1-3: Final (BOS). 2.58-1.39: xG (BOS-OTT). 11-4: HD (BOS-OTT). +1.58: Philips GSAx Game 2 of the Boston-Ottawa semifinal was Game 1 in a funhouse mirror, Boston controlled the chance battle by an even bigger margin than the night before, and Ottawa took the road win anyway. Boston outshot the Charge 31 to 21 , outchanced them 2.58 xG to 1.39 , won the high-danger battle 11 to 4 . Ottawa scored three times on a combined 0.13 xG of recorded shots (excluding the empty-netter). Gwyneth Philips was the difference for the second straight game: 30 saves on 31 shots for +1.58 GSAx . Across the two-game series so far Philips is +2.67 GSAx on 59 shots faced. The series is tied 1-1 heading to Ottawa for Game 3. Three goals on 0.13 xG of real shots The two non-EN Ottawa goals graded 0.03 xG and 0.10 xG . Ronja Savolainen 's point shot at P1 13:16 was the lowest-quality goal scored against Boston this postseason, a sub-30-foot wrister with no traffic. Fanuza Kadirova 's wrister at P2 1:54 sat at the median for an unscreened mid-slot try. The third was an empty-netter from Gabbie Hughes at P3 18:30. Of Ottawa's 21 recorded shots, only one, a Brianne Jenner look at 0.12 xG , graded above 0.10. The Charge didn't generate a high-danger chance. They didn't need to. Two recovery wristers found space behind Frankel that the model would put under each shot's individual ceiling. Boston's 11 high-danger looks, all stopped The Fleet's offence was the team-version of Game 1: heavy, persistent, top-10-percentile in chance creation. Alina Müller had two separate looks above 0.20 xG, a P1 1:56 tip (0.22 xG) and a P2 18:54 wrister (0.22 xG) that came three seconds before Abby Newhook 's 0.18 wrister. Ella Huber added two more above 0.14. Five of Boston's looks above 0.13 xG all hit Philips's pad or glove. The lone goal Boston scored, Megan Keller at P2 19:50, 0.02 xG, was the team's lowest-quality shot of the night that found net. Across the chance distribution there was a clean mismatch: Boston manufactured the playoff process, Ottawa converted the playoff variance. Goaltending, decomposed Gwyneth Philips stopped 30 of 31 live shots worth 2.58 xG, finishing at +1.58 GSAx . The single goal she allowed was the Keller 0.02 xG point shot, a goal she'd stop nine times out of ten and the only blemish on what is otherwise a 30-of-30 line on shots between 0.05 and 0.25 xG. Two-game series totals: +2.67 GSAx , save percentage sitting near .967. That's the best two-game playoff start in PWHL history by save-process metrics, and the workload was non-trivial in both games. Aerin Frankel faced 20 live shots worth 1.39 xG , allowed two for -0.61 GSAx (the empty-net goal isn't counted on either side of the ledger). Both goals against were under 0.10 xG; on a normal night a goalie below the median process would land closer to -1.0 here. The two goals against weren't bad goals, they were the kind of recovery shots that find net once a postseason and both happened to find it on the same night. The game in cumulative xG The scoring timeline P1 13:16 · Ronja Savolainen · xG 0.03 · BOS 0, OTT 1. P2 1:54 · Fanuza Kadirova · xG 0.10 · BOS 0, OTT 2. P2 19:50 · Megan Keller · xG 0.02 · BOS 1, OTT 2. P3 18:30 · Gabbie Hughes (EN) · xG 0.00 · BOS 1, OTT 3 Top scoring chances BOS P2 18:54 · Alina Müller wrist · xG 0.22. BOS P1 1:56 · Alina Müller tip · xG 0.22. BOS P2 18:57 · Abby Newhook wrist · xG 0.18. BOS P1 18:42 · Ella Huber slap · xG 0.15. BOS P3 5:01 · Ella Huber wrist · xG 0.14. BOS P3 12:51 · Alina Müller tip · xG 0.14. OTT P2 5:24 · Sarah Wozniewicz wrist · xG 0.12. OTT P1 16:39 · Fanuza Kadirova wrist · xG 0.12. OTT P2 10:43 · Michela Cava wrist · xG 0.11. OTT P2 14:01 · Rebecca Leslie wrist · xG 0.11. OTT P1 11:56 · Fanuza Kadirova wrist · xG 0.10. OTT P2 1:54 · Fanuza Kadirova wrist · xG 0.10 (goal) What this read isn't Not a Boston process problem. A 2.60 xG home playoff game with 31 shots and 11 high-danger chances is championship-level offence. The Fleet generated more total xG in this loss than they did in their Game 1 win, which already topped the postseason chart. They're shooting 1.16 percent across the two games on a series xG total of 5.67; expected goals at league-average finishing puts them roughly five-for-the-series instead of three. The shot mix is fine; the bounces aren't.. Not a Frankel cratering. Two goals on shots worth a combined 0.05 xG is just variance. Across the two games she's roughly at expectation; the gap on the scoreboard is on the other end of the rink.. Not a road-team gameplan worth admiring. Ottawa scored on shots their own forwards would tell them they don't usually take. The Charge will look at this game and want to manufacture more 0.15+ xG looks for Game 3 even if it loses them volume, variance won this game, process won't win the series.. Not a series referendum either way. The series is 1-1 with two GSAx swings explaining 95 percent of the scoreboard variance. Game 3 in Ottawa is the real first read on whether Boston's chance-quality edge is sustainable or whether Philips can keep stealing. 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.. 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/310 . The series projection, including how Game 2 moves Boston's and Ottawa's series-WP, lives at /playoffs/bracket .