Prompt file imported from itzcole03/A1Betting7-13.2 (.github/prompts/tennis.prompt.md). Copyright stays with the author.

Tennis Data & Modeling Prompt

Goal

Build, validate, and deploy an end-to-end Tennis betting prediction pipeline using real data, with a focus on accuracy and actionable recommendations.

Data Flow

• Ingest real Tennis data from SportRadar, TheOdds, PrizePicks, etc.
• ETL: Clean, normalize, and store in database (PostgreSQL/SQLite).
• Feature engineering: Extract player, match, surface, and context features.
• Model: Train/test ML/LLM models for Tennis props and outcomes.
• API: Expose predictions via BetAnalysisResponse (with enriched_props, confidence, etc.).
• Frontend: Display Tennis props, predictions, and explanations in PropOllamaUnified/PropGPT.

Focus

• Validate every step with Tennis-specific data and edge cases.
• Tune models for Tennis stat distributions and betting markets.
• Document lessons learned for future sports.

Next Steps

• Integrate and test live Tennis data feeds.
• Build and validate Tennis feature engineering pipeline.
• Train and evaluate Tennis-specific models.
• Expose Tennis predictions via API and frontend.