Game Excitement Index

How exciting was every NBA game?

The Game Excitement Index (GEI) quantifies how thrilling a basketball game was by tracking how much the win probability swung back and forth throughout the contest. A game where the lead changes hands repeatedly and the outcome stays in doubt until the final possession scores much higher than a wire-to-wire blowout.

Originally developed by Luke Benz for college basketball, this adaptation applies the same framework to the NBA using ESPN's play-by-play win probability estimates across 0 games.

Methodology

Imagine taking the win probability curve and stretching it out flat like a piece of string. The longer the string, the more the game's outcome swung back and forth, and the more exciting it was to watch. A blowout produces a short, flat string; a back-and-forth thriller produces a long, tangled one.

Formally, GEI sums the absolute changes in win probability between consecutive plays, normalized to a standard 48-minute game:

GEI = (2880 / t) × Σ |pi − pi-1|

The normalization by game length (t, in seconds) prevents overtime games from ranking artificially high simply for having more plays.

Distribution

Games
0
Mean GEI
0
Median GEI
0