3-2: Final OT (MIN). 3.08-1.81: xG (MIN-TOR). 35-24: Shots (MIN-TOR). +0.08: Kirk GSAx On the chance-quality split and the penalty ledger, this was a comfortable Minnesota night. The Frost outshot Toronto 35 to 24 , outchanced them 3.08 to 1.81 in xG , and took zero penalties to Toronto's three. The score-adjusted Corsi finished 52 percent in Minnesota's favor. And yet the game needed overtime, where Kelly Pannek finally converted Minnesota's territorial pressure into the win 148 seconds in . The one 0.33 chance Toronto needed Toronto generated one high-quality look the whole night and buried it. Emma Woods 's wrister at P1 18:12 , graded at 0.30 xG , was the single biggest scoring chance in the game for either team. It tied the score after Denisa Křížová had opened the scoring eight minutes earlier on a 0.09 xG backhand. The only other Toronto shot above 0.14 xG was Daryl Watts 's 0.12 xG go-ahead wrister at P2 8:04 . Two goals on shots worth a combined 0.45 xG was the entire Toronto offensive night; the rest was volume the Frost's goalie absorbed. Minnesota's power play: one missed look, then two goals Minnesota drew 3 power plays against zero for Toronto, generated 8 shots worth 1.20 xG , and converted twice. The turning point was a 76-second stretch in the third. Sidney Morin missed a 0.27 xG backhand at P3 4:06, the best Minnesota scoring chance at that point in the game. Seventy-two seconds later Taylor Heise tied it on a 0.10 xG slap shot. The big PP chance went begging, then a smaller one went in. That's variance on samples this small. The power-play unit did what it was supposed to do: build enough volume that the model's expectation kept climbing past one goal of actual return. The overtime winner came on another power play. Kelly Pannek 's snap shot at P4 2:28 graded at 0.20 xG , the second-best Minnesota chance of the night after Morin's miss. Abby Hustler had been the team's most persistent threat all game, with four separate looks above 0.15 xG that stayed out. Minnesota built 3.08 xG of pressure and needed it all; Pannek's shot was the one that finally cashed. Goaltending, decomposed Raygan Kirk faced 35 live shots worth 3.08 xG , allowed three, and finished at +0.08 GSAx . That's the goaltending line of a team that got heavily outplayed territorially; the positive GSAx is what kept Toronto in a game they were losing on chance quality. Of the three goals she allowed, none were classic high-danger one-timers. Křížová's 0.09, Heise's 0.10, and Pannek's 0.20 are all in the mid-quality band where save percentage generally ends up close to the expectation. Maddie Rooney had a quieter workload at 24 shots worth 1.81 xG , giving up two goals for -0.19 GSAx . Save-process flat, which is fine on 24 shots and a dominant Minnesota territorial night. The Woods goal was a 0.33 xG look that was going to score often; the Watts goal was a 0.12 shot that got through. Nothing in her line suggests the game was decided at her end. The game in cumulative xG The scoring timeline P1 10:11 · Denisa Křížová · xG 0.07 · MIN 1, TOR 0. P1 18:12 · Emma Woods · xG 0.30 · MIN 1, TOR 1. P2 8:04 · Daryl Watts · xG 0.08 · MIN 1, TOR 2. P3 5:18 · Taylor Heise (PP) · xG 0.11 · MIN 2, TOR 2. P4 2:28 · Kelly Pannek (PP) · xG 0.25 · MIN 3, TOR 2 Top scoring chances MIN P3 4:06 · Sidney Morin backhand · xG 0.27. MIN P4 2:28 · Kelly Pannek snap · xG 0.25 (goal). MIN P2 19:14 · Abby Hustler wrist · xG 0.22. MIN P1 13:10 · Abby Hustler tip · xG 0.19. MIN P1 8:39 · Abby Hustler wrist · xG 0.15. MIN P2 4:32 · Kelly Pannek wrist · xG 0.13. TOR P1 18:12 · Emma Woods wrist · xG 0.30 (goal). TOR P4 0:26 · Daryl Watts backhand · xG 0.14. TOR P2 5:27 · Sara Hjalmarsson wrist · xG 0.12 What this read isn't Not a claim Toronto played poorly defensively. Giving up 35 shots and 3.23 xG is a rough night at the back, but the Sceptres blocked their share (Corsi finished nearly 50-50), and Kirk's +0.08 GSAx was the reason the game ever reached OT.. Not a knock on Maddie Rooney . A -0.19 GSAx on 24 shots is save-process flat, well within the noise of a single game. She made the saves the model expected her to and gave up the ones the model expected her to give up.. Not a referendum on Toronto's five-on-five offense. One 0.33 xG Woods chance in 60 minutes is a small sample, but it's also the kind of outlier that single games are full of. Expect regression in both directions across the season.. Not a model failure. A game where the team with 1.28 more xG wins in OT is exactly where the model's probabilities end up on a split like this. The model gives Minnesota a meaningful edge; the result lands on the expected side of the ledger. 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/71 .