NHL Advanced Stats — GAR, WAR, Expected Goals
Hockey Alchemy is an advanced NHL analytics platform: Goals Above Replacement (GAR), Wins Above Replacement (WAR), expected goals (xG), RAPM, Elo-based power rankings, Stanley Cup odds, and prospect pipeline coverage for the 2025-26 season.
Top Skaters by GAR — 2025-26
| # | Player | Team | Pos | GAR |
|---|---|---|---|---|
| 1 | Connor McDavid | EDM | C | 38.8 |
| 2 | Nathan MacKinnon | COL | C | 29.7 |
| 3 | Nikita Kucherov | TBL | RW | 29.3 |
| 4 | Macklin Celebrini | SJS | C | 29.0 |
| 5 | Cole Caufield | MTL | LW | 27.8 |
| 6 | Jason Robertson | DAL | LW | 27.7 |
| 7 | Quinn Hughes | MIN | LD | 27.6 |
| 8 | Nick Suzuki | MTL | C | 25.1 |
| 9 | Tage Thompson | BUF | RW | 24.5 |
| 10 | Lane Hutson | MTL | RD | 23.2 |
| 11 | Brandon Hagel | TBL | LW | 22.9 |
| 12 | Dylan Guenther | UTA | RW | 22.9 |
| 13 | Alex DeBrincat | DET | LW | 22.6 |
| 14 | Evan Bouchard | EDM | RD | 21.7 |
| 15 | Jack Eichel | VGK | C | 21.1 |
| 16 | Moritz Seider | DET | RD | 20.5 |
| 17 | Leon Draisaitl | EDM | C | 20.1 |
| 18 | Mark Scheifele | WPG | C | 19.6 |
| 19 | Jack Hughes | NJD | C | 19.5 |
| 20 | Clayton Keller | UTA | LW | 19.5 |
Top Goalies by GSAx — 2025-26
| # | Goalie | Team | GSAx |
|---|---|---|---|
| 1 | Jeremy Swayman | BOS | 38.1 |
| 2 | Logan Thompson | WSH | 34.3 |
| 3 | Ilya Sorokin | NYI | 33.4 |
| 4 | Jakub Dobes | MTL | 19.9 |
| 5 | Scott Wedgewood | COL | 19.6 |
NHL Teams
Eastern Conference
Western Conference
Recent Analysis — The Lab
The 2026-27 NHL Schedule, by Team: Who the 84-Game Grid Favors
The NHL's first 84-game season since 1993-94, by team impact: the rest edges that actually tilt the standings, full strength-of-schedule for all 32 teams, how the identical divisional structure shapes every slate, plus opening night and five nights outdoors.
What to Expect With Jim Hiller
Toronto tried the Cup-pedigree hire and Auston Matthews cratered. What Jim Hiller's actual Kings numbers — and our metrics — say to expect from the Leafs' next coach.
Distance and Angle Aren't Enough: Why Our xG Model Splits the Ice Into Zones
A pure distance-and-angle expected-goals model looks reasonable — and quietly misprices the whole ice. We map exactly where it goes wrong, then show how zone, rush, and prior-event features lift out-of-sample AUC from 0.70 to 0.84.