Not a client project, not a job — just 18 seasons of ball-by-ball data and a question I couldn't leave alone after RCB finally won it. This is what happens when curiosity gets a SQL query.
Aggregating 18 seasons of ball-by-ball data in SQL and Python, I quantified a phase-wise bowling shift: RCB's powerplay economy dropped from 9.47 to 8.52 runs per over — 0.76 below the IPL average. Across the league stage, that's roughly 85 runs saved, and it traces directly to the eventual 6-run final margin.
Built a cost-efficiency scoring model in Python — cost-per-wicket, cost-per-run — across every squad pick from the auction. It surfaced two under-priced match-winners bought for a combined ₹4.1Cr against a single ₹12.5Cr anchor signing that returned 22 wickets. The framework was designed to be reusable as a pre-auction evaluation tool, not just a one-off finding.
Cross-tabulating home vs. away performance surfaced a pitch-independence signal: bowling economy was better away (9.22) than at home (9.35), while the away win rate jumped from 33% to 85% year-over-year. That reframes how a franchise should think about retention priorities — home-ground advantage wasn't actually where RCB's edge was coming from.