3-2 (OT): Final (MTL). 1.82-1.82: xG (MTL-OTT). 8-6: HD chances (MTL-OTT). 1-0 MTL: Walter Cup Final The Walter Cup Final opened with the most even game the model has graded all postseason. Expected goals (xG) is the chance a shot becomes a goal, added up across every attempt: Montréal finished at 1.82 , Ottawa at 1.82 . That gap, four hundredths of a goal, is as close to a coin flip as a hockey game gets. Ottawa played it like the slightly better team, won the shot-attempt share, and led 2-1 inside the final four minutes on two Rebecca Leslie goals. Then Nicole Gosling , a defender, tied it with three seconds left in regulation, and Abby Roque won it early in overtime. Montréal leads the best-of-five 1-0 out of a game that, by every process number, could just as easily have gone the other way. Five goals, and not one was a good bet Five pucks went in, and the model priced every one of them as unlikely. The five goals came on shots worth 0.61 xG combined . To put that in scale, a clean look from the middle of the slot is often worth more than 0.20 on its own, so this was five goals out of a little more than half of one expected goal. The night's single best chance, Leslie's opener, did beat Ann-Renée Desbiens . So did the third-best, Roque's overtime tip. But the looks the model liked next, Hayley Scamurra 's wrister, a Marie-Philip Poulin tip, a Michela Cava chance, all stayed out. Three of the five goals came on shots valued between 0.07 and 0.11 xG, the kind of attempt that scores maybe once in ten or twelve tries. When that many low-probability shots find the net in one game, the scoreboard stops tracking which team played better. The chances, on the rink 5-10: P1 shots. 8-8: P2 shots. 11-7: P3 shots. 2-0: OT shots Read the period split above and the swing is clear. Ottawa pushed first, outshooting Montréal in the opening period; the game leveled in the middle; and Montréal carried the third, when it was chasing the game. The Charge also won 59 percent of faceoffs and 54 percent of shot attempts . None of those edges is large, and none of them is nothing. Ottawa was the marginally better team and has a one-goal overtime loss to show for it. The goals, in order Rebecca Leslie opened the scoring at P2 16:56 with the best shot of the night, a wrister worth 0.20 xG , unassisted. Abby Roque answered at P3 12:12 off feeds from Nadia Mattivi and Laura Stacey , a 0.07 xG shot, to make it 1-1. Leslie restored the Ottawa lead at P3 15:56 , a 0.07 xG look set up by Sarah Wozniewicz . That is where the game sat, 2-1 Ottawa, with the clock under a minute. Three seconds Nicole Gosling is a defender. She scored the biggest goal of Montréal's season so far, at P3 19:57 , with three seconds left in regulation. The shot graded 0.11 xG , a one-in-nine look, off assists from Maureen Murphy and Marie-Philip Poulin . It did not tie a game so much as rescue a series position: a regulation loss puts Montréal down 0-1 with the next game in hand-wringing range. Instead the building got overtime. Abby Roque ended it at P4 2:29 , tipping a Laura Stacey shot past Gwyneth Philips for a 0.16 xG goal, her second of the night. Each team got two goals from one player. Roque's second is the one that counts in the standings. The game in cumulative xG Win probability: Montréal fell to 10 percent The live win-probability model spent most of the night unable to separate the teams. It read near 50 percent through two periods. Then Leslie's second goal put Ottawa ahead 2-1 with time running down, and Montréal's win probability fell to about 10 percent . The Gosling goal with three seconds left is the near-vertical line back to even. Overtime did the rest. Goaltending Gwyneth Philips stopped 23 of 26 and finished at -1.18 GSAx . GSAx, goals saved above expected, compares the goals a keeper allowed to the summed xG of the shots she faced; a negative number means more went in than the model expected. Read on its own, -1.18 looks like a poor night: the three goals against were worth just 0.34 xG combined , so the model says she should have had them. But two of the three reached her through traffic the model cannot see, Roque's overtime tip and Gosling's shot through bodies in the final seconds. Philips did not lose this game. She had a slightly below-par night in a game where slightly below par was enough to lose it. Ann-Renée Desbiens stopped 23 of 25 for -0.18 GSAx , a par playoff line. She allowed Leslie's two goals, the 0.20 xG opener and a 0.07 xG follow-up, and was not made to steal anything: 25 shots worth 1.82 xG is a manageable Game 1 workload. Neither goalie won or lost this game. That is the point of the night. The scoring timeline P2 16:56 · Rebecca Leslie · xG 0.20 · MTL 0, OTT 1. P3 12:12 · Abby Roque · xG 0.07 · MTL 1, OTT 1. P3 15:56 · Rebecca Leslie · xG 0.07 · MTL 1, OTT 2. P3 19:57 · Nicole Gosling · xG 0.11 · MTL 2, OTT 2. P4 2:29 · Abby Roque · xG 0.17 · MTL 3, OTT 2 Top scoring chances MTL P3 12:44 · Hayley Scamurra wrist · xG 0.19. MTL P4 2:29 · Abby Roque tip · xG 0.17 (goal). MTL P1 2:44 · Marie-Philip Poulin tip · xG 0.16. MTL P3 19:08 · Laura Stacey default · xG 0.13. MTL P2 7:34 · Alexandra Labelle snap · xG 0.13. MTL P3 19:57 · Nicole Gosling wrist · xG 0.11 (goal). OTT P2 16:56 · Rebecca Leslie wrist · xG 0.20 (goal). OTT P3 7:03 · Michela Cava wrist · xG 0.14. OTT P1 5:10 · Brianne Jenner tip · xG 0.13. OTT P1 11:29 · Brianne Jenner backhand · xG 0.13. OTT P3 2:46 · Brianne Jenner tip · xG 0.12. OTT P1 19:21 · Gabbie Hughes wrist · xG 0.10 What this read isn't Not a Montréal statement. The Victoire were three seconds from losing Game 1 on home ice. They won a 50/50 game on the bounces, which counts the same in the standings as winning a blowout, but the process was even at best: 1.82 xG to 1.82. The series lead is real. The performance gap behind it is not.. Not an Ottawa collapse. The Charge generated 1.82 xG, the marginally better number, won the faceoff and shot-attempt battles, and led for most of the third period. They lost on a defender's 0.11 xG shot with three seconds left and a tip in overtime. That is variance, not a structural problem to fix.. Not a goalie story. Neither Gwyneth Philips nor Ann-Renée Desbiens posted a GSAx that explains the result. The game turned on five low-probability shots and which net they found, not on a goaltender stealing or surrendering it.. Not a series referendum. Game 1 of a best-of-five is a single data point. It moves Montréal's series win probability and it earns the Victoire the first of the three wins they need. It does not tell you who the better team is, because tonight the game itself could not tell you that either. 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 count.. Win probability : the in-game model behind the trace updates after every event from score, time remaining, and game state. The shot map and cumulative-xG charts read straight from the play-by-play.. Numbers regenerate on every build. A PBP re-sync or xG recalibration reshapes every figure and every chart 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/318 . The series projection, including how this Game 1 result moves Montréal's series win probability, lives at /playoffs/bracket .