The GSAx Paradox
The analytics community's preferred goaltending metric is less reliable than the one it was designed to improve. 550 consecutive-season pairs across 18 years explain why.
Human-performance researcher applying information theory, physiology, and decision science to sport.
Founder, Beyond the Metric. PhD in Human Kinetics (Exercise Physiology). Clinical Exercise Physiologist & High-performance specialist. 41+ peer-reviewed publications in human performance, exercise physiology, and applied sport science.
The analytics community's preferred goaltending metric is less reliable than the one it was designed to improve. 550 consecutive-season pairs across 18 years explain why.
1,869 goalie-seasons. 18 years. The year-to-year correlation for save percentage is r = 0.33. Only 11% of the variance is real. The rest is a mirage.
17 drafts. 4,006 picks. One question: how good is the screening decision?
What the 180 Seconds After a Goal Against Actually Look Like
Biathlon’s Hidden Dimension
What the Best Team in Hockey Tells Us About the Limits of Any Single Metric
Why the NHL's Most Disciplined Offenses Are Its Worst
Eight seasons. 633,318 shots. One finding survived every statistical correction. Several others collapsed.