Detailed Depth Breakdown

Understanding Depth Metrics

Each bar shows how evenly a team distributed shots, scoring chances, and ice time across their lineup. Higher depth means more players contributed, rather than relying on a few stars. The core concept behind each component of the Total Depth Index (TDI) is a Gini coefficient. The Gini coefficient is a statistical measure of dispersion typically used to capture inequality in economic resources, but it can be used to capture dispersion in any instance where there is a finite resource (e.g., total shots taken or time on the ice) distributed across a fixed number of actors (i.e., skaters on the ice).

For example, between 2010 and 2025, the game with the most unevenly distributed shots was played by the New York Islanders in 2014 against the Washington Capitals. In this game, the New York Islanders rostered a full 18 skaters and took 11 shots. These 11 shots came from six players, resulting in a shot-on-goal gini coefficient of 0.783. Conversely, the game with the most evenly distributed shots was played by the Columbus Blue Jackets in 2016 against the Colorado Avalanche. In this game, the Columbus Blue Jackets also rostered a full 18 skaters and took 37 shots-on-goal. These 37 shots came from 17 players, resulting in a shot-on-goal gini of 0.158. Thus, taking depth as the complement of these shot inequalities (i.e., 1 - shot-on-goal gini), the New York Islanders had a shot depth of 0.217, and the Columbus Blue Jackets had a shot depth of 0.842, respectively, reflecting very shallow and very deep shooting distributions. Notably, shot depth is not problematically correlated with volume metrics like total shot counts or total expected goals.

One thing I have noticed, anecdotally, when the high depth team loses it is often due to under and overperforming goaltending. For example, in the October 9, 2025 Habs win over Detroit. The Red Wings had higher depth relative to the Habs (.331 vs. .240 by the end of the game). But the Habs won 5-1. If you look then at the goaltending, you will see the Habs Dobes let in 1 against 3.238 xG (+2.24). Detroit's Gibson, on the other hand, only played 37 minutes and let in 5 against 1.483 xG (-3.52).

The Four Components

Note: These metrics show depth in this specific game. For season trends and historical analysis, I use a standardized Total Depth Index (TDI) calculated using structural equation modeling factor scores. Learn more about the methodology →