The AI has consistently ranked KalPa ahead of SaiPa for the past few months. Throughout the season, it also predicted KalPa, Lukko, and Ilves to reach the playoffs and semifinals. At the same time, SaiPa's meteoric rise to the championship battle surprised both the AI and other forecasters. Get to know the predictions on Digia.com (in Finnish).
The prediction is based on a mathematical model that calculates the most likely winner of each game and thus arrives at the final championship outcome.
Over the entire season, the AI's success rate is approximately 69%, meaning it correctly predicted about two-thirds of all games. For the playoffs, the success rate is 75%, or three out of four games correctly predicted.
AI Outperforms Betting Offices and Professionals
Last season, 2023-24, AI outperformed betting offices and professionals with a success rate of 76%. Then, as well as this season, one would have made a sweet profit if they had consistently bet according to the AI's predictions.
Last season, the AI achieved numerous successes in predicting the top positions in the regular season, playoffs, and finals. This season, the AI faced innumerable variables, the most significant of which was the change in the league format.
"In the new league system, some teams faced each other less frequently, at least four times, which provides the AI with relatively little historical data. Last season, all teams faced each other roughly the same number of times. Considering this, I am pleased with how well the finals were predicted," says Hilla Tilhi, who was responsible for the prediction at Digia.
Tilhi reminds us that rankings can depend on minimal factors, even a single game or goal. For example, in the bronze game, the AI correctly predicted a tie in regular time, but in overtime, Ilves scored the first goal, not Lukko. The real strength of the AI lies in identifying trends from large data sets.
AI is a Good Tool for Management When You Know Its Strengths and Limitations
Digia implemented the AI prediction for the Liiga season outcomes to illustrate the potential of AI as a tool for decision-making and management.
The key to utilizing AI is understanding what predictions are based on. One should not blindly trust AI; when used correctly, it is an excellent tool for decision-making, forecasting, and management.
"There is a lot of hype and misconceptions about AI, but in reality, it is a daily tool like any other technology, and it needs to be used correctly. As with any technology, understanding the strengths and limitations of AI is key to its effective use. Combining machine-generated insights with human intuition often produces the best results," says Juhana Juppo, CTO, Digia.
More information:
Juhana Juppo
Chief Technology Officer, Digia Plc
tel. +358 40 172 2859
juhana.juppo(a)digia.com