Jets pull away from the Penguins

Pittsburgh and Winnipeg are indistinguishable on paper — the model gives the Penguins 51% win probability in what amounts to a dead heat. The only meaningful difference is that the Penguins own the underlying play.

Pittsburgh's 51.0% 5v5 expected goals share against Winnipeg's 48.0% represents a marginal edge that's their entire advantage. Their Corsi sits at 50.0% to Winnipeg's 49.0%, confirming they control possession and generate scoring chances slightly more often. Pittsburgh's PDO of 1.012 suggests they've gotten lucky with bounces; Winnipeg's 0.998 suggests the inverse — both are due for regression.

The goaltending matchup adds uncertainty. Both C. Hellebuyck and T. Jarry are on back-to-backs, which could mean tired starters or substitutions. In a game this close, that uncertainty matters.

Rickard Rakell is Pittsburgh's critical forward. He has 13 goals on 12.1% shooting with 1.01 xG per 60, pairing volume with efficiency; his three high-danger goals show he scores from the areas that count. Winnipeg's Kyle Connor counters with 25 goals and 30 assists at 13.0% shooting. At 0.83 xG per 60, his production comes from a diverse range of ice and situations — 25 goals without extreme danger concentration reflects a complete scorer. Nino Niederreiter adds secondary scoring at 0.87 xG per 60 with six goals and six assists, including three high-danger goals.

Caleb Jones is Pittsburgh's defensive weak point; Colin Miller plays that role for Winnipeg. Neither team generates scoring from high-danger situations — both stand at 0% HD goal rate and 0% rebound goal rate — meaning both have relied on secondary scoring and goaltending rather than front-of-net dominance.

This game swings on a power play, a goal-mouth bounce, or a line matchup hidden from box-score readers. The model's 51% edge for Pittsburgh isn't confidence — it's nearly a shrug. Expect a tight, low-event game where one team converts marginally better.