Discussing the 2020 NHL Draft
To conclude their annual draft coverage, Colin and Ary update readers on how they evaluate prospects, talk draft strategy, and set expectations for the Senators’ draft class
It’s finally time. As the first round of the 2020 NHL Draft begins tonight, we thought it was time to reflect on the long year it’s been for prospect coverage here at Silver Seven.
This year’s coverage spanned over 40 articles and 9 exclusive podcast episodes, including 11 long-form player profiles on the top players in the draft, well over 100 draft-eligible skaters highlighted in our region-specific pieces from around the world, and countless interviews, clips, and commentary from scouts, analysts, and players.
This wasn’t our initial plan, but the more we thought about the scope of our coverage, the more we wanted to rise to the occasion for the biggest draft in franchise history. Through the trades of superstars and fan-favourites, combined with their original seven picks, the Ottawa Senators had stockpiled 16 selections. At the time of writing, that number is down to 12, and we’ll surely see further movement in the coming hours. But there’s an energy around the team; a mix of nervousness and excitement.
While we’ll have lots to keep you busy on the site today, we wanted to bring back a reader favourite: a book-end piece where we discuss how we evaluate prospects, share our thoughts on draft strategy, and soak in the final moment of reflection before Pierre Dorion, Trent Mann, and co. are on the clock.
Note: If you’re looking to see how our thoughts have grown over the last two years, click these links (2018, 2019).
This year, you’ve expanded your coverage of draft-eligible players. What’s changed about your prospect evaluation during this time?
AM: Here’s what I wrote back in 2018:
I’m a huge fan of risk-takers, and with all positions, I think players are only worth drafting if you think that they have top-six, top-four, or starting goalie potential. With forward tools, I value those who are the linchpins of their teams, and use statistics such as even-strength primary points (looking only at goals and first assists) to see who are the primary drivers of offence. It means a lot to me if you can do more with a worse supporting cast — good players on bad teams — or get put into a top role as a first-year player on a strong team. I also tend to favour players with later birthdates (Kotkaniemi, Wahlstrom) over those with earlier ones (Zadina, Tkachuk). This, of course, means that I have some blind spots. It’s hard to evaluate European players playing in non-Swedish/Finnish leagues due to lack of information. I was also a bit skeptical of a pick like Alex Formenton, who despite having a decent first-year in the OHL, was buried on London’s fourth line. I’ve learned to try and recognize good “programs” — London, Sault Ste. Marie, Penticton (BCHL), Boston College/University, etc. — and put some weight in the talent evaluators of those perennially successful franchises recruiting certain players, but like all talent evaluators, I’m a work-in-progress.
What’s stayed the same? The overall strategy of using draft picks exclusively on players who have top-six forward, top-four defence, or starting goalie potential.
That’s not to say that a basic level of positional sense is important — a ‘floor’ that allows a prospect to win the trust of their coaching staff and play prime minutes is key to ever allowing a risky game in any zone to translate — but I’m a fan of any team, especially a rebuilding one, swinging for the fences. To use a baseball analogy, you want to get yourself as many plate appearances as you can to try to do something big, and that’s what Pierre Dorion has done in stockpiling picks in the top-70.
Where I’ve tried to grow in terms of my analysis is in teasing out context, using a combination of data and video to pick out translatable skills. When looking at the tracking data that we were able to present thanks to the work of people like Mitch Brown, Finlay Sherratt, Will Scouch, and Lassi Alanen, I always have to remind myself that micro-stats measure events, not outcomes or ability. It’s easy to look at a player like Jérémie Poirier, see his great zone exit numbers, and think that he’s suddenly fantastic at that and his defensive-zone game must not suffer as much. However, we have to put those events into context and ask if we’re seeing the results that we want. Is Saint John outscoring their opponents when Poirier is on the ice? Nope. Do zone exits indicate in-zone defensive ability? Nope. Does the way that we commonly see zone exits tracked indicate how Poirier is generating them? Nope. On the flipside, we can see a player with poor boxscore numbers like Dylan Holloway and question his overall offensive ceiling. Without putting those numbers into the context of Holloway’s team environment and the structure Wisconsin had in place this year, we’re going to fundamentally misunderstand what he might bring to the game.
