Montreal controlled the shot sheet—55% of attempts, 27 shots to San Jose's 22—yet lost 4-2. The Sharks won because their chances mattered more. They finished with 57% of expected goals (4.84–3.65), a gap that explains the final score better than any single play.

Shot quality tells the story. The Sharks generated 0.091 xG per shot; Montreal managed only 0.059. Both teams scored twice off the rush: Montreal on 27 chances, San Jose on 26. But when play slowed and chances became rarer, the Sharks made the better looks count.

Both A. Nedeljkovic and J. Dobes outperformed their expected goals against. The Sharks goaltender saved 1.65 goals above expected (0.926 SV%, 25 of 27 saves), while Montreal's J. Dobes saved 1.84 above expected (0.857 SV%, 18 of 21). The Canadiens underperformed their xG by 1.65 goals, wasting volume. The Sharks underperformed by only 0.84. Montreal's wastefulness and San Jose's efficiency decided the game before either goalie's exceptional performance could matter much.

Macklin Celebrini's snapper from the wing in the third period (17:57) proved decisive. He finished with two goals and an assist in 22:11 of ice time. C. Graf added a goal and two assists in 17:10. N. Dobson led Montreal with two assists in 21:29, but could not shift the underlying story: the team that generated better chances and wasted fewer of them won.

The model projects the Canadiens at 52% win probability—a genuine pick'em by any measure. Montreal's narrow edge traces to recent form: they're 6-1-3 over their last ten games, while San Jose limps in at -7 goal differential across the same span. Yet season-long structure sketches a tighter picture. Both teams sit at 48.0% expected goals share, and Montreal's possession advantage (49.0% Corsi%) barely outpaces San Jose's 47.0%.

Goal luck splits the teams. Montreal's PDO of 1.018 indicates they're running hot—outperforming their underlying expected output. San Jose at 0.998 sits near baseline, having absorbed some adverse variance. Regression is historical; Montreal's efficiency cushion won't hold.

The game hinges on playmaking and finish. Cole Caufield (0.97 xG/60, 33 goals, 16 assists, 1 high-danger goal, 20.0% shooting) is Montreal's crux—a sniper who converts high-quality chances into goals at an elite clip. Oliver Kapanen (0.83 xG/60, 19 goals, 12 assists, 4 high-danger goals, 17.8% shooting) deepens their offensive deck.

San Jose's attack relied on Pavol Regenda (1.10 xG/60, 5 goals, 1 assist, 2 high-danger goals, 16.1% shooting) as a centerpiece—but he's marked as likely unavailable. That absence forces Igor Chernyshov (0.91 xG/60, 6 goals, 8 assists, 1 high-danger goal, 15.0% shooting) into the primary playmaking role. Chernyshov's offensive profile is sound, but his defensive play remains a liability.

The loss of Regenda lowers San Jose's ceiling. Montreal carries form momentum and a goal-luck edge, but the underlying metrics resist any compelling narrative. At 52%, the model hedges. It should.