Polymarket Crypto/15M Analytics

Goal

Build a service that collects all current and resolved events from the crypto/15M section, stores their lifecycle metadata, and computes analytics per event:

• start time
• end time
• final resolution (if resolved)
• min/max probability (price) during the event window

Architecture (High-level)

1. Market discovery (Gamma API)
   • Periodically fetch markets.
   • Filter by category/section: crypto/15M.
   • Persist events + market metadata.
2. Price streaming (CLOB WebSocket/RTDS)
   • Subscribe to tokens for discovered markets.
   • Persist price time series.
3. Analytics processor
   • Compute min/max price per event over [start_time, end_time || now].
   • Store aggregated analytics.
4. API service
   • Expose endpoints to query events and analytics.

Data model (MVP)

• events: core lifecycle info (start/end/resolution/status) + token_id
• price_history: time series price points by token/market
• event_analytics: precomputed min/max/last price

Implementation plan (detailed)

1. Scaffold project
   • FastAPI app with typed models.
   • Config layer for API base URLs + DB URLs.
2. Gamma client
   • Fetch markets via Gamma API.
   • Normalize fields into Event records.
3. CLOB client
   • Fetch current price for token.
   • (Later) WebSocket/RTDS streaming.
4. Persistence layer
   • Abstract repository interfaces (event repo, price repo, analytics repo).
   • Start with in-memory implementation; swap for Postgres.
5. Collector
   • Scheduled job: fetch markets, upsert events, track tokens.
6. Price ingestion
   • Poll (temporary) + store price points; later upgrade to streaming.
7. Analytics
   • Compute min/max over event window; update aggregates.
8. API endpoints
   • /events?category=crypto/15M
   • /events/{event_id}
   • /events/{event_id}/analytics

Running (local)

python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
uvicorn app.api.main:app --reload