League Strength and NHL Equivalency

Professional hockey draws talent from a fragmented global pipeline — major junior, NCAA, AHL, SHL, Liiga, KHL, and lower tiers. Unlike the more centralized paths in other sports, there is no single developmental highway. Front offices need a consistent way to compare scoring across environments. NHL Equivalency (NHLe) models attempt to translate production onto a common scale, with the NHL fixed at 1.0.

The basic idea is straightforward: if a player produces at a certain rate in another league, how many points would that translate to over a full NHL season? Public analysts have refined the approach over the years, moving from simple direct comparisons to more sophisticated network models that account for indirect paths through intermediate leagues.

This piece summarizes the established public work on the topic and lays out the theory that NHLe is most reliably used as an index of league strength — the relative difficulty of generating offense in a given environment — rather than a direct formula for projecting an individual player’s NHL point total.

Public Implementations

The most widely referenced network-based approach comes from CJ Turtoro. His Network NHLe (NNHLe) work models leagues as nodes in a graph and player movements between leagues as weighted edges. Indirect paths allow factors to be calculated even for leagues with few or no direct NHL transitions. Volume weighting across multiple paths dampens the impact of small samples or outliers.

Patrick Bacon built a detailed public NHLe implementation that draws heavily on the network method and applies it to prospect evaluation, with additional layers for projections. Other analysts, including those at Hockey Prospecting and various independent modelers, maintain their own factor sets and tools.

These efforts all start from the same core problem: raw points per game do not mean the same thing in every league. Stronger environments (better goaltending depth, tighter systems, higher pace) suppress scoring relative to weaker ones.

The Theory: League Strength, Not Direct Projection

NHLe works reasonably well as a rough translator when applied in aggregate. A player putting up strong numbers in a high-factor league is doing something meaningful. The same raw production in a low-factor league carries less signal about NHL readiness.

The more durable insight, however, is that the numbers primarily describe the environment, not the individual. A 0.60 factor for the KHL does not mean “every 50 KHL points equals 30 NHL points for this specific teenager.” It means that, on average across many transitions, the KHL environment has permitted roughly 60% as much offense per unit of production as the NHL.

Treating the output as a precise per-player formula runs into several structural problems: small samples, age and development effects, role and usage differences, team quality, and the simple reality that counting stats are noisy at the individual level. The signal stabilizes when you step back and look at relative league difficulty.

This is why the most useful application is contextual weighting of production rather than direct translation into an NHL point projection. In our own draft rankings, production from stronger leagues (SHL, Liiga, KHL) receives more credit than equivalent raw numbers from major junior, with additional adjustments for age relative to competition and role. The NHLe framework supplies the rationale for those weights.

The same principle applies when evaluating historical production or cross-league comparables. The tool shines at showing which leagues have historically been efficient pipelines and which have been more insulated. It is less reliable as a calculator that spits out an exact future NHL stat line for any one player.

Application in Our Work

The league-strength lens is the explicit foundation for the production-in-context rules used in the 2020–2024 draft class rankings (see Dataset 2). The same logic appears in our early look at the 2027 class. Production is never treated as raw; it is always interpreted against the strength of the league and the player’s role within it.

References

The public work referenced above includes:

• CJ Turtoro — Network NHLe writeup and tools: cj-turtoro.shinyapps.io
• Patrick Bacon (TopDownHockey) — Detailed NHL Equivalency and prospect projection models
• SJHockey Stats — Revamped NHLe model
• Hockey Prospecting (Byron Bader) — Large public database using standardized/equivalency-adjusted scoring
• Earlier foundational contributions from Gabriel Desjardins and others on basic league equivalencies

These coefficients are the full set used in our draft evaluation work.