How to Interpret the Data

The app includes a lot of numbers, but most of them can be grouped into a few simple questions. 1. Results: what happened? These are the most familiar measures: - Goals - Wins - Points - Save percentage - Special teams outcomes - Scorelines They matter, but they are not enough on their own. Results can be noisy, especially in hockey. 2. Process: how did it happen? This is where metrics like: - Corsi - Fenwick - xG - shot location - zone context - chance distributions - on-ice event rates become important. Process metrics help answer whether performance is sustainable. A team winning despite weak shot quality may be living on finishing or goaltending. A player with modest point totals but strong chance-driving metrics may be better than the box score suggests. 3. Context: against whom, with whom, and in what role? This is where hockey gets more complex. A player's numbers are shaped by: - Teammates - Opponents - Usage - Zone starts - Special teams role - Score state - Game state - Team system The app tries to keep that context visible instead of flattening everything into one number. 4. Isolation: what is the player or team actually contributing? This is where RAPM and lineup tools become especially useful. They are designed to go beyond "what happened while this player was on the ice" and move closer to "what effect did this player have?" No model can solve that perfectly. But the right tools can get you much closer than raw plus-minus, point totals, or on-ice goal share.