Pittsburgh stole this one. The Penguins beat Florida 5–3 despite the Panthers controlling 59% of expected goals at 5.26–3.60 — a heist built on elite goaltending and precision finishing in clutch moments.

The Panthers dominated by volume. They held 68% of shot attempts and 9 high-danger chances to Pittsburgh's 5. But volume without conversion is just noise. Florida managed only 12 shots on goal from all that territory; Pittsburgh converted at an astonishing rate in danger areas, finishing 2-for-5 on high-danger chances while Florida managed 3-for-9.

T. Jarry turned the outcome decisively. He stopped 34 of 37 shots, outperforming his expected save total by 2.26 goals — the kind of performance that swings a game when you're a heavy underdog in the metrics. S. Bobrovsky faced a lighter workload (12 saves on 17 shots) but underperformed his expected goals by 1.40, surrendering key chances when the game was in reach.

Sidney Crosby's wrist shot from a sharp angle midway through the first period opened Pittsburgh's advantage. Crosby finished with 2 goals and an assist over 19:06; E. Karlsson's 2 assists in 22:27 provided continuity on the back end.

Florida demonstrated precision in specific areas — their rush game was lethal, converting 3 goals from 44 rush attempts — but they couldn't sustain pressure when it mattered most. Even their dominant power play (2 goals on 19 shots) couldn't overcome Pittsburgh's elite goaltending and clinical execution in transition.

The model sees this as a pure toss-up: 2.6 goals projected for Pittsburgh, 2.6 for Florida, with the Panthers sitting at exactly 50% win probability. There is no edge here.

The Penguins have the momentum. They're 8-2-0 in their last 10 games, which translates to a pace that beats the rest of the league. That form advantage matters—even against a capable team, Pittsburgh enters with rhythm.

But the underlying metrics suggest the Panthers are no easy target. Pittsburgh leads in shot quality with an xG% of 51.0% to Florida's 49.0%, a gap that matters over time. Florida counters with territory—a Corsi% of 52.0% to Pittsburgh's 50.0%—which means they're controlling play even when generating fewer dangerous chances. The luck factor tells a story too: Pittsburgh's PDO sits at 1.012, a shade above neutral, while Florida's 0.989 suggests some regression risk.

Pittsburgh's top line is carrying the load through Rickard Rakell, who has 13 goals and 13 assists while shooting at 12.1% on chances that average 1.01 expected goals per 60 minutes. That's volume with precision. The x-factor is Filip Hallander, who generates 1.28 xG/60—elite creation—though he carries 0 goals and 3 assists, a stat line that signals he's either snake-bitten or due. Hallander is likely out, recalled from conditioning for further evaluation.

Florida counters with Sam Bennett, who has 18 goals and 19 assists on 0.97 xG/60, a clinical finisher at 13.1% shooting. Brad Marchand (0.93 xG/60, 17G 9A) is their secondary threat, operating at an eye-popping 19.5% shooting rate. That kind of conversion cannot hold; regression is due.

The breakdown point for Pittsburgh: Caleb Jones anchors a defense that has generated 0% of its offense from high-danger chances and 0% from rebounds. That's not just conservative—it's vulnerable. Florida's mirror image, 0% on both counts, creates an equivalent problem. The difference is Marchand can still pull a game loose with one shift; Pittsburgh lacks that wildcard now that Hallander is on the shelf.

This remains exactly what the model said: a genuine coin flip.