Capitals Prospect Assessment With The ‘NHLe Player Comparison Tool’

The whole concept of this piece will focus on NHLe. If you’re unfamiliar with NHLe (NHL Equivalency)… it’s a way of extrapolating a player’s production (points per game) from one league to expected production (points per game) in the NHL.

For example, the translation in the OHL is 0.3.  Roughly speaking, for every point the player scores in the OHL, it’s worth approximately 1/3 of a point the NHL when they transition the following year. If a player has 110 points in 82 games in the OHL, his NHLe would be 32.  Based on the average of all the players who have gone from the OHL to the NHL in the following year, we can estimate the player would score 32 points over 82 games in the NHL in the first year after transitioning.

European Men’s (EM) leagues typically have much higher translations than junior leagues like the OHL, WHL, QMJHL, USHL, etc. It’s much harder to score in an EM league, as these are, by and large, professional adult players, many of which having played in the NHL previously.  That is to say, a point is worth much more in an EM league than a junior league. The translation for the Swedish league, for instance, is about 0.8. If a player registers the same 110 points in 82 games in the SHL, his NHLe would be 88, which would be extraordinarily rare and indicative of a superstar (e.g., Elias Pettersson).

So Many Leagues
NHLe’s provide a great way to gauge what you might expect from your favorite prospect who’s making the jump to the NHL shortly. But they also provide a way of standardizing players across leagues.

There are approximately 50 leagues a player could come from before making it to the NHL (there’s many more leagues than that worldwide, but 50 high-level leagues that an NHLer would realistically come from 99% of the time).

How can you rank all of these players and keep track of them?  One way is through NHLes. And after accounting for progression, position, age, height, etc., we can now put the players on one level playing field, and through a lens of offensive production, can see them all on the same scale. When we can see the players on this same level like this, amazing insights can be discovered.

A Framework For All
My work, which can be found at, is focused on NHLe and putting players on the same playing field.  I have built a model that follows the player for five years, from their pre-draft year (e.g., D-1) to three years after their first eligible draft year (D+3).

The tracking looks at how the player’s NHLe evolves (increases, consistencies or decreases over time) and what that means.  The model goes back thirty years and is very predictive of future NHLers and Stars. It accounts for the player’s age, position and era they played in, as these can play massive factors.

Even players that are six months apart but look identical otherwise can mean a world of difference. From there, I’ve built-out NHLer and Star probabilities, based on the 30 years of history. Simply put… what is Player X’s chances of making the NHL or becoming a Star, based on every player that has come before them that looks like him.

Looking At Capitals Prospects
Now that we have a framework, let’s look at a few Washington Capitals prospects through the model and see what it tells us about the players. For this exercise, I’ve chosen four recent draftees: Connor McMichael, Brett Leason, Aliaksei Protas and Alexander Alexeyev.  I’ve chosen these four players as they’re all recent picks and are extremely different. There’s a wealth of insights to garner from each one.


The snapshot above provides a look at the player’s NHLe progression, Star and NHLer probabilities at the time the player is drafted as well as after 5 years of development (or as of right now if the player doesn’t have five years of history yet) and their closest comparables when drafted and after 5 years of development.

Connor McMichael was drafted in the 1st round of the 2019 draft, 25th overall. If you’ve been tracking McMichael’s 2019-20 season (his D+1 season), you’ve probably felt some optimism, perhaps even thinking ‘the Capitals have drafted a game-breaker!’  But you’ve probably also had some uncertainty about McMichael’s season and what it all means. ‘Is this a lot of points for a player in the OHL at his age?’ ‘How many other players have looked just like this and been busts?’ In McMichael’s case, you should be incredibly optimistic about him and what he might become.

When he was drafted, he was on the older side, being a mid-January born player.  He was older than 80% of everybody drafted in 2019 in their first true draft year. His production (an NHLe of 26) is pretty typical of a late 1st round selection. Forwards like this don’t tend to develop into Stars.  They develop, at a gradual pace, and become effective NHLers (generally middle six players). Every once in a while, though, a player like this will have a follow-up season that will completely change their outlook, and how the model perceives them.  McMichael, to this point, has an NHLe of 54 in his D+1 and is one of these players.

That production is absolutely extraordinary and unexpected. It has taken his chances of becoming a Star from 12% to 43% in one season. What’s more is that the progression is so atypical that the model (based on 30 years of history) wouldn’t be able to fully account for how big of a change this is. This is due to how rare this is.

Going from an NHLe in the 20s, as an older player, in one’s draft year to one in the 50s in one’s D+1 has really only happened four times in 30 years, out of nearly 4,000 players. Only three NHLers look like this! Those players, featured in the comparison portion of the snapshop of the tool above, are Alexei Yashin, Mikko Rantanen and Connor McMichael.  Aleksi Heponiemi is the other player who also reached this extraordinary feat and he’s still developing in the Florida Panthers’ system. Jeremy Roenick, Craig Janney and Miroslav Satan are also close but differ slightly. Anybody would recognize most or all of those names.