CC: I agree on the importance of looking at contextual factors involving points, and while we’ve certainly highlighted those factors in past years, and there are some fantastic models out in the public that do a lot of the heavy lifting, I feel like there’s some further details that we’ve come to recognize.
One of the big ways I found this affected my coverage this year was being able to better suss out the specific intricacies of each league, and what prospects can do within them to manipulate their opponents. The extra time certainly helped with that, although our new angle of looking at a specific region each week opened my eyes a lot more to the giant differences between the SuperElit and MHL, for example. Similarly to how we know that QMJHL defensive systems tend to be more relaxed than the WHL, MHL structures tend to be way more focused on just getting pucks to the net from wherever a player can; this means that players who generate high-danger chances are all the more notable. I found it helpful when trying to isolate a player’s skill set — will they be able to keep doing what they’re doing once they have to move up in the ranks, or are they merely playing to the extra flaws of their young opponents?
In general, I feel that the importance of isolating a player’s skillset has been amped up in my analysis this year, as that’s what will ultimately lead to a cohesive development plan. Every pick has some level of risk, so getting as clear a picture as possible of where a prospect currently stands makes the bigger questions a lot easier: how much risk is worth mitigating? Which traits do we value in prospects, and how do they factor in for that risk scale?
Those are giant questions, and I think my answer to them has stayed fairly consistent from a year ago, maybe even a tad more amplified. Hockey is still a strong-link game, where you’re better off constructing a roster by maximizing high-end talent rather than elevating the floor. Taking the big, risky swings are worth it, even though statistically some of those swings will inevitably fall flat by not filling in the necessary gaps. Betting on raw talent level, which to me encompasses a player’s cerebral approach to the game along with phenomenal puck skills, are next-to-impossible traits to coach through development that have a high impact on out-playing opponents. Those are the players I value most in the draft.
AM: Splitting out our coverage ‘regionally’ this year was a new idea that we tried, and I felt like it really helped me put players into context. I think we can get so stuck with comparisons and while I get the appeal for fans, I often think we’re more likely to shoehorn a prospect into a role or comparison because of it and set inappropriate expectations. I’ve actively tried harder this year to compare players within teams (i.e. Tyson Foerster versus Evan Vierling versus Ethan Cardwell; or Emil Andrae versus Anton Johannesson) or within leagues, rather than across leagues. Obviously, the latter is something that many say is the hardest part of scouting and why I respect people who humble themselves by putting out rankings — figuring out how to weigh these varying factors is always the secret sauce.
There’s a unique situation this year where a goalie (Yaroslav Askarov) is potentially available to be selected at 5th overall. Is taking a goalie high in the draft a chance that could be worth taking?
CC: I think my ranking of Askarov at 35th made it particularly clear that I just hate goalies. Kidding, obviously, goalies can be and often are the most important player on a given team. Heck, the Jack Adams award is essentially given every year to the coach who rode the hottest goalie. Even last season, Connor Hellebuyck’s 9.1 Standings Points Above Replacement (SPAR, from Evolving-Hockey) was more than any individual skater.
But, and this is a big but, predicting which goalie is going to be elite in any given season is next to impossible. I discuss this in-depth in the full profile on Askarov (jump to the ‘Drawbacks’ section), but there’s only been one goalie since the dawn of hockey’s data era that has defied the odds by stringing together numerous consistently elite seasons, and even then there’s some question given the shoddy shot location data from his home arena.
The point is that goalies have an inherently limited ceiling compared to skaters. Askarov could be the best goalie in the world for one year, but the degree of certainty that he’ll stay elite the following season is significantly smaller compared to elite skaters, especially forwards. There’s plenty of elite forward prospects in this draft class that would be a gem to have slip to 3rd or 5th overall in most years, so spending that chance on a goaltender is a move that would leave me scratching my head.
