What to Expect With Jim Hiller

2026-06-30 · Maple Leafs · Jim Hiller · coaching · Kings

Toronto tried the Cup-pedigree hire and Auston Matthews cratered. What Jim Hiller's actual Kings numbers — and our metrics — say to expect from the Leafs' next coach.

Toronto just ran the prestige play — and it didn't work. Two years ago the Maple Leafs hired a Stanley Cup-winning head coach in Craig Berube, imported the pedigree, and waited for the culture-change bump. What they got instead was the worst two-year stretch of Auston Matthews' career and no closer to a Cup. On June 17, 2026, GM John Chayka made the pivot official, naming Jim Hiller the 41st head coach in franchise history after a five-week search that reportedly reached as many as 55 candidates — no ring, a recent head-coaching run in Los Angeles, and a familiar face from four seasons on Toronto's bench alongside Mike Babcock, back when Auston Matthews and William Nylander were breaking into the league.

So the question isn't what Hiller's résumé implies. It's what his actual track record — and the cautionary tale of the coach he's replacing — tell us to expect. Here's the case, in our numbers.

The pedigree trap

Coaches get hired on reputation far more than on any measurable record of improving a roster. A ring is the ultimate reputation, and it travels — as a story, not as a guarantee. The uncomfortable historical pattern: Stanley Cup-winning coaches almost never win another Cup with a different team. Repeat champions overwhelmingly did it with the same franchise; winning again somewhere new is rare enough that Scotty Bowman is the exception people reach for. Pedigree is a bet on the past, and the past doesn't transfer with the coach.

Toronto just learned that the expensive way — and their best player is the clearest evidence.

The Matthews decline under Berube

Auston Matthews went from a 69-goal, 107-point season in his last year under Sheldon Keefe to back-to-back career-worst campaigns under Berube. He was hurt in both Berube seasons (67 and 60 games), so to keep the comparison fair we'll use wins above replacement on a per-82-game basis — a rate that isn't distorted by missed time. His WAR/82 fell from the mid-6s under Keefe to ~4.0 under Berube:

From ~6.5 to ~4.0 — roughly a one-third drop in per-game impact, injuries controlled for.

The important part is where the decline came from. This isn't only a finishing slump — his underlying chance creation collapsed too. Individual expected goals (a system-and-deployment signal, not luck) went from an enormous 52.7 to 38.3 to 32.6. And his power-play production, once a pillar, cratered:

A goal-scorer generating one-third fewer expected goals and half the power-play points is not a player "declining" so much as a player being used differently — a system story.

How Berube deployed him: less power play, more penalty kill

Here's the deployment half of that system story — and it's the clearest evidence that this was a coaching choice, not just aging or luck. Berube didn't cut Matthews' overall ice time; he sat right around 20:30 a night every season, Keefe or Berube. What changed is the kind of minutes. Berube pulled him off the power play, gave him real penalty-kill work for the first time, and stopped sheltering him with offensive-zone starts:

Every lever moved toward defense. His power-play time fell roughly a quarter from his Keefe norm (3:22 down to 2:31) — the same power play whose point production cratered from 28–29 down to 12. His penalty-kill minutes, essentially zero under Keefe (six seconds a game in 2022-23), roughly tripled. And his offensive-zone starts, once a lavish 81%, dropped to barely above half.

That reframes the chance-creation drop. A sniper who loses a quarter of his power-play minutes and gets roughly half as many offensive-zone faceoffs will generate fewer expected goals almost by construction — you can't score from the penalty box or from your own blue line. His individual-xG decline (52.7 down to 32.6) is partly aging and partly injury, but it is also, measurably, a coach choosing to use his best goal-scorer as a two-way center. The offense followed the assignment.

And that's the lever Hiller inherits. If he restores Matthews toward a sheltered, power-play-heavy, offense-first deployment, some of the "decline" could reverse without Matthews changing a thing — which is exactly why how a coach deploys his stars is the number worth watching, not just whether he owns a ring.

The on-ice tracking backs this up. Using AllThreeZones manual tracking (a sampled set of Matthews' games — 109 under Keefe, 57 under Berube), his transition game flattened under the new system: fewer controlled entries, a lower carry-in rate, and fewer rush shots and chances per game.

A carry-in rate sliding from ~66% to ~56%, with rush shots and chances each down roughly a quarter, is the fingerprint of a player doing more dump-and-retrieve and less rush offense — exactly the north-south identity Berube's teams are built around. (AllThreeZones covers a sample of games, so treat these as directional, not full-season totals.)

