In recent weeks, there has been a subtle change to the narrative regarding how Canucks management discusses their team. Before the season started, all management seemed to talk about was the return of the Canucks to the playoffs. We were told we were retooling, not rebuilding.

However, of late, management’s focus is becoming more apparent, specifically when it was brought to light that Benning was unwilling to part with Jared McCann, Jake Virtanen or Bo Horvat as part of a deal to land Evander Kane. While the Canucks are very much intent on making the playoffs (Francesco does love that playoff revenue after all!), they appear to be just as focused on preparing to be successful for 3 to 4 years from now in a world without the Sedins. Maybe, just maybe they’ve come to realize that the window to compete for a championship with this core has closed.  

What is all the more interesting is that the 2015 draft class looks to be one of the strongest draft classes in recent memory, even beyond Connor McDavid, who looks to be a generational player.  With a very strong chance the Canucks will be able to pick up a top tier prospect when they select in June, much focus will be on who is available in the Canucks range, likely to be somewhere between 10 to 20 overall. 

This brings to the much discussed Lawson Crouse of the Kingston Frontenacs. After playing for team Canada after the World Junior tournament, Crouse has seen his draft stock skyrocket on most draft charts, as high as 3rd overall on ISS’ prospect rankings, behind only the aforementioned McDavid and Jack Eichel. But is he really deserving of this high ranking though?

Traditional scouts rave about Crouse’s skating, puck handling, hockey IQ, and defensive game, not to his mention his NHL ready 6’4/212lb frame. However, while the scouts sing his praises, the analytics crowd have been wondering if he’s so good then why are his points totals so unimpressive (currently 34 points in 42 games this year)? ESPN’s Corey Pronman, a long-time supporter of utilizing data as part of player assessment, weighed in last week explaining why he thinks Crouse should be a top 10 pick despite his unimpressive boxcar statistics.

While the analytics community has made significant strides in expanding our quantitative understanding of the game in recent years, the same can’t be said about evaluating prospects outside the NHL, due in large part to the lack of data maintained by most leagues. That said, we do know that height, age, and CHL points-per-game are statistically significant variables when projecting NHL points-per-game. 

With that in mind, I thought I’d dig a bit deeper into Crouse’s historical comparables, as well as a few other draft eligible CHLers, who share a common trait. They were all projected top 10 picks in the preseason who have since been bumped out of the top projections by Crouse: Mathew Barzal (Seattle Thunderbirds), Daniel Sprong (Charlottetown Islanders), Travis Konecny (Ottawa 67’s) and Nick Merkley (Kelowna Rockets). 

To evaluate these players, I’ve used the data set scraped by Josh Weissbock which I discuss in this article on the importance of size in prospect evaluation. In study I found that 8% of CHLers eventually became NHL regulars, defined by over 200 NHL games (note: Jonathan Willis had a nice summary of the reasoning behind the 200 game benchmark over at Oilers Nation). 

I also decided to include the player’s 16 year old cohort in the analysis. When looking at data from the same (1987-88 to 2008-09 seasons. I found there is actually a higher correlation between a players 16 year old season points-per-game and the NHL points per game (.486), than for their 17 year-old season (.41). Here is a breakdown of the number of 16 year olds who made it to the NHL, sorted by points-per game: 

Some interesting notes on the table above. First, the number of players in this sample is significantly lower than the study on 17 year olds (1,435 players versus 3,355). The second interesting point is that the number of 16 year old CHLers who made it to the NHL (12%) was significantly higher than the number of 17 year old CHLers who made it to the NHL (8%). This implication is that there is a higher talent bar to make it to the CHL at the younger age, which intuitively makes sense. Otherwise, we still see a steady increase in the probability a player will make it at the NHL level as their points per game increases, consistent with what we saw for 17 year-olds.

The one other change I made is to use age-adjusted points per game (AAPPG), as opposed to points-per-game, since we now know that age is a statistically significant variable (more on AAPPG can be found here). For reference, player AAPPG can now be found on Josh Weissbock’s excellent chlstats.com. 

To ensure each cohort had a comparable size (in this case 30), I used the player height in centimeters as the baseline, and included players in a range of 3 centimeter taller and shorter in the sample (literally, give or take an inch). Once the height cohort was established, I created the AAPPG sample of 30 players with the players actual AAPPG equating to the median. For example, Nick Merkley is 178 CM tall and had an AAPPG of 1.0 pts/gm in this 16 year-old season, equating to a height range of 175-181cm and an AAPPG cohort of 0.93-1.23 pts/gm. 

