clock menu more-arrow no yes mobile

Filed under:

The Bargain Bin Pt. 1: An Analytical Breakdown of 192 North American 2017 Draft Eligible Prospects. Who would be a good steal for the Red Wings?

Who is outperforming their draft ranking? Who is underperforming? Let’s look at the numbers.

Portland Winterhawks v Vancouver Giants Photo by Rich Lam/Getty Images


Initial disclaimer before you start reading this: this is going to be a very long article. Grab yourselves a snack and/or a coffee and brace yourselves.

Rather than talking about a single prospect today, I am going to be dedicating this article to providing an analytical perspective on a large sample of 2017 draft eligible CHL and USHL prospects, both ranked and unranked. While discussing which prospect the Red Wings should take in the first round (and even the second round) is great, as you know, there are seven rounds in this draft, and the Wings just happen to have 11 picks this year. That means there is a lot of opportunity for the Wings to land players later in the draft that may end up becoming very good NHLers. Look no further than players such as Johnny Gaudreau, Colton Parayko, Mark Stone, Brendan Gallagher, Jamie Benn, and yes, Pavel Datsyuk, for examples of how later rounds of the draft will occasionally have a player that eventually blows up and ekes out a very respectable (and, in some cases, brilliant) NHL career. There is no telling if any of the players in the later rounds of this draft can ever carve a career out for themselves in the way that any of the players that I just mentioned have/had, but there are still a number of players that should be getting discussed that are projected to go in the later rounds, and are perhaps even unranked on NHL CSS’s final rankings. The purpose of the present article is to evaluate, using analytics, which prospects are outperforming their draft ranking, and which prospects are underperforming their ranking. In this article, I will be focusing specifically on forward prospects. In a follow-up article, I will be focusing on defensive prospects. Based on these analyses, we may be able to conclude which prospects may end up being the best options in the later rounds of the draft.

Any of the analyses that I have performed will be methodologically explained, and the data, as well as the code that I use to run these analyses can be found here. I wanted to keep all of the materials including my data and scripts accessible for those of you who wish to replicate the methods that I used for yourself, I will have the code that will allow you to do so included in the script “Scouting code 2”. All of the data in this article will be completely open access. All of the analyses conducted in this article were done using R statistical software (version 3.2.5) , which, if you haven’t downloaded yet, you can find here.



The present analyses used data collected from 192 draft-eligible North American prospects. All of these prospects played in either the CHL or the USHL at some point over the course of this year. Of these prospects, 182 were ranked on NHL CSS’s final rankings for North American skaters, and 10 were unranked. This analysis included 66 defensemen and 126 forwards.

Many prospects in leagues such as the OJHL (Nick Campoli, Matthew Kellenberger, Bryce Misley, Phil Lagunov, Finn Evans), the CCHL (Cam Crotty, Bobby Dow), any of the high school leagues (Jonathan Dugan, Sam Huff, Keaton Pehrson, Benton Maass, Tyce Thompson, Noah Ganske, Santeri Hartikainen, Neil Shea, Corey Andonovski, Ryan O’Connell, Jayson Dobay), the USPHL (Michal Stinil), the BCHL (Kale Howarth), the AJHL (Cale Makar and Ian Mitchell specifically), and NCAA (Carson Meyer, Patrick Khodorenko, Jack Ahcan, Denis Smirnov, Ryan Poehling, Joshua Wilkins, Colt Conrad, Luke Martin) are unfortunately omitted from these analyses, as no source has actually compiled some of the vital statistics that I am analyzing. Even a few prospects such as Antoine Crete-Belzile in the QMJHL had to be omitted because their stats had to be compiled. This will have to be treated as a limitation of these analyses. Regardless, there was still a good enough sample size to evaluate that this shouldn’t matter too much.


5v5 P1/60. Along with each prospect’s name and position, I compiled several other variables for this analysis. I looked strictly at 5v5 production. I compiled three measures of 5v5 production: G/60, A1/60, and P1/60 (goals and primary assists/60). All of these stats were derived from I was primarily focused on P1/60, and this served as my dependent variable in all of my analyses. If I were to go into the separated G/60 and A1/60, the article would probably take far too much time to actually read. If you wish, I will also provide scripts that can be used to run these analyses too, and we can discuss these details within the comment sections. For the time being, however, looking at these statistics was a bit beyond the scope of the present article. Other 5v5 measures that I additionally collected included GF%Rel, QoT, and QoC.