AM: I don’t have much to add to what Colin is sharing, or what was written in the profile on Askarov. What I will say, though, is that I get why the Senators are entertaining the possibility with their two high picks, and why, despite everything we know about goaltenders, Askarov is ranked in the top-15. He represents something that’s extremely difficult for all probabilistic endeavours such as this because he’s an outlier. I don’t think many have seen performances like Askarov’s before, so we get into uncharted territory and many are rightly scared to be proven wrong by a goaltender who could theoretically provide the most value out of any player in this draft.
Why I wouldn’t use a top pick on Askarov is simple: Sens fans should know all too well about the many market inefficiencies that exist around goaltenders among NHL teams, and I’d rather spend an ‘expensive’ asset like a first-round pick on a player who has a top-line or top-pair ceiling that — unless you play for crappy ownership — usually isn’t available on the trade or free agent market. Take Craig Anderson, the best goaltender in franchise history. Ottawa acquired him in 2010 for a younger, struggling Brian Elliott, and watched Anderson guard the crease for the next decade. The three best goalies since 2018? Robin Lehner (ouch), Ben Bishop (double-ouch), and Anton Khudobin — three players who have played for four, three, and three teams respectively over the last five seasons. Scouting the goalie market in Sweden, Finland, and Russia also represents a potential source of pro depth or quality backups that with enough time, talent, and opportunity, could turn into a NHL starter and provide value without prohibitively harming the cap.
Do you approach high draft picks (like #3 & #5) different than picks in the middle of the first-round? Or is it best-player-available no matter what?
CC: Strategically speaking, I think there has to be some form of difference in approach. Of all the odd and unique things happening with this draft class, having two top-five picks separated by only one slot is something the NHL hasn’t seen since the Sedin twins. If the Senators like Raymond ahead of Stützle, for example, but feel certain that the Red Wings have them ranked oppositely with someone like Perfetti or Rossi in between, then they may as well take Stützle with the hope that Detroit passes on Raymond.
On a broader scale, it’s still really just an extra layer to trying to snag the best players available. What that means to different teams and scouts will vary drastically, which is why I find the consolidated rankings particularly helpful for gauging reaches and steals. But there’s still very much a change in dynamic moving from the first round to the seventh, by which point the disparity in opinions stretch from one galaxy to another.
AM: I’m in agreement. It makes total sense for the Senators, given their rare situation, to take their relative pick position into context. One of the fascinating things about this draft class is just how bunched up the team’s picks are. There’s #3 and #5, but #28 and #33; #59, #61, and #64. As you get later and later into the draft — and I count #28 as late — you have to take timelines into context whereby the player you’re drafting will likely not be in the NHL or AHL until two years later, and are likely three or four years out from NHL minutes if at all. If the teardown of the Senators over the last 36 months has shown anything, it’s that a team’s situation can vastly change over a short period.
What are some difficulties in evaluating young prospects?
AM: I started to touch on this in an earlier answer, but I think accounting for the varying differences in coaching styles, systems, and league-structure can make it extremely difficult to rank players applying their trade in vastly different environments. Arguably, that was a big difference in the seasons of Lucas Raymond and Alexander Holtz — the former playing for one of the SHL’s deepest teams in Frölunda and the latter had a bigger role on rival Djurgårdens. It’s why in-person scouting, video scouting, and data will always go hand-in-hand when evaluating prospects — all are important. Lacking quality data sources, especially in fringe leagues, blacks out an entire set of information.
I learned a ton this year about different development systems, such as the DEL mandating that each club must have least two U23 German-born players on their roster every year, and how that contributed to the playing time of Tim Stützle, Lukas Reichel, and John-Jason Peterka; or how Sweden’s J20 SuperElit (now called J20 Nationell) is split into a pre-Christmas and post-Christmas format — vastly impacting point totals depending on the quality of competition faced. How should we factor this into our player evaluation, especially for environmental factors that might be beyond a specific player’s control?