Generally, players that hit NHLes to that extent by their D+1 have already had more impactful seasons previously and had every indication that’s where they were going next.  For instance, a consensus Top 5 younger pick that had an NHLe of 40 in their draft year. An older player with an NHLe in the mid-20s in his draft year to over 50 in his D+1 is a unicorn. What is all this telling us about McMichael?

McMichael is doing something extremely rare and the company that has done so have all turned into Stars in the NHL. McMichael, regarded by many scouts as a player with a sky-high hockey IQ, has gone from a player that you could reasonably assume would become a reliable NHLer in a few years to a player with rare superstar potential, possibly already ready for the NHL.  I would expect McMichael to be in the NHL next year and, while he may not produce immediately, my research would suggest it will be there, in short order.


The snapshot above provides a look at the player’s NHLe progression, Star and NHLer probabilities at the time the player is drafted as well as after 5 years of development (or as of right now if the player doesn’t have five years of history yet) and their closest comparables when drafted and after 5 years of development.

Brett Leason, drafted in the 2nd round, 56th overall, was taken as an overager (i.e., not his first eligible draft year; his first eligible draft year was in 2017). Leason produced an NHLe of 40 the year he was drafted. On the surface, this would seem extraordinary. While it’s certainly a significant NHLe, it’s coming at a later age, which is a really big factor.  If Leason had put up numbers like that in his first eligible draft year, given his size, he would have been drafted in the Top 5 and would rank very favorably in the model. Doing so later drops the player down dramatically. In this case, Leason is doing so two years after his first eligible draft year.

The other thing that you notice with Leason is his results prior to hitting the 40 NHLe in his D+2 were not strong.  For three straight years he had an NHLe of 13 or less. There are hundreds of players that are drafted with numbers like this and their success rate is extremely low (10 to 12%).  The D+2 numbers in Leason’s case are encouraging but I ask myself “where is this coming from? where was it before?”.

Leason is from Western Canada and grew up in the Western Hockey League. The Western Hockey League only has younger players (16 to 21 years old). Leason is a gigantic man, standing 6’4’’and 200 pounds, and by the 2018-19 season, he would have towered over many of his peers. We often see this, especially with larger players in the CHL. The player starts to dominate their competition once they hit 19 or 20. They dominate when they’re much bigger and mature than most around them. We can see now that Leason is playing against players more his age and size in the AHL, he’s really struggling, and his production is back in a similar range to where he was prior to the 2019 outlier year.

Leason should have definitely been drafted in 2019. He was producing impressive numbers and he was young enough that he was worthy to take a shot on, especially given his size. However, he would have been a player I would have targeted past the second round as, the snapshot above indicates, the model was suggesting a lot of warning signs and as a result didn’t have him ranked very high.  It’s also important to note that Leason is currently in his 5th year of development. At this point, even with big strides, it’s diminishing returns.  He is what he is at this point… a long shot.


The snapshot above provides a look at the player’s NHLe progression, Star and NHLer probabilities at the time the player is drafted as well as after 5 years of development (or as of right now if the player doesn’t have five years of history yet) and their closest comparables when drafted and after 5 years of development.

Aliaksei Protas, taken in the 3rd round, 91st overall, is a prototypical 3rd round pick.  Protas had an NHLe in the teens when he was drafted. At that point in the draft, that’s mostly what’s left. Like Leason, Protas is huge, standing at 6’6’’, which is an intangible that is also coveted.  Protas was a typical low probability late-round pick, but Protas has begun tipping the scales on that low probability.

Protas, in his D+1 in the WHL, is currently amassing an NHLe of 38 which is strong growth. To put that into perspective, you should be thrilled with a late 1st rounder jumping up to an NHLe in the high 30s in their D+1. When a deep draft pick does it, so early, it’s found money. While Protas, to this point, is not trending to be a Star, he is now rising towards becoming an NHLer.

A player doubling their NHLe between their DY and D+1, from an NHLe in the 10s to one in the 30s, is often a good sign. It puts them on the radar of the team that drafted them and shows that they can start to produce more when their responsibility on their team grows. His chances of becoming an NHLer are now nearly a 50/50 shot and will only continue to grow if his NHLe continues to progress.

To confirm further whether or not this is a true depiction of players that looked like him, I went back to my database for another point of validation. I looked at every player that looked exactly like him (older player that went from an NHLe in the 10s in their DY to an NHLe in the 30s in their D+1). There have been 57 developed cases (i.e., we can say with certainty that enough time has passed that they’ve made the NHL or not by now). Of the 57 cases, 28 of them played at least 100 games in the NHL, which works out to 49%.  The model is giving an accurate representation of his odds.

As luck would have it, based on all we know of Protas to this point, one of his closest comparables is current Capitals’ wrecking-ball, Tom Wilson. This is not to suggest they play the same way, although they are nearly identical in size.  This suggests that out of all the players in the model, the player’s NHLe progression/age most resembles those NHLers. If you look at all of the closest comparable they’re mostly NHLers that contributed in a middle-six role and had/are having good NHL careers. What are we looking at with Protas?

