The approaches routinely used to model the outcomes of football matches are characterized by strong assumptions about the dependence between the number of goals scored by the two competing teams and their marginal distribution. In this work, we argue that the assumptions traditionally made are not always based on solid arguments. Although most of these assumptions have been relaxed in the recent literature, the model introduced by Dixon and Coles in 1997 still represents a point of reference in the betting industry. While maintaining its conceptual simplicity, alternatives based on modelling the conditional distributions allow for the specification of more comprehensive dependence structures. In view of this, we propose a straightforward modification of the usual Poisson marginal models by means of thoroughly chosen marginal and conditional distributions. Careful model validation is provided, and a real data application involving five European leagues is conducted. The novel dependence structure allows to extract key insights on league dynamics and presents practical gains in several betting scenarios.
Modelling dependence in football match outcomes: Traditional assumptions and an alternative proposal
Schiavon, Lorenzo
;
2024-01-01
Abstract
The approaches routinely used to model the outcomes of football matches are characterized by strong assumptions about the dependence between the number of goals scored by the two competing teams and their marginal distribution. In this work, we argue that the assumptions traditionally made are not always based on solid arguments. Although most of these assumptions have been relaxed in the recent literature, the model introduced by Dixon and Coles in 1997 still represents a point of reference in the betting industry. While maintaining its conceptual simplicity, alternatives based on modelling the conditional distributions allow for the specification of more comprehensive dependence structures. In view of this, we propose a straightforward modification of the usual Poisson marginal models by means of thoroughly chosen marginal and conditional distributions. Careful model validation is provided, and a real data application involving five European leagues is conducted. The novel dependence structure allows to extract key insights on league dynamics and presents practical gains in several betting scenarios.I documenti in ARCA sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.