Forensic Hockey
All methods

Adjusted Plus-Minus (RAPM)

Validated model

Tested on data the model never fit, with the accuracy published.

What it means

A skater's even-strength impact on chance rates with linemates and competition regressed away.

How it's computed

Ridge regression over every 5v5 stint; the penalty is chosen by cross-validation.

What's validated - and what's not

Cross-validated with split-half reliability checks. Still a shrunk single-season estimate - differences of a few hundredths are noise.

Coverage

2010-11 to present (needs shift data).

Provenance

Macdonald, “A Regression-Based Adjusted Plus-Minus Statistic for NHL Players” (2011).

Ridge regression over our 5v5 stints; the penalty is chosen by cross-validation, with split-half reliability checks.

Our research ledger - every source we tested, adopted, or rejected →

Tiers are a promise, not a style. Reconciled is checked against the official NHL record; Validated is tested on data the model never fit; Experimental is a derived construct without out-of-sample validation - read it as informed opinion. When in doubt we take the lower tier.