Finally, I have a neuroscience degree and work in mental health. Every year, I’m amazed at the amount of ink spilled talking about the psychology of a player in very vague terms, likely measured with flawed tools and not appropriately put into context by those with knowledge of psychometric properties and people who understand sound research methodologies. I absolutely think that a player’s “character” matters, and I hope that teams in a multi-million dollar industry are being careful to measure these constructions in a way that is reliable and valid.
CC: Oh boy, where do we start? Ary brought up so many fantastic points, and for me it boils down to juggling and weighing the dozens of factors that need to be taken into account — what we see, what we read, what we hear, what we uncover in the data, and then ultimately deciding how to weight everything while confronting our personal biases. In each of those aspects there’s so much to dissect, but I think there’s a few that stand out for me.
Watching games, for one, is something I don’t do very often (get your jokes in), a lot of it because of the physical restriction of watching hours of footage through countless streaming platforms. At the same time, relying on your eyes can only take one so far, introducing a mountain of additional biases worth confronting. That isn’t exactly unique to watching prospects, but it gets amped up by a factor of ten when younger players are much less refined, trying new things, and overall just exude a lesser degree of consistency, even for the very top players. This isn’t to dismiss the importance of watching prospects in the slightest, but it’s meant that most of the profile analysis comes from synthesizing information from the more keen watchers, who still generally tend to have a disparity of opinions. While I believe that learning from others’ opinions should be an important step regardless, missing the initial watch phase still leaves room for missing out on key points of analysis.
The data aspect, as Ary alluded to, is and has always been tricky for prospects given the limited resources compared to the NHL. I can’t emphasize enough how spoiled we were this year, though, and I’ll link to a bunch here because they all deserve immense recognition — Dave MacPherson’s stats website Pick224, Will Scouch’s manually tracked data used for his detailed video reports, Mitch Brown’s manually tracked data for the CHL and USDP, Lassi Alanen’s manually tracked data for Finnish prospects, Mikael Nahabedian’s NHLe player cards... this is invaluable information that definitely helped shape our opinions on this year’s draft class. I’m sure all of these people also recognize the tremendous gap in information that still exists between junior leagues and the NHL, with basic things such as shot attempts being limited to only players spending games in European pro leagues. It’s a whole other beast weighing all the other contextual factors. I don’t envy the job of a scout — it’s a daunting process, albeit one I enjoy tackling. When it comes down to discussing a bunch of volatile 17-year-olds, there comes a wall where any scout is forced to shrug their shoulders a bit and just bet on the information available.
How do you view this year’s draft class in comparison to others?
CC: There was a lot of hype heading into this year’s draft class. Lafrenière was already the CHL MVP at 17 years old, the amount of talent brewing in the OHL was evident, and records were being broken in Sweden, Finland and Germany. In terms of the high-end skills available in this draft class, it’s unlike any one I’ve covered to date. To this day I’m still convinced there’s around double the amount of ‘top-five’ talent-level players if they were to be slotted into most other years, and the fact that the players are so distinct from one another has easily made this the most fun year to cover as well.
However, will we be seeing another class like 2003 where seemingly every team was rewarded with at least a couple quality NHLers? I wouldn’t place my bets on it happening, as prospects are still very volatile with plenty of room needed for improvement. The depth doesn’t ‘trickle down’, as I’ve referred to it on previous occasions. That said, it shouldn’t devalue the worth of a 2nd or 3rd round pick — as it is with every year, there will inevitably be steals to be had, some predictable and others completely unpredictable. Ottawa is a team that has publicly stated that it’s rebuilding from the ground up, so that process starts at the draft.
Looking ahead, the best way I can describe the 2021 draft class is perplexing. There’s no consensus #1 overall pick — I’ve seen around five different players ranked at the top from various outlets, with so many others in contention. There’s lots of phenomenal defenceman available too, who I can’t wait to cover when I inevitably write a ‘20 in 2021’ post (hold me to it, y’all).