Draft Ranking. In this analysis, the independent variable was the prospect’s ranking on NHL CSS. The range of rankings for ranked prospects was between 1-217. Due to some prospects not having a ranking, and the statistical software which I use being rather finicky when it comes to missing cells, unranked prospects were given a ranking between 218-227 for the sake of being able to run analyses on them. Forward and defensive prospects were separated into different groups, and separated into quantiles on the basis of their draft ranking. For forwards, six quantiles (or sextiles) of 21 prospects were created. In such an organization, the smaller a player’s sextile, the higher their draft ranking. Sextile 1, for example, would encompass the 21 highest ranked forward prospects, while Sextile 2 would encompass the 22nd-42nd highest ranked forward prospects. Defensemen were also divided in the same manner, with six sextiles of 11 defensemen. The purpose of doing this manipulation was to establish comparison groups for all of the prospects within the sextile. Doing this quantile manipulation is based on two premises. First is the idea being that we should theoretically be able to evaluate prospects on the basis of how they perform compared to prospects in a close vicinity to them in the rankings. Thus, prospects were treated as cohorts on the basis of these rankings, and all statistics were computed on a sextile-by-sextile basis. Second, is the idea that there are significant differences between players across sextile. Players that are ranked higher on NHL CSS’s rankings can be assumed to generally be better players than players ranked lower. While this doesn’t always hold true, it is a general trend, and players ranked lower can be thought of more as outliers.


The forwards and defensemen were separated, and were subdivided into six quantiles each. I then ran a series of analyses in which I regressed P1/60 on both the player’s draft ranking. The result of doing this can be seen here:

Notes: The R2 statistic indicates that roughly 11% of the variance in 5v5 P1/60 is accounted for by draft ranking. While the majority of the variance within P1/60 is due to other factors, this is still a modest correlation, and the prediction was statistically significant (r = .33, p < .0005 with 1 and 124 degrees of freedom.)

As can be seen, there was a statistically significant negative relationship between P1/60 and NHL CSS draft ranking.

After determining that there was, in fact, a relationship between P1/60 and draft rankings for forwards, the next step was to estimate the mean, standard deviation, standard error, and 95% Confidence Intervals for all of the sextiles.

After estimating these parameters, through a rather... lengthy series of ifelse() loops in R, I computed each player’s standard score. The formula for this calculation appears as such:

zij = (xi – μj)/σj.

zij denotes the standard score for player i in sextile j. This is calculated by subtracting each player’s P1/60 stat (denoted by xi) from the mean of sextile j (μj), and dividing that by the standard deviation of that sextile (σj). The result of this calculation produces six unique standard scores for each player. What the standard score tells us is how many standard deviations from the mean a particular player’s P1/60 stat lies within the sample. This stat is an interger, and can vary both positively and negatively. The more positive this score is, the more standard deviations the score lies above the sample mean. In other words, a higher z-score means this player is outperforming the sextile that they’re being compared to, while a lower z-score means they are underperforming. The purpose of this calculation was to compare each player’s performance to not only their own sextile, but to all of the other sextile as well. The value obtained from doing this can be used as an estimate of the statistical significance between an observation and the sample, depending on what alpha level you’re using as your criterion of significance. A score of 1.64, for example, denotes that the difference is significant at the .10 (10%) level, meaning there is a 10% probability of incorrectly rejecting the null hypothesis (i.e., there is no statistically significant difference in the observations). A score of 1.96, on the other hand, would be significant at the .05 level, meaning that there is only a 5% probability of incorrectly rejecting the null hypothesis. Such a stat is thus quite useful in the process of hypothesis testing.


Sextile 1 Comparison

Using these scores, I made a series of six z-score plots as a function of sextile to evaluate each player’s productivity to every other sextile. First, comparing all forwards to Sextile 1:

Notes: The red lines represent the median ranking and a z score of 0 (meaning the player matched the mean P1/60 for Sextile 1). Any of the players below the horizontal red line underperformed. Any of the players above the red line outperformed the Sextile 1. The four dotted horizontal green lines represent scores that are at an alpha level of .10 and .05, respectively.

Based on comparisons to Sextile 1, there are several major observations we can make. Prospects such as Nick Suzuki, Gabe Vilardi, Eeli Tolvanen, Shane Bowers, Jaret Andeson-Dolan, Nikita Popugaev, and Kole Lind were all either around the average for the 1st round, or were slightly higher. However, these players were within a standard deviation of the average. There were several notable overperformers and underperformers, which I will go over.

Biggest Performers in Sextile 1:

  • Nico Hischier (#2 on NHL CSS) - it should come as no surprise to anybody, but Hischier was one of the most statistically impressive prospects in the entire sample. Any of the players who outperformed him in this sample did so on a smaller sample size, or were overaged players that were passed on in previous drafts.
  • Owen Tippett (#7 on NHL CSS) - As with Hischier, the only players to outperform him had a smaller sample size of games, or were overaged. Another interesting note: most of Tippett’s P1/60 was driven by goalscoring. For players that played as many games as he did this season, he was the best pure goalscoring prospect in the entire draft.
  • Kailer Yamamoto (#17 on NHL CSS) - Why nobody gives Kailer Yamamoto’s offensive capabilities more respect is beyond me. Yamamoto is one of the most electrifying talents in this draft, and is ranked criminally low in the first round for someone with as much gamebreaking talent as he has. One of the best offensive catalysts in the entire draft.
  • Cody Glass (#6 on NHL CSS) - One of the best playmakers in the entire draft. If Glass is around when the Wings pick at 9th overall, it would be very difficult to pass on him. Given that WIIM has recently selected him in a recent mock draft, perhaps Glass would be the center that this franchise needs moving forward.

