Backcheck
Recap · NJD at CAR

Hurricanes run the Devils off the ice on the underlying numbers

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

By the ModelJun 16, 2:12 PMxG: xgboost-0.758Edited for clarity
Away
NJD
New Jersey Devils
3
Home
CAR
Carolina Hurricanes
6
FINAL
Process vs result

The underlying story

2.33Expected goals5.09
69% share for the better side
22Shot attempts (SOG)33
0High-danger chances9
45Corsi (all attempts)73
Score-adjusted: 40% / 60%
Source · NHL play-by-play · XGBoost xG

Carolina's 6–3 victory over New Jersey landed on the scoreboard, but the truly damning statistic lay underneath: the Devils generated zero high-danger chances yet still netted three goals. This paradox—complete slot dominance by the Hurricanes paired with a seemingly competitive final score—reveals how lopsided the underlying match truly was.

Carolina controlled 69% of expected goals (5.09–2.33) and 60% of shot attempts (33–22), a depth of dominance the scoreline barely captures. New Jersey overperformed their 2.33 xG to score three; Carolina overperformed their 5.09 xG to score six. Both teams exceeded expectation, but only one had created the circumstances to begin with.

Carolina's attack was surgical. They generated 0.070 expected goals per shot compared to New Jersey's 0.052, a quality edge that compounded their volume advantage. The Hurricanes scored three goals from low-danger positions—spots most teams fail to cash. Off the rush, Carolina converted 5 goals on 36 chances while New Jersey converted 3 on 25. Both teams found success in transition, but only one dictated play across sixty minutes.

Goaltending played a supporting role. F. Andersen stopped 19 of 22 shots (0.864 SV%), outperforming expectation by 0.67 goals. J. Markstrom faced heavier traffic, saving 27 of 32 (0.844 SV%), and performed to expectation given the volume differential.

The competitive scoreline masked Hurricanes dominance across nearly every dimension. Carolina didn't just win; they ran the Devils off the ice, and the underlying numbers leave no doubt.

How the game was won

The danger ladder

shotsgoalsexpected
High dangerslot & crease
NJD0 / 0 · 0.00 xG
CAR1 / 9 · 2.47 xG
Medium dangermid-range
NJD3 / 17 · 1.89 xG
CAR2 / 16 · 1.78 xG
Low dangerperimeter & point
NJD0 / 28 · 0.44 xG
CAR3 / 48 · 0.83 xG

CAR went 3-for-48 from low danger on 0.83 expected goals — 2.2 above expected.

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

Finishing vs expected

expectedactual
NJD
3 G · 2.33 xG · +0.7
CAR
6 G · 5.09 xG · +0.9
01234567
Goalies · GSAx (goals saved above expected)
J. MarkstromNJD
+0.1
F. AndersenCAR
-0.7

NJD finished +0.7 against expected, CAR +0.9 — F. Andersen (CAR) allowed 0.7 more goals than expected.

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

Shot diet

goalsshots
NJD
SNAP2/10
WRIST0/13
SLAP0/1
BACKHAND1/3
TIP IN0/4
WRAP AROUND0/1
DEFLECTED0/0
OTHER0/13
CAR
SNAP1/26
WRIST1/10
SLAP1/8
BACKHAND0/5
TIP IN2/4
WRAP AROUND0/0
DEFLECTED1/1
OTHER0/19
NJD
CAR
Off the rush
3 G on 25 · 1.93 xG
5 G on 36 · 3.80 xG
Off rebounds
0 G on 1 · 0.04 xG
0 G on 2 · 0.32 xG

Snap shots carried the volume: NJD went 2-for-10, CAR 1-for-26. CAR scored 5 of theirs off the rush.

