The (Data) Story So Far
We’re well past the halfway point of the hockey season, and it feels high time to share some findings through the tracked data I’ve generated through this point. As of time of writing (Feb 10, 2024), I’ve watched 324 performances of players spanning around the globe from the likely top end of the draft right down to “what was I thinking jotting this name down at 2am on a Tuesday”. So far this season has been… a mixed bag, but an exciting one nonetheless and I thought it was time to share some deeper findings.
The data I track is strictly focused on events where the player is directly involved, and entirely collected at 5v5. The idea is essentially to provide a stable backbone of what is actually happening on the ice, and how much the targeted player factored into all the various goings on in a hockey game. Blueline transitions are vital, but how those transitions are generated is vital to understand, and how a player succeeds or doesn’t succeed is even more pivotal to explore. Shots are great, but where are they coming from? How often is their team shooting? Are they capable of distributing the puck to others who then shoot it themselves? Are players getting involved in their own end breaking cycles, causing turnovers and getting play moving? Raw production data simply does not capture this, and relying solely on such data is simply scratching the surface of what’s out there in the NHL Draft landscape.
This data is not what I base my final judgements on, I should note. If I did, Vsevolod Gaidamak and Benjamin Baumgartner would’ve been in my first rounds as some of the best high danger shot generators in the universe. What it does do, is give me a memory bank to go back to and see exactly who is doing what and how, with the video archive providing vital context to get to the qualitative bit. I honestly don’t think I would feel comfortable doing this work without doing things how I do it. I can jump from player to player easily, I’m agile, and over the years I’ve honed in on what I believe are the core aspects and meaningful events that occur in a hockey game. In the simplest terms, you are supposed to take the puck from your opponents, take it towards their net, and put the thing in the net. That’s really all there is to it. Defending, Transition, Offense. Defending, Transition, Offense. Defending, Transition, Offense. Over and over and hopefully you win.
My raw tracked data is locked behind my consulting business Fractal Hockey, so unfortunately I cannot share specific metrics and information publicly, but I felt that it would be prudent to give a peek behind the curtain and point out some intriguing data and video focused findings thus far, as many key players are well past the point where I feel comfortable about my read on them.
Luke Misa: Analytics King
Over the 2023 offseason I built a tool to take raw data inputs from my tracking and filter them down into text fragments to help tell the story the data tells. In almost every case I’ve dug into, they “feel” right, relative to my thoughts on the player in general, and while there are some standouts, the one that appears to rise above the rest, especially relative to their public valuation, is Mississauga Steelheads winger Luke Misa.
Luke Misa’s (Mississauga #33) directional changes and playmaking skill on display. Cross-ice vision and forechecking are key strengths of his offensive game.
Almost across the board, Misa is a standout. Massive involvement and efficiency in offensive transitions (OTS%), with elite puck carrying ability, with 34.6% of his transitions coming through carries, maintaining control on 92.9% of attempts. Once the puck is in the offensive zone, Misa is among the leaders in my work in attempting passes to the slot (DPass/60) and creating dangerous chances for teammates (DShA). Misa is also a significant standout when it comes to generating his own shots from scoring areas (iDSAT/60). Combining all of this puts him in rare company in my work over the years including Zach Benson, Logan Cooley, Seth Jarvis, Matvei Michkov, Will Smith, and Conor Bedard among others. As a share of his team’s shot attempts, nearly 20% of them are taken by Misa from scoring areas which puts him again in rare company, but combining >10 dangerous shot assists is something I haven’t seen since starting to track shot assists more granularly. The story continues to be impressive off the puck, as Misa is again right near the top of the class in the offensive zone, generating turnovers and gains of possession as often as names likely to be selected in the top 15 picks such as Zayne Parekh, Konsta Helenius, Berkly Catton, and the big boy Macklin Celebrini.
At the end of the day, this is a player who at worst is highly capable of transporting pucks up the ice with speed and dual-threat offensive creation that is both dangerous off his own stick as well as involving teammates to create dangerous chances of their own. In my view, a second to third round valuation is bordering on absurd, at least on paper. So why the relatively poor outlook on Misa’s future? Setting aside the common “smol” argument people throw around, there are legitimate gaps in Misa’s game that could threaten how effective he could be in the NHL.
