NHL coverage in PlayCaller's DaaS API splits the same way NFL, NBA, and MLB do: a raw data layer available in the free Developer Sandbox, and a validated intelligence layer on top of it.
The raw data layer — 8 endpoints
GET /v1/nhl/games— scores and scheduleGET /v1/nhl/standings— standings by divisionGET /v1/nhl/teams— all 32 NHL teamsGET /v1/nhl/teams/{team}/roster— team rosterGET /v1/nhl/players— player list with filtersGET /v1/nhl/players/{id}— player detailGET /v1/nhl/players/{id}/stats— player game logGET /v1/nhl/depth-chart/changes— lineup/rotation change signals (Growth+)
The first 7 are included in the free 14-day Developer Sandbox — real production data, no credit card.
A real confidence model, not a borrowed number
NHL confidence is its own validated model, not NFL's model relabeled for a different sport. It's walk-forward validated at AUC-ROC 0.6688 with a 36.96pp quartile lift. It's reachable through the same /v1/feed intelligence dossier NFL, NBA, and MLB use — the model underneath is sport-specific, the endpoint shape is consistent across all four.
Scope, honestly stated
There's no NHL heat model yet — that field returns null for NHL players rather than a number borrowed from another sport's model. If a heat signal specifically is what you need, NFL is currently the only sport that has one.
Ready to build? Start a free 14-day Developer Sandbox — no credit card, API key in under a minute.