Uma abordagem baseada em passeios aleatórios para identificar o melhor piloto de Fórmula 1 da história

Abstract

In the world of motorsports, when we are interested in identifying the best driver, we generally only consider those who have raced in Formula 1, the pinnacle of the global four-wheeled sport. However, the definition of what it means to be the best driver is subjective, as it can involve different factors, whether technical or emotional. In this work, we propose to evaluate the performance of Formula 1 drivers through an exact and objective approach. To do this, the similarity of the performance of two drivers is measured based on metrics that take into account variables that evaluate both the driver's talent and the quality of their cars. From these similarity measures, we will build a directed graph with weights, where each vertex represents a driver and each edge is weighted by a function of the similarity measure between the vertices that defines it. Finally, we will find the stationary distribution over this graph, and the best Formula 1 driver is the one with the highest distribution. Based on this decision rule, we also obtain a ranking of driver performance.