Backcheck
Team diagnostic
2025-26last seasonAs of Sun, Sep 20

Minnesota Wild

5-on-5 · model-estimated

Wild: a possession-driving team (67.9% 5v5 xG) with league-average goaltending.

How to read this

This is one team through the publication's three layers. The identity row separates process — the chances a club creates and allows (, ) — from luck, which is what measures. Process predicts the future; luck predicts a correction.

Everything below follows that split: regression flags mark places where results have outrun (or trailed) the underlying play, talent tiers and the player table rank the roster by a model-estimated , and roster needs compare each position group against league average. The proxies rank things usefully but share credit with linemates — read orderings as solid and exact magnitudes as approximate.

PROCESS
chance creation and possession — xG%, Corsi. The repeatable part.
RESULT
goals, wins, special-teams conversion. What the standings see.
LUCK
PDO, finishing vs expected, GSAx — the part that mean-reverts.
GOLD/RED
regression flags: gold = watch it, red = act on it
Fine print · verbatim from the model
  • WAR / Impact are model-estimated proxies, not cap-validated
  • Roster-need gaps are ordinal — ranked correctly, magnitudes relative
Identity · possession-driving
67.9%
xGoals %
This team controls play outright.
48.5%
Corsi %
0.04
xGF / 60
1.008
PDO
Luck is about neutral; the record is earned.
Strengths
  • Drives 5v5 play (67.9% xG share).
  • Elite top-end talent: Matt Boldy, Brock Faber.
Weaknesses
  • Leaky penalty kill (80% kill rate).
Talent tiers · model impact
Elite02
  • Matt Boldy
  • Brock Faber
Top07
  • Kirill Kaprizov
  • Quinn Hughes
  • Ryan Hartman
  • Vladimir Tarasenko
  • Jonas Brodin
  • Jared Spurgeon
  • Jeff Petry
Middle12
  • Mats Zuccarello
  • Marcus Johansson
  • Joel Eriksson Ek
  • Bobby Brink
  • Danila Yurov
  • Yakov Trenin
  • Marcus Foligno
  • Jake Middleton
  • Nick Foligno
  • Michael McCarron
  • Nico Sturm
  • Robby Fabbri
Depth03
  • Ben Jones
  • Daemon Hunt
  • Zach Bogosian
Rate-first

Special teams

Power play
0.00 xGF/60 · 25.2% SH%
7.56GF/60
Penalty kill
0.8 Kill%
7.33GA/60
Source · MoneyPuck 5-on-5 + ST splits
Team GSAx +0.0

Goaltending

Above expected
GoaltenderGP
Filip Gustavsson5090.410+0.0
Jesper Wallstedt3591.570+0.0
Source · GSAx = xGoals − goals, computed in-app
Roster needs · ranked
01
Bottom-6 forward depthhigh
team 52.0 vs league 81.9
-30
02
Depth defensehigh
team 62.5 vs league 87.3
-25
03
Top-6 forwardhigh
team 66.6 vs league 86.1
-20
04
Top-pair defensemanhigh
team 77.1 vs league 91.5
-14

Gaps are ordinal — needs are ranked correctly; treat the magnitude as relative, not absolute.

Top players · adjusted impact
PlayerGPTOI/GP
Matt BoldyL7614.2+80.50+4.3
Kirill KaprizovL7817.2+68.70+4.1
Quinn HughesD7422.2+71.90+3.3
Brock FaberD8020.4+80.20+3.0
Mats ZuccarelloR5914.6+59.60+2.2
Ryan HartmanR7614.5+69.60+2.2
Marcus JohanssonL7513.0+56.40+2.0
Vladimir TarasenkoR7512.2+63.10+2.0
Joel Eriksson EkC7013.2+58.00+1.9
Bobby BrinkR6812.9+53.10+1.2
Schedule load · logistics

How hard is this team’s season, logistically? This grades the schedule itself — travel, rest, and clustering — separate from how good the team is.

Logistics, not talent

Schedule load

DEPTH
Load index
39/ 100 · pctile vs league

The 20th-heaviest schedule in the league84 games, 1.3 days average rest.

Total travel
46,990
road miles flown
10
Back-to-backs
2 games, 0 days off
11
3-in-4s
3 games / 4 nights
14
4-in-6s
4 games / 6 nights
5
Longest road trip
2,627 mi
Rest-days distribution
0d10
1d52
2d16
3+d5

Nights of rest before each game · 0 days = red, 3+ = green

Rest advantage

19 games with more rest than the opponent · 20 with fewer · 44 even.

Hardest stretches
  • 6 games in 11 days with 4,480 travel miles

    Nov 21Dec 1

  • 6 games in 10 days with 3,526 travel miles

    Feb 11Feb 20

  • 6 games in 10 days with 3,491 travel miles

    Oct 6Oct 15

Source · NHL schedule · arena coordinates · rest & travel model
By the Model5-on-5 aggregates · MoneyPuck* WAR / Impact are model-estimated proxies, not cap-validated