In late 2019, Quintessa reported the development of a novel algorithm (“N-Estimates”) to assess the overall standard, or “rating”, of a sporting team. As well as calculating a team’s rating, the approach can be used to predict upcoming results. It has previously been used to predict scores for UEFA Euro 2020, the FIFA World Cup in 2022 and UEFA Euro 2024. Now, we present predictions for the group stages of the upcoming 2026 World Cup.

The N-Estimates algorithm uses a mathematical model that is tuned to historical results and considers effects such as home advantage and opposition strength. Using this, the algorithm derives a probability distribution of the performance of a team on a given day, from which the mean “rating” is also calculated. When predicting match results, the distributions of the two teams are converted to a probability distribution of goal difference. An appraisal of both teams’ recent results is also undertaken, allowing the probabilities of different scorelines to be calculated. This calculation has been altered since Euro 2024: previously, it was based on the numbers of goals scored and conceded by each team; in the newest version of the algorithm, the key statistic is the average number of goals per game (regardless of which team scored) in recent matches involving either of the two teams. The most likely scoreline (accounting for the probabilities of both the number of goals and the goal difference) is taken as the predicted match result.

This information is displayed in the prediction plots below. The x- and y-axes represent potential numbers of goals scored by the "home" and "away" teams respectively, with the home wins below the diagonal black dashed line and away wins above (home advantage is applied to the host nations Canada, Mexico and USA; for all other matches the designation of a “home” team is arbitrary). The colour scale represents the normalised probability of each result, with a high probability plotted in orange and a low probability in blue. The most likely result is highlighted by the black crossed-circle.

In the opening stage of the tournament, teams will be competing in twelve groups of four. Each team will play the other three teams in their group, giving a total of 72 results to predict across all groups. Predictions will be added throughout the group stage to allow the results of earlier matches to be taken into account. The algorithm’s predicted group winners are Mexico, Switzerland, Brazil, Turkey, Germany, the Netherlands, Belgium, Spain, Norway, Argentina, Portugal and England. Group I is particularly interesting here, as it contains two teams that are highly-rated by the algorithm, Norway and France. While France are on good form going into the tournament (with seven wins from their last nine non-friendly matches, mostly by two goals), the algorithm rates Norway higher, due to their recent record (having won their last 10 non-friendly matches, winning eight of those by three or more goals including an 11-1 victory over Moldova and two 3-goal victories over Italy). Is the algorithm right in thinking they will convert that momentum into a group-topping performance? Watch this space!

Quintessa is not affiliated in any way with FIFA. Its application of the N-estimates algorithm to FIFA World Cup 2026 is an independent and non-commercial endeavour.