The Capitals won 2–1 and controlled 59% of 5v5 expected goals (5.08–3.55). Yet that superiority masked an inefficiency: they underperformed their xG by 3.08 goals—a gap that would normally cost you a game, not win it.

The difference was shot quality. The Capitals averaged 0.092 expected goals per attempt; the Devils managed 0.052. That disparity meant Washington could afford some waste. They generated seven high-danger chances to New Jersey's three. They crashed for rebounds and created chances from angles the Devils' defense had surrendered. J. Allen (0.963 SV%, +4.08 above expected) kept the game close with elite goaltending, but even he couldn't suppress the volume of dangerous looks.

L. Thompson's own work—0.968 SV%, +2.55 above expected—was excellent and occasionally necessary. Yet the game's arc was set by the Capitals' ability to generate offense from structure and positioning, not just cycle possession.

Jesper Bratt's snap shot (P3, 19:17) was the spark, breaking through first. It wasn't enough. The Capitals' xG dominance was too large to overcome through goaltending alone. Both teams underperformed their chances—the Devils by 2.55 goals—but when one side controls shot quality as thoroughly as Washington did, goaltending can only do so much.

The model projects a 2.5–2.7 goal margin with Washington at 52% win probability. This is about as close to a pick'em as you'll find.

Five-on-five play is dead even. New Jersey and Washington both post 50.0% xG%, and Corsi splits 51.0% to 49.0%—barely distinguishable. Neither team dominates at even strength.

The regression angle favors New Jersey. Washington's PDO of 1.018 indicates they're running hot; New Jersey's 0.971 indicates they're running cold and due for correction. The luck gap should tighten.

Timo Meier leads New Jersey at 1.08 xG per 60 minutes with 13 goals and 3 high-danger chances. Jack Hughes adds 0.88 xG/60 and 18 goals at an 11.5% shooting percentage—efficiency exceeding his chances. Alex Ovechkin contributes 19 goals at 0.97 xG/60 and 13.8% shooting for Washington, while Anthony Beauvillier registers 0.89 xG/60 and 14 goals.

Both offenses share a structural weakness: neither generates high-danger chances or rebound goals efficiently. Second chances matter less; primary scoring opportunities matter more. The margin for error compresses.

New Jersey's defensive vulnerability is Johnathan Kovacevic. Washington's is Declan Chisholm. The team that prevents its opponent's best players from getting clean looks at those two defensemen probably wins.

The model isn't wrong. It's recognizing two evenly matched teams where variance and execution separate winners from losers. Coin flips are rare in the NHL. Trust them.