Where Is Zayne Parekh’s Production Coming From?
Zayne Parekh is on track to be the first OHL defenseman to hit 100 points in their NHL Draft season since that Bruce Cassidy did it back when Billie Jean was first hitting the airwaves. My own experience watching Parekh, one of my most heavily watched players thus far, has been extremely perplexing, however. The data ingested isn’t necessarily poor but it doesn’t really seem to line up with what you would expect from a potentially record-setting offensive defenseman. In some areas, there are definitely areas of concern that could negate what offensive advantages he may create at 5v5.
Possessionwise, Parekh is losing the shot attempt battle. He touches the puck a ton and has an enormous load of pass attempts to go with it (PassAtt/60), with just over two attempts per minute of 5v5 ice time. Of all those attempts, over a third result in a loss of possession. Generating offense is a curious area, generating 8.5 shot attempts per 60 minutes of ice time, but the real interesting it is the raw volume of shot attempts Parekh puts on net himself. Parekh currently registers over 21 shot attempts per 60 minutes off his own stick, putting him well ahead of the next challenger Carter Yakemchuk down at 17. 40.8% of Saginaw’s shot attempts taken with Parekh out there are his, and removing dangerous attempts leaves a whopping 31.8% of their total attempts being perimeter shots from Parekh. These numbers would put him well ahead of any NHL defenseman with regards to total shot attempts relative to team shot attempts. Dougie Hamilton has been the only defender this year to pass 20 individual shot attempts, and those still account for just 27% of New Jersey’s attempts with him on the ice. Parekh is in uncharted waters when projecting to the NHL in the shooting regard.
The major strengths and areas of improvement for Parekh (Saginaw #19) with his mobility to draw defenders away from pass options and generate off of perimeter shots. On the other hand, rush defense and board battles are… A work in progress.
The real trouble comes in the rest of the picture. Parekh is clearly one of the most highly involved transition defenders in my data set, largely driven by his long stride, covering stretches of ice easily to snag loose pucks in low to moderate pressure situations. In those situations, he’s generally quite effective in at least turning play around and moving things offensively again, driving his massive rate of pass attempts and overwhelming preference for passing pucks across bluelines (PassPref). Otherwise, Parekh is a player who, at least to me, brings a lot of concern. Many high point defenders have been drafted in recent years, only to underachieve potential, and in my view, the commonality is the lack of high end skating ability combined with an aversion to putting in that extra “oomph” to use a scientific term in close quarters and shutting down opposing rushes and cycles. This to me has been Parekh’s most serious weakness this year that his offensive output may mask for now. The video below displays some of the more… questionable defensive plays I’ve seen out of Parekh this year that brings me significant pause when it comes to valuing him as much as his raw numbers would indicate.
All of this having been said, Parekh is a confident player constantly looking to generate offense using misdirection, pressure and his shot to get pucks into space and on net. All of those tools especially come out on the power play, and Parekh does bring strong, volume oriented offensive output clearly shown in the data. The sustainability of all of this is called into question when incorporating his play in his own end and his reliance on massive shot attempt volumes that likely factors into why he isn’t locked in at the top end of NHL Draft boards.
What Exactly Are You Getting in Macklin Celebrini?
There has already been endless talk about Macklin Celebrini and we haven’t even seen a single NHL team eliminated from playoff contention. I’m of the belief that drafting first overall is usually no guarantee of a franchise turnaround considering the sheer difficulty of playing in the NHL in the 2020s, but it certainly helps to have your pick of the litter. We’ve seen some surprises in recent years, we’ve seen first overall picks struggle to translate, and we’ve seen franchises reborn through the name selected first. The 2024 NHL Draft to me is characterized by a somewhat limited selection of players that you can clearly identify as potential major difference makers, but flying pretty high above a sea of uncertainty filled with potential role players and interesting development potential. At the top end, Macklin Celebrini reigns supreme as my #1 player in the NHL Draft, but I do believe he has some competition, and this can be indicated through both the video and the data I’ve tracked to this point.
Celebrini (Boston University #71) displaying his scoring area instincts, playmaking and forechecking/defensive skill.