Sextile 1 Underperformers

  • Michael Rasmussen (#5 on NHL CSS) - I have never been shy about my mortal terror at the thought of drafting Michael Rasmussen. These analyses confirm what I’ve been saying about him all of this time. In fact, Rasmussen’s production was so awful, he was in the negatives when compared to every. single. sextile. The single poorest-producing forward in the Sextile 1.
  • Casey Mittelstadt (#3 on NHL CSS) - How does somebody with so much skill produce so poorly at 5v5 in the USHL? I can only offer some tentative guesses as to why he did so poorly. First, his team was terrible in general. Despite playing many fewer games than everyone else, he ended up 6th on the team in scoring. Second was his small sample size. He only played 24 games. However, this, and his poor showing at the scouting combine recently are very unflattering. It remains to be seen whether these results were just an anomaly, and whether he will perform as he’s advertised in the NCAA.
  • Nolan Patrick (#1 on NHL CSS) - Make no mistake. Patrick is an excellent prospect, and he does a lot more than just produce points. In fact, if you were to check the file All stats.xlsx (not .csv; the .xlsx file has the ability to sort and filter data at your convencience) he led all 192 prospects in this sample in GF%Rel, suggesting that he’s been a major driving force for the Wheat Kings whenever he steps on the ice. However, given the fact that he’s ranked #1 on NHL CSS, and he’s being outperformed quite drastically by players such as Nico Hischier, Owen Tippett, Cody Glass, Kailer Yamamoto, Nick Suzuki, and others, these results must be noted. His offensive production at 5v5 was surprisingly... average for someone ranked so highly. In fact, from a production standpoint, he would only get out of the negatives in Sextile 3.
  • Alex Formenton (#29 on NHL CSS) - The London Knights forward had very.... meh production, although, from an analytics perspective, he was very good at driving shots from the low slot, where one would expect generate the most dangerous scoring chances, as can be seen here. I have never really been sold on Formenton

Other Top Performers

There were several players in this sample in other sextile that drastically outperformed even players in Sextile 1.

  • Josh Norris (#34) - A Michigan commit, Josh Norris was the leading scorer for the USNTDP U18 team this year. One of the things that stands out about him is his skating. He has a long, powerful stride, and can reach high speeds quickly. He has a very good one-timer, and gets a very quick release on his shot. His vision and passing abilities are nothing to sneeze at either. The cherry on top of it all is the fact that, from what I’ve observed, he’s money in one-on-one situations like in the shootout. The only caveat, along with the next two prospects I will discuss below, is the incredibly small sample size of games he played for which I could quantify this data. He played less than 30 games in the USHL, and this data is based on that sample. There’s no telling if he would have been able to sustain this level of production had he played a larger sample of games. Nevertheless, Norris would probably be one of the best options available in the 2nd round.
  • Evan Barratt (#65) - While Barratt has not received as much attention as Norris, Barratt has done very well for himself this year, and was 2nd among all eligible forwards in 5v5 P1/60. While lacking any physical gifts (such as skating ability, size, strength), he is a player that seems to always produce when he’s on the ice, and more than makes up for these average physical attributes with how cerebrally he plays the game. The way he thinks the game out is amongst the best in the draft. I could see him blossoming into a very good player one day, as long as he’s given enough time and patience to work on his strength and conditioning. As with Norris, however, he only played a very small sample size of games, and it is unknown whether this scoring pace would be sustainable over a longer period of time. For that reason, I would say Barratt would be a solid option to gamble on between the 3rd and 4th rounds. Regardless, keep an eye on him at Penn State next year.
  • Jacob Tortora (#165) - Jacob Tortora led all prospects in 5v5 P1/60. The small winger models his game after Johnny Gaudreau, and it’s definitely evident when you watch him. He has a lot of speed and shiftiness to his game, much like his idol. Loves to shoot, and scores a lot of very inventive and creative goals. he has also grown on the defensive side of the game. Seems like a kid that’s always eager to further develop his skill and become better at the game. I still think he needs to find that extra gear to his game, and work on his strength, as with many other prospects. Tortora will be following Johnny Gaudreau’s example and play college hockey at Boston College next year, where he will have the benefit of being in a great development program there. All-in-all, a very promising young player that can probably be drafted with a later-round pick. I could see him going between the 4th-6th rounds, and I think his set of skills would be absolutely ludicrous for someone picked that low. As with Norris and Barratt, however, Tortora had only a small sample size of games, and his numbers should be taken with a grain of salt.
  • Mason Shaw (#55) - The best first-year eligible playmaker in the draft. The only reason he’s ranked lower is because of his size and his skating. I would personally not gamble on him until the 3rd round or afterwards, as that combination of size and skating issues could be an issue with him translating to the NHL. Otherwise, he is somebody that you should know about, and should be paying close attention to going into June 24th. And don’t let the fact that he shares the same last name with human septic tank Andrew Shaw discourage you. There is no relation.
  • Ivan Lodnia (#36) - Vanya Lodnia had an excellent year with the Otters, albeit with a rather disappointing playoffs. Nevertheless, he was one of the best performers in the OHL all year for a forward prospect. Within the 2nd sextile of forwards, the only forward to produce better than him at 5v5 with more than 30 games played was Mason Shaw. If you look at the All Stats.xlsx data which I provided, you should also notice that Lodnia led all first-year eligible OHL prospects in high-danger shots/60 (a statistic which I calculated based on the number of high danger shots, the average TOI per game that they played, and the number of games that they played). The fact that he had a disappointing playoff showing should probably affect his draft stock a bit, but if Lodnia can be landed between the late 2nd and early 3rd, he’s a great steal.
  • Artur Tyanulin (#166) - I have already gone over Artur Tyanulin in quite some detail here. Tyanulin is a 3rd-time eligible forward, and is probably one of the best overaged players in the entire draft. Definitely not somebody that you’d gamble on in an early round. But if he’s there between rounds 6-7, he’s a very good bargain.
  • Vladimir Kuznetsov (unranked) - In my opinion, the best overaged player in this sample. He is not even ranked on NHL CSS, but was 2nd only to Vitalii Abramov for a player his age in scoring in the QMJHL this year (and yes, this is including 2016 3rd overall pick Pierre Luc-Dubois). Why he was passed on last year is beyond me, but I don’t think teams should be making the same mistake again.
  • Petrus Palmu (unranked) - Petrus Palmu is absolutely disgusting. A 3rd-time eligible forward, the only reason he might not become an NHLer is his diminutive size. Other than that, he has an absolute howitzer of a shot, and has the speed and smarts that make him a force to be reckoned with on every shift.

