The Penguins won a game they had no business winning. Tampa Bay dominated the expected-goal battle 5.47–3.27, controlling 63% of xG at 5v5 — yet Pittsburgh escaped with a 4-3 victory.

This was a triumph of goaltending and variance. T. Jarry made 37 saves on 40 shots at 0.925 SV%, stopping 2.47 goals above expected. J. Johansson made 27 saves on 31 shots at 0.871 SV%, allowing 0.73 goals above expected. That swing decided the game.

The Lightning controlled everything else. They held 56% of shot attempts (40–31 on goal) and led in high-danger chances 9–5. Tampa Bay generated 7 rebound opportunities—evidence of net-front aggression—yet converted none. They were also lethal on the rush, scoring 3 goals on 38 rush chances, but underperformed their xG by 2.47 goals.

Pittsburgh's edge came from efficiency on the rush: 4 goals on 26 rush chances. The Penguins were also dominant on the power play, scoring twice on 8 shots with 0.75 expected goals from that unit.

Evgeni Malkin provided the turning point, wrist shot from the slot in the third period at 17:17. He finished with 2 goals and 1 assist in 14:58.

The model calls this a pick'em: 2.5-2.7 goal differential with the Lightning at 55% to win. Neither team has a clear edge.

Tampa Bay owns the underlying play. At 54.0% 5v5 expected goals and 53.0% Corsi, they control possession in a way Pittsburgh—which sits at 51.0% xG and 50.0% Corsi—simply doesn't. Expect territorial tilt in the Lightning's favor.

That said, Tampa's dominance isn't as wide as it appears. A four-point gap in xG% and three-point lead in Corsi% is meaningful but not overwhelming. Pittsburgh has the ability to stay close despite the possession disadvantage. What tips this toward Tampa is recent form: the Lightning are 8-2-0 in their last 10 games, a momentum advantage Pittsburgh doesn't match.

The real complication is Tampa's underlying metrics are buoyed by unsustainable luck. Their PDO sits at 1.019 compared to Pittsburgh's 1.012. That extra heat isn't skill; it's variance working in their favor. Expect some regression.

Vasilevskiy faces a back-to-back start. Monitor whether Tampa rolls with the backup or runs its starter tired—a fatigued Vasilevskiy is still elite, but it's worth tracking.

Players to Watch:

Rickard Rakell leads Pittsburgh's attack at 1.01 xG per 60 minutes with 13 goals, 13 assists, and three high-danger goals on a 12.1% shooting rate. Filip Hallander (1.28 xG/60) is likely out with injury evaluation. Watch Rakell to carry load.

For Tampa, Brandon Hagel generates 1.10 xG per 60 with 23 goals, 26 assists, and a 15.8% shooting rate backed by two high-danger goals. Anthony Cirelli (0.95 xG/60) converts at 17.0% with five high-danger tallies. Both will be featured in a game Tampa expects to control.

Both teams leave scoring chances on the table—each converts high-danger chances at 0%, which is unsustainable in either direction. Defensively, Pittsburgh's weak link is Caleb Jones; Tampa's is Victor Hedman. These vulnerabilities won't swing the game alone.

The Lightning should win this on merit. They'll dominate possession and generate better scoring chances. But the gap is narrow enough that any variance—a hot night from Rakell, a soft goal from Vasilevskiy—flips the outcome. That's why the model sees it as a flip.