Fair caveat: Matthews was injured in both Berube seasons (67 and 60 games), so this isn't a clean "the coach broke him." But the drop shows up on a per-game and a chance-quality basis, not just in the counting totals — the bump the pedigree hire promised simply never arrived. See Auston Matthews' full profile for the live card.

The initial reaction

The reaction to the Hiller hire split about where you'd expect: one camp saw an uninspiring retread — no ring, a Kings tenure that ended on a downswing — while another saw a details-and-structure coach who already knows this market and this core. As usual, the reaction is a referendum on reputation. The more useful exercise is to look at what his teams actually did.

Hiller's résumé — what the Kings actually did

Hiller took over the Kings on an interim basis midway through 2023-24 and ran the bench through 2025-26. The team-level picture, before and during his tenure — note the 5v5 expected-goals share peaking at an elite 55.4% in his first full season, then regressing to league-average 50.9%:

Three things stand out, and they define the profile:

• He inherited a good team. McLellan's Kings were already a strong possession club (104 points, 52.2 xGF% in 2022-23). Hiller didn't build this from nothing — important context for any "he transformed them" claim.
• His best team was genuinely excellent at 5v5 — and elite on the penalty kill. The 2024-25 group posted a 55.4 xGF% and killed penalties at 81%+, after an 84% PK in the half-season he took over. Structure and special-teams defense are his calling card.
• The power play was a weakness, and year two regressed. His power play sat at a below-average 17.3% both full seasons (McLellan's had hit 24.5%), and in 2025-26 the whole thing slipped — xGF% back to 50.9, the PK down to 74.7, faceoffs under 50%.

And it isn't only our numbers. As The Athletic noted at the hire, that 2024-25 Kings team allowed the second-fewest goals and second-fewest shots in the entire NHL — trailing only a Winnipeg club backstopped by a Vezina-winning Connor Hellebuyck — and posted the league's best five-on-five expected-goals-against rate. For a Toronto team whose defensive lapses have defined its playoff exits, that defensive ceiling is the single most relevant line on Hiller's résumé.

The 2025-26 dip: mostly goaltending and roster

Before pinning that regression on the coach, look at what changed around him — because the two biggest factors had nothing to do with coaching. Start in net. Darcy Kuemper went from a .922 save percentage and a 2.02 goals-against average in 2024-25 to .891 and 2.78 in 2025-26 — a roughly 30-point save-percentage collapse from the same starter. That alone drags a team from comfortably-good to lottery-adjacent, and no bench boss fixes it.

The blue line, meanwhile, lost its most valuable piece — and it was a young one. Los Angeles' two best defensemen by GAR in 2024-25 were its youngest: Jordan Spence (9.4) and Brandt Clarke (10.2). Spence was traded away; Vladislav Gavrikov's shutdown minutes walked in free agency; and the veterans brought in to fill the gap — Brian Dumoulin (1.7 GAR) and Cody Ceci (−3.4) — were downgrades. The aging forward core slipped on top of it, Anze Kopitar most of all.

That said, not all of it was out of Hiller's hands — and it's worth being honest about the part that was. He rode two negative-value veterans hard: Ceci at 17:33 a night (−3.4 GAR) and Joel Edmundson at 18:28 (−3.2). The Dumoulin-Ceci pairing played 78 games together and was outscored 23–36 at 5-on-5 — though its 52.8% expected-goals share says the goaltending behind it makes the pair look worse than it actually played. So read the dip as mostly goaltending and roster attrition, with a fair critique that Hiller leaned on the wrong veterans — not as a coaching collapse. That the 5v5 structure still held around a league-average 50.9% xGF% with that roster and that goaltending is arguably a point in his favor.

The style: forecheck over flash

Team results tell you how good; microstats tell you how they play. Using AllThreeZones manual tracking (zone entries/exits, forechecking, and shot origins, on a sampled set of games), Hiller's Kings developed a clear identity — and it isn't run-and-gun:

The direction is consistent: more forechecking pressure (up to ~17 a game), fewer rush shots (15+ under McLellan down toward 9), and fewer controlled zone entries (carry-in rate 54% → 46%). That's a dump-and-forecheck, north-south team that hunts pucks back on the wall rather than carrying them in off the rush. Some of the 2025-26 extremity is the weaker roster — less skill to carry the puck cleanly — but the identity was already set in 2024-25.