Sextile 2 Comparison

Note: See the figure above for an explanation of this figure.

The results of looking at Sextile 2 revealed many similarities in the data compared to Sextile 1. Besides perhaps Jonah Gadjovich, Kyle Olson, Morgan Geekie, and Lane Zablocki, there weren’t any other players that deviated very far from the sextile average positively that haven’t already been talked about. However, there were definitely a few underperformers in this data:

  • Maxime Comtois (#30) - It has been a horrendous year for Comtois, a player who was once projected to go 3rd overall in the draft in very early mock drafts near the start of the season. Comtois possesses excellent vision, good size, and solid two-way abilities, but his season did not see him produce very good results, and he has continued to slide down the rankings. One of the biggest issues for him is his speed and skating, and it is definitely something he needs to improve on. I can see him projecting as a 3rd line center that can provide a lot of offensive upside. At the midterms, he was #15. Since then, he has fallen 15 spots, and he may end up finding himself slipping to the 2nd round now, as taking him in the 1st may be a bit more of a gamble.
  • Alex Chmelevski (#43) - At face value, there’s quite a bit to like about Chmelevski. He’s got quick hands with great puckhandling skills, with short deceptive strides in his skating. However, when you look at what he’s done this season, it leaves you scratching your head. He had struggles to score this season, having an extended drought of about 14 games at one point going into February. Otherwise, he’s had great showings at the Top Prospects game, and was a beast at the Ivan Hlinka tournament for the US last year. One of his issues, however, is that he relies on the powerplay to score his goals a lot. At one point, more than half of his goals were on the PP, which can be concerning. He needs to work on his consistency, as well as his defensive game. He’s also a bit of a perimeter player, and needs to be able to generate chances from inside the low slot more.
  • Mackenzie Entwistle (#44) - Entwistle is a player that I’ve heard quite a bit of praise about over the course of the last season. Scouts love his two-way game, and he brings a combination of size, hockey sense, and skating. He’s very strong on the puck, useful on the PK, and has been utilized heavily in those situations, and is an excellent forechecker. He also had an excellent tournament for Canada at the U18s in April, where he was probably one of their best forwards. However, when discussing what he does from a production standpoint, there have definitely been more productive forwards this season that reside in the same sextile as him.
  • Stelio Mattheos (#38) - Like Maxime Comtois, Mattheos is another player that was at one point touted as one of the top prospects in the 2017 draft. Ranked 23rd at the midterm rankings, Mattheos has continued to slide further and further down the rankings as the year has went on. Not the flashiest of skaters, and his acceleration needs some work. Needs to also work on his lower body strength. But he’s very smart, sees the ice very well, and has good playmaking abilities. His biggest weakness, however, is his goalscoring. His goalscoring is probably amongst the weakest of any forwards ranked in the top 40 this year.
  • Nate Schnarr (#47) - Yet another player that was, at one point, ranked much higher at the midterms, but fell down the rankings as the season went on. He was 32nd at the midterms, but dropped 15 spots on the final list. Guelph was a complete trash fire this year, and you can’t necessarily evaluate his production without taking how his team did. Schnarr was someone that played very heavy minutes on the PK for Guelph, and was buried by very tough competition. In fact, Schnarr faced the toughest minutes of any forward in the OHL this year, and played more time shorthanded than any other OHL forward prospect. That being said, his pro upside is rather limited.