Source · Backcheck xG model · NHL shot types
Three stars · computed, not voted
S. Aho
#20 · C ·
0 G · 2 A · 3 SOG · 21:39 TOI
S. Jarvis
#24 · R ·
2 G · 0 A · 3 SOG · 21:12 TOI · +1
L. Hughes
#43 · D ·
0 G · 2 A · 2 SOG · 20:46 TOI
Ranked from the box score — points first · Backcheck
What the model flagged
  • NJD generated higher-quality looks than the shot total suggests.
  • CAR generated far more dangerous chances — 0.070 xG/shot vs 0.052. Quality over quantity.
  • CAR scored 3 goals from low-danger spots — the opposing goalie needs to have those.
  • CAR was lethal off the rush — 5 goals on 36 rush chances.
  • NJD was lethal off the rush — 3 goals on 25 rush chances.
Before the game — the model's pre-game read
Preview · NJD at CAR · 7:30 PM ET

The Hurricanes are the model's lean over the Devils — and regression is coming

By the ModelJun 16, 2:11 PMEdited for clarity
Win probability

The model's lean

58%
CAR
Model favorite
42%
58%
NJD50CAR

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

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

Tale of the tape

NJD← edge
edge →CAR
50.0%5v5 xG%56.0%
51.0%Corsi%60.0%
2.45xGF / 602.98
2.50xGA / 602.31
1.66Goals for / gm2.26
2.23Goals against / gm1.95
0.971PDO — luck, not skill0.984

CAR holds 6 of 6 process categories — and PDO (0.971 vs 0.984) says CAR has run hotter.

Source · MoneyPuck season tables · 5v5

Carolina projects as the favored side at 58% win probability, with a model range of 2.4–2.9 goals against New Jersey. The bet is fundamentally about territory: the Hurricanes control 5v5 play decisively—56.0% expected goals share versus 50.0%—and translate that dominance into shot volume, leading at 60.0% Corsi. Against a Devils team that's posted 3-2-2 in their last 10, the edge goes to the team that dictates where play happens.

The catch is that both clubs are running cold on luck. Carolina's 0.984 PDO sits below neutral, meaning the Hurricanes are generating chances that aren't converting at sustainable rates. New Jersey's 0.971 tells the same story. For Carolina, this is clarifying—a team already controlling the game will look even more dangerous once results normalize. For New Jersey, it's the inverse: a club barely holding even at 51.0% Corsi needs its goaltending and conversion luck to hold tighter than it currently does.

The Hurricanes' territorial edge flows through their best creators. Jackson Blake (1.06 xG/60, 14G 19A, 10.9% shooting) generates chances at high frequency. Seth Jarvis (0.98 xG/60, 17G 12A, 11.9% shooting) has converted at an unsustainable clip—regression will come for that shooting rate. On the other side, Timo Meier (1.08 xG/60, 13G 15A, 6.7% shooting) is the most dangerous weapon but carries the lowest shooting percentage, a setup for positive variance. Jack Hughes (0.88 xG/60, 18G 17A, 11.5% shooting) has been prolific but will see regression from volume, not skill.

The tactical vulnerability for each side runs through their defensive anchors. Johnathan Kovacevic's pairing defines New Jersey's weakness; Charles Alexis Legault serves the same role for Carolina. Neither team has found secondary scoring through rebounds or loose play—both sit at 0% of goals from those sources—which concentrates the game's value in primary-chance conversion and territorial control.

That concentration favors Carolina. The Hurricanes don't need regression luck to cut the Devils down; the model already prices in their advantage. New Jersey's path requires an upset-sized swing on PDO coupled with neutralizing Carolina's best line—a two-part gamble when the first part isn't guaranteed.

Three Stars

  • Jackson Blake (1.06 xG/60, 14G 19A)
  • Timo Meier (1.08 xG/60, 13G 15A)
  • Seth Jarvis (11.9% shooting, 17G 12A)
Computed danger · scouting

Players to watch

NJD
Timo Meier13G 15A
1.08 xG/60 · 3 HD · 6.7% sh
Jack Hughes18G 17A
0.88 xG/60 · 1 HD · 11.5% sh
CAR
Jackson Blake14G 19A
1.06 xG/60 · 2 HD · 10.9% sh
Seth Jarvis17G 12A
0.98 xG/60 · 1 HD · 11.9% sh
Source · MoneyPuck skaters