2-1: Final OT (BOS). 1.77-1.74: xG (BOS-VAN). 7-7: HD (BOS-VAN). +0.74: Frankel GSAx This was as even a regulation game as the season has produced on the chance-quality split. Boston and Vancouver finished within 0.02 xG of each other ( 1.77 to 1.74 ), matched at 7 high-danger chances each , and went 0-for-7 combined on the power play. Three goals total in regulation and overtime, all on shots the model graded under 0.13 xG. Shay Maloney 's wrister 41 seconds into overtime decided it. Three goals on 0.30 combined xG Haley Winn 's 0.06 xG wrister at P3 3:19 broke a scoreless first 43 minutes. Hannah Miller answered for Vancouver at P3 16:34 with a 0.04 xG attempt, the lowest-quality scoring shot of the night. Shay Maloney won it 41 seconds into overtime on a 0.12 xG attempt. Combined: 0.30 xG of expectation across three scoring shots . None of the three was a true high-danger one-timer; the model's read on the chance-quality split was always going to leave the result up to which mid-quality look found space first. Even special teams, no special-teams goals Boston had 3 power-play opportunities and converted 0. Vancouver had 4 and converted 0. Seven combined PP opportunities yielded zero goals and washed out the special-teams ledger entirely. Both PP units generated reasonable volume but neither produced the kind of in-tight high-danger one-timer that converts at high rates. The decision came at 5-on-5 and in OT. Goaltending, decomposed Aerin Frankel stopped 25 of 26 live shots worth 1.74 xG, finishing at +0.74 GSAx . The lone goal she gave up was Miller's 0.04 xG attempt; she stopped each of Vancouver's seven high-danger looks. The above-zero GSAx on a 26-shot workload was the small edge that gave Boston the chance to win in OT. Kristen Campbell stopped 25 of 27 live shots worth 1.77 xG, finishing at -0.23 GSAx . The Winn 0.06 xG wrister and Maloney 0.12 xG OT winner were both shots in the kind of mid-quality range that score sometimes; the save process line came out essentially flat against expected. The two ends of the rink played close to even on goalie performance, with Frankel a bit ahead. The game in cumulative xG The scoring timeline P3 3:19 · Haley Winn · xG 0.07 · BOS 1, VAN 0. P3 16:34 · Hannah Miller · xG 0.05 · BOS 1, VAN 1. P4 0:41 · Shay Maloney (OT) · xG 0.17 · BOS 2, VAN 1 What this read isn't Not a claim either team deserved to win. The chance-quality split was within the noise of a single-game estimate. The result fell on Frankel's small save edge and a Boston OT goal that arrived 41 seconds in.. Not a single-game indictment of Kristen Campbell . Two goals on shots in the 0.04 to 0.12 xG range with essentially-zero GSAx is a clean save process line. The losing variable wasn't her save process; it was a tied chance battle that needed an OT goal somewhere.. Not a claim either special-teams unit is broken. 0-for-7 combined on small samples is variance; both PP units produced reasonable volume, neither broke through.. Not a model failure. A near-even chance-quality game ending 1-1 in regulation with an OT decision is well within the variance the model's probability outputs allow. 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/72 .