In case you haven’t seen this yet, Micah Blake McCurdy released his Magnus Prediction Model over at HockeyViz.com. You can see Matthews’ graphic here:
The plots on the top row represent offence (you want a positive isolated number), while the plots on the bottom row represent defence (you want a negative isolated number). Each column refers refers to the part of the results attributed to the column heading. As you can see, Matthews’ isolated impact allows his team to generate 12.4% more threatening offence when he is on the ice, while his team gives up 2.6% more threatening offence. Matthews’ teammates contributed to offence 15.3% more threatening than league average, but defense that gives up 8.4% more threatening offense to the other team.
You can see a quick summary of each player’s isolated impact in the chart below:
Hockeyviz.com Isolated Impacts (Since ‘16-’17)
There’s a few things to point out here. First, Toronto’s 1-2-3 punch up the middle looks to be incredible, as Matthews, Tavares, and Kadri all boast incredible numbers. I heard many people call out Tavares for his defence in the past, but this model put the majority of the blame on his teammates.
There are a few surprises, as Marner looks fairly mediocre in terms of isolated impact, while Hyman looks outstanding. It’s tough to know what to make of this. Marner may be one of those players who “breaks the model”, as it does not fully appreciate his skillset. The model loves James van Riemsdyk, and it’s possible that Marner’s isolated results are dragged down a bit too much from this quality of teammates effect (perhaps the reverse is true with Hyman).
Josh Leivo looks like a star, but it’s tough to put much stock into a small sample, and I am sure that these results are bound to regress at least a little bit. Tyler Ennis looks like a nice addition for just $650k, while Kasperi Kapanen looks like a player who could fit nicely on a NHL checking line.
HockeyViz.com Isolated Impacts (Since ‘16-’17)
Connor Carrick stands out to me on the back-end, as hockey twitter gives him a ton of credit for his on-ice results, but this model gives most of that credit to his teammates and usage. Nevertheless, he’s one of the few Leafs defenders with an encouraging defensive impact.
Moving on from Roman Polak looks to be a wise move, while Martin Marincin has the weirdest results I have ever seen. It looks like Marincin is an offensive superstar who can’t play defence, so maybe this is just an issue of having a small sample size (Leivo’s results were similarly strange in a small sample).
You can check out Micah’s work here, and I will post the link again down below. Micah provides a detailed explanation of how to interpret these charts towards the end.
In Case You Missed It
The Leafs will be looking for redemption tonight, as Monday’s 5-1 loss was simply dreadful.
Yegor Korshkov on injured list until mid-December- Katya Knappe
Korshkov’s struggled with injuries in his young career, and he will miss a few months of action yet again.
The Furies bid farewell to a number of familiar faces, in hopes that new blood will lead to a change in fortunes.
Back to Excited Episode 31: Reports from the Preseason- Acting the Fulemin
Arvind and Acting the Fulemin share their thoughts on Toronto’s preseason thus far. This podcast is worth checking out.
Where should Rasmus Sandin play? - Katya Knappe
I think he will play for Rogle, but I selfishly want him to play for the Marlies. Either way, I can’t wait to watch him at the World Juniors.
From the Branches
Montreal Canadiens place Michael McCarron, four others on waivers- Jared Book, Habs Eyes on the Prize
Michael McCarron and Hunter Shinkaruk are former first round picks, while Byron Froese and Rinat Vailev are former Leafs.
This is a great article, and if you even don’t want to read about this model, you can scroll to the bottom to see the results. This is such a good tool.
Have a good day!