The Rangers' 3-1 victory looks convincing until you check the underlying numbers. Expected goals split 3.60–3.50 in their favor—a virtual dead heat—yet they won by two. Variance, not dominance, explains the margin.

The Rangers generated quality over volume. They held 45% of shot attempts (22 on goal to the Panthers' 27), but converted at a far higher rate: 0.077 expected goals per shot versus 0.049. When both teams generated six high-danger chances each, the Rangers buried their opportunities while the Panthers didn't.

The Panthers underperformed their xG by 2.50 goals—they squandered chances despite out-shooting the Rangers. Mackie Samoskevich's wrist shot from the slot at 19:19 was Florida's lone breakthrough, but it proved insufficient against the Rangers' lethal rush game. New York scored twice off 22 rush opportunities.

Shesterkin was the difference-maker. He stopped 26 of 27 shots (0.963 SV%), outperforming expected performance by 2.50 goals. Bobrovsky made 19 of 21 saves (0.905 SV%) and saved 1.60 goals above expected, but one dominant outing doesn't overcome a collapse in shot quality.

The Rangers' efficiency rewarded execution. A. Fox delivered one goal and one assist over 23:52, while J. Miller chipped in an assist. M. Samoskevich's goal represented Florida's token response to a night when they simply couldn't finish.

• A. Fox (NYR) — 1G, 1A in 23:52
• J. Miller (NYR) — 1A in 17:44
• M. Samoskevich (FLA) — 1G in 17:43

The model projects this at 2.4 to 2.5 goals with New York at 52 percent win probability. This is about as close to a pick'em as you'll find.

Florida's underlying numbers are solid. The Panthers control Corsi at 52.0%, a meaningful territorial advantage. Their expected goals share sits at 49.0%—only a point behind New York's 48.0%. But Florida is -5 on goals over its last 10, meaning puck luck or finish is working against them. The PDO gap tells the story: Panthers at 0.989, Rangers at 1.009, a gap suggesting regression is coming. New York's 52 percent edge lives entirely on recent variance, not structural superiority.

Sam Bennett is Florida's engine at 0.97 expected goals per 60, with 18 goals, 19 assists, and 6 high-danger chances. His 13.1 percent shooting rate is sustainable and marks him as the matchup Florida must control. Brad Marchand, normally dangerous, has been out 23 days and likely won't play. Florida's offense rides Bennett.

New York counters with Adam Sykora, who owns 0.91 xG/60 but only 3 goals and 1 assist on a 20.0 percent shooting rate—a gap that almost always corrects. Gabe Perreault is more reliable: 0.79 xG/60, 8 goals, 11 assists, 2 high-danger chances, and a 13.1 percent shooting rate matching Bennett's. Perreault represents New York's baseline; Sykora's hot hand its ceiling.

D. Tarasov is on a back-to-back, which could mean a backup start or tired legs—either way, it matters.

The possession arrow points Florida; the possession defense favors New York. Bennett in isolation is more dangerous than any Ranger. Sykora's shooting is borrowed time. The Rangers' 49.0 percent Corsi signals they'll chase, which on the road is vulnerable.

This is what 52 percent looks like: a game decided by execution and hot hands, not scheme. Neither team owns a structural claim to victory. The advantage exists, but it's razor-thin.