Dallas Stars

What the Quant Says

238 goals, 21st in the league, off 2.641 expected goals per 60 at even strength — 14th in hockey.

At five-on-five it scored 147 against an expectation of 169.2. Twenty-two goals of finishing lost, the second-worst mark in the league. This is the most important number on the page and it is not a matter of who took the shots: the chances were the fifteenth-best in hockey and the goals were the twenty-first.

Two men beat it. Jason Robertson scored 41 on 220 shots at 18.6 per cent, the 21st-highest conversion rate of the 393 qualified forwards, and Roope Hintz 37 on 213 at 17.4 per cent, the 91st percentile. Between them, 78 of the club's 238 goals.

Everyone else missed. Tyler Seguin shot 11.0 per cent, the 46th percentile, from a median of 27 feet at the 72nd — further out than he used to shoot. Jamie Benn shot 9.9 per cent, the 36th percentile, for 18 goals in 82 games, the fewest in any season of his career over sixty-nine games. Denis Gurianov shot 7.6 per cent, the 69th-lowest of 393. Radek Faksa shot 5.6 per cent, the 30th-lowest, on 4.40 shots per 60, the 17th-fewest. Joel Kiviranta scored once on 59 shots — 1.7 per cent, the 2nd-lowest of the 393 qualified forwards.

Joe Pavelski had 81 points at 37 with 61 per cent of his shots from the slot at a median of 19 feet. Mason Marchment arrives with 3.78 even-strength points per 60, the 3rd-highest of 393 — a figure earned on a Florida club that led the league in points, chance share and goals.

What the Scout Says

The top line is one man's. Jason Robertson holds 54 to 60 per cent of the chances without every team-mate on the roster; Joe Pavelski without him falls to 48.9 and Roope Hintz to 47.7. Both of those are fragments — 462 attempts and 370 — and both point the same way.

Robertson is inside, not outside, and the shot chart says so before the finishing does. So is Hintz: he gets to the same places Pavelski does and arrives at them a great deal faster. Pavelski does not carry the puck and he does not need to.

Below that line the shooting is honest and the results are not. Radek Faksa goes to the right place and almost never gets a shot away when he is there. Luke Glendening shoots from the right place and there is nothing behind it. Joel Kiviranta's shots are not bad and one of them went in — at fifty-nine attempts that is a coin landing the same way and not a shooting problem.

Denis Gurianov shoots from outside, which for a man with this much speed is the wrong answer.

Mason Marchment is the one arrival with a real offensive case, and it is a five-on-five case: he covers ground without going fast over it, which for a 6-5 winger is the forechecking profile rather than the rush one.

What the Quant Says

246 goals against, 13th in the league — off 2.551 expected goals against per 60 at even strength, 10th.

At five-on-five it conceded 159 against an expectation of 162.4. The defence did what its chances said it would; the offence did not.

Miro Heiskanen played 24:53 a night, the 93rd percentile among defencemen, at 53.6 per cent of the chances. His median shot travels 44 feet, the 10th-shortest of the 202 qualified defencemen, and 18 per cent come from the slot at the 90th percentile — a defenceman who gets inside rather than shooting from the point. His 23.1 mph is the 6th-fastest of the 192 defencemen the tracking covers.

Ryan Suter played 23:39 at 37 with 1.42 hits per 60 — the 11th-fewest of 202 — and 2.72 blocks, the 14th-fewest. His 20.7 mph is the 6th-lowest top speed of 192. The most-used defensive minutes in Dallas belonged to a man in the bottom fifteen of the league at both of the things the box score counts as defending.

Esa Lindell took 0.215 penalties per 60, the 10th-lowest of the 202 qualified defencemen — twelve minutes in seventy-six games at twenty-two a night, most of it against the other side's best. He covers 7.93 miles per 60, the 11th-lowest of 192.

Jani Hakanpää recorded 11.48 hits per 60, the 7th-most of 202, at 6-7 and sixteen minutes a night. Joel Hanley played 11:56, the fewest of any qualified defenceman, and took 2.06 shots per 60, the 3rd-fewest.

