Exploring and modelling team performances of the Kaggle European Soccer database

Abstract

This study explores a big and open database of soccer leagues in 10 European countries. Data related to players, teams and matches covering seven seasons (from 2009/2010 to 2015/2016) were retrieved from Kaggle, an online platform in which big data are available for predictive modelling and analytics competition among data scientists. Based on both preliminary data analysis, experts’ evaluation and players’ position on the football pitch, role-based indicators of teams’ performance have been built and used to estimate the win probability of the home team with the binomial logistic regression (BLR) model that has been extended including the ELO rating predictor and two random effects due to the hierarchical structure of the dataset. The predictive power of the BLR model and its extensions has been compared with the one of other statistical modelling approaches (Random Forest, Neural Network, k-NN, Naïve Bayes). Results showed that role-based indicators substantially improved the performance of all the models used in both this work and in previous works available on Kaggle. The base BLR model increased prediction accuracy by 10 percentage points, and showed the importance of defence performances, especially in the last seasons. Inclusion of both ELO rating predictor and the random effects did not substantially improve prediction, as the simpler BLR model performed equally good. With respect to the other models, only Naïve Bayes showed more balanced results in predicting both win and no-win of the home team.

Predicting Wins, Losses and Attributes’ Sensitivities in the Soccer World Cup 2018 Using Neural Network Analysis

“…In particular, predictability has become possible according to players information through the analyses of data from past-recorded matches. For instance, the machine learning algorithms that have been employed within soccer research are K-Nearest neighbors [33,34], Neural networks [25][26][27][28]30,35,36], Decision Trees [37][38][39] and Bayesian networks [29,31,32,40,41].…”

The Personality Panorama: Conceptualizing Personality through Big Behavioural Data

“…The inclusion of diverse indicators, extracted from structured and unstructured data at a more or less granular level, allows for fine‐tuned hypothesis formulation and testing (Carpita, Ciavolino, & Pasca, 2019; Landers et al, 2016). Recent work has shown, for example, that studies that include a wider variety of behavioural indicators can more accurately and comprehensively predict personality traits than those with less varied data (Azucar, Marengo, & Settanni, 2018).…”

“…Many symptoms of depression and other psychopathologies are already observable on social media using current methods (Birnbaum, Ernala, Rizvi, De Choudhury, & Kane, 2017; Guntuku, Yaden, Kern, Ungar, & Eichstaedt, 2017; Hassanpour, Tomita, DeLise, Crosier, & Marsch, 2019; Reece & Danforth, 2017; Stanton et al, 2017; Thorstad & Wolff, 2019); it would be a grave mistake to ignore the reduction in human suffering that could be afforded by faster and more accurate diagnosis and treatment. But a utilitarian framework such as this may be insufficient, not just because the distribution of winners and losers is dramatically uneven, but also because it conflicts with a Kantian view in which one does not use others as means to an end (Celie & Paris, 2019; Tunick, 2014). Privacy is fraught, for secrecy can destroy as well as elevate human dignity (Martin, 2019).…”

The role of passing network indicators in modeling football outcomes: an application using Bayesian hierarchical models

“…Furthermore, Ievoli et al (2021) used network indicators to model the probability of winning the game for a team. From a slightly different perspective, Carpita et al (2019) included the a priori evaluation of players' abilities (involving passing skills) in predicting the win, without using passing network information.…”