It is time to view the 2019 NHL draft through the eyes of Mr. Potato (who is similar but legally distinct from Mr. Potato Head.)
First, the ranking:
1. Kaapo Kakko
2. Jack Hughes
3. Samuel Fagemo
4. Arthur Kaliyev
5. Bowen Byram
6. Ville Heinola
7. Alex Turcotte
8. Jakob Pelletier
9. Nathan Légaré
10 .Thomas Harley
11. Dylan Cozens
12. Brayden Tracey
13. Trevor Zegras
14. Ryan Suzuki
15. Mikko Kokkonen
16. Raphaël Lavoie
17. Kirby Dach
18. Ronnie Attard
19. Alex Beaucage
20. Cam York
I have not followed the coverage to this draft very closely, so I cannot say how this ranking compares to the "consensus," with a few exceptions at the very top.
The potato's preference for Kakko over Hughes isn't too surprising, as the model has long had a certain affinity for players playing in men's leagues. The biggest lesson I have learned from this project is that teams have historically undervalued players who can play in the SHL/Liiga/etc even if their production is not incredibly high.
I do think there is a certain concern here, in that Hughes realistically put up about as high a point total as could be expected for the league he is in, and I've wondered if I need to implement some sort of capping system, to scale production against some sort of practical maximum. If 112 points in 50 games is basically the most anyone can expect to get in a league, then maybe he should get more credit for hitting it than the model currently gives him. But that's a project for another day and I don't think Kakko over Hughes is all that controversial.
What is controversial, most likely, is the #3 choice, and it's a good illustration of why you shouldn't let a potato make picks. Even if you want Fagemo, he can almost certainly be had in later rounds, whereas the potato is going to shock the world by taking him in the top of the draft. It will be interesting to see how that plays out.
Yes, the potato factors in that it is Fagemo's 2nd go-around (see the Introduction,) but players who play full-time in the SHL and even put up relatively modest production (25 P in 42 GP for Fagemo,) are virtual locks to become NHL players (Lindholm, Franzen, Backstrom,) and the penalty for being a year older just doesn't hurt him enough to keep him out of the top-3. Johan Franzen, interestingly enough, was 24 years old when he was drafted out of the SHL with pretty similar production. Again, I wouldn't take him in the 1st round but, but I would roll the dice in the second round even if he's projected to go later, as I think the benefit out-weighs the risk even at pick #35 or so.
Arthur Kaliyev would likely be my pick for the Canucks at #10, based on the data. I feel like he is the guy that the group-think might be allowing to fall this year. His production is excellent, he's not small, he's not old, and he's not playing on a team where he's benefiting from an obvious superstar. He lead the Bulldogs in points by a margin of 23, with the second-place producer being 19-year-old Matthew Strome. Overall, Hamilton was not a great team, and guys who score 50 goals, 100 points in the OHL pretty much never miss unless they are playing for a juggernaut that has a future NHL star (like Dane Fox on Connor McDavid's Erie Otters)
My immediate go-to explanation for why players like Kaliyev are lower-ranked than their production would indicate is skating. This, I believe, is the biggest weak-spot of the potato. Some players can produce big numbers in juniors with poor skating and there is just no way to account for this with the current data available to me. When I read scouting reports of some of the potato's big misses historically, this is a somewhat common theme: "He needs to work on his skating."
Is this what will hold Kaliyev back? It's hard to say. The consensus of scouting reports seem to say that while his top-end speed isn't great, he is strong on his skates and has above-average edge-work. But how do I quantify this? Does top-end speed have better projectability, or is it edge-work that's more important? These are the sort of things I would be looking into trying to collect data on. I think robust data around skating ability is the single biggest factor that teams should be looking into if they are not already.
For now, the potato has no choice but to dismiss it as intangible noise. After all, it is just a potato, and cannot take things into account unless they are factored into the basic formula it uses to sort players. As such, Kaliyev looks like a stud and rates as the 4th best player in the draft.
This draft looks to be an interesting one for the potato. Of the 15 or so drafts I have in my database, it appears likely to be the one with the most divergent top-5. It will be fascinating to see what it is about some of these players that the scouts don't like, and track their progress over the next 5 years.
Scouts: What do they know? Do they know things?? Let's find out!