John Klingberg's 1,336 five-on-five minutes have gone to Anaheim, and the right side behind Hakanpää is now Colin Miller, whose 100.9 mph is the 15th-hardest of the 649 skaters the tracking covers, and Thomas Harley and Nils Lundkvist, who have 59 NHL games between them.

What the Scout Says

Miro Heiskanen carries the puck out himself and he loses it more than anyone on this blue line. He is not covering more ground than his peers; he is covering it faster, and the two are not the same measurement. He defends by getting there first, and the box score records almost none of it.

Ryan Suter is the hardest read in the chapter. A man playing the most defensive minutes in Dallas finished in the bottom fifteen of the league at both of the things the box score counts as defending. Either he is positioning himself so well that neither is required, or he is not getting there. Nothing in the repo separates the two, and the phrasing matters more than it looks. What the repo does separate is the company he keeps: eighteen points of chance share, and the variable is entirely which forwards the coach sends out.

Esa Lindell stays between the puck and his own net and does not chase, and nothing here measures gap control or how he defends the rush, which is the whole of the argument for him.

Jani Hakanpää does not take pucks off people; he takes people off pucks. A stick that long changes what a forward can do on the wall, and the repo records hits and blocks, which are the events that happen when the reach has already failed.

Joel Hanley is the eighth defenceman on most nights and the value is that a club can play him for twelve minutes in March and lose nothing it can measure.

What the Quant Says

12th in the league at plus-3.4 goals against the chances faced — slightly better than the shots deserved, on a club that conceded the 13th-fewest.

Jake Oettinger went 30-15-1 with a .9144 and a 2.53 across 48 games on 1,331 shots — 15th of the 53 goaltenders who played 25 games or more. A good season and not a remarkable one.

Then 285 shots in seven playoff games against Calgary: 272 saves, 13 against, .9544 with a shutout, in a series Dallas lost in overtime of the seventh. Calgary took 60 per cent of the expected goals and 60 per cent of the unblocked attempts across those seven games. He faced 40.7 shots a night for a fortnight and the series went the distance because of it.

A fortnight of .9544 is the least repeatable measurement in hockey. The two regular seasons are the useful record: .9113 on 710 shots as a rookie and .9144 on 1,331 last year, which is consistency rather than a leap, and consistency is the safer bet.

Anton Khudobin has gone to Chicago and Braden Holtby is off the roster. Scott Wedgewood took the net for the last eight games of the season and returns as the backup with a career .9058 in 77 games — a .9097 across New Jersey, Arizona and Dallas last year, which is an ordinary number posted behind the 32nd- and 30th-ranked crease situations in the league.

What the Scout Says

Jake Oettinger is 6-6 and 225, which is at the top of the range the position now selects for, and the reason a lot of the .9144 is simply that he covers the net.

Nothing in this repo measures a goaltender's glove, his push or how deep he plays. What it measures is shots and goals, and at 1,331 shots the number is stable enough to say he was better than the league average and not by much. The playoff figure is the opposite: 285 shots is a fortnight, and it happened, and it is not a forecast.

The useful sentence about him is the boring one. A young goaltender whose second season matches his first is a much safer bet than one whose second season is a leap, and Dallas has the first kind.

The seventh game was level after regulation and he lost it in overtime. Nothing in the data says how a 23-year-old carries that into a next season, and this book will not pretend otherwise.

Scott Wedgewood is 6-2 and 201, which is small by the standard the position now sets, and it is one reason it took him until 30 to get a season. The .9097 says he can hold twenty-five starts; the seventy-seven career games say nobody should be surprised either way.

What the Quant Says

14th in the league at 6.809 expected goals per 60, over 4:55 a night — 17th in power-play time and 14th in what it did with it. Squarely average at both.

Joe Pavelski took 2:44 a night, the 79th percentile among forwards, and 31 per cent of his points came on it. Jason Robertson took 2:41 at the 76th percentile and only 27 per cent of his points came there — he is not a power-play creation, which is the most reassuring thing about a 41-goal season.

