. The model also sees Denmark doing well, by winning its group and losing to Portugal in the quarter-finals.
But perhaps aware of the risk of getting egg on their faces – or probably trying to keep clients happy all over Europe – the analysts acknowledge “it is difficult to assess how much faith one should have in these predictions”, even with “sophisticated statistical techniques”. That's just due to the unpredictable nature of football, they say.
Indeed, the statistics may have been upended just hours before the report was finalised: Kevin de Bruyne, one of Belgium's star players, was injured on Saturday night in the last club game before players departed for international duty, putting his participation in doubt.