Backcheck
Recap · VAN at PIT

Penguins run the Canucks off the ice on the underlying numbers

The Penguins earned it — the game and the expected-goals battle both.

By the ModelJul 11, 12:32 PMxG: xgboost-0.758Edited for clarity
Away
VAN
Vancouver Canucks
1
Home
PIT
Pittsburgh Penguins
5
FINAL
Process vs result

The underlying story

3.24Expected goals5.20
62% share for the better side
24Shot attempts (SOG)26
3High-danger chances13
69Corsi (all attempts)65
Score-adjusted: 54% / 46%
Source · NHL play-by-play · XGBoost xG

The Penguins' 5-1 rout looks understated compared to what the underlying metrics reveal. Pittsburgh controlled 13 high-danger chances to Vancouver's 3, a margin that dwarfs the final tally. At 5v5, the Penguins' expected goals crushed Vancouver's 5.20 to 3.24—a 62-38% split that was never in doubt.

The shot attempts told a different story. Pittsburgh took 26, Vancouver 24, a near-even split that masked deeper control. But volume doesn't matter when Penguins' chances averaged 0.080 expected goals per shot versus the Canucks' 0.047. Quality overwhelmed quantity.

Vancouver wasted what chances they generated. K. Lankinen finished with a 0.808 save percentage, stopping just 21 of 26 shots. More damning: the Canucks underperformed their expected goals by 2.24, squandering high-danger looks at a 92% clip against Lankinen's own glove. A. Silovs proved the answer, stopping 23 of 24 for a 0.958 save percentage—2.24 goals above expected. He stole a blowout that the underlying numbers had already scripted.

Pittsburgh was lethal at the margins. They converted two goals from low-danger spots, the kind of finishing that separates dominance from catastrophe. Their rush attack produced 2 goals on 25 chances. Vancouver generated four rebound opportunities and converted none; Pittsburgh created five and buried one.

J. Brazeau struck with a wrist shot from the high slot at 5:56 of the third period, opening the rout. Brazeau finished with 1G, 2A in just 11:50. K. Letang added two assists, while A. Mantha posted 1G, 1A. Pittsburgh's supporting cast delivered in full.

This wasn't a scoreline that flattered Pittsburgh. The underlying numbers revealed an even more total dominance.

How the game was won

The danger ladder

shotsgoalsexpected
High dangerslot & crease
VAN0 / 3 · 0.82 xG
PIT1 / 13 · 3.47 xG
Medium dangermid-range
VAN0 / 12 · 1.52 xG
PIT2 / 13 · 1.31 xG
Low dangerperimeter & point
VAN1 / 54 · 0.91 xG
PIT2 / 39 · 0.42 xG

PIT went 1-for-13 from high danger on 3.47 expected goals — 2.5 left on the table.

Source · Backcheck xG model · NHL play-by-play
The luck layer

Finishing vs expected

expectedactual
VAN
1 G · 3.24 xG · -2.2
PIT
5 G · 5.20 xG · -0.2
0123456
Goalies · GSAx (goals saved above expected)
K. LankinenVAN
+0.2
A. SilovsPIT
+2.2

VAN finished -2.2 against expected, PIT -0.2 — A. Silovs (PIT) saved 2.2 goals above expected.

Source · Backcheck xG model
What they shot — and what went in

Shot diet

goalsshots
VAN
WRIST0/31
TIP IN0/4
SNAP0/4
SLAP1/7
BACKHAND0/3
OTHER0/20
PIT
WRIST4/26
TIP IN1/9
SNAP0/4
SLAP0/0
BACKHAND0/3
OTHER0/23
VAN
PIT
Off the rush
1 G on 28 · 2.18 xG
2 G on 25 · 3.46 xG
Off rebounds
0 G on 4 · 0.40 xG
1 G on 5 · 1.07 xG

Wrist shots carried the volume: VAN went 0-for-31, PIT 4-for-26. PIT scored 2 of theirs off the rush.

