Boston Bruins

What the Quant Says

222 goals and 27th in the league in creating chances at even strength, 2.33 expected goals per 60. This was the problem, and it was the problem before the deadline sales made it worse.

David Pastrnak had 106 points — 4th in the league — on a club that scored 222 goals, and his on-ice goals were 65-54 against an expected 50-49. Morgan Geekie scored 33 on 150 shots, a 22 per cent rate that is the least repeatable number in this edition. After those two: Elias Lindholm 47, Pavel Zacha 47, Casey Mittelstadt 40 across two clubs.

Finishing was 3.1 goals below expectation, 18th, so the club did not get help there either. What the summer added — Viktor Arvidsson, Tanner Jeannot, Sean Kuraly, Michael Eyssimont — is a bottom six, not a scoring line.

What the Scout Says

Pastrnak and a series of men who are useful next to somebody better.

The release is the whole of the club's offence and he takes a quarter of his attempts from inside 20 feet and a third from beyond 40, which is a man shooting from wherever the shift allows because most shifts allow nothing. The playmaking has become the larger half — he holds pucks on the wall and finds a seam most shooters never look for.

Morgan Geekie has a quick release from a big frame and a genuinely improved sense of where to be, and converted at a rate nobody sustains. Pavel Zacha is 6-4 with better hands than the frame suggests and creates by holding the puck a beat longer than a defenceman expects; there is no rush game. Elias Lindholm's reads are still good and the ability to drive a line has gone, if it was ever his rather than his wingers'. Casey Mittelstadt holds the puck below the goal line and holds it a beat too long, which against a good forecheck is how a skilled centre reaches the 19th percentile.

Matthew Poitras sees a play before it is available and delivers into space rather than onto a stick, and weighs 189 pounds, which is the entire obstacle. Viktor Arvidsson is 5-10 and plays as though he is much larger, goes to the net front and takes cross-checks and stays.

What the Quant Says

272 goals against and 18th in even-strength expected goals against at 2.48 per 60, which is the least bad number on this page and still not good.

Charlie McAvoy played 50 games at 23:40 a night and posted 50.5 per cent, the best share on the blue line; he was hurt in February and did not return. Mason Lohrei finished minus-43 — the worst rating in the league — with 33 points and a 48.5 per cent chance share, which is a young defenceman being played through his education on a club with no alternative. Nikita Zadorov posted 49.2 per cent over 1,371 minutes and finished plus-25 on a team with a minus-49 goal differential, which is the strangest single number in this edition.

Brandon Carlo was traded to Toronto in March. What is left is McAvoy's availability, Lohrei's development, and Andrew Peeke playing 18 minutes because somebody has to.

What the Scout Says

A blue line with one real defenceman, one very interesting problem, and a great deal of weight.

Charlie McAvoy is the whole structure when he is available: the best skater in the group, a genuine fourth attacker on a rush, and a defender who finishes checks and defends his own net front. 46 penalty minutes in 50 games is part of the style and the availability is the rest of the story.

Mason Lohrei has top-pair offensive tools — he carries out of trouble, sees the seam from the line, and put up 28 assists on a club that scored 222 goals — and does not play to 6-5 at all. He does not clear a crease, cannot be relied on to win a battle, and attempts the difficult play at the wrong moment, which is how the puck ends up in the slot behind him. 16 penalty minutes in 77 games is not discipline, it is absence.

Nikita Zadorov supplies everything Lohrei does not: he clears the crease, hits constantly, and is the reason opponents do not enjoy the building. Andrew Peeke boxes out, ties up sticks, blocks shots and takes 15 penalty minutes a year doing it, which is the most impressive line in his record. Henri Jokiharju escapes a forecheck cleanly and moves the puck early and cannot defend a net front.

What the Quant Says

Jeremy Swayman went 22-29-7 with an .8922 across 58 games. He had stopped .916 or better in each of his first four seasons.

The club's goaltending came in 5.3 goals worse than the chances faced, 20th in the league — the first time in a decade Boston has been on the wrong side of that number, and the single largest change from the season before. Joonas Korpisalo stopped .8932 in 27 games, which is to say the two goaltenders had the same season.

Swayman missed the entire training camp in a contract dispute that became public, then signed a large deal and had the worst year of his career. How much of the .8922 is that is unanswerable, and it is the only question about this roster that matters more than the scoring.

What the Scout Says

The position that decided the season, in the direction nobody expected.

Jeremy Swayman's game is athletic and reactive rather than positional — quick, tracking well through traffic, making saves in sequence better than almost anyone at his best. What went wrong last season looks technical rather than physical: deeper in the net than he had been, square less often, worse rebound control. That is the profile of a goaltender who missed a training camp, which is exactly what happened, in a contract dispute that became public and unpleasant. There is no record before this season of him being fragile and a considerable record of the opposite.

Joonas Korpisalo is 6-4 and plays deep and square and relies on size rather than quickness. The lateral movement is the weakness and always has been — a puck moved across the slot beats him more often than it should — and the rebound control is the worse of the two, which behind this club is the wrong pair of flaws. Eleven seasons, five organisations, no evidence of a goaltender who unravels and none of one who strings three good months together.

What the Quant Says

17th at 6.97 expected goals per 60, over 4:52 a night. League-average, which on this roster counts as a success.

Pastrnak led it with 9 power-play goals, Geekie 4, Mittelstadt 3 across two clubs, Zacha 2. The unit is Pastrnak and four other men, and it produces at a median rate because he is good enough to make that work. It is also the phase least damaged by the deadline, since almost everything runs through the one player who was not traded.

What the Scout Says

One player, and a unit built to give him the puck.

Pastrnak on the off wing is the whole thing: the one-timer has to be respected from the moment the puck enters the zone, and the late change of angle beats a goaltender who has read the entry correctly. Morgan Geekie gets to the net front and can hold a position there. Pavel Zacha works from below the goal line and is the closest thing to a second distributor.

What the unit lacks is a defenceman who forces a decision. McAvoy walks the line and is dangerous when available; Mason Lohrei sees the seam from the point and does not have the release to punish a killer who ignores him. Casey Mittelstadt is the summer's answer at the half-wall and holds it too long. A unit that is 17th with the best shooter in the league on it is being carried by one man.

What the Quant Says

14th at 6.57 expected goals against per 60, over 4:36 a game — a middling workload. The best of the four phases and it is not close.

Zadorov and Peeke clear the front of the net, Lindholm and Zacha take the draws, Mark Kastelic and John Beecher supply the legs. The concern is the volume being brought in: Kastelic took 106 penalty minutes in 61 games, Tanner Jeannot 89, Michael Eyssimont 63, and Zadorov 145, and a club that has just signed three more of that type is choosing to spend more time shorthanded, not less.

What the Scout Says

The one phase where the roster's construction actually fits.

Nikita Zadorov and Andrew Peeke both clear the front of a net properly, which is the specific skill a kill is built on, and Zadorov's reach ends plays that a smaller defenceman would have to chase. Elias Lindholm takes the hardest defensive-zone draws in the league without complaint. Mark Kastelic and John Beecher both skate well enough to pressure a breakout, and Beecher's straight-line speed makes a careless first pass expensive.

The problem is arithmetic rather than personnel. Zadorov took 145 penalty minutes, Kastelic 106 in 61 games, Michael Eyssimont 63 and Tanner Jeannot 89 on other clubs, and all four are on this roster. A kill that is 14th in rate and asked to work more often than it was last season will not stay 14th.