Asha Latha Maanepalli

IPL Crunch '26 — Ball-by-Ball Match Intelligence

IPL generates massive amounts of ball-by-ball data — but most analysis stops at scoreboards and highlights. The real question nobody asks: what actually separates winning teams from losing ones, phase by phase, delivery by delivery? I wanted to go beyond "who won" and find the structural patterns behind why they won — using 289,000 deliveries across 19 seasons to find something that would hold up statistically, not just look good on a chart. Process Started with data cleaning in MySQL — standardized date formats, handled nulls in win columns, then engineered a Phase_Type column using CASE-WHEN logic to classify every delivery into Power Play, Middle Overs, or Death Overs. Ran phase-level aggregations comparing winners vs losers on run rate and wickets. Then moved to player-level analysis — separate queries for batters and bowlers with minimum ball thresholds to filter noise. In Power BI, the real learning was building a Season dimension table to create relationships between the main table and three imported datasets — phase data, batter data, bowler data. Wrote DAX measures instead of static numbers, cleaned data types in Power Query, and synced slicers across all three pages so the report feels like one connected story. Results Death Overs produce a 1.6 run rate gap — the largest of any phase. Winners score 18% faster while losing 39% fewer wickets in those overs. Toss impact is statistically zero at 50.5%, but the decision made after winning it isn't — teams choosing to field win 10% more often. Across 19 seasons, the pattern holds consistently, which means this isn't a trend. It's a structural truth about how IPL matches are won. Reflection The dataset has venue and city columns I never used. Captains don't randomly choose to field — they're reading dew, moisture, humidity, and pitch behavior. Testing whether those environmental instincts actually translate to wins is the analysis I'd build next. I'd also track Orange Cap and Purple Cap holders through phases — connecting player efficiency to Death Over performance specifically. And partnership analysis would complete the picture that wicket counts alone can't tell.