Explainers
How wRink’s numbers work, what they tell us, and where their limits are.
All explainers
16 explainers · A–ZContract judgment
How contribution, role and contract terms inform the judgment of a deal.
Coverage and missing data
What a missing number means, and why an available estimate may cover only part of a season.
Expected goals (xG)
How scoring chances become an expected-goal total, and what the recorded play can leave out.
Finishing
How we estimate a shooter’s contribution beyond his chances while accounting for a limited sample.
Goals above expected
The gap between actual goals and expected goals from the same attempts.
Goals saved above expected (GSAx)
How a goalie’s goals allowed compare with the expected goals from the shots he faced.
Model tests
How we test whether the model adds useful information on data it did not learn from.
Penalties
How drawing and taking penalties contribute to a player’s value.
Process
How we estimate a skater’s five-on-five contribution to creating chances and limiting them.
Ratings and percentiles
How to read a player’s standing among comparable skaters without treating every small gap as meaningful.
Shrinkage
Why uncertain estimates move toward average, and why stronger evidence earns more weight.
Special teams
How power-play and penalty-kill contributions are measured in their different roles.
Uncertainty
How much weight to put on a player estimate, and why small differences deserve caution.
WAR $
How WAR becomes dollars on today’s salary-cap scale.
Wins above average (WAA)
How the components measure a player’s contribution against average performance.
Wins above replacement (WAR)
How player contribution is expressed in wins beyond a depth replacement.
The full method: How our numbers are made