Celebrini will certainly be a player that should have little issue translating to the NHL as a well rounded offensive centre. While to some he may not carry the ideal frame for an elite NHL centre, the way he’s wired to play the game certainly goes quite a ways to overcome those concerns. In my tracking he may not be the most physical player throwing himself into bodies and in front of pucks in his own end, but he is regularly anticipating the flow of play without the puck, causing turnovers in the offensive zone and breaking up cycles with well timed stick checks and pass interceptions in his own end. There isn’t a particular concern to me of Celebrini being one of those young players who only wants the puck on his stick and solely drive offense.
The skill level on display with Celebrini is also highly noteworthy in it’s practicality. Is he carving through opponents at 5mph at times like 2024 counterpart Ivan Demidov? No, not really. In fact, a teeny tiny criticism I’d make of Celebrini is trying to take on too much pressure on his own, skating cross-ice into double, triple and even quadruple coverage and can lose possession. As mentioned before however, Celebrini doesn’t give up on plays where he may lose possession. He’s right back on the horse, lifting sticks and getting into passing lanes to get the puck moving towards the opposing net again. In the data, this also seems to be a bit of an overinflated concern as he’s carrying pucks up the ice 24% of the time in my data and successful 90% of the time, more than any other transition type.
Where Celebrini does and likely will thrive down the road is in the offensive zone. His defensive transition data is strong, with good stick check involvement at 48% and successful on 63% of those transitions. There is a bit of a lack of physical intensity and strength, but it’s perfectly okay to see gaps in the games of elite players, especially when there are so many bright spots to point at. The text analysis fields above indicate that Celebrini is indeed a dual-threat offensive player, and the proverbial eye-test would agree. Whether defenders like it or not, Celebrini gets to scoring areas and shows a strong willingness to do so many, many times over. He’ll fire passes into the slot, slither around the boards behind the net, snag rebounds, wrap pucks around the net for scoring chances, fake wraparounds, and find ways to battle his way to the slot with and without the puck to gain higher danger scoring chances. What Celebrini certainly isn’t is a volume scorer. He picks his spots well, and takes shots from the perimeter 32% of the time which is impressive for an NCAA freshman to say the least. Ivan Demidov may get chances from closer more often, but the MHL is not the NCAA and Cayden Lindstrom may do the same in my data, but he’s also blessed by god with physical strength and confidence. Celebrini feels a little bit more refined in his decisionmaking and gives me no doubt that there is a willingness and ability to create offense in a manner fitting for the NHL.
So what is a team getting in Macklin Celebrini? I don’t see him on the echelon of a Hughes, MacKinnon or McDavid who will take whatever team he ends up with turn into a serious threat to the league, but I would certainly take a player like Brayden Point on my team any day. A major part of a contender one day, especially with some improvements to his strength under pressure, but the processing skill, shot selection and ability to generate dangerous offense in a variety of ways bodes well for his NHL future.
Ivan Demidov: Your Next Favourite Superstar… Maybe.
If you haven’t yet, go to YouTube and watch some Ivan Demidov highlights if you can find them. It doesn’t take long to see how unbelievably entertaining this player has been throughout the season in the Russian junior league. At time of writing, Demidov has put up 57 points in 28 games, but looking deeper things get increasingly outrageous. 27 have come in his last 10, 18 of them at even strength. He’s worked his way to a 50% involvement percentage in team offence, a 47% involvement in team offence at even strength, but the data I have tracked on him to this point has been absolutely bananas. I will say that in Demidov’s particular case, my first tracked game was a KHL performance, which we’ll get to later, but other than that I have three games of data of his. This number will obviously grow and if anything changes I’ll certainly factor that in, but for now, the sample is small, but so ridiculous and fun that we’ve got to chat about this guy. The image above also factors in the tracked KHL game so the metrics are slightly different due to the inclusion of that game in the database.
