Research

Questions worth answering rigorously.

Research Interests

  • Causal Inference
  • Bayesian Statistics
  • Machine Learning
  • Generative AI
  • Natural Language Processing
  • Multimodal AI
  • Data Science
IN PROGRESS

Does the Undercut Pay Off?

Pit-Stop Strategy & Race Performance in Formula 1

An investigation into whether the "undercut" — pitting before a rival to gain track position through fresher tyres — causally improves race outcomes, or whether the apparent benefit is confounded by pace, tyre state, and track position at the time of the stop.

  • Causal Inference
  • Bayesian Analysis
  • Statistical Modeling
  • Machine Learning

Research Pipeline

  1. 01Research QuestionDoes the undercut causally improve race position?
  2. 02DataLap-level Formula 1 race and pit-stop data
  3. 03Causal DesignIdentifying and adjusting for confounders
  4. 04Statistical AnalysisEstimating the strategy's effect
  5. 05Machine LearningSupporting predictive analysis
  6. 06ResultsEvidence-based read on undercut effectiveness
  • Framed as a causal inference problem rather than a pure prediction problem — the goal is to estimate an effect, not just fit a model.
  • Also featured on the Projects page as an applied case study.