Without further ado, here are the players I looked at:

Top NHLers in Merkley’s height/age adjusted Pts/GP cohort are as follows:

Prior to the 2014-15 season, not many scouts projected Merkley as a top 20 pick, despite him having an excellent 16 year-old season (58 points in 66 games). That began to change this year, as he’s the WHL’s highest scoring draft eligible player this year (79 points) by a fairly wide margin. He’s known as an excellent playmaker, with high hockey IQ, and an excellent work ethic. Other than his size (5’10), scouts note he needs to work a bit on his defensive game. 

23 of out 60 CHLers (38%) who were a similar size to Merkley and produced at a similar rate went on to play 200 NHL games. On average, these 23 NHLers produced at a 0.55 points-per-game rate, which is roughly the average for a second-line NHL player. 

Top NHLers in Barzal’s height/age adjusted Pts/GP cohort are as follows:

Prior to the season 2014-15 season, many scouts projected Barzal as a potential top 5 pick, due to his elite skating and playmaking. However, a knee injury sidelined him in November, and his draft stock has taken a hit as a result, making him a potential steal if he slides to the Canucks.

25 of out 60 CHLers (42%) who were a similar size to Barzal and produced at a similar rate went on to play 200 NHL games. On average, these 25 NHLers produced at a 0.53 points-per-game rate, which is roughly the average for a second-line NHL player. 

Top NHLers in Sprong’s height/age adjusted Pts/GP cohort are as follows:

After scoring 68 points in 67 games last year, much was expected from the Dutch winger, but so far he hasn’t been good, but less dominant than expected (64 points in 55 games), and his draft stock has slid. This skating, puckhandling and passing is described as elite, and he’s also known to be defensively responsible. As you can see above, he has very similar cohort comparables to Barzal, with a high cohort success rate. If he does slide to the later end of the first round, Sprong will be a steal. 

28 of out 60 CHLers (47%) who were a similar size to Sprong and produced at a similar rate went on to play 200 NHL games. On average, these 28 NHLers produced at a 0.54 points-per-game rate, which is roughly the average for a second-line NHL player. 

Top NHLers in Konecny’s height/age adjusted Pts/GP cohort are as follows:

Konecny had an unexpected slow start to his 2014-15 season, playing for a horrible Ottawa 67’s team. Unfortunately for Konecny’s stat lines, the 67’s don’t really have many other players worth worrying about for the opposition, which I think is part of what you see reflected in Konecny’s stats. Scouts rave about his skating, puck-handling, hockey IQ and his shot, and note that he’s not afraid to mix it up despite his small stature. His production has been increasing of late, so it will be interesting to see how his AAPPG cohort %’s change by the end of the year. 

19 of out 60 CHLers (32%) who were a similar size to Konecny and produced at a similar rate went on to play 200 NHL games. On average, these 25 NHLers produced at a 0.53 points-per-game rate, which is roughly the average for a second-line NHL player. 

Top NHLers in Crouse’s height/age adjusted Pts/GP cohort are as follows:

On the positive side, Crouse’s skating, puck handling and shooting all look to be at the elite level, which is why pundits such as myself struggle to understand why he doesn’t score more than he does. Clearly, scouts drool over the potential to draft the next Todd Bertuzzi, with good reason. 

19 of out 60 CHLers (25%) who were a similar size to Crouse and produced at a similar rate went on to play 200 NHL games. On average, these 25 NHLers produced at a 0.55 points-per-game rate, which is roughly the average for a second-line NHL player. 

While there has been a lot of discussion around whether Crouse warrants the rank ISS/CSS and other project for him, the reality is that with roughly a 25% chance of success based on his historical cohorts, he’s most definitely a first round pick. The art of drafting comes down to trying to figure out whether he’s going to be part of the 25% in his cohort that make it, or the 75% that don’t, and by the eye test there is definitely a lot to like about him. 

However, looking at the NHLers in each prospects cohort, we see a lot of commonalities. Without a doubt, there are lots of good players listed, but very few stars. On average, each prospect projected to be a second line player, based on the players in their cohort, if they are successful in overcoming the odds to make it to the NHL in the first place. 

So, is Crouse really a top 10 pick? It’s actually difficult to say, but his average upside is nearly identical to a group of players ranked a handful of picks below him. History also tells us that Crouse has the largest bust potential of the players I looked at too, with around 75% of CHL players that produce in a similar manner and who are of a similar size missing the NHL. His well-rounded game arguably gives him a greater than ~25% shot at becoming an NHLer, however it’s tough to say how much this increases his chances as we have seen solid defensive projections amount to absolutely nothing before.

With the Canucks unlikely to pick in the top ten, and players like Barzal, Merkley, Sprong, and Konecny likely to slide as a result, I sure hope Crouse is taken in the top-10 though!