2-3: Final (VAN). 5-8: HD (TOR-VAN). 1.96-1.68: xG (TOR-VAN). -0.04: Maschmeyer GSAx This was a low-event game where the chance-quality split and the territorial split disagreed. Toronto generated more xG (1.96 to 1.68) on a smaller shot volume; Vancouver carried play at 5-on-5 (53% Corsi) and produced 8 high-danger chances to Toronto's 5. Three goals in 8:19 of the second period settled the scoring, and the decisive sequence was a 45-second window with two power-play goals from opposite teams. Two opposite power plays, 45 seconds apart At P2 8:09 Blayre Turnbull tied the game on a 0.17 xG 5-on-4 backhand, her second of the night. Toronto then took a penalty, and Vancouver's resulting power play produced a Mannon McMahon tip-in at P2 8:54 (45 seconds later) on a 0.07 xG attempt. Two consecutive opposite power plays converted in under a minute is a rare line-item; in a 60-minute regulation game where neither team had drawn more than two PP opportunities, both went 1-for-2, and the swing came out a goal in Vancouver's favour. McMahon's was the game-winner; nothing scored across the next 31 minutes of game-clock. Five goals in the same xG band All five regulation goals came on shots the model graded between 0.07 and 0.19 xG. None were in-tight one-time high-danger looks; none were broken-coverage scrambles. Five mid-quality conversions in 60 minutes is a normal two-team scoring rate by chance quality, and a slightly above-rate finishing day on combined volume of 56 shots. Goaltending, decomposed Emerance Maschmeyer stopped 25 of 27 live shots worth 1.96 xG, finishing at -0.04 GSAx . The Toronto chances she gave up were both above 0.16 xG, the kind of mid-quality looks any goalie is expected to concede at this rate; her save line on everything else kept the game close enough for the special-teams swing to decide it. Elaine Chuli faced 29 live shots worth 1.68 xG and conceded three. Her GSAx came out at -1.32 , on the wrong side of expected. None of the Vancouver scoring shots she saw was over 0.19 xG, but they all found space. On another night, one of the three holds out and the result turns the other way. The game in cumulative xG The scoring timeline P1 2:32 · Blayre Turnbull · xG 0.16 · TOR 1, VAN 0. P1 7:50 · Hannah Miller · xG 0.19 · TOR 1, VAN 1. P2 0:35 · Izzy Daniel · xG 0.18 · TOR 1, VAN 2. P2 8:09 · Blayre Turnbull (PP) · xG 0.16 · TOR 2, VAN 2. P2 8:54 · Mannon McMahon (PP) · xG 0.08 · TOR 2, VAN 3 Toronto's looks above 0.08 xG Every Sceptres attempt the model graded above the 8% scoring-probability threshold. Two of them went in (both Turnbull goals). The rest are the chances that Maschmeyer turned away or that missed the net. The ceiling is around 0.20 xG, which is the xG distribution of an offence that didn't get a true high-danger look all night. P2 15:48 · Blayre Turnbull · xG 0.27. P1 2:32 · Blayre Turnbull (goal) · xG 0.16. P2 8:09 · Blayre Turnbull (goal) · xG 0.16. P3 14:03 · Blayre Turnbull · xG 0.16. P1 18:35 · Emma Maltais · xG 0.15. P3 6:41 · Emma Gentry · xG 0.10 What this read isn't Not a claim Toronto deserved a different result. The chance-quality split was within 0.4 xG and Vancouver took the high-danger battle 8 to 5. The model's read is consistent with a tight one-goal game that could go either way, and on this night it went Vancouver's way.. Not a claim Elaine Chuli played poorly in a structural way. A -1.3 GSAx night against three goals all under 0.20 xG is mostly variance landing on her side. The save process was much closer to expected than the headline number suggests.. Not a claim either special-teams unit is on a hot streak. Two opportunities each is a tiny sample; both teams converting once is mostly noise around the rate of an average power play. The 45-second window is the kind of single-game pattern that doesn't repeat.. Not a model failure. A near-even xG game ending 3-2 in regulation with no high-danger conversion outliers is exactly the kind of variance the model's probability outputs allow on small samples. 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.. 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/91 .