Roope Hintz took 2:32 and 33 per cent of his points came on the man advantage, the 85th percentile — a higher dependence than Robertson's and the number to watch if the finishing regresses.

Miro Heiskanen took 2:18 from the point, the 84th percentile among defencemen, and Ryan Suter 2:11 at the 80th — with 41 per cent of Suter's points coming there, the 93rd percentile. Strip the power play out of Suter's season and 0.70 even-strength points per 60 is left, the 23rd percentile.

John Klingberg's power-play time has gone to Anaheim. Mason Marchment arrives having taken 0:52 a night, the 33rd percentile, with 4 per cent of his points on it — everything he produced in Florida, he produced at five-on-five, and he is not the answer here.

Colin Miller's 100.9 mph is the one new argument for the second unit, and 93 per cent of his attempts came from beyond forty feet.

What the Scout Says

The unit is Pavelski at the net, Robertson and Hintz on the flanks and Heiskanen at the point, and the interesting thing about it is how little of the scoring depends on it.

Robertson is not a power-play creation — twenty-seven per cent of his points came on the man advantage, which is thoroughly ordinary, and it is the most reassuring thing about a 41-goal season. Hintz's thirty-three per cent is the number to watch, and it sits alongside a 17.4 per cent conversion rate that is a good season and a lucky one at once.

Ryan Suter is the second unit's whole shape and the reason to be careful with him: strip out the man advantage and the offence is close to nothing. He shoots rarely and from distance, which is the profile the position had twenty years ago.

Colin Miller's shot is as pure a point shot as the data records — 93 per cent of his attempts came from beyond forty feet and none from inside twenty. That is a specific job on a second unit and it is the only argument for him.

Mason Marchment produced everything he produced at five-on-five, which is the rarest way for a 47-point season to be built and the most likely to travel. It is also the reason he will not fix this unit.

What the Quant Says

3rd in the league at 5.286 expected goals against per 60, over 4:47 a night — behind only Carolina and Calgary. On a club that was 21st in goals and 31st in finishing, this is the best thing it did.

Esa Lindell killed 2:38 a night, the 86th percentile among defencemen, and Jani Hakanpää 2:16 at the 72nd. Luke Glendening killed 2:13, the 96th percentile among forwards, and Radek Faksa 1:51 at the 87th. Four men, and the unit is theirs.

Glendening's skating file is the shape of the job: 8.81 miles per 60 is the 10th-lowest of the 365 forwards the tracking covers, and 12.5 bursts over 20 mph per 60 is the 22nd-most. Almost nobody covers less ground and almost nobody sprints more often — very short shifts taken at a dead run.

Faksa blocks 3.36 per 60, the 97th percentile among forwards, and takes 1.09 penalties per 60 at the 82nd — the one cost in an otherwise clean unit.

Miro Heiskanen killed 1:38 and Ryan Suter 1:34, both the low forties among defencemen — the club's two biggest minute-eaters are used sparingly here, which is a deliberate choice about workload and leaves the kill to the specialists.

Nothing in the four names above has changed. The third-best penalty kill in hockey is the one part of this roster that returns intact.

What the Scout Says

Four men and a structure, and it is the best thing this club does.

Luke Glendening's file is what a penalty-killer looks like when the tracking finally measures one. Almost nobody in the league covers less ground and almost nobody sprints more often: very short shifts taken at a dead run, which is exactly what a fourth-line penalty-killer is for and is not something the eye would have separated from "he works hard". Fifteen penalty minutes in eighty-two games from a man who hits that often is the finding.

Esa Lindell takes on the hardest assignment in the sport and does not hand the other side a power play for it. Twelve minutes in seventy-six games, at twenty-two a night, is the single most valuable thing on his page.

Radek Faksa is on the ice to kill penalties and he creates them — 1.09 taken per 60 at the 82nd percentile, on a unit whose other members are near the floor for fouling. The blocks are elite and the discipline is the cost.

Jani Hakanpää's forty-three minutes in eighty games is not an indiscipline problem; it is the arithmetic of a man that size finishing every check.

None of the four has moved. Whatever else this roster has lost, the third-ranked kill in hockey returns whole.