The Golden Knights' 3-2 win over the Stars masked a far more dominant underlying performance. Vegas controlled 62% of expected goals (3.98 to 2.41)—the kind of xG edge that suggests the scoreline flattered Dallas.

The Stars managed only 31% of shot attempts (15 to 33), generating 38% of expected goals despite losing by one. C. DeSmith kept Dallas in it with 30 saves on 33 shots (0.909 SV%), performing +0.98 above expected—an outlier performance. A. Hill stopped 13 of 15 shots (0.867 SV%) for Vegas.

But Vegas won on efficiency. The team scored 3 goals on 34 rush chances while generating 7 high-danger opportunities to Dallas's 5. Reilly Smith's snap shot from the doorstep made it 2-1 Vegas—a moment that epitomized the match: Vegas attacking from scoring areas; Dallas defending in a hole.

The Stars fought back by attacking the net, generating three rebound opportunities and converting one, but couldn't overcome the underlying deficit. M. Heiskanen led Dallas with an assist in 26:08, while Vegas's M. Marner (22:14) and S. Theodore (21:28) both chipped in assists.

DeSmith's strong night prevented the margin from expanding, but even a 0.909 SV% cannot erase a 33-15 shot disadvantage and a 7-5 edge in high-danger chances. The underlying numbers confirmed what the score should have been: a Vegas blowout.

The model projects a 2.4-2.3 Golden Knights victory with Vegas at 51% win probability—as close to even odds as these things come.

Dallas enters hot: 7-1-2 over their last ten games. Vegas, by contrast, is underwater at -6 goal differential in the same span. That recency shouldn't fool you. The underlying data runs the other direction.

Vegas dominates possession—54.0% expected goals and 52.0% Corsi at 5v5, versus Dallas's 51.0% xG% and 47.0% Corsi%. Expect the Golden Knights to control play, which they'll need to overcome Dallas's efficiency edge. The Stars' PDO sits at 1.024, running hot enough to expect regression. Vegas checks in at 0.983, due for positive regression from the other direction.

Goaltending presents a variable: both J. Oettinger and A. Schmid are on back-to-backs, so backup duty or tired starters remain in play.

Vegas's offensive depth will be tested. Tomas Hertl carries 0.88 xG per 60 with a team-leading 5 goals and 17 assists, hitting at 4.3%. Brandon Saad, who otherwise rates at 0.91 xG/60 with a 4.1% shooting touch, is marked likely inactive, reshaping what Vegas can throw at Dallas.

Dallas counters with Jason Robertson—0.96 xG/60, 25 goals, 21 assists, and a lethal 12.4% shooting rate. Michael Bunting (8 goals, 7.6% shooting) adds secondary depth at 0.83 xG/60.

Defensively, Vegas's weak link is Rasmus Andersson; Dallas's is Tyler Myers. Both teams sit at 0% high-danger and rebound goal rates, suggesting structured defending that eliminates the scramble game.

The verdict lands on process: Vegas has the underlying play but Dallas has the recent form. With goaltending uncertainty and Dallas riding hot, this genuinely teeters. The model's 51% for Vegas carries legitimate conviction, but only just.