We are sitting at the three-quarter pole of the NHL season, with most teams having played somewhere in the vicinity of 20 games. At this point, we have lots of data to work with. Personally, I incorporate seasonal data after a couple weeks, but generally speaking, once we get to 20 games, teams and players are who they are for the season. Of course some things will change; injuries, coaching changes, lineup changes, COVID shutdowns, the Olympics, and dozens of other issues, there are lots of reasons why performance will change for some players/teams the rest of the way. But most players and teams won't vary wildly from what they've done so far, which is why the proverbial cutoff of American Thanksgiving is a popular one in the NHL: it happens to coincide when stats stabilize.
With that in mind, here are some stats from the first six-or-so weeks of the season, and what it could mean for the rest of the year. Let's start with the most obvious team:
New York Islanders
A little over a week ago, I wrote about the Islanders and their awful start to the season. Everything I said there holds up, and there's an added wrinkle: a COVID shutdown. Now, I am not one to believe in excessive coincidence, and I find it odd that the three players named in the article as having very poor starts – Chara, Bailey, Lee – were the first three players put on the COVID list before the team's shutdown (or among the first, I can't quite remember the order they hit the list). Were these players playing with COVID for weeks leading up to the shutdown? That is nearly impossible, so the likely answer seems to be that these guys were playing poorly, and *then* caught COVID. I have to say, that doesn't make me feel good about a turnaround.
The team was playing poorly – 27th by shot share, 15th by expected goal share – in their 17 games before the team was shutdown. At time of writi