South Korea national team players look dejected after the Group A second match of the 2026 North and Central American World Cup group stage between South Korea and Mexico ends 0-1 at Guadalajara Stadium in Mexico on the 18th (local time)./Courtesy of News1

Korea's national soccer team lost 0-1 to co-host Mexico in Group A's second match of the 2026 North and Central America World Cup at Guadalajara Stadium in Mexico on the 19th. After beating the Czech Republic 2-1 in the first match, Korea is now at one win and one loss, and whether it advances from the group stage will hinge on the final match against South Africa.

Meanwhile, an international research team that included Achim Zeileis, a statistics professor at the University of Innsbruck in Austria, compiled the tournament's trajectory by bringing together Machine Learning assessments of participating teams' strength, player value, recent form, and betting market evaluations, and released the results on The Conversation on the 9th (local time).

The researchers gathered data such as results of A matches over the past eight years, international betting firms' odds, players' goal contributions, transfer market value, and the Fédération Internationale de Football Association (FIFA) rankings to estimate each team's strength.

They then used a Machine Learning technique called "random forest" to predict World Cup match outcomes. In a random forest, multiple decision models each predict results, which are then combined to produce final probabilities. Reflecting the official bracket and FIFA regulations, the researchers simulated the entire World Cup schedule 100,000 times.

Before the tournament, Korea's probability of reaching the round of 32 was 83.68%, ranking about 16th among the 48 participating countries. In other words, the model viewed Korea's chances of clearing the group stage as relatively high.

Mexico, also in Group A, had an 83.96% chance of advancing to the round of 32, 0.28 percentage points higher than Korea. While the gap between Korea and Mexico was not large on the prediction table, Mexico won the head-to-head and secured a round of 32 berth first. The Czech Republic's probability of reaching the round of 32 was 72.53%, and South Africa's was 31.73%.

This World Cup differs from previous tournaments. With the field expanding from 32 to 48 teams, the top two from each of the 12 groups (24 teams) plus the eight best third-place teams first reach the round of 32. Teams then must win in the round of 32 to advance to the round of 16.

The researchers put Korea's probability of reaching the round of 16 at 41.31%. That indicates the chance of clearing the group stage and then winning in the round of 32 to make the round of 16. It ranked about 17th among all 48 teams, assessing Korea as an upper-mid-tier team capable of aiming for the round of 16. Mexico's probability of reaching the round of 16 was 42.66%, 1.35 percentage points higher than Korea's, while the Czech Republic's was 31.92% and South Africa's was 8.87%.

Korea's probability of reaching the quarterfinals was 15.34%, and of reaching the semifinals was 4.83%. The probability of reaching the final was 1.60%, and the probability of winning the title was 0.48%. Korea was assessed as a team with a solid chance of making the round of 16, but with probabilities dropping quickly deeper into the tournament.

Meanwhile, the No. 1 favorite to win the World Cup, according to the researchers, was Spain at 14.5%. England and France followed at 12.4% each, and Germany at 11.2%. Portugal was at 8.9%, and Argentina at 8.2%, placing them among the title contenders.

Zeileis said, "We correctly picked the United States to win the 2019 Women's World Cup, but we did not rank Argentina, the 2022 Men's World Cup winner, or Spain, the 2023 Women's World Cup winner, as the top favorite," adding, "Predictions are ultimately about probabilities. We cannot be 100% accurate in picking winners, but we will be better than a fortune-telling octopus."

References

The Conversation (2026), DOI: https://doi.org/10.64628/AAI.53qqmc6k5

※ This article has been translated by AI. Share your feedback here.