Net_Man-Games-Lost_Money: A new metric to measure NHL Coach and GM Performance.
The traditional way we evaluate NHL coaches and general managers lacks a critical component, and some organizations have made poor decisions as a result. We judge coaches on win-loss records and GMs on raw point totals, treating injuries as random acts that excuse poor performance or mask true coaching talent.
But in a hard-cap universe, player availability is an organizational capability. By utilizing Net_MGL_Money – a metric developed by Probility AI that tracks a team’s game-by-game injury deficit against its opponent weighted by player Annual Average Value (AAV) – front offices can finally separate dumb luck from actual execution.
Here is how Net_MGL_Money provides an objective, data-driven framework to evaluate bench bosses and front executives.
Part 1: Why the “Injury Excuse” Doesn’t Hold Up
In the NHL, front offices often write off injuries as bad luck. But recent data suggests that health is a choice – or at least an outcome of how a team is run.
A Better Way to Measure Health
Absolute injury counts are irrelevant when assessing or predicting game outcomes. A team may be healthy, but if their opponents are even healthier, they still face a disadvantage.
A metric called Net_MGL (Net Man-Games Lost) solves this. It calculates the difference between Team A’s injuries compared to Team B for every game.
- A positive number means Team A has more healthy players available than Team B.
- A negative number means Team A is shorthanded compared to Team B.
But that’s not the whole story. Losing a star player with a $10 million cap hit hurts much more than losing a rookie making the league minimum. The evolution of Net_MGL is Net_MGL_Money. This comparable version weights injuries by salary producing a measure of the impact injury likely has on a team’s roster. Now the game-by-game evaluation looks like this:
- A positive number means Team A has more healthy dollars available than Team B.
- A negative number means Team A is skating a team with less total AAV compared to Team B.
Net_MGL_Money can also be expressed as a percentage of each team’s roster spend (and availability on any given evening) against the cap. It’s not just which team is putting more money on the ice, but which team is putting more of its money on the ice for any given game.
This metric isn’t everything, but it shifts the odds. In the 2024-25 season, 56% of playoff teams had a positive health advantage:
- Seattle had the best health in the league but still missed the playoffs.
- Colorado had the second-worst health deficit and still made it.
We can use that to evaluate team performance. When a team like Seattle misses the playoffs despite being the healthiest in the league, it suggests the problem is tactics, talent, or roster construction. When Colorado makes the playoffs despite having more money in the stands than their opponents over the course of a full season, well, tip of the hat to the coaching staff.
Part 2: Bench Evaluation: Measuring Coaching Value Above Replacement
When a team wins games, it can be hard to tell if the coach is a tactical genius or if they are simply driving a Ferrari. Net_MGL_Money acts as a weight-distribution scale for a coach’s record.
The Overachievers (High Negative Net_MGL_Money)
A coach who consistently wins while carrying a deep negative Net_MGL_Money is getting their roster to systematically over-perform against superior, healthier competition. When a lineup is stripped of millions in cap-weighted talent and still secures a playoff berth, that is a direct reflection of a coach’s structural systems, deployment strategies, and locker-room culture.
Look at Kris Knoblauch, formerly Head Coach in Edmonton. He navigated a persistent manpower deficit to back-to-back Stanley Cup Finals. Edmonton showed him the door in 2026 after a first-round playoff exit. The Net_MGL_Money framework argues that Knoblauch generated massive organizational value that standard analytics and raw point totals understate. If you are an NHL team looking for a new head coach, we suggest giving Kris a call.
The Underachievers (High Positive Net_MGL_Money)
Conversely, a positive Net_MGL_Money means the coach consistently matched up against depleted, cheaper rosters. If a team possesses a massive health-and-dollar advantage across an 82-game season and still misses the postseason, the blame should focus on the bench, not the training room.
Dan Bylsma in Seattle enjoyed the league’s largest relative manpower advantage during the 2024-25 season but failed to convert it into a playoff spot. He was fired in April 2025. The Kraken have one of the most sophisticated analytics departments in the NHL. Did it provide general manager Ron Francis with data-backed metrics to inform that decision? Likely. These decisions are not easy and gaining access to key measurables supports good decision-making. In this case, Net_MGL_Money strips away the injury excuse and rightly forces scrutiny onto tactical quality and asset deployment.
Part 3: Front Office Evaluation: Tracking Roster Durability and Acquisition Errors
While coaches manage the players they are given, General Managers control the checkbook. A GM’s job is to maximize the on-ice return of every single dollar under the salary cap. Net_MGL_Money exposes exactly where asset allocation turns into dead cap space.
Exposing the “Fragility Premium”
Some players are elite when healthy but carry an established track record of missing time. If a GM routinely signs or trades for high-cap, low-durability assets, their team will show a persistently negative Net_MGL_Money.
This metric acts as an executive scorecard:
- It highlights whether a GM is built to withstand depth shocks.
- It quantifies the literal cap cost of roster fragility.
- It separates a GM who builds a resilient, deep roster built to win in the playoffs from one who builds a top-heavy house of cards that collapses at the first sign of physical adversity.
Better Data, Better Decisions
The goal of data like that produced by Probility AI isn’t just to look backward and assign blame – it is to alter how front offices make decisions in real time.
Real World Example: Workload and Crease Protection:
Positional analysis within the Net_MGL_Money framework reveals that goaltender injury curves between playoff and non-playoff teams permanently diverge around game 15. By mid-season, non-playoff teams suffer goalie injuries at 1.5 times the rate of qualifiers.
For a GM and coach, this insight dictates operational strategy. It quantifies the exact return on investment for high-end backup goaltenders and dictates predictive, data-driven rest cycles for the primary netminder to protect the crease before the mid-season cliff.
Maximizing the Championship Window:
Historical data proves a strict “injury ceiling” exists for Stanley Cup champions. With rare exceptions that relied on pre-2024 LTIR loopholes, nearly every modern champion won with fewer than 300 total injuries and kept their total injury cost under $10 million.
Net_MGL_Money provides the predictive boundaries for a championship roster structure. By integrating cap compliance with relative health analytics, NHL executives can move away from reactive “bad luck” narratives and start proactively managing durability as a core competitive advantage.
Conclusion: Make Availability a KPI
In a cap league, “injuries” aren’t noise—they’re an input cost. Net_MGL_Money turns availability into a measurable, game-by-game handicap, so coaching impact and roster construction can be evaluated on a level surface. That means fewer narratives, more signal: who wins despite a real deficit, who underperforms with a built-in advantage, and which organizations keep turning cap dollars into missing games.
For hockey ops, the implication is blunt: if you can’t separate performance from availability, you can’t evaluate coaching, pro scouting, or roster construction with confidence. Treat Net_MGL_Money as an organizational diagnostic – then build incentives, workflows, and roster strategy around reducing the most expensive kind of waste: cap dollars in street clothes.
- Add Net_MGL_Money to every postgame and quarterly review so results are graded against opponent-adjusted availability.
- Use it to audit acquisition decisions: identify contracts/trades that systematically increase “fragility premium” exposure.
- Set a season-long availability target (in dollars and % of cap) and align sports science, workload management, and depth planning to hit it.
In short, build player availability into the architecture of your next championship team.
Interested in learning more? hello@probility.ai
@Probility’s CDO @QuintonKrueger will be presenting the full academic findings underlying development and deployment of Net_MGL_Money at the 5th International Ice Hockey Analytics Conference – LINHAC 2026, June 2-4, 2026, Linköping, Sweden.
