Ottawa decimated Toronto's expected goals—72% to 28%—yet still only won 3-1. The Senators generated 5.31 xG against Toronto's 2.05, controlled shot attempts 38-20, and won high-danger chances 6-3. But they underperformed their xG by 2.31 goals, meaning quality opportunities went unfinished.

Toronto's D. Hildeby (0.946 SV%, +3.31 above expected) saved 3.3 goals beyond his expected workload. Despite the final score, he was the game's most consequential player, keeping the Maple Leafs in a game they should have lost by larger margins.

The underlying edge came from shot quality. Ottawa created chances at 0.076 xG per shot against Toronto's 0.049. The Senators were lethal on the rush—2 goals on 33 chances—and created 4 rebound opportunities, though they converted zero.

Where the Senators pulled clear: special teams. They scored twice on 9 power-play shots generating 0.88 xG—a dominance that broke through when even-strength play was tightly contested.

Dylan Cozens's wrist shot from the slot at 19:54 of the third was the turning point.

The model projects Ottawa winning 61% of these matchups, with a final score likely landing between 2.2 and 2.8 goals per side. That's a lean toward the Senators rooted in measurable territorial dominance.

Ottawa has won 5 of its last 10 games on a 5-3-2 pace, while Toronto has stumbled to -15 goal differential over the same stretch. Recent form matters in hockey, and the Senators are playing the better game.

The underlying numbers confirm it. At 5v5, Ottawa generates 56.0% of expected goals versus Toronto's 46.0%—a 10-point gap that reflects who controls play. The Corsi split reinforces that: Ottawa owns 53.0% of shot attempts to Toronto's 45.0%, suggesting a territorial tilt throughout the evening. Toronto's 1.002 PDO (marginally fortunate) and Ottawa's 0.993 (unlucky) suggests regression; the Senators will get their chances and convert them more efficiently than recent results indicate.

Brady Tkachuk is Ottawa's engine at 1.19 xG/60, carrying 11 goals and 20 assists on a 7.5% shooting rate. Dylan Cozens adds depth at 1.00 xG/60 with 13 goals, 15 assists, and 9.5% shooting. Both have room to exceed their current output if this model holds.

For Toronto, the situation is more complicated. Auston Matthews has generated 1.10 xG/60 this season with 17 goals and 13 assists, but he hasn't played in 34 days and is likely unavailable. John Tavares will carry more offensive burden than usual: he's at 0.91 xG/60 with 15 goals and 24 assists on a hot 12.4% shooting rate, but high shooting percentages are unsustainable. Once Tavares' luck normalizes, Toronto's power-play reliance becomes even more exposed.

Defensively, Toronto has surrendered opportunities without capitalizing on its own: a 0% hard-goal rate and 0% rebound-goal rate suggest they're not getting second chances. Bo Groulx represents a weak point in the lineup. Ottawa shows the same limitation with Nick Jensen as the defensive vulnerability, but a team that controls play and gets more chances doesn't need to be perfect at converting the difficult ones.

This is a matchup where the model is translating structure into probability. Ottawa has the territorial edge, the hotter team, and the players generating quality looks. Toronto is undermanned without Matthews and riding an unsustainable shooting percentage. The Senators' 61% projected win rate reflects a gap in actual game control, not randomness.