Scorecards tell you what happened. IQ tells you what it means for the rest of the tournament.
The big-picture rating. How valuable has this player been across the entire season so far?
The last 4 matches only. Picks up hot streaks and cold spells faster than Season IQ.
When Form IQ runs ahead of Season IQ, you're Rising. When it falls behind, you're Flagged.
A player at the 50th percentile of their role lands at 82. The best in the tournament push into the mid-90s. Players having a rough time drop into the 70s. It reads like a quality grade, not a raw stat.
Scored per player, updated after every match. Players are always compared within their role group.
A Bayesian rating that blends where a player stood coming into the season with how they have performed so far, ranked against peers in the same role.
Batters: runs per match, average, strike rate, 30+ scores, runs in wins. Bowlers: wickets per match, economy, strike rate, dot ball %, death bowling.
A short-memory rating built from the last 4 matches only. Picks up hot streaks and cold patches faster than Season IQ.
Exponentially weighted so the most recent match counts far more than the oldest. Anchored to entering baseline to prevent artificial dips early on.
A label based on the gap between Form IQ and Season IQ. Rising means recent form is running well above their season average. Flagged means the opposite.
Requires 4+ matches so small sample noise does not trigger flags.
One word that tells the story of a player's tournament. Breakout means they arrived with a low baseline and jumped. Only available after 6+ matches.
Derived from: entering baseline, season delta, first-half vs second-half split, and volatility.
The pre-season starting point built from the last 3 IPL seasons, weighted toward the most recent. This is the Bayesian prior that Season IQ shrinks toward early on.
Players with limited IPL history start at 82 (the scale midpoint).
How does this player perform under pressure? Tracks performance in tight chases, death overs, and powerplay collapses.
Labels: PRESSURE PLAYER, PRESSURE VULNERABLE, or PRESSURE NEUTRAL based on squeeze performance vs overall rating.
Scored per team using results and roster composition. Updated after every match the team plays.
The team-level equivalent of Season IQ. Built entirely from team results — points, margins, run rates — not averaged player scores.
Points percentage, net run rate, batting run rate, bowling economy, and playoff performance.
How dangerous is this team right now? Blends the season record with the live form of their best players.
Season Team IQ plus the current form of the team's core players (top contributors by season and form rating).
Breaks the squad into 5 departments (top order, middle order, pace, death, spin) and scores each one. Spots imbalances like an elite top order paired with weak death bowling.
Each department scored by the average Season IQ of its players, weighted by importance to match outcomes.
Is the team on the way up or sliding? Combines player trajectory, win margins, and quality of opposition.
Requires 3+ matches.
Is the team picking the right eleven? Compares every player selected with bench alternatives in the same role.
Penalises picking flagged players over rising bench options and IQ gaps between starter and bench.
Computed fresh for each match. Single-game snapshots, not cumulative.
After each match, one player gained the most Season IQ and one lost the most. The two players the match mattered most for.
Pure delta ranking across all players in the match.
Did a player earn their spot today? Compares actual performance against what their IQ predicted.
Based on the gap between actual match performance and IQ-based expectation.
The raw per-match contribution score that feeds into Season IQ and Form IQ. You won't see this number directly, but it powers everything underneath.
Batting: runs, strike rate, run share, not-out bonus, death overs. Bowling: wickets, economy, dot ball pressure, death bowling.
A short explanation for counter-intuitive IQ movements. Why did a player score 60 but their IQ dropped? The delta story answers that.
Example: a batter scores 60 but their IQ drops because their profile is still below the cohort leaders on rate metrics.
Every match generates batting impact and bowling impact scores from runs, strike rate, wickets, economy, dot ball pressure, death bowling, and win contribution.
Impacts accumulate into per-match rate metrics (runs per match, wickets per match, and so on) so that players who have played different numbers of matches can be compared fairly.
Season IQ uses Bayesian percentile scoring within role cohorts. A batter is always ranked against other batters. A bowler against other bowlers. Early in the season, the score stays close to the historical baseline. As matches pile up, live data takes over.
There is no mode switch or sudden cliff. Confidence ramps smoothly from the first match to mid-season. One formula, no breaks.
Every player is assigned one role. All IQ comparisons happen within that role group.
Scored on run output, consistency, strike rate, and contributions in wins. Top-order batters get a slight anchor bonus.
Scored on wicket-taking, economy, dot-ball pressure, and death bowling quality. Lower economy is rewarded.
Scored across both batting and bowling, with a bonus for dual-skill contribution. The stronger discipline carries more weight.
Primarily scored on batting output against the non-bowler pool, with a small component for keeping dismissals.
Before the first ball of the season, every player receives an entering IQ built from their last three IPL seasons, weighted toward the most recent one. Players with no IPL history start at 82.
For IPL 2026, the baseline window is 2023, 2024, and 2025. This approach was validated by replaying the entire IPL 2025 season using 2022-2024 baselines and passing 64 benchmark assertions.
Free, open-data ball-by-ball cricket records under the ODC-By 1.0 license. Used for building entering baselines and validating the model.
Live scorecards, lineups, and ball-by-ball data during the IPL 2026 season.