Drivers of Success: A Bayesian State-Space Model to Disentangling Latent Driver and Constructor Abilities in Formula One

2026-08-05

AI Summary

This study addresses the challenge of disentangling the dynamically evolving contributions of drivers and teams to Formula 1 race outcomes, which are jointly determined yet difficult to separate. The authors propose a Bayesian state-space model that jointly leverages qualifying fastest lap times and race finishing positions within a unified framework to infer latent time-varying abilities of both drivers and teams, explicitly accounting for the influence of starting grid position on race performance. By incorporating zero-centered constraints and a time-varying structure for driver–team combinations, the model uniquely enables effective separation and quantification of their dynamic effects. Empirical analysis using Formula 1 data from the hybrid era (2014–2021) reveals that team ability exhibits greater volatility and often dominates race outcomes, whereas driver ability remains relatively stable. Bayesian inference is performed using weakly informative priors and the No-U-Turn Sampler, enabling temporally resolved latent variable estimation at the Grand Prix level.

Abstract

Formula One outcomes reflect the joint contributions of drivers and constructors, but these contributions are unobserved and vary over time. We propose a Bayesian state-space model that disentangles dynamic driver and constructor abilities using two observed outcomes: fastest qualifying lap times and race rankings. Both outcomes depend jointly on latent driver and constructor states that evolve at the Grand Prix level, while the race equation additionally accounts for starting-grid position. The decomposition is supported by constraints that center the driver and constructor abilities at zero, together with variation in driver-constructor assignments over time. Bayesian inference is performed using the No-U-Turn sampler under weakly informative priors that treat driver and constructor abilities symmetrically. Applying the model to the Formula One hybrid era from 2014 to 2021, we find substantial heterogeneity in both driver and constructor abilities. Driver abilities are generally more stable over time, whereas constructor abilities exhibit greater variation and, for many driver--constructor combinations, contribute more strongly to observed performance.

Problem

• Formula One
• latent abilities
• driver performance
• constructor performance
• state-space model

Innovation

• Bayesian state-space model
• latent ability disentanglement
• Formula One performance analysis
• dynamic hierarchical modeling
• No-U-Turn sampler