Projects→Dota2 Betting Analytics Pipeline
STATUS
● Production
CLIENT
Personal / Research
YEAR
2025–2026
ROLE
Builder & Architect
STACK
PythonSupabasePostgreSQLOpenDota APIStratz APIPinnacle/BettingIsCool APIPandas
EVIDENCE LINKS
  • ● Live demo unavailable (Private)
Note: Where client systems are private, screenshots are anonymized and architecture is shown without protected data.

Dota2 Betting Analytics Pipeline

End-to-end Dota2 match data pipeline: 50,000+ OpenDota accounts scraped, league enrichment via STRATZ API, Pinnacle odds mapped with fuzzy timestamp matching, CLV analysis.

Dota2 Betting Analytics Pipeline

Problem

No existing tool combined OpenDota match data with Pinnacle odds at scale for Closing Line Value analysis. Manual correlation was infeasible across 50k+ accounts.

Implementation Decisions

  1. SQLite for resumable progress tracking — 50k accounts scraped across multiple sessions without restart
  2. Fuzzy ±300s timestamp matching between Valve match IDs and Pinnacle fixtures
  3. Tiered API backoff (30s → 2min → 30min) respects free-tier rate limits without missing data
  4. Supabase upsert for idempotent odds writes — reruns produce zero duplicates