Colombia’s 2026 FIFA World Cup Chances: What 50,000 Simulations Predict
Colombia’s 2026 FIFA World Cup prospects look stronger than many fans might expect. A statistical model developed by Credicorp Capital simulated the tournament 50,000 times and found that Colombia wins the title in roughly 3% of scenarios, placing the Tricolor among the competition’s top contenders. That figure may sound modest at first glance, but it takes on more weight in context: only seven other teams in a 48-nation field have higher probabilities.
Colombia sits alongside the Netherlands (3.5%), ahead of Germany (2.8%) and Norway (1.6%), and firmly within the statistical elite of a tournament where most teams carry odds below 1%. While Spain remains the clear favorite, the model suggests Colombia enters the World Cup not as a long shot, but as a legitimate contender capable of shaping the tournament.
How the World Cup simulation model works
The Credicorp Capital model rests on three methodological pillars that work together to translate football history into probability, starting with the Elo rating system, a method originally developed for chess and later adapted for football that scores every national team based on the outcomes of all their recent matches, giving more weight to victories over strong opponents and penalizing losses to weaker ones, so that the number assigned to each team at any given moment represents its accumulated competitive track record in a single comparable figure.
The second pillar involves Poisson distributions, a mathematical tool that uses each team’s historical scoring and defensive averages to estimate how many goals they are likely to score and concede in any specific match, producing not a single predicted result but a full range of scorelines with attached probabilities, so the model can calculate not only who is likely to win but how likely a 1–0 is compared to a 2–1 or a draw.
The third and most computationally intensive element is the Monte Carlo simulation, a process that runs the entire tournament, all 104 matches across all phases, 50,000 separate times, each time introducing slightly different random variations within the probability ranges the model established, so that instead of a single deterministic prediction, the output is a distribution of outcomes built from tens of thousands of imagined tournaments, each one consistent with what the data considers realistic. Credicorp then calibrated the raw statistical output against the prices available on international prediction markets such as Polymarket and Kalshi, using the collective judgment of people placing real money on tournament outcomes as a cross-check against the model’s own calculations.
2026 World Cup favorites according to the data
Across all 50,000 simulations, Spain won the title 33.7% of the time, a margin so large it sets the Spanish team in a category of its own and reflects a combination of FIFA ranking, recent tournament performance, squad depth, and the consistency of its playing system under Luis de la Fuente. France follows at 18.5%, nearly half of Spain’s probability but still nearly double that of England (11.5%), which sits third, and Argentina (9.3%), the defending champion, which the model rates fourth despite holding the world title entering the tournament. Portugal, Brazil, and the Netherlands cluster between 3.5% and 5.8%, with Colombia at 3% sitting just below the Netherlands and above Germany (2.8%) and Norway (1.6%).
The consistency of these rankings across independent methodologies is striking, because the Opta supercomputer, Goldman Sachs’s separate economic and statistical model, and DataFactory’s analysis all converge on the same general hierarchy, Spain leads, France and Argentina follow, and Colombia lands between 2.1% and 3.2% across all three systems, consistently occupying the ninth or tenth position among title contenders regardless of which methodology produces the number. That convergence across different analytical teams using different data inputs and different computational approaches suggests the ranking reflects something real about the comparative strength of these squads rather than an artifact of any single model’s assumptions.
Colombia’s projected path to the knockout stage
For Colombia specifically, the simulation projects a tournament journey that begins with a second-place finish in Group K, behind Portugal but ahead of Uzbekistan and the Democratic Republic of the Congo, and then extends into the knockout rounds through a victory over Croatia in the round of 16 before ending in a quarterfinal exit at the hands of Spain, the same team the model identifies as the tournament’s dominant force. The group draw for Colombia placed the Tricolor in a bracket that the models consider manageable but not straightforward, because Portugal, the group favorite, arrives at the tournament with one of the best-ranked squads in the world and with a transition underway from the Cristiano Ronaldo era toward a generation built around players such as Vitinha, Bruno Fernandes, and Rafael Leão.
Colombia’s matches in Group K will take place in Miami and other U.S. venues, with the critical match against Portugal scheduled for June 27 at Hard Rock Stadium in Miami, a stadium Colombia’s supporters know well and one that the model treats as a potential tipping point, since the two teams’ respective Elo ratings produce a genuinely competitive probability distribution for that match rather than a decisive Portuguese advantage. The cities Colombia could visit in the knockout rounds, however, depend on group-stage results from other brackets, and the model projects a round-of-16 venue in Toronto, more than 2,400 kilometers from Miami, followed by a potential quarterfinal confrontation with Spain, meaning that Colombia’s geographical journey through the tournament could span three countries and several time zones before the path ends.
Rafael Castellanos, the managing director of asset management at Credicorp Capital, who led the study, acknowledged directly that statistical models carry an inherent ceiling when applied to football, because the sport contains a component of randomness, individual brilliance, and collective momentum that no algorithm has yet learned to fully capture.
Colombia’s squad, built around players such as Luis Díaz, Jhon Arias, James Rodríguez, Richard Ríos, and Jhon Córdoba, holds the individual quality to produce performances that exceed what historical averages would project, and the tournament format, which compresses entire seasons of competitive pressure into a few weeks, historically rewards teams that peak at the right moment rather than teams that simply dominate the regular statistical record.
The 3% probability the models assign to Colombia means that in a fully random sample of 100 equally likely tournament outcomes, Colombia lifts the trophy three times, and those three outcomes exist not as statistical noise but as genuine branches of the probability tree that the model built, branches that depend on the kinds of performances that happen on football pitches rather than inside computer simulations, and that have materialized for teams ranked far lower than Colombia at the start of previous World Cups.