Protas is a player that started with an outside chance of making the NHL to one that is now making a strong case that he will be an NHLer.  This all revolves around what happens in the next few years. Does Protas continue to get better or at least hold steady and produce at a similar rate?  If he does, he’ll likely at least get a taste of the NHL. If he falls back down in the next few years, his odds will decrease drastically. Right now, Protas appears to be a very solid 3rd round selection.


The snapshot above provides a look at the player’s NHLe progression, Star and NHLer probabilities at the time the player is drafted as well as after 5 years of development (or as of right now if the player doesn’t have five years of history yet) and their closest comparables when drafted and after 5 years of development.

Alexander Alexeyev, drafted 31st overall in 2018, is a two-defenseman that is well on his way to a lengthy career in the NHL. At the time he was drafted, Alexeyev was an older player (older than 80% to 90% of first year eligible defensemen drafted) and had an NHLe of 20. In the forward world, that’s slightly above average.  In the defenseman world, that’s in the 90th percentile! However, the fact that he’s an older defenseman, rather than say a defenseman with the same numbers who was born in the Summer, is the biggest factor in dropping his score.

On the other side, Alexeyev also had impressive numbers in his D-1, registering an NHLe in the 10s. This helps his case. Many defensemen that look like this are drafted earlier in the 1stround, in fact – examples include: Morgan Rielly, Seth Jones, Aaron Ekblad, Noah Hannifin, Charlie McAvoy and Jay Bouwmeester.  The model gave him approximately a 50/50 shot of being an NHLer at the time he was drafted. You’ll notice that after another year and a half of development Alexeyev is mostly in the same position. This is due to him holding steady at an NHLe in the low 20s over that time. He’s neither improving nor decreasing his score.

It’s also important to note there’s more intangibles with defensemen that the model doesn’t account for than forwards.  For one, defensemen are meant to produce offense but they are much more reliant on to reduce offense the other way. Therefore, there are defensive attributes which aren’t picked up by this model (e.g., abilities to exit the zone, defend the blueline, shot suppression, gap control, etc.). The defensemen model performs well but does function a bit differently and has different weightings from the forward model. As well, where the player is taken is really important, and especially with defensemen.

If you have two players that look identical in the model, and one is taken in the 1st round and one is taken in the 7th round, I would lean towards the one taken in the 1st round in a big way.  The model assumes all players are equal and doesn’t factor in round, but 1st rounders are much more likely to make the NHL than deeper round picks.  This is especially true for defensemen.  Approximately 55% of players that look just like Alexyev make the NHL (similar to what the overall model suggests). But if we only look at the ones drafted in the 1st round, that number increases to 80%.  What are we looking at with Alexeyev?

His numbers and very gradual, status quo, development from his draft year (where he was already impressive) suggest he’s probably not a defenseman that produces a huge quantity of points and runs the first powerplay unit in the NHL. His numbers, position taken and consistent production across leagues, however, do suggest a player that will make the NHL in a year or two and become a fixture on the backend, likely chipping in 20 to 30 points per year. Additionally, his comparables are very favorable.  Ultimately, he looks like a dman that should play in the Top 4.

By Byron Bader

A word about and the model…

I mentioned earlier the model and the website. The model goes all the way back to 1990 and includes the 2020 eligible draftees. The tool pulls up a snapshot just like the ones used in this model for each player. This model and tool have been validated and are highly accurate and is something that NHL hockey operations personnel have started to utilize recently.

The site is a membership-site ($24.99 yearly subscription).  However, until January 31, 2020, there’s a 20% discount (coupon code: HP-Launch) as part of the new launch. I encourage you to register, put in the coupon code and try it out! We provide a 3-day free trial and cancellation takes only a few seconds.  You have nothing to lose so please give it a try… it will change the way you see prospects!

Follow myself or Hockey Prospecting on twitter: @ByronMBader and @HckyProspecting

***PS: If you’re curious what the dynamic-duo, Ovechkin and Backstrom, look like in the modelthey appear like nobrainer superstars, as you might expect.

About Jon Sorensen

Jon has been a Caps fan since day one, attending his first game at the Capital Centre in 1974. His interest in the Caps has grown over the decades and included time as a season ticket holder. He has been a journalist covering the team for 10+ years, primarily focusing on analysis, analytics and prospect development.
This entry was posted in News, Propsects and tagged , , , , , , , , , , , . Bookmark the permalink.

3 Responses to Capitals Prospect Assessment With The ‘NHLe Player Comparison Tool’

  1. Anonymous says:

    From watching all of his games in Hershey, Leason is definitely the biggest question mark. True, he was a slow starter in the WHL, but there are younger players with fewer games under their belt that have already passed him by. His shot has been all over the place.

  2. Anonymous says:

    Fantastic analysis – thank you for this piece!

  3. Pingback: Who Could Be Moved At The Trade Deadline? Player Trade Ratings For The Capitals’ Organization | NoVa Caps

Leave a Reply