Colorado won 8-2, but expected goals said barely half a goal of advantage: 3.24 to 2.60. The Avalanche overperformed their xG model by 4.76 goals—a chasm between scoreboard and prediction.

The underlying play favored Colorado but not decisively. They held 52% of shot attempts and won high-danger chances 7–4. Their shot quality mattered: 0.051 xG per shot versus Ottawa's 0.039, a clear edge in converting attempts to dangerous chances. But goaltending tilted everything. S. Wedgewood (29/31, 0.935 SV%) played to expectation. M. Sogaard (11/16, 0.688 SV%) collapsed—3.47 goals worse than expected. L. Meriläinen (15/18, 0.833 SV%) was 1.28 under expected.

That's where the game was won. Colorado scored five low-danger goals, shots goaltenders should stop, plus six more on the rush from 28 attempts. Ottawa crashed the net on five rebound chances and converted zero. Colorado crashed on three and converted zero. Volume and finishing on marginal opportunities decided the blowout.

The power play was secondary: two goals on 15 shots, 1.47 expected goals. Dominant but not decisive.

By the numbers: Colorado played well, Sogaard imploded, and Colorado buried chances that barely register in the model. The 6-goal gap is where underlying metrics and results diverged.

Models project this as a 2.5–2.7 goal game with Colorado at 54% win probability—genuine toss-up territory.

The underlying metrics explain the stalemate. Ottawa owns 56.0% expected goals and 53.0% Corsi; Colorado counters with 57.0% in both. Neither team has pulled decisively ahead at even strength. The gap lies in execution: Colorado's PDO of 1.021 means they're running hot and facing regression, while Ottawa's 0.993 suggests untapped upside if finishing normalizes.

Colorado's Corsi advantage (57.0%) guarantees a territorial tilt, but it hasn't yet produced the xG gap that would make them clear favorites. Ottawa can blunt that zone time by limiting high-danger chances. Neither team scores from rebounds or scrambles—both sit at 0%—so this matchup hinges on quality finishes, not chaos.

The goaltending factor: S. Wedgewood's back-to-back could mean a backup start or an exhausted starter. Fatigue compounds the scoring variance already embedded in close games.

Colorado's attack centers on Artturi Lehkonen (1.11 xG/60, 18 goals, 6 high-danger goals at 14.9% shooting). That finishing rate won't hold, but his high-danger output signals genuine quality. Ross Colton (1.10 xG/60, 7 goals, 1 high-danger goal, 4.9% shooting) underperforms his chances—Colorado's inverse problem.

Ottawa counters with Brady Tkachuk (1.19 xG/60, 11 goals, 2 high-danger goals at 7.5% shooting), generating elite volume while underperforming, leaving room to run. Dylan Cozens (1.00 xG/60, 13 goals, 1 high-danger goal, 9.5% shooting) profiles as steadier—hitting his expected value.

Defensively, Nick Jensen (Ottawa) and Nazem Kadri (Colorado) become liabilities under sustained forechecking pressure. In tight games, those breakdowns cascade.

Colorado owns the territorial edge and the hot hand, but neither advantage is large enough to overcome the variance inherent in close, even hockey. That's why 54% feels right.