AI Accurately Predicts Finnish Ice Hockey League Regular Season Performance

Digia's developed artificial intelligence model has shown remarkable accuracy in its predictions for the Finnish ice hockey league, Liiga. The model successfully predicted the regular season winner and the second-place team. Furthermore, nine out of the ten teams that qualified for the playoffs were correctly identified in the forecast.

The AI estimated a two-point difference between the top two teams, whereas the actual difference ended up being three points. While many forecasts matched eight teams, Digia's model correctly predicted nine.

Contrary to other predictions, Digia's AI had previously forecasted Jukurit to finish in 7th place, while other forecasts ranked the team much lower. Jukurit ultimately placed fifth. Kärpät exceeded expectations, as the AI predicted them for 10th position, but the team finished in fourth.

Digia's Chief Technology Officer, Juhana Juppo, explained that the model is based on historical game data but acknowledged that surprises can still occur. For instance, a mid-season coaching change for Kärpät led to a shift in playing style that the model did not fully anticipate. During the model's development, hundreds of factors were analyzed, and approximately 20 variables with the most significant impact on the outcome were selected.

Juppo emphasized that AI is a tool whose benefits and limitations must be understood. It is best suited for situations with numerous variables where the goal is to identify the key influencing factors. Digia leverages AI solutions to enhance its customers' business operations across various industries.