FisayoSangolade/athlete-recovery-analysis

Exploring what drives athletic recovery using biometric and lifestyle data — AltSchool Africa Data Science Semester 2 # Athlete Recovery Analysis ## What's this project about? Every athlete wants to recover faster. But what actually drives recovery? This project analyses a synthetic dataset tracking ~300 athletes over 28 days, looking at sleep, heart rate variability (HRV), stress levels, training load, and how recovered they felt each day. **Central question:** What actually drives an athlete's recovery score? ## What's in this repo? - `FisayoSangolade.ipynb` — full analysis notebook - `athlete_recovery_synthetic.csv` — dataset used ## Key Findings 1. Sleep is the strongest predictor of recovery — more sleep consistently meant better recovery, peaking at 86.4 for athletes sleeping 9+ hours 2. HRV is positively correlated with recovery (statistically significant) 3. Stress independently reduces recovery — even good sleep can't fully cancel high stress 4. Training load alone does not predict recovery ## Tools Used Python, Pandas, NumPy, Matplotlib, Seaborn, SciPy ## Context Semester 2 exam project — AltSchool Africa Data Science Programme