Taken as a whole, Demidov has found himself participating in 50% of SKA-1946’s offensive transitions, maintaining control of them 83% of the time constantly receiving pucks across the defensive blueine and springing into the offensive zone with the puck under his control. Nobody in the draft can match that. The first game tracked, Demidov controlled 22 of 26 offensive transitions. He is averaging a controlled entry into the offensive zone every Simpsons Intro’s length(ish) of 5v5 ice time. Ridiculous. He’s passing the puck over twice a minute, completing 77% of his passes, with a pass directed at the slot every 70 seconds, or the length from the start of the Simpsons intro to the point where they sit on the couch. That rate has also increased in each successive game, but that may just be the selection of games. He’s completing these passes every 2:30 of ice time. He’s taken 27% of his team’s shot attempts, 72% of those from scoring areas, and set up another 27% of everyone else’s. Across the board, Demidov has been an absolute nightmare offensively over these games, and if recent video is any indication, that hasn’t stopped.
A common refrain from players like this is “but what about the defense” and with Demidov, the numbers nowhere near as tremendous, but they’re at least adequate, if not strong. His 10% involvement in defensive transitions is slightly below average but he’s heavily involved with his stick checks and finds strong rates of success in those situations. Has he thrown a single check in transition? No. Did he only throw a hit in the most recent game with data? Yep. This is a look into some of the pitfalls but this also isn’t what you’re drafting Ivan Demidov for anyway.
Behold some of the dual-threat creativity and remarkable confidence Demidov (SKA #11) displays. Keep in mind the reliance on lower pace movement and moments of overconfidence that were major factors in his KHL struggles.
The catch to all of this is that Demidov has an area that needs improvement that looms over all of this. No, it isn’t just that the MHL’s quality of play is an adventure to itself, but something much more to do with how Demidov’s game will actually translate to higher levels. Through all the insane results, Demidov actually started his season somewhat slow. I was among those who thought he was ready for a KHL introduction after last season, and while Demidov cracked SKA’s lineup, the KHL is not the MHL. Demidov’s skating stride and puck handling often relies on changing his angles and maneuvering his body around the puck to find passing lanes and cross up defenders. The issue is what seems to be short leg extensions in his stride that hamper him avoiding in transition. Opponents can close on him quickly, and he is a player who wants the puck, and wants to hang onto it. In the MHL, he turns himself around turning that forechecker into mist with a skill move, or uses the speed he does have to beat the weaker competition. His tracked KHL game did involve him in transitions at a similar rate, but he was relying more on northbound passes under pressure that were turned over, and he wasn’t given a chance to really use his stickhandling and evasion. The higher pace of play left him behind on a few defensive transitions, and settling pucks was an issue. It simply felt as though the speed and physicality of the game snuffed his ability to play how he likes to play. His early MHL results were good, but his level of dominance wasn’t there compared to what I’ve seen, and his ability to control the puck and change angles has improved, as well as what seems to be more of a willingness to be in motion for pass receptions and finding open ice a little more has led to huge gains for Demidov over the season. In my view he’s gotten more surgical, calculated and mobile, which I am fascinated to see at higher levels. It is far past time for Demidov to get an extended look against men somehow until the end of the season. This level isn’t a challenge, but he’s a player who I believe needs to be challenged. If this guy hits, he’ll be a lot of people’s favourite players, but I could see a world where things don’t quite come far enough to reach the potential he’s shown over the last two years.
As I’ve said, data is just information. It helps tell a story of some kind of event that is transpiring, but it doesn’t fully explain everything or capture the full experience of the event itself. These players have stood out in some manner on paper, and the video has provided fascinating context to that information. It also felt prudent to simply put out there what exactly the consensus 1st overall selection actually does out there, and if there may be any competition out there that isn’t being considered as much as it should. Misa drives tremendous results, but does display some potential limitations at higher levels. Parekh scores a ton but beyond raw point production on a powerhouse OHL team, there are issues lying in the data. Celebrini is highly talented, projectable and diversified in his abilities, but to me a question remains if there is any threat to his label as #1 overall this year. If anyone is going to really challenge Celebrini, it may be Demidov simply for the raw superstar potential of this player. Sure there may be issues now, but with the right patience and guidance, where could he be in 5 years? If you had a pick of a Brayden Point-type player or a Artemi Panarin-type highlight reel machine? The data suggests there’s a discussion to be had, and the video seems to lead me to the same conclusion. The next few months could be fascinating, so hold on to your butts.
