This is the first edition of the weekly mailbag segment, where I answer your questions that you sent to me through email (jfreshhockey@gmail.com) and on Twitter.
Jake K: I'm wondering what the analytics say about the performance of Penguins rookie D Pierre Olivier Joseph. The media and eye test seem to indicate that he's off to a very promising start but I'm wondering if the analytics match the narrative and the eye test. Thanks.
There’s good reason to be very pleased with POJ’s performance so far, especially since he wasn’t expected to be a big impact player this season. It is worth noting, however, that his eye test has probably been a bit inflated by a tremendous amount of on-ice puck luck; his even strength expected goals for percentage is 51% (perfectly solid) but his goals for rate is 80%. Once that regresses a bit I think we’ll have a bit fairer an assessment of his play.
Jeff O: Rakell's all situations shooting percentage is down for the 4th year in a row. Is he in an overall decline?
Unfortunately, I think that might be the case. Rakell’s age curve has been pretty dramatic and if his shooting problems keep getting worse I wouldn’t be very optimistic. His offensive play-driving has been remarkably consistent since he entered the league (just a bit above-average) but his finishing is over the hill - almost literally:
(these timelines can be found on my Patreon)
Jordan M: With Rielly and Muzzin already locking down the top two left defenceman spots and with Sandin seeming ready for NHL duties, how much longer should the Leafs hold onto Dermott? What's his realistic ceiling?
Jordan is a beauty for giving me the opportunity to plug my deep dive on Travis Dermott which I published for EP Rinkside on Sunday.
Matthew H: My question is which young players are displaying the best progression based on the underlying numbers, indicating that they’ve reached the next level. Also in the same lane, how would you identify growth that can be sustained v one off good seasons in younger players particularly those on ELC?
An easy answer is Cale Makar, who’s had a Norris-calibre season so far for the Colorado Avalanche. Connor Garland and Andrew Mangiapane have also been crushing it.
In terms of sustainable vs. on-off good seasons, the first thing you have to look at are percentages. For example, Dominik Kubalik was bound to see some regression this season because, well, he shot 19% last year. And you just don’t do that. After that you’d want to see who they played with, whether they were the driver on their line, and then deeper eye test stuff would probably get you some of the rest of the way there.
Daniel D: Which types of shots are underweighted or overweighted the most by xG? I remember you discussing this in the Shea Theodore piece a little.
This is something I’ve touched on quite a lot because it’s one of the more interesting topics in hockey stats. An incomplete list off the top of my head…
Underweighted: Breakaways, odd-man rushes, rushes, shots from cross-ice (royal road) passes, shots where the goalie is moving, shots where the goalie is screened, one-timers, shots off plays with multiple passes
Overweighted: Shots off the cycle, shots where the goalie is square to the shooter, net-mouth scrambles, rebounds directly into pads
Joseph R: Are the Flyers actually this good?
Probably not, no. The’ve been one of the luckiest teams in the league in terms of significantly outperforming their expected goals for and against
Jack: When talking about players who are labeled as good at counterattacking, does that necessarily mean they're effective defensively, or could they be a liability on a macro level, but are effective on a micro level by creating scoring chances from said counterattacks on the rush?
No, it doesn’t in my books. Many players who rank very highly in takeaways have horrible defensive impacts because it’s just a small part of defending. You can be checked out defensively and then put in a lot of effort when it comes to the more “exciting” stuff of stealing pucks so that you get to go on rushes. The counter-attack chance registers as a positive thing offensively, and getting the puck out of the defensive zone registers as a positive thing defensively at a macro-level but if the other team counter-counter-attacks or if you’re otherwise getting caved you’re still having a negative impact on your own end.
Goofball M: Do you ever look into whether the skaters play better “for” a particular goalie?
Two people asked this question. Generally what I care about for goalies is purely how they perform with all else held constant, including what their teammates in front of them do. But especially if you are a fan of a team or covering them, I think there is the chance to learn a lot from checking out those stats. If you would otherwise use save percentage and think it’s airtight because two guys are playing behind the “same defence,” I would recommend that you either a.) stop using save percentage or b.) take a look at those differences. You can do this using NaturalStatTrick’s “Line Tool,” HockeyViz’s environment distiller, and EvolvingHockey’s xfSv%.
Nik L: What are your thoughts on EA Sports NHL ratings for players
I do not envy the scout(s) that have to make the player ratings for thousands of players from a dozen leagues. That being said, the first thing that I do when I get the new game is chuck on a few podcasts and change essentially every player rating in the NHL and a lot of prospect ones as well.