4-0: Final (BOS). 2: Toronto HD. 0.94: Toronto xG. +0.94: Frankel GSAx Toronto produced 0.94 xG across all 60 minutes at home. The model logged 2 high-danger chances on the night, the lowest single-game total in this series of reviews. Boston's offence wasn't running away with it either, finishing at 2.16 xG, but they had 6 high-danger looks of their own and converted 4 of them on shots the model priced between 0.02 and 0.25 xG. Aerin Frankel closed the back end with a clean shutout. Four Boston goals on 0.61 combined xG The Fleet's four scoring shots came at 0.19, 0.02, 0.13, and 0.25 xG . Combined: 0.61 xG of expectation against four actual goals, which works out to 1.85 times expected on a per-shot basis. The 0.02 xG goal in the second period was Liz Schepers 's second of the night, a wrister the model said scores roughly one time in fifty. Adding her 0.19 xG opener, Schepers finished the night on 0.24 combined xG and two goals . Across a season that conversion rate sits much closer to 1.0; in a single game on two shots it can land anywhere. Toronto's chance-quality ceiling The Sceptres' attack had the shape of a team that struggled to get inside. Eighteen total shots produced only two attempts the model graded as high-danger , and the average xG-per-shot ran around 0.052. Compare that to a typical PWHL game where the average shot xG is closer to 0.07-0.08; Toronto's distribution sat below average across nearly every attempt. They went 0-for-2 on the power play and didn't generate a single PP shot the model graded above 0.10 xG. Goaltending, decomposed Aerin Frankel stopped all 18 live shots she faced for 0.94 xG of workload. That is +0.94 GSAx , an above-average single-game line on a thin volume rather than an outlier-shaped night. Toronto's chance distribution did most of the work for her: only two in-tight looks, both turned aside without much drama. Elaine Chuli faced 23 live shots worth 2.16 xG and conceded four. Her GSAx came out at -1.84 . Three of the four goals against her were under 0.20 xG, including one under 0.03; that is mostly variance landing on her side of the ledger rather than a structural save problem. On a night where the team in front of her generated essentially no offence, the GSAx number reads more negative than the underlying performance was. The game in cumulative xG The scoring timeline P1 11:38 · Liz Schepers · xG 0.22 · BOS 1, TOR 0. P2 10:54 · Liz Schepers · xG 0.02 · BOS 2, TOR 0. P2 14:32 · Jessie Eldridge · xG 0.13 · BOS 3, TOR 0. P3 14:14 · Sophie Shirley · xG 0.23 · BOS 4, TOR 0 Toronto's six best looks Every Sceptres attempt the model graded above the 5% scoring-probability threshold. The list is short and the ceiling is low: nothing above 0.20 xG. That is the shape of an offence that didn't generate a true high-danger chance all night. P2 12:40 · Emma Maltais · xG 0.12. P3 3:48 · Blayre Turnbull · xG 0.11. P3 15:23 · Emma Maltais · xG 0.10. P3 12:58 · Savannah Harmon · xG 0.09. P1 5:58 · Jesse Compher · xG 0.08. P2 19:05 · Maggie Connors · xG 0.07 What this read isn't Not a claim Elaine Chuli played a losing line. Three of the four goals against her were under 0.20 xG, one under 0.03. The headline GSAx number reads worse than the save process underneath. The team-level offensive profile is the larger story.. Not a claim Toronto's offence is broken. One quiet game on chance generation against a structured Boston defence is one data point. The right test is whether this xG-per-shot profile shows up across the next several games or normalises back to league-average.. Not a claim Boston dominated. The Fleet generated 2.26 xG, which is a normal regulation-game offensive output. Their finishing rate ran above expected; the 4-0 score line is finishing variance plus the shutout, not a possession blowout.. Not a claim the model is wrong. A 60-40 chance-quality favourite winning 4-0 with above-average finishing and an above-average save total is exactly the kind of single-game outcome the model rates as most likely under the conditions. 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/89 .