L. Thompson stole this one. The Hurricanes edged the Capitals 3-2 in a shootout despite commanding the underlying numbers — and Thompson's 37-save performance (0.949 SV%, +3.09 above expected) was the difference.

Carolina dominated the 5v5 play. They controlled 62% of expected goals (5.09–3.08), held shot attempts to 39% (25–39), and created the more dangerous chances (6 high-danger shots to Washington's 7). B. Bussi was solid in goal (23/25, 0.920 SV%, +1.08 above expected), but the Capitals' goalie was phenomenal — 100% on high-danger chances, a wall when the Hurricanes pressed hardest.

Where Carolina stumbled: they couldn't finish. 0 for 7 on high-danger looks. They also underperformed their underlying metrics (2.09 below xG), wasting the quality they created.

Washington, meanwhile, squeezed everything from their limited opportunities. Off the rush — where they found just 28 chances — they scored twice. Carolina had 36 rush attempts and also scored twice, a sign of how marginal Washington's efficiency was everywhere except between the pipes.

Seth Jarvis's snap shot from the doorstep decided it in the shootout, but Thompson nearly stole that too.

The model projects a CAR 2.8 - 2.6 WSH game with the Hurricanes at 54% win probability. This is about as close to a pick'em as you'll find.

Carolina dominates the underlying metrics. The Hurricanes post a 56.0% xG% against Washington's 50.0%, and drive territorial control at 60.0% Corsi to the Capitals' 49.0%. Expect them to dictate pace and possession.

Washington's strength is recent form. They're 8-1-1 in their last 10 games—the far hotter team—and they enter with three days of rest to Carolina's one. B. Bussi is on a back-to-back, which muddies that advantage; watch for either a tired starter or a backup in goal.

The regression signals work in opposite directions. Washington's 1.018 PDO indicates they're running hot and overdue for negative variance. Carolina's 0.984 PDO means they're running cold and due for positive regression. Neither team has a luck buffer to spare.

Offensively, the talent is close but tilts different ways. Jackson Blake leads Carolina with 1.06 xG/60, 14 goals, and 19 assists. Seth Jarvis matches the creation rate at 0.98 xG/60 and has 17 goals and 12 assists. Washington counters with Alex Ovechkin at 0.97 xG/60, 19 goals, and 17 assists—slightly behind in expected output but ahead in actual results. Anthony Beauvillier adds depth at 0.89 xG/60.

The critical unknowns: Can Carolina contain Ovechkin without exposing Charles Alexis Legault as a weak link? Can Washington manage the Blake-Jarvis combination while Declan Chisholm's on the ice?

This is competing edges at their purest—elite skill and upside for Carolina against current form and rest for Washington. The model splits the difference, and so should you.