Sextile 3 Comparison

In a rather dramatic fashion, the results of the Sextile 3 comparison produced some very decisive results, and there were a number of players that outperformed and underperformed the 3rd sextile. As can also be seen, Nolan Patrick is now finally in the positive values, and the only 3 players from Sextile 1 that are still in the negatives are Alex Formenton, Casey Mittelstadt, and Michael Rasmussen (hint: neither Rasmussen’s nor Mittelstadt’s situation change for any of the sextiles).

Best Performers in Sextile 3

  • Tyler Steenbergen (#70) - A 2nd-time eligible center playing for the Swift Current Broncos, Steenbergen has an excellent shot that is lethal in close and has a lot of accuracy. A tad undersized at 5’10” and 180 lbs, he skates very smoothly and possesses great hands. He is very mobile, and is always zipping around the offensive zone with the puck on his stick and putting on a show. Possesses great one-on-one moves, and the shiftiness needed to overcome his deficit in size, which allows him to outmaneuver larger players effectively. He’s a threat every time he has the puck on his stick in the offensive zone. Shows a lot of great leadership traits, he’s someone that I could see setting an example for his teammates. A very patient and accurate passer, he keeps his feet moving and gets himself into position very well.
  • Kevin Hancock (#91) - Another 2nd-time eligible forward that has seen his play explode in his draft +1 year, to the point where he probably could be taken. Here’s a list of players who he outscores in terms of P1/60 at 5v5 that are in the same age group as him: Cliff Pu, Dmitry Sokolov, Janne Kuokkanen, Will Bitten, Jack Kopacka, Anthony Salinitri, Michael McLeod, Kyle Maksimovich, Luke Kirwan, Givani Smith, Adam Mascherin, Drake Rymsha, Jonathan Ang, Nathan Bastien, and Logan Brown. That’s a pretty impressive list for him. Hancock has a wicked wrist shot with a lot of power on his release, and possesses a great one-timer. He is just as good at setting up his teammates as he is at scoring, and has an excellent offensive tool set. Has great hands, and a great curl and drag to his game. He was recently voted as one of the best defensive forwards in the OHL by a coaches’ poll. He’s going to make whatever team lands him very happy if he can be drafted in a later round I think. Personally, I think he should have been drafted last year. But you know as they say, hindsight is always 20/20.

Underperformers in Sextile 3

  • Pavel Koltygin (#83) - The Drummondville Voltigeurs forward was 3rd on his team in scoring, despite being the youngest of those 3 players, on what was a pretty terrible team. This may have affected his numbers. Koltygin possesses a lot of skill with the puck, and is quick to bounce on the opportunities that he’s given. He’s got a good release, and a good backhand shot. His wrist shot can catch goalies off guard with how fast he gets it off. His footwork is also very good, and he’s got great east-west movement, which allows him to slip through defenders effectively. A player with a lot of finesse, and the speed to burn, accelerate, and change directions on defensemen. His mobility on his skates allows him to score some pretty highlight reel goals, going coast-to-coast sometimes. Very gifted player, although his defensive game needs some work. Surprisingly good at faceoffs, and is one of the best faceoff men in the Q. Much of his woes could be that he could have just been facing very difficult deployment.
  • Michael Pastujov (#80) - A Michigan commit that has experienced major setbacks in his development over the last two seasons, when he experienced a shoulder injury that made him miss the rest of the season back in 2015. His production in the USHL is rather troubling, however. A lot of his assists are secondary assists, and his goalscoring rate was rather underwhelming. However, he was pretty good at the U18s. Overall, he’s a very tough player to get a good read on.
  • Shaw Boomhower (#112) - This guy has Columbus Blue Jackets pick written all over him, given their predilection towards drafting players with the most redneck-sounding names you can think of (think: Boone Jenner). Boomhower is a player that I have a difficult time thinking of reasons to justify why he’s even ranked by NHL Central Scouting. He doesn’t excel at scoring goals. He is very bad at setting his teammates up. He produces at one of the worst paces of any forward that I have looked at this year, and doesn’t show any signs of being good on the defensive side of the puck. And to top it all off, he takes a lot of penalties, racking up 102 PIM this year. He is not somebody I would want to draft, given that the league is moving away from players like him, and I no longer see someone with his playing style having a niche within the league.

