An Introduction to NHLe and the NHLe Thresholds Analysis (aka the NHLe Stars Model)

NHL Equivalencies (NHLe) are a concept that have been around some fifteen years.  The idea, developed by Gabe Desjardins and expanded on by hockey analytics gurus like Rob Vollman, originated in the early 2000’s.  The idea is simple: how much can we expect a player to score in the NHL, in their first season, coming over from a junior, college or Euro feeder league.  

For instance, a highly-touted first round pick that scores 1.5 ppg in the OHL in his 18-year-old season, who is expected to make the jump to the NHL. How many points can we expect him to score in the NHL? The answer is 37, as on average, for each point a player scores in the OHL, it is worth approximately 0.30 points in the NHL. 123 points over 82 games in the OHL (the NHLe equation always converts everything to 82 games based on the point-per-game pace) is worth 37 points over 82 games in the NHL. 

NHLes provide a guideline for how a player new to the NHL can expect to produce in their first season in the NHL. But also, when applying the overall translations from each league to each player, season over season, they provide a clean way to standardize player scoring and rank them.

Furthermore, you can take the NHLes and group players by NHLe thresholds (i.e., NHLe in the 20s, NHLe in the 30s, NHLe in the 40s, etc.) and start to determine if hitting certain thresholds, at certain ages, is indicative of making the NHL and producing in the NHL.Turns out they are incredibly telling of who makes the NHL and who becomes an impact offensive producer (i.e., a Star). This is the fundamental concept behind this analysis.

This is a concept I started to develop a few years ago at FlamesNation (Predicting Future NHL Scoring Success with NHLe Thresholds). The original analysis consisted of a small sample of players. The results from this brief analysis used the original translations developed by Desjardins and suggested, in large part, players that record high NHLes and do so the earliest tend to make the NHL and are most likely to produce in the NHL. A couple questions still persisted:

1) Equivalencies have evolved quite a bit since then, do translations change if you’re purely looking at younger players?

2) What does the whole population look like?

This analysis, an effort one year in the making, includes every player drafted between 1990 and 2019 (in the first seven rounds) – which, in total after removing players that were drafted twice, includes over 5,000 players.

The translations I use, while similar to the originals, do differ. I’ve also built translations for leagues where they were previously non-existent. All told, I’ve included over 40 league translations in the analysis. The analysis follows the players along for five years of development, through their D-1 year (pre-draft year) until their D+3 year (third year after their true draft year), essentially tracking them from Age 16/17 to 21/22.

I group the players into six groups – Star (0.7+ career ppg; 200+ games played), Average Producer (0.4 to 0.69 career ppg; 200+ games played), Replacement Producer (0 to 0.39 career ppg; 200+ games played), 100 Gamer (100 to 200 games played; drafted between 1995 to 2011), Bust (less than 100 games played; drafted between 1995 to 2011) and Developing (drafted between 2013 and 2019 and haven’t played enough games yet to be deemed a 100 Gamer or Bust)

Next, I’ll start to speak to the standardization NHLe approach and a new metric that derives from it all, what I call the Standardized Production Index (SPI).

Leave a Reply