BT1437 tennis performance-indicator reproducibility package: reconstructed player-level datasets, analysis code, and figure/table outputs

This dataset is the reproducibility package for a manuscript on court-specific key performance indicators (KPIs) in men's singles tennis (2019–2025), based on Match Charting Project (MCP) source data. It contains the materials needed to reproduce the reconstructed player-level datasets, the 18-feature analytical dataset, the main analysis outputs, supporting-information tables, and the publication figures. Package contents:1.data/raw_mcp_cache/ : cached Match Charting Project source files used during reconstruction.2.data/processed/ : clean, ASCII-named copies of the reconstructed player-level datasets.3. code/ : reconstruction and analysis scripts, including portable figure scripts.4. outputs/tables/ : main tables, figure data, and supporting-information tables (S1–S9).5.outputs/figures/ : generated figures (Fig1–Fig4, S1–S3) in PNG/SVG/PDF/TIFF where available.6. docs/ : data dictionary, provenance and licensing notes, and a Data Availability statement template. Primary datasets (recommended for citation and reuse):1. reconstructed_player_level_2019_2025_27_features.csv : 284 player-match records from 142 matches, with 27 raw/directional performance indicators plus metadata.2.composite_18_features_player_level.csv : 284 player-match records with the 18 composite features used in the main analysis. Software environment: Python 3.13.5 with scikit-learn 1.9.0, numpy 2.4.4, pandas 3.0.2, scipy 1.17.1, matplotlib 3.11.0, and seaborn 0.13.2. Dependencies are listed in requirements.txt. Source data are derived from the Match Charting Project; users must comply with the original Match Charting Project license terms. See docs/provenance_and_license.md for details.