Personally, though, I’m already more excited for the 2022 class, which at this time features three players who very well may be in a Lafreniére-like tier — Shane Wright, Matthew Savoie and Brad Lambert. The next two years of USNTDP talent is also shaping up to be miles better than the 2020 group... dare I say rivalling 2019? There’s still so much time for things to change, though, which I guess is part of the fun in covering prospects.
AM: At the end of our equivalent article on the 2019 draft, I said this:
I am all onboard the 2020 NHL Draft train, and it appears that Dorion and co. are as well, with the team stockpiling seven picks in the first three rounds in that draft compared to four in 2019. I think the team will be able to get three decent pieces at 19, 32, and 44, and I’d be fine swapping later-round picks for next year if there’s a team that really has a player they like on Saturday. In general, I’m of the opinion that there isn’t a significant difference between most draft classes, and it’s more up to the scouts and development staff to make the best choice given the talent available.
One hundred thousand words later (for real) and I present the more nuanced opinion: like Colin, I think there’s going to be an above-average number of “stars” from the top-end of the draft, but I don’t think there will be the depth that was seen in 2003, 2009, or even 2015. Byron Bader recently mused that there will be approximately 58 NHLers to come out of the draft and “would ballpark there to be 9 to 11 stars in the 2020 draft as well.” The Athletic’s Corey Pronman used time-on-ice to factor in the potential quality of NHL players coming out of each draft and found that, on average, for the 2004-09 NHL drafts “I had an average of 9.7 high-end players and 14.7 very good players, making 24.3 combined. Thus, the definition conforms to the draft value curve used by Schuckers and others, which shows a steep drop around the 10 slot, and a flattening around the 25 slot.” On average, that’s around 25 players who are top-six forwards, top-four defencemen, or starting goaltenders — the majority at the top of the draft class. With Pierre Dorion’s group graduating 44% of his picks to play at least one NHL game since 2008, and only eight other teams having as much draft pick value as the organization has stockpiled for 2020, I expect that it’ll be a long time until we see another draft class like this one for the Ottawa Senators.
What should we be hoping to see from the Senators in this draft? Are there any strategies we their scouting team should employ?
CC: I really hope they take some big risks with at least a few of their many picks, swinging on potential high-end talent while daringly taking on the task of helping them fill out the more teachable parts of their game. Not only for all of the reasons outlined earlier on why these are player I think bear the most value at the draft, but as an organization, high-end players are what they need. They’ve finished in the bottom two for three straight seasons — while they will almost certainly come away with two phenomenally gifted players with their two highest picks, there’s a lot of filling out to do with the roster if they want to build a team with true depth of talent.
I dream of the day the Sens have a centre corps comparable to what the Lightning have in Point, Stamkos, Cirelli, Gourde and Paquette. They’re not buoyed by a couple game-breaking talents, but every single one has the potential to inflict major goal-scoring damage. Betting on high-level talent at the draft is the first and most important step towards attaining that.
AM: I hope that the team drafts two high-end forwards at the top of the class, opts to make most of their picks in the top-70, and swing for the fences to bolster the system with talent and depth for the next five years.
If they are using their pick and/or prospect capital on NHL-ready assets, I hope they prioritize “helping” teams in a cap crunch by taking on a young player a cap team cannot afford, or opting to trade for high-end talent that might need a change of scenery, like they successfully did in the Kyle Turris for David Rundblad swap.
Has the way you evaluate prospects changed over the last year? What makes prospect evaluation and analysis hard for you? Let us know in the comments!
Finally, this marks the end of our pre-draft coverage for 2020. Thank you for reading and discussing with us. Any feedback would be extremely helpful in shaping our processes for the future, and you can feel free to pass that on in the comments section or via Twitter. You can find us at @carteciel (Ary) or @CudmoreColin (Colin).