A Data-Driven Analysis of Formula 1 Car Races Outcome

Abstract

AbstractThere are a range of factors that affect the outcome of Formula 1 (F1) car races. Today, it is reasonable to say that F1 races are first won at the factory, and then on the track. F1 teams accumulate enormous amounts of data during races. In this paper, we propose a data-driven approach to identify the most important factors that contribute to the overall points scored by each driver in a F1 season. We perform a correlation analysis along with a principal components analysis (PCA) to identify the factors that are closely related. Furthermore, using PCA, we efficiently reduce our 21 input variables into a lower-dimensional subspace, that can explain most of the variance in our data and which is easier to comprehend. We obtain 5 years (2015–2019) of data explaining the F1 car characteristics from a publicly available website https://www.racefans.net/. We use this web-scrapped F1 race study to understand the impact of the different car features on the total points scored by a driver in the season. To the best of our knowledge, our work is the first of its kind in the area of F1 car races.