4-3: Final (OTT). 2.15-3.43: xG (OTT-SEA). 9-7: HD (OTT-SEA). 0.04: Winner xG Seattle controlled the shot count but not the high-danger count. The Torrent put 35 shots on net to Ottawa's 27 and finished at 3.43 xG to 2.15 . Ottawa still had 9 high-danger shots to 7 , which is the story of a team getting fewer attempts but better ones. The Charge converted four of their 27 shots; Seattle converted three of 35. Fanuza Kadirova 's power-play wrister at P3 18:14 , graded at 0.04 xG , was the game-winner with 1:46 left. The HD-to-shot ratio went Ottawa's way Ottawa's 9 high-danger shots on 27 total is a 33 percent rate from the slot. Seattle's 7 on 35 is 20 percent. That gap is where an Ottawa xG/shot of 0.080 came from, against Seattle's 0.098. The shot-volume gap was real, but the chance-per-shot gap partly offset it. Alexa Vasko 's 0.25 xG opener at P1 14:15 was the single best Ottawa look of the night; Seattle had four separate shots above 0.18 xG that didn't go in. Eldridge ran the Seattle power play Jessie Eldridge scored both Seattle goals, both on the power play, both on shots the model graded at 0.24 xG. The Torrent went 2-for-4 with 10 shots and 1.49 xG across four opportunities, which is efficient, in-tight PP work. Ottawa answered Eldridge's 2-1 goal 42 seconds later when Emily Clark restored the two-goal lead, but Eldridge's second goal at P2 14:47 cut it back to one, and Alex Carpenter equalized at P3 10:08 . The Torrent rode power-play offense back from a two-goal deficit to a one-goal deficit to a tie, then gave up a late PP of their own. The 0.04 xG winner Ottawa's fourth goal came on its fourth power play of the night. Fanuza Kadirova scored at P3 18:14 on a wrister the model graded at 0.04 xG , roughly a one-in-twenty-five shot. That's the kind of number that makes the result feel thin, because on the chance-quality split Seattle had the better of it. Ottawa's PP finished 1-for-4 on 3 shots and 0.22 xG , which is a low-event PP line across eight minutes of time. The winning goal was the only PP shot of note the unit produced, but it happened to be the one that cashed. Goaltending, decomposed Gwyneth Philips faced 35 shots worth 3.43 xG and allowed three goals for +1.43 GSAx . Slightly above expectation on heavy volume. Both Eldridge PP goals were 0.24 xG shots, about the rate that a mid quality PP one-timer converts; Carpenter's tying goal wasn't matched to a shot row with a graded xG in the feed. The save-process line across 35 shots reads as a steady night on a very heavy workload. Hannah Murphy stopped 23 of 27 live shots worth 2.15 xG, finishing at -1.85 GSAx . That's two goals above expectation on a lighter workload, and it was the other side of the numerical coin in a game where Ottawa's four goals came on shots averaging 0.14 xG each. Murphy's save process is a small sample in any one game, but on 27 shots she gave up more goals than the model expected she would. The game in cumulative xG The scoring timeline P1 14:15 · Alexa Vasko · xG 0.28 · OTT 1, SEA 0. P1 19:02 · Fanuza Kadirova · xG 0.12 · OTT 2, SEA 0. P2 9:13 · Jessie Eldridge (PP) · xG 0.25 · OTT 2, SEA 1. P2 9:55 · Emily Clark · xG 0.14 · OTT 3, SEA 1. P2 14:47 · Jessie Eldridge (PP) · xG 0.24 · OTT 3, SEA 2. P3 18:14 · Fanuza Kadirova (PP) · xG 0.04 · OTT 4, SEA 2 Top scoring chances SEA P3 15:28 · Natalie Snodgrass shot · xG 0.27. SEA P2 9:13 · Jessie Eldridge shot · xG 0.25 (goal). SEA P2 14:47 · Jessie Eldridge shot · xG 0.24 (goal). SEA P3 0:46 · Julia Gosling shot · xG 0.23. SEA P2 9:11 · Julia Gosling shot · xG 0.21. SEA P2 14:46 · Julia Gosling shot · xG 0.17. OTT P1 14:15 · Alexa Vasko shot · xG 0.28 (goal). OTT P3 3:12 · Kateřina Mrázová shot · xG 0.15. OTT P3 11:11 · Brooke McQuigge shot · xG 0.14. OTT P2 4:08 · Rebecca Leslie shot · xG 0.14. OTT P2 9:55 · Emily Clark shot · xG 0.14 (goal). OTT P1 19:02 · Fanuza Kadirova shot · xG 0.12 (goal) What this read isn't Not a claim Ottawa played the better game on process. On shot volume and xG, Seattle did more to generate offense. Ottawa won on a better HD-to-shot ratio and by converting a 0.04 xG look late.. Not a sign Ottawa's power play has turned a corner. 1-for-4 on 3 shots and 0.22 xG in eight minutes of PP time is low-event work. The winner went in; the process didn't meaningfully change.. Not a knock on Hannah Murphy as a goalie. A -1.85 GSAx night on 27 shots is one game's variance. The two 0.24 Eldridge shots at her end didn't go.. Not a model failure. A game where one team generates 1.22 more xG and loses on a 0.04 PP goal is exactly where the chance-quality probability calibration allows single-game results to land. Methodology xG model : PWHL-calibrated logistic regression on 16,709 non-EN shots. Same fit as the methodology post . Features: distance, angle, rebound (≤3 s), penalty shot, power play, overtime, scorer-tagged quality chance, seven shot-type dummies. Empty-net shots are priced outside the main model (MoneyPuck / NST convention).. GSAx : summed xG on shots faced minus goals allowed, computed per-goalie from the PBP feed. Empty-net shots and goals are excluded from both sides of the ledger. Overtime shots are included.. Numbers regenerate on every build. A PBP re-sync or xG recalibration reshapes every figure in this post on the next static export. Full play-by-play, shift overlays, and the per-player game log for this game are at /games/68 .