Simulation Of 100,000 World Cup Matches Predicts A Clear 2026 Winner, And Suggests The Bookies May Be Wrong

So, how do you go about predicting the results of football (soccer, if you must insist on using horrible words) events? 

There is, of course, the old tried and tested method of asking an octopus, gorilla, capybara, elephant, or orangutan. It doesn't really matter which animal you pick, as the outcome (assuming no pre-training or bias for certain flags) is going to be 50/50. That's generally not too helpful.

Instead, if you want a better prediction based somewhat on football performance, then the first step is to gather as much information as possible about the football teams involved. This is what a team of international researchers set out to do.

They collated aspects including historic data on all matches played over the last eight years, weighted so that more recent games are given more influence, as well as the performance of individual players both at international and club level, and their market value.

This was then incorporated alongside the predictions of 24 international bookmakers to represent the expert opinion. Finally, all the information was fed into a machine-learning algorithm, which combined the data and used it to make predictions on the number of goals each team might score. 

"Using the forecasts from the machine learning algorithm yields the predicted number of goals for both teams in each possible match," the researchers explain. "The predicted number of goals for the two teams in each match can then be plugged as expectations into two independent Poisson distributions, from which we can compute the probability that a certain match ends in a win, a draw, or a loss." 

"The same can be repeated in overtime, if necessary, and a coin flip is used to decide penalties, if needed."

Running 100,000 simulations enabled the statisticians to make predictions about how likely each football team is to win the overall competition. Compared to previous World Cup tournaments that the international team also worked their math on, the team says there is more uncertainty about who will win the trophy this time around. 

One particular confounding factor is that there are simply more teams in the 2026 competition, and this added an additional knockout round. By the time we even get to the knockout stage, the researchers notes, there are 495 potential permutations of teams.

Even so, the scientists were still able to make their predictions. They found that Spain was the most likely winner overall, with a 14.5 percent chance of winning the soccer (if you must insist on using that horrible word) competition. 

England fans may be equally happy and very unhappy to know that they came out joint second alongside France on 12.4 percent. Germany was ranked fourth most likely to win at 11.2 percent, followed by Portugal and Argentina.

While it's unsurprising to find these teams in the top spots, there were significant differences between these statistical predictions and those predicted by the gambling industry.

"Comparing our forecasts to those based only on the bookmakers odds, it is striking that Germany is ranked 4th, closely behind the three top teams, while it is only ranked 7th by many bookmakers," the researchers wrote. "Conversely, Brazil and Argentina are typically ranked higher by the bookmakers but perform worse in our machine-learning-calibrated simulation."

The scientists caution that all their predictions, including Jordan being the least likely team to win, are probabilistic and are, of course, not guaranteed. In previous years, they successfully predicted the outcome of the 2019 women's World Cup, but their top pick failed to win the 2022 men's competition or the 2023 women's competition. Though it should be said that the third pick in 2022 – Argentina – did go on to win.

"In any case, all of this means that the probabilistic forecasts leave a lot of room for surprises and excitement during the 2026 FIFA World Cup," the researchers concluded. "But what is absolutely certain is that we look forward to an entertaining tournament as football fans (much more than as professional forecasters)."