This project provides a fully reproducible analysis of Google Trends data for "Cricket Australia" from 2014–2024. It demonstrates how to: Import and clean monthly Google Trends data. Explore time series patterns, including trends, seasonality, anomalies, and autocorrelation. Fit multiple forecasting models (ARIMA, ETS, TSLM) on historical data and evaluate their predictive performance. Forecast search interest for 2025 using all models and visualize combined historical + forecast data. Interpret results and discuss limitations, including the absence of match- or team-specific context—spikes in search interest are often driven by visits from countries like India or England. The document is designed for complete novices, with clear step-by-step instructions, commented code, and reproducible analysis. It serves as a portfolio-quality example of combining sports data, R programming, and statistical modelling for time series forecasting.

This project provides a fully reproducible analysis of swimming events from the Tokyo 2020 Olympics using the SwimmeR R package. It demonstrates how to: Import and clean official Omega-format results PDFs. Visualize performance patterns across events and genders using violin plots, jitter plots, and histograms. Build and evaluate regression models to explore relationships between reaction time, split times, stroke type, and gender. Conduct athlete-level analyses, including split comparisons for Ariarne Titmus’ 400m freestyle races. The document is designed for complete novices, with clear step-by-step instructions and commented code, allowing anyone with the dataset to fully replicate the analysis. It serves as a portfolio-quality example of combining sports data, R programming, and statistical modelling.