Other Outperformers

  • Nick Swaney (#138) - A 3rd-time eligible forward, Swaney possesses a great shot and has decent playmaking abilities. A University of Minnesota-Duluth commit. Personally, I think he would be someone that should only be taken in a very late round, given that he is in his final year of eligibility.
  • Skyler McKenzie (#176) - A 2nd-time eligible forward playing on the Portland Winterhawks, McKenzie is a pint-sized firecracker that injects a lot of energy into any line that he’s on. He was 2nd in the WHL amongst draft and draft +1 forwards in 5v5 G/60, 2nd only to Matt Philips. To put that in perspective, he had a higher scoring pace than Kailer Yamamoto, Brett Howden, Dillon Dube, Sam Steel, Noah Gregor, Tyler Steenbergen, Eetu Tuulola, Ty Lewis, Morgan Geekie, and Tyler Benson. Plays with a very serious edge to his game, despite being only 5’8” and 161 lbs. He has an awesome wrist shot, which makes him very dangerous around the faceoff dots down low, and can handle the puck at very high speeds easily. A fantastic skater that gets a lot of acceleration with few strides. He’s more of a shooter than he is a playmaker, but he has developed into one of the best goalscorers in the WHL, and should be one of those players that you should be entertaining as a late round pick.
  • Ryan Hughes (#169) - Yet another forward from the Portland Winterhawks, I have already gone over Ryan Hughes’ profile in quite some detail here. Despite being ranked 169th by NHL CSS, Hughes is definitely a standout in this sample. For a detailed overview, please see the profile in which I have already covered him.
  • Jayden Halbgewachs (#168) - Yet another 3rd-time eligible forward, playing with the Moose Jaw Warriors this year. He’s extremely undersized, but has blossomed offensively into a great goalscorer. An excellent skater that plays a fast-paced game. He’s smooth on his skates, keeps a wide stance, and accelerates quickly on his edges. Has a strong stride and can elude defenders with speed and agility. Moves well through traffic and explodes down the ice when given space. He has a good shooting arsenal, and can get good acceleration on the puck without holding onto the puck very long. He’s not afraid to shoot either, and is dangerous in the slot. He’s god at using defenders to change the angle of his shots and deceive goalies. His hands are great, and he shows good stickhandling and patience with the puck, and limits his turnovers. I hear a lot of comparisons to Brayden Point.
  • Linus Weissbach (#139) - A very speedy winger playing with the Tri-City Storm this year in the USHL. A 2nd-time eligible forward, he has quick feet and can handle the puck at high speeds. Undersized at 5’9” and 154 lbs, but he’s so incredibly mobile that he can cut through defenses like a hot knife through butter. He’ll be playing for the Wisconsin Badgers next year.
  • Oliver Castleman (#164) - Oliver Castleman may be one of the most underrated prospects in this draft, and I can’t figure out why people aren’t talking about him. He is the youngest player in the draft. Had he been born a day later, he’d only be eligible in 2018. He was 9th in the entire OHL in P1/60 at 5v5 for first-time draft-eligible players. What makes him so intriguing is the fact that he had by far the worst QoT of anybody in the top 10. He was in the bottom half in terms of QoT, and yet he thrived. To put this in perspective, the mean P1/60 for a player that had the same QoT or worse as him was 0.63 P1/60, and he was at 1.98 P1/60. Here are a list of players he had better 5v5 production than, despite having worse quality teammates (and being younger than): Matt Strome, Adam Ruzicka, Jason Robertson, Zach Gallant, Alex Formenton. Did I also mention that his P1/60 pace was better than quite a few of the other players projected to go in the first round, like projected top 2 pick Nolan Patrick, for example? And he’s ranked at 164th, which screams bargain. It is a huge surprise to find someone that makes such good use of their ice time and is so productive when given so little to work with. Plays on the PK, so it’s very clear that his coach does count on him to fill that role. He’s an aggressive backchecker that makes great use of his stick to pressure and muscle forecheckers off of the puck and force turnovers, and is very good on the transition. Has great two-way skills, and shows the ability and willingness to apply pressure to attacking blueliners, get in their shooting lanes, and block their shots. This can often result in breakaways for himself. He has soft hands in tight and a good backhand. He favors his backhand quite heavily, and puts himself in situations where its effective, and shows a lot of patience with the puck on his stick, being able to outwait the goaltender very often. His wrist shot is nothing to write home about at this moment, however. He doesn’t have very good velocity on it, and I rarely see him beat a goalie clean with it (after reviewing months worth of his tape, I might add). Shows a good scoring touch on the breakaway, however. A strong skater, he has good balance, and pivots on his skates very well. Shows a lot of mobility in attacking north-south, and able to enter the neutral zone with speed on the rush, and catch defenders flat-footed if they’re not careful. A number of his goals he had defenders flat-out beat, either through a passing play, or from making very smart reads on the play and intercepting the puck around the blueline. When he sets up shop on the forecheck in the offensive zone, he seems to be in his element deep in the offensive zone. He’s someone I’d be very happy with grabbing in one of the very late rounds.
  • MacAuley Carson (unranked) - The fact that someone like Shaw Boomhower is ranked by NHL CSS on the final rankings, but someone like Macauley Carson isn’t has to be one of the dumbest injustices that I can think of. Carson is an excellent first-time eligible prospect, possessing good size at 6’1” and 205 lbs, and plays a hard-nosed two way game that you like to see in a prospect. He’s a strong kid with a good shot, great hockey sense, useful on the PK, and is a beast in terms of 5v5 production. He brings a lot of versatility to the table, and plays a style of game that should translate very well to the next level. He combines all of this with an excellent work ethic, and he shows a ton of leadership qualities. He was originally a winger, but has been moved to center this year. Here’s the kicker about MacAuley Carson: He was 6th in 5v5 P1/60 in the OHL draft class, behind only Suzuki, Tippett, Lodnia, Gadjovich, and Vilardi. He was actually ahead of Robert Thomas, Matt Strome, Jason Robertson, and Morgan Frost. He was also 3rd in GF%Rel. He finished the year with 30 goals, which put him behind only Tippett, Suzuki, Robertson, Gadjovich, and Strome, who are all ranked by NHL CSS. The kid is a coach’s dream, and it is incredibly asinine on NHL CSS’s part that they didn’t even rank him on their final list. I’d take Carson in a heartbeat in the 3rd round. He may be even more underrated than Oliver Castleman is.
  • Zach Solow (#183) - An undersized center from Naples, Florida, who apparently hates the cold. Stands at 5’9” and 181 lbs. His goalscoring abilities are subpar, but he more than makes up for it with his fantastic playmaking abilities and solid two-way game. I really like his hockey IQ. Very dynamic player, that may not score goals a lot, but can compliment high scoring wingers with his vision and smarts. A lot of teams will shy away from him because of his size, but he’d be worth considering to gamble on in a later round.

