JPAR and Ranking Exploration
JPAR Calculation Data
The actual event results and intermediate values entering the JPAR calculation. The default filter shows only member-resolved rows that received a JPAR update in this cutoff run.
Note: Some ranking systems are currently disabled pending further testing.
JPAR Ranking System Comparison
Compare JPAR with alternative ranking systems across final ranks, historical drift, score distributions, and one-to-one rank plots. Click a person row to highlight them throughout.
Ranking system descriptions
Drift Percentiles Over Time
Lines show p10, p25, median, p75, and p90 of each system's as-of score distribution after each event. Median is black.
Score Distributions
Histograms use the final person-level scores for each system, clipped to the 1st and 99th percentiles to keep long tails readable.
Rank Comparisons vs JPAR
Each scatter plot compares original fiducial JPAR rank on the x-axis against another system's rank on the y-axis. Lower rank is better, so closer to the upper-left is stronger.
Post-event rank in each system for the selected person.
Custom Rank Comparison
Choose any two enabled ranking systems for a direct one-to-one comparison at the currently selected ranking cutoff.
Note: Some ranking systems are currently disabled pending further testing.
Puzzler Data
Inspect one puzzler's incoming system score against what happened next. Rank comparisons are genuinely pre-event. Converted time estimates deliberately use the observed event distribution and are diagnostic, not leakage-free forecasts.
Systems shown on both scatterplots
Diagnostic predicted vs actual time
How predicted times are derived
Pre-event predicted vs actual rank
Selected Event
Prediction History
Event Predictions
Compare the ranking systems immediately before an event with the entrants' actual finish order. Incoming ranks are recalculated within each event's eligible field, so the one-to-one line represents a perfect prediction.
Systems to compare
Prediction accuracy
Plot style
Entrants in Selected Event
Entrants with an incoming JPAR appear first. Competitors without one are retained at the bottom and excluded from JPAR's prediction metrics.
Note: Some ranking systems are currently disabled pending further testing.
Raw vs Adjusted JPAR by Event
Raw event JPAR is computed from the event mean; adjusted event JPAR applies JPAR's calibration multiplier. The divergence between their medians is the calibration intervention for that event. Green shows the as-of median latest JPAR after each event.
Event Impacts in JPAR
JPAR's adjustment factor is
event mean time / anchor-implied mean time. Below 1.00 inflates every participant's raw event JPAR (a relative penalty); above 1.00 lowers it (a relative boost). These are calibration shifts, not a claim that a puzzle itself was objectively easy or hard.Median JPAR delta =
adjusted_event_jpar - raw_event_jpar: a positive value increases JPAR scores (worse); a negative value decreases them (better).Largest relative JPAR penalties
Largest relative JPAR boosts
Cumulative JPAR Calibration Impact
Negative final calibration effect means JPAR ended lower (better) than the same running update using raw event JPAR; positive means it ended higher (worse). Click headers to sort.
Note: Some ranking systems are currently disabled pending further testing.