Source · Backcheck xG model · NHL shot types
Three stars · computed, not voted
J. Brazeau
#16 · R · PIT
1 G · 2 A · 2 SOG · 11:50 TOI · +1
K. Letang
#58 · D · PIT
0 G · 2 A · 2 SOG · 17:02 TOI
A. Mantha
#39 · R · PIT
1 G · 1 A · 2 SOG · 13:33 TOI · +1
Ranked from the box score — points first · Backcheck
What the model flagged
  • VAN generated higher-quality looks than the shot total suggests.
  • A. Silovs (PIT) stopped 2.2 goals above expected.
  • Open game — 16 high-danger chances combined (PIT 13, VAN 3).
  • PIT generated far more dangerous chances — 0.080 xG/shot vs 0.047. Quality over quantity.
  • PIT scored 2 goals from low-danger spots — the opposing goalie needs to have those.
  • PIT was lethal off the rush — 2 goals on 25 rush chances.
  • PIT created 5 rebound opportunities (1 converted) — crashing the net effectively.
  • VAN created 4 rebound opportunities (0 converted) — crashing the net effectively.
  • VAN underperformed xG by 2.2 — wasted quality chances.
  • K. Lankinen was a wall on HD chances — 92% SV on 13 high-danger shots.
  • A. Silovs (PIT) saved 2.2 goals above expected — stole the show.
Before the game — the model's pre-game read
Preview · VAN at PIT · 7:00 PM ET

The Penguins are the model's lean over the Canucks — with PDO regression in play

By the ModelJul 11, 12:31 PMEdited for clarity
Win probability

The model's lean

58%
PIT
Model favorite
42%
58%
VAN50PIT

PIT by 16 points of win probability — a clear lean. Projected score VAN 2.42.9 PIT; 18% chance it's tied after 60.

Source · Backcheck Poisson projection · season scoring rates
Season profile · edge from the spine

Tale of the tape

VAN← edge
edge →PIT
44.0%5v5 xG%51.0%
46.0%Corsi%50.0%
2.22xGF / 602.66
2.81xGA / 602.55
1.60Goals for / gm2.45
2.56Goals against / gm2.12
0.974PDO — luck, not skill1.012

PIT holds 6 of 6 process categories — and PDO (0.974 vs 1.012) says PIT has run hotter.

Source · MoneyPuck season tables · 5v5

The projection: Pittsburgh 58% to win, with a median final score of 2.9 to 2.4. The case rests entirely on structural dominance—the Penguins control both quality and territory.

Pittsburgh owns a 51.0% 5v5 expected-goals share against Vancouver's 44.0%, the kind of shot-quality gap that usually finds the scoreline. They're also winning possession at 50.0% Corsi, meaning more time in the attacking zone and fewer defensive scrambles. This isn't luck; it's territorial control.

The secondary edge belongs to Vancouver's regressing fortune. The Canucks' PDO sits at 0.974—measuring the ratio of goals scored to shots and conceded goals to shots faced—meaning they're scoring below model expectation while bleeding more than their underlying play should allow. That ratio normalizes. Pittsburgh's 1.012 PDO suggests they've run hot, but the underlying 51.0% xG affords them room to regress without collapsing.

Rickard Rakell anchors Pittsburgh's attack at 1.01 expected goals per 60 minutes, with 13 goals and 13 assists on 12.1% shooting. Filip Hallander, who generates 1.28 xG/60, is out for further evaluation. Rakell isn't flashy, but his volume and conversion describe a player operating within an efficient system.

Vancouver's depth splits between two engines. Filip Chytil generates 1.03 xG/60 with 3 goals and 0 assists despite a 15.0% shooting rate—a mismatch suggesting either variance or deployment away from high-danger looks. Drew O'Connor has accumulated 13 goals and 11 assists at 0.85 xG/60, meaning his goal total is genuine—he's shooting 11.8%—yet the underlying play doesn't justify the assist volume. His linemates are finishing.

Neither team generates high-danger or rebound goals at any measurable rate. Vancouver's 0% HD goal rate and 0% rebound-goal rate signal scoring arrives from mid-range, not the crease. Pittsburgh shares this deficit. That invites variance—a power-play surge or goaltending spike from either side could alter the script.

Defensively, Pittsburgh's Caleb Jones and Vancouver's Victor Mancini flag as weak links, but territorial advantage—the Penguins' core edge—neutralizes matchup quality. Pittsburgh's structural lead and the Canucks' pending PDO regression should hold even if this game tightens closer than 58% implies.

Computed danger · scouting

Players to watch

VAN
Filip Chytil3G 0A
1.03 xG/60 · 1 HD · 15.0% sh
Drew O'Connor13G 11A
0.85 xG/60 · 3 HD · 11.8% sh
PIT
Filip Hallander0G 3A
1.28 xG/60 · 0 HD · 0.0% sh
Rickard Rakell13G 13A
1.01 xG/60 · 3 HD · 12.1% sh
Source · MoneyPuck skaters