Sextile 4 Comparisons

Besides some of the players that I mentioned above, only three players really stick out in Sextile 4.

Best Performer compared to Sextile 4

  • Maxime Fortier (#181) - Technically, Fortier is in the Sextile 5, but he’s the only forward that hasn’t been talked about that seems to be outperforming other players in this sextile. Fortier is a 2nd-time eligible forward prospect playing for the Halifax Mooseheads in the QMJHL. I’m going to be honest. I still have no idea why he was passed up in the last draft, when he scored 31 goals and 46 assists in 68 games. This year, he not only showed an improvement in his offensive production, scoring 32 goals and 55 assists in the same number of games, but his defensive abilities showed some improvement as well. Fortier possesses lightning quick feet, and he’s effective on both the PP and the PK, where his skating abilities are very useful at initiating the breakout and transitioning out of the defensive zone on the rush. He’s very dangerous on the rush, as he has a very good set of hands and a good shot. He has a very fiery personality, and is the type of guy you can count on to energize any line that he’s on. Reminds me a little bit of Andreas Athanasiou. Might not have the same level of puck control, but I see a lot of similarities when watching them. While he’s not producing at the same level as other draft +1 players like Vitalii Abramov or Vlad Kuznetsov, he’s a very solid pick for someone this late. Great skater and works very hard, but has average hockey sense. Regardless, he’s one of the best overage players eligible in this draft


  • Cedric Pare (#146) - For some reason that is far beyond my comprehension, Cedric Pare rose from 203rd at the midterms to 146th in the final rankings, for a net gain of 57 spots. It is beyond me why he has risen. He has incredibly limited pro upside, and is not someone I would recommend drafting.
  • Skyler Brind’Amour (#157) - Again, another player that rose from 202nd to 157th between the midterm and finals. Brind’Amour had the single worst P1/60 of any of the prospects in the sample. However, this comes with a serious caveat: he only played 8 games. Now, in those 8 games, he did not look particularly good, scoring only 1 goal and being a -10 and all, but to give you a proper evaluation of him on such a ridiculously limited sample would not be fair to him. It is literally impossible to evaluate him based on this data, and I have not actually had the chance to view him, given such a limited viewing opportunities. He’s Rod Brind’Amour’s kid, so there is an element of pedigree in why he would get selected in the draft. Look for him at Michigan State, as he has committed to play for the Spartans starting in the 2019-2020 season, much like his father did.

Sextile 5 Comparisons

Sextiles 5 and 6 are a bit trickier to compare, because these sextiles actually had a higher average P1/60 than Sextiles 3 and 4, partly due to the fact that there were a a larger number of overaged players in these sextiles. Further, many of the players who do stand out in this sextile have already been discussed.


  • Andrei Grishakov (#174) - This Calgary Hitmen left winger saw himself rise 23 spots since the midterms, but there really isn’t much that stands out about him in any sense. His offensive abilities are below average, and his defensive abilities are only slightly above average. There are definitely players with a lot more upside to take this late in the draft than him.
  • Nicolas Guay (#195) - Guay is a defensive center that is used primarily in a checking role to shut down other opponents, and is counted on to play big minutes in defensive situations. While he doesn’t produce, his 14.14 GF%Rel is quite impressive, especially when you consider the fact that the Drummondville Voltigeurs were a horrible team this year. However, his pro upside and offensive ceiling are extremely limited. I don’t see him developing into an NHLer.
  • Cole Purboo (#189) - This Windsor Spitfires centre is one player that I would be completely surprised if he ever makes it as anything more than an ECHLer. His upside is extremely limited. He plays a gritty style of hockey, battling hard in front of the net, looking for those greasy goals, and that’s where he’s at his best. However, he generates so few shots that he’s rarely involved offensively. I don’t recommend getting him.

