FIFA World Cup 2022 Prediction and Actual Results with Logistic Regression Algorithm: An Evaluation

The FIFA World Cup is the most prestigious tournament in the world, held every four years, the FIFA World Cup features 32 (thirty-two) countries competing against each other for the World Cup. In this study, a model was proposed to predict the results of the 2022 FIFA World Cup with the Logistic Regression algorithm using FIFA baseline ranking variables and match history. The results of this study will show the probability of match results from the Group matches, Round of 16 matches, to the Final matches. Then, evaluate it by comparing it with the actual results in the 2022 FIFA World Cup. The dataset is taken from FIFA.com, which includes the history of matches of all countries from 1872 to 2022. Then, from the history of the match results, it is entered into the head-to-head table of matches between countries in the Group stage, thus producing a new dataset. From the new dataset, predictions are then made using Logistic Regression, which will produce the probability of winning, losing, and drawing from each match. The conclusion obtained from this study, after conducting data sets of training and testing in the 2022 FIFA World Cup, can predict matches with FIFA ranking variables and match history (head-to-head) between countries. Germany became the winning country of the FIFA World Cup 2022 with a probability of winning 0.342 in the Final Matches, and the actual matches’ accuracy from Round-of 16 Matches until the Final Matches was 61.29 percent.