The Golden Knights won 5-3 despite the underlying numbers suggesting a near-even game. Expected goals split 3.75–3.33 in Vegas's favor, but the scoreboard diverged—the Knights outscored their model by 1.25 goals while Columbus underperformed.

A. Schmid (VGK) outperformed his xG model by 1.20 goals. J. Greaves (CBJ) underperformed by 1.25. The split in goaltending efficiency decided the game.

Vegas dominated the chance battle. High-danger opportunities favored the Knights 6–4. The visitors generated 0.074 xG per shot versus Columbus's 0.052—quality over volume. The Knights' rush attack was lethal, converting 5 of 25 chances; Columbus scored twice on 27 rush attempts, a sharp efficiency gap.

The Knights' power play supplied two goals on nine shots, an expected value of 1.74 goals. That created a ceiling Columbus couldn't reach even with a solid offensive night. Vegas also scored two low-danger goals—exactly the kind of mistakes that shift close games.

Brett Howden's wrist shot from the slot in the third period (16:14) was the turning point, the moment Vegas pulled clear.

The model projects a CBJ 2.4-2.6 VGK game with the Golden Knights at 54% win probability—essentially a flip of the coin. Vegas holds a slim edge in underlying metrics: 54.0% xG% and 52.0% Corsi% to Columbus's 52.0% xG% and 51.0% Corsi%. The gap is marginal enough that neither team can afford to assume superiority; execution will matter far more than raw talent.

Vegas is running cold. Their PDO sits at 0.983 while Columbus sits at 1.003—a full twenty-point gap that suggests the Knights are substantially due for positive regression if their underlying play remains sound. That regression probability matters more than any single score differential in determining this matchup's outcome.

Players to Watch

CBJ

• Boone Jenner — 0.98 xG/60, 11G 21A, 2 HD goals, 9.6% shooting
• Mathieu Olivier — 0.89 xG/60, 12G 9A, 1 HD goals, 13.3% shooting

VGK

• Brandon Saad — 0.91 xG/60, 3G 6A, 1 HD goals, 4.1% shooting
• Tomas Hertl — 0.88 xG/60, 5G 17A, 1 HD goals, 4.3% shooting

The strategic battleground centers on two conflicts. Jenner drives Columbus's entire offensive system and must be neutralized; Olivier appears due for regression and can't be Columbus's primary offensive source against Vegas. For Vegas, Saad hasn't converted his underlying play into results, while Hertl's assist total masks a deeper scoring drought. Vegas's offensive variance is the real vulnerability.

Defensively, the game hinges on structural gaps. Conor Garland represents Columbus's pronounced weak link in his own zone, while Rasmus Andersson's positioning leaves Vegas exposed to forechecking chaos. Both teams have posted 0% high-danger and 0% rebound-goal rates, suggesting either risk-averse play or unusually clean defensive positioning. Whoever exploits these vulnerabilities first dictates the tempo.

The deciding factor is regression. Vegas's 54% win probability is built on the assumption that their PDO normalizes—that cold shooting ends. If the Knights' underlying play holds and results align, they win outright. Columbus has the metrics to compete and is not outmatched, but closing a PDO gap while defending Garland's weakness is a tougher equation than waiting for variance to swing your way. Give this one to Vegas on regression math.