Sextile 6 Comparison

Best Performers in Sextile 6

  • Denis Mikhnin (unranked) - Mikhnin was a big surprise this year playing for the Rimouski Oceanic in his first year playing in North America, where he scored 20 goals and 19 assists in 59 games, despite playing more of a middle 6 role than a role on the top line. Mikhnin’s P1/60 was 1.91, which was better than some of the players in Sextile 1. Further, his 18.72 GF%Rel is one of the best of any forward that I’ve seen in this draft. It would be a huge surprise to me if he gets drafted, due to not being very well known, but he’s someone that I think is a lot better than he’s given credit for.
  • Sean Dhooghe (unranked) - Probably the smallest player I’ve ever seen in my life at only 5’3” and 139 lbs. Dhooghe is pretty much a pure goalscorer, who possesses an incredibly good wrist shot. When he’s on his game, he can be a force out there, and is very dangerous on the transition, as he’s a fantastic skater, and is able to burn up ice very quickly. He recently played the role of hero for Team USA at the U18s, where he scored several critical goals to help lead the team to gold, and finished the tournament with an impressive 9 points in 7 games, including 3 goals, and a +8. A lot of people in the scouting community have embraced the kid despite his size because of his ability to elevate his game when his team needs him the most. I would love the Wings to grab him, partly because I’d love making the “Dhooghe Howitzer” pun for the next decade or so, much to the chagrin of the rest of the league. He will be playing for the Wisconsin Badgers starting this upcoming Fall.


  • Logan Cockerill (#214) - Cockerill is the 3rd lowest ranked prospect on the entire NHL CSS ranking list, and it may be for a good reason. His hands are very quick, and I’m very impressed with his ability to handle the puck, but nothing really stands out about his production when you look at it, and his pro upside seems to be rather limited.
  • Josh Paterson (#205) - Paterson’s draft stock has plummeted since the midterms (where he was ranked 163rd), falling a total of 42 spots in NHL CSS’s final rankings. The Saskatoon Blades were by no means a good team, and one could partly excuse Paterson’s production if it were only based on that fact, but he has such poor offensive numbers while playing against poorer QoC, that I don’t think there’s much prospect for him to play in the NHL.
  • Otto Makinen (#211) - This Soo Greyhounds center is primarily used as a shutdown center, but demonstrates very little pro upside. Part of that may be his deployment, but he lacks the tools that one looks for in a pick. It’s not all bad though. He has some pros. He does a lot of dirty work that makes his teammates better. He’s good in battles along the boards, and comes away with the puck a lot. Plays a gritty effective game and helps get the puck to his teammates who are in a position to make a play. His attention to detail is good, and he does a lot of the little things right, like win faceoffs, play defense, and establish possession for his team. However, his skating is very average, and his lateral movement isn’t very good. He’s more effective in tight spaces where he can leverage his strengths and his smarts to establish positioning than when he’s in the open ice. When doing research on him, I discovered that he was actually ranked #13 on NHL CSS’s list last year at some point, but fell off and wasn’t drafted. This will be his second time being eligible. He projects as a third or fourth liner, but could be effective in that role if put in a position to succeed.
  • Kyle MaClean (#209) - MaClean isn’t a good player by any stretch of the imagination, but a lot of that may also be because his coach has put him in a very bad position. He faces the 6th highest QoT in the entire OHL for a forward, and no other player has QoT as low as him while facing that kind of competition amongst forwards. That being said, he’s only -2.03 GF%Rel, which isn’t terrible given his circumstances. However, his pro upside is basically nonexistent.

Limitations, Future Directions, and Conclusions

First and foremost, thank all of you for being such awesome readers and getting to this point in what is, admittedly, an extremely detailed statistical jargon-ridden article. I hope that you have taken at least a few things away from all of this, and leave this article more informed about many of the prospects than you were when you had just started it.

This article represents the culmination of several months worth of compiling data and reviewing prospects from NHL CSS’s rankings list. There are a lot of things that I can see being done with this data in the future, such as looking at how strongly measures of productivity such as P1/60, G/60, A1/60 predict future performance across seasons. However, there are some admitted limitations of this article. First, while I would have loved to report data on other prospects such as Cale Makar, or any of the prospects in the NCAA, no such data has been compiled for them. Second, I realize that there are quite a few prospects who only played a small sample size of games that were discussed. In future analyses, I will probably end up compiling this data and including it as factors in my analyses to try to get some semblance of an idea of how many games it takes for a player’s productivity to stabilize. Finally, while I did not include player age as a metric, this probably would have been a good detail to include. However, this has been a very ambitious project, and this is really only the start of what I hope will lead to more follow-up work.

In my next article, I will be expanding the talk to discuss, from an analytics perspective, how the draft-eligible defensemen measure up to each other. I can’t promise that this article will be out by the time that draft day rolls around, but you can expect to be hearing more from me around draft day, when we will finally find out what prospects the Wings have added. For me, it will be like a kid waking up on Christmas morning to finally be able to talk in certain terms about what kind of prospects we have landed, and I am itching for Friday and Saturday to come.

Anyways, that concludes this article. I would like to apologize for not doing this article sooner, as I have been up to my eyeballs in busywork in my professional life, and had to put off writing this for a while. But I’m hoping that I will have Part 2 written sooner, rather than later.