4-1: Final (MIN). 2.48-2.04: xG (VAN-MIN, full game). 10-6: HD (VAN-MIN). +1.48: Rooney GSAx On shot volume and chance quality in the in-tight zone, Vancouver was the better team at home ice. The Goldeneyes outshot Minnesota 32 to 22 , had 10 high-danger shots to 6 , and finished at 50.8 percent Corsi. The scoreboard went the other way because Maddie Rooney stopped 31 of 32 live shots for +1.48 GSAx , and because Kelly Pannek 's 0.40 xG power-play goal with 1 second left in the first period made it 3-0 and closed the realistic comeback window. The first-period buzzer-beater Kendall Cooper opened the scoring at P1 4:57 on a 0.09 xG wrister. Britta Curl-Salemme made it 2-0 at P1 17:07 on a 0.09 slap. At P1 19:13, Vancouver's Hannah Miller took a holding minor, and Minnesota spent 46 seconds of the resulting power play working the puck around until Pannek found a finish at P1 19:59 , a shot graded at 0.37 xG . That's a high-quality in-tight chance that scores about one time in two and a half. It scored. The buzzer sounded a second later and Vancouver went to the room down 3. Vancouver's HD edge The Goldeneyes generated 10 high-danger shots , including four separate chances above 0.17 xG: Mannon McMahon 's 0.21 snap at P1 13:08, Tereza Vanišová 's 0.19 slap and Gabby Rosenthal 's 0.19 snap both on the Vancouver power play at P2 4:53 and 4:46, and another McMahon 0.17 slap on the same PP. None of the four went. Vancouver's PP overall produced 5 shots for 0.62 xG across 2 opportunities, the kind of volume that usually cashes at least once. Michelle Karvinen 's 0.10 xG tip at P2 12:35 was the single Vancouver shot Rooney didn't stop. Rooney, decomposed Maddie Rooney faced 32 shots worth 2.48 xG and allowed one live-play goal for +1.48 GSAx . That's well-above-expectation save-process across a heavy workload. Nine of Vancouver's ten high-danger chances stopped, including the three on the second-period power play that the model graded at a combined 0.55 xG. When the Vancouver offense was at its most dangerous, Rooney was at her steadiest. That's the pattern that produced a one-sided scoreboard on a close underlying game. Kristen Campbell stopped 18 of 21 live shots worth 2.04 xG, finishing at -0.96 GSAx . Save-process below expectation, mostly on Pannek's 0.40 PP shot getting through. Two mid-single-digit xG shots (Cooper 0.09 and Curl-Salemme 0.09) also beat her. Across three regulation goals on 2.18 xGA of live workload, her line runs about a goal below the model's save expectation, with the buzzer-beater doing most of that work. The game in cumulative xG The scoring timeline P1 4:57 · Kendall Cooper · xG 0.08 · MIN 1, VAN 0. P1 17:07 · Britta Curl-Salemme · xG 0.09 · MIN 2, VAN 0. P1 19:59 · Kelly Pannek (PP) · xG 0.37 · MIN 3, VAN 0. P2 12:35 · Michelle Karvinen · xG 0.11 · MIN 3, VAN 1. P3 19:26 · Britta Curl-Salemme (EN) · xG 0.00 · MIN 4, VAN 1 Top scoring chances (excluding empty-net shots) VAN P2 4:53 · Tereza Vanišová slap · xG 0.20. VAN P1 13:08 · Mannon McMahon snap · xG 0.19. VAN P2 4:46 · Gabby Rosenthal snap · xG 0.19. VAN P2 4:41 · Mannon McMahon slap · xG 0.18. VAN P3 13:35 · Claire Thompson backhand · xG 0.15. VAN P3 18:20 · Hannah Miller tip · xG 0.14. MIN P1 19:59 · Kelly Pannek default · xG 0.37 (goal). MIN P1 4:07 · Kelly Pannek snap · xG 0.21. MIN P1 19:23 · Grace Zumwinkle tip · xG 0.20. MIN P1 6:59 · Peyton Anderson tip · xG 0.14. MIN P3 4:14 · Katy Knoll wrist · xG 0.12. MIN P1 1:06 · Grace Zumwinkle wrist · xG 0.11 What this read isn't Not a dominant Minnesota process win. Vancouver outshot, out-HD'd, and won Corsi. Minnesota won the game by converting efficiently and getting a +1.48 GSAx night from Rooney. The underlying play was closer than the final.. Not a full indictment of Kristen Campbell . A -0.96 GSAx on 21 live shots is one game's variance, and the Pannek 0.40 chance was going to score more often than not. The regulation Curl-Salemme goal was a lower-quality slap that happened to go.. Not a sign Vancouver's power play is broken. Five shots and 0.62 xG on two opportunities is strong process. The finishing didn't land; the unit's chance-creation was healthy.. Not a model failure. A 4-1 scoreline on a 3.14-to-2.54 xG split with the empty-netter included is well inside the variance the model's probability outputs allow, especially when the save-process line lands on one side. 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/61 .