🔬 Deep Analysis

Comprehensive audit identifying 15+ player ID mismatches between Cricsheet data and IPL 2026 squad files. Resolves duplicate IDs from franchise trades (MI↔LSG) and ensures data integrity across all analytics pipelines.

Analysis of where batters typically bat in the order (entry points). Validates our batting position classifications and ensures OPENER, MIDDLE_ORDER, and FINISHER tags align with actual match data.

Exploratory data analysis that determined optimal thresholds for player tags using only 2023-2025 IPL data. Establishes baselines for SPECIALIST vs VULNERABLE, PP_BEAST vs PP_LIABILITY, and other phase-specific tags.

Systematic comparison of league-wide baselines against our player tag criteria. Validates that SPECIALIST and VULNERABLE thresholds are meaningful relative to overall IPL performance distributions.

Deep-dive into Pro Football Focus's play-by-play grading methodology. Explores how context-aware evaluation and position-specific metrics can be adapted for cricket analytics.

Analysis of KenPom's college basketball analytics — adjusted efficiency, tempo-free stats, and the "Four Factors." Foundation for our planned CricPom team rating system.

Opponent-adjusted, venue-normalized, phase-aware T20 efficiency rating system. 5-factor tournament weighting (PQI, CI, Recency, Conditions, Confidence) with sigmoid-based sample size scoring. 231 IPL 2026 players rated across 14 T20 tournaments.

Geometric mean formula weighting 14 T20 tournaments across 5 factors: Player Quality Index, Effective Conditions Index, Recency decay, Conditions Similarity, and Sample Confidence. Produces per-team weighted SR, avg, economy, boundary%, and dot% composites.

Sigmoid-based confidence scoring: 1 / (1 + exp(-0.02 × (matches - 100))). Players with 200+ matches get 95%+ confidence. Under 30 matches = low confidence. Applied to all 231 CricPom player ratings and pressure performance metrics.

Three-metric cluster validation: silhouette score (cohesion vs separation), Davies-Bouldin index (cluster similarity), and Calinski-Harabasz index (between/within variance ratio). Tests K-means against random baseline to confirm clustering is meaningful, not noise.

Consecutive dot ball and boundary sequence analysis for bowling pressure and batting resilience profiling. Rates teams Elite/Strong/Average/Weak based on sequence length thresholds. Identifies clutch performers and choke risks via SR delta under pressure.

Product requirements document for our K-means clustering model. Defines the 6 batter archetypes (EXPLOSIVE_OPENER to FINISHER) and 7 bowler archetypes (PACER to PART_TIMER).

Creative descriptions and narrative framing for each player archetype. Makes technical clusters accessible to fans through cricket storytelling and real-world player examples.

Rolling consistency analysis across IPL 2023-2025. Tracks coefficient of variation in runs scored, single-digit failure rate, and form trajectory by season. Separates "reliable anchors" from "streaky match-winners."

Measures how batting pairs amplify each other's performance. Synergy index compares partnership run rates against individual averages. Year-wise trends show which combinations are improving or declining.

How every IPL batter performs against pace, off-spin, leg-spin, and left-arm spin. Identifies PACE_SPECIALIST, SPIN_SPECIALIST, and VULNERABLE_VS_SPIN tags based on 2023-2025 data with statistically significant sample sizes.

How every IPL bowler performs against left-handers vs right-handers. Reveals asymmetric matchups — bowlers who dominate one handedness but struggle against the other. Critical for batting order optimization.

Win/loss records for every team at every IPL venue, with year-wise breakdown. Identifies home fortress effects, away vulnerabilities, and neutral venue performance patterns across 2023-2025.

Tracks bowler performance under pressure — economy and strike rate in death overs, when defending small totals, and in consecutive dot-ball sequences. The cricket equivalent of "clutch" performance.

How bowlers distribute their overs across powerplay, middle, and death phases. Grouped analysis reveals captaincy patterns — which bowlers are trusted at death, which are powerplay-only, and who bowls through all phases.

How every IPL batter performs across low, medium, and high pressure bands (2023-2025). Segments strike rate, boundary percentage, and dot ball frequency by match situation to reveal who thrives and who wilts under pressure.

Composite pressure performance ratings for batters and bowlers. Measures the delta between overall and pressure-situation strike rates, boundary percentages, and dot ball rates. Assigns CLUTCH, STEADY, or FADES ratings based on statistical thresholds.

Complete specification of our Predicted XII algorithm. Covers constraint satisfaction (overseas limits, balance requirements), scoring weights, impact player selection, and tie-breaking rules.

Domain expert review of our clustering model outputs. Andy Flower's validation of player archetypes and recommendations for threshold adjustments based on cricket expertise.