Forensic Hockey
All methods

Goals Saved Above Expected (GSAx)

Experimental

A derived construct without out-of-sample validation - read as informed opinion.

What it means

Goals a goalie saved compared with what an average goalie would concede on the same shots.

How it's computed

Expected goals faced minus goals against.

What's validated - and what's not

Built on the holdout-validated xG model but never validated goalie-specifically - and it cannot tell a hot goalie from cold opposing shooters; the shot model sees location, not shooter talent.

Coverage

2010-11 to present.

Provenance

Ryder’s shot-quality-neutral save % (2004); Schuckers, goaltending evaluation (2016).

Built on the holdout-validated xG model. Honestly: never validated goalie-specifically, and it cannot separate a hot goalie from cold shooters.

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.