Not necessarily a reply to you directly but on the analytics discussion. This is from the Athletic's 16 Stats article re: the Jets:
Is the team better offensively? No, they’re earning the same amount of chances and still struggling to score at five-on-five.
Is the team better defensively? No, they’re allowing 0.1 more expected goals against per 60 compared to last season.
Is the team’s power play improved? No, they’re scoring one fewer goal-per-60 and generating 1.3 fewer expected goals-per-60. To their credit, they make up for that with a slightly improved penalty kill giving up 0.8 fewer expected goals against per 60.
Is Connor Hellebuyck just playing out of his mind again? Yes, he has a .938 save percentage and has saved 12.4 goals above expected in just 10 games.
Ah yes the fun with aggregate numbers so early in the season when the sample size is not statically significant.
First two games the Jets saw off the expected goals and come out with 1 win and one loss.
They then steal a win against Colorado in which they get crushed in expected goals.
They then get crushed in expected goals and amd lose to Vegas
They slightly get beat by TO in that metric and lose to the Leafs.
They won the expected goal battle and beat the Blues.
They get crushed by the Kings and Helle steals the game.
We crush the Coyotes and win the game
They get crushed by the Knights and lose the game.
We crush each of our last three games and win each game.
Really Helle has only stolen us 2 of our 8 wins. We have been full marks for the other wins.
So the Athletic jumping in on aggregate values to talk about a team this early in the year isn't adding much value given this team has won their matchups more games then its lost their matchups.