Leverage & clutch scoring

What we measure

Leverage asks a simple question: when a player's offense arrived, was the game still in the balance? A point scored to tie a game late is worth remembering; a point scored up four goals in garbage time is not, even though the box score treats them the same. We separate the two by tagging each goal and each point with the leverage of the moment it was scored, then reporting the share of a player's production that landed in the tightest situations - their leverage ratio - alongside a plain-language tier.

How we approximate leverage

True leverage would weight each point by the win-probability swing at the exact instant it was scored. The public feed doesn't hand us a live win-probability model, so we approximate it from something it does give us cleanly: the score margin at the moment of each goal. A goal scored in a tied or one-goal game - or in overtime - is treated as high-leverage; a two-goal game is medium; a three-plus-goal margin is low. Crucially, this is judged event by event, not from the final score. A goal that ties the game while trailing by one counts as high-leverage even if the team goes on to win 6-1. Shootout deciders are set aside, since they aren't scored inside the run of play. It's a proxy, and we label it as one - but it's a proxy grounded in the state of the game as each goal happened, not in how the night happened to end.

How the tiers stay honest

The temptation with a clutch metric is to pick a round threshold - say, 'a Closer earns 55% of their offense in tight games' - and let it stand. That would be misleading here, because most goals in hockey are scored in tied or one-goal states to begin with. Across recent seasons roughly 69% of all league points land in high-leverage situations. That's a property of the sport, not a mark of any individual's clutch skill, and an absolute cutoff would simply flag almost everyone as a Closer.

So we grade every player against that season's league base rate instead. A Closer is a player whose high-leverage share runs meaningfully above the league norm for their season; Empty Calories runs meaningfully below it, concentrating production in blowout time; Balanced sits in the middle. The band that separates the tiers is about six-tenths of a standard deviation of the leverage-ratio distribution in either direction - a genuine top-and-bottom-quartile split, not an arbitrary line. Because the comparison is season-relative, a Closer badge means the player beat their peers in the same scoring environment, not that they cleared a fixed number that the era's style of play would inflate or deflate on its own.

Coverage and tier distribution

The metric spans 17 regular seasons (2009-2025), covering 9,068 qualified player-seasons across 1,717 distinct players. A player must have at least 10 total points in a season to be tiered, so a handful of well-timed goals can't manufacture a clutch label. Under the season-relative rule, the three tiers split as a real distribution rather than piling everyone into one bucket:

Game-winning goals, overtime goals, and games played are shown alongside for context. Those come straight from the boxscore and are not run through the leverage proxy - they're there to round out the picture, not to derive the tier.