It also matches how Hiller describes himself. At his introductory press conference he called himself a "data guy," open to analytics, and said "skating is the first chain in competing." A coach who prizes skating and pressure, tracked out over hundreds of games, looks exactly like the forecheck-first team the microstats describe.

His own framing, though, leans less on system than on buy-in. Introduced in Toronto, Hiller played down X-and-O dogma — "there's a pretty standard template across the league that all coaches play with," he said, adding that he couldn't yet tell you "exactly what system we're going to play," but wanted "a style of play that the players will be excited to play." Sportsnet framed the hire plainly as one made for the players — the message being that the game "has to be fun" again after two grinding years. Hold both ideas at once: his Kings results were structured and defense-first, yet his stated priority is engagement. "Fun," in Hiller's mouth, most likely means a group that buys in and plays fast — not a green light for run-and-gun.

Deployment: youth up, aging stars managed

His ice-time usage is that of a coach who develops. Hiller trimmed Drew Doughty from a 26-minute workhorse toward the mid-24s, leaned on a heavy shutdown pair (Gavrikov–Anderson at 22-23 minutes), and — most tellingly — promoted young players into real roles. Quinton Byfield climbed to a top-line load (18.6 then 20.0 minutes, ~10-12 GAR), rookie Alex Laferriere earned 16-18 minutes (11.7 GAR), and young defenseman Brandt Clarke jumped from a depth role to nearly 20 minutes a night — and rewarded the trust, leading all Kings defensemen at 14.0 GAR in 2025-26.

For a Toronto team that has often buried its youth and overplayed its veterans, a coach who trusts young players and manages his aging stars' minutes is a meaningful — if quiet — change in approach.

The Toronto connection

Hiller isn't a stranger to this building — he spent time on Toronto's coaching staff before Los Angeles. That cuts against the culture-shock risk: he knows the market, the pressure, and the core. After the Keefe-to-Berube whipsaw, "a coach who already understands this team" is a real, if unglamorous, point in his favor.

What to expect

Put it together and the projection writes itself — with one open question:

• Don't expect a pedigree miracle. The Berube experiment is the local proof that a ring doesn't transfer. Toronto already paid for that lesson.
• Do expect structure, a strong penalty kill, and a heavier forecheck. Hiller's best Kings group was a top-tier 5v5 and PK club that forechecks hard and defends the rush. For a Leafs team that has defended and killed penalties poorly at the worst times, that's the most plausible upgrade — with a more north-south, dump-and-chase feel than Toronto is used to.
• Expect the kids to play. Hiller handed young players (Byfield, Laferriere, Clarke) real minutes in Los Angeles and managed his veterans' workloads. Toronto's younger players — and its aging stars' ice time — should both feel that.
• Watch the power play. Toronto's stars live on the man advantage, and Hiller's 17.3% power play is not what this roster needs. If the PP doesn't improve, the ceiling doesn't move much — and Matthews' PP-point collapse is exactly the wound that needs healing.
• The real unknown is the roster, not the system. His 2025-26 dip was mostly a downgraded, aging Kings roster, so the question isn't whether his structure works — his best team's numbers say it does — but whether it lifts this roster the way it needs.

The single biggest lever on whether this works is Matthews. A system that lifts team xGF% and gets him back to generating grade-A chances is precisely what a bounce-back requires. If Hiller's structure travels the way his Kings numbers suggest it can, the Leafs' best player is the first place it should show up.

Track it yourself: Auston Matthews' profile, the Maple Leafs team page, and our methodology for how GAR and expected goals are built.

Data: Hockey Alchemy metrics from NHL public play-by-play and shift data, regular season, pulled June 2026. GAR is our Goals Above Replacement model; xGF%/CF% are 5v5 expected-goals and Corsi share; PP%/PK% are full-strength special-teams rates. Individual expected goals (iXG) measures a skater's own shot-quality generation. Zone-entry, forecheck, and shot-origin microstats are from AllThreeZones (Corey Sznajder) manual tracking, aggregated over a sampled set of games per season — so those per-game rates are directional estimates, not full-season totals. Coaching-tenure attribution is by season; mid-season changes (Kings, 2023-24) are noted. Figures are a point-in-time snapshot as of publication.

More from The Lab

How Our Expected Goals (xG) Model Works

A transparent look inside Hockey Alchemy's xG model: four situation-specific XGBoost models, 53 features, 16 seasons and 1.6M shots, with cross-validated log loss, AUC, and calibration.

Introducing the Hockey Alchemy Chrome Extension

GAR, tier, and contract value for any NHL player — in a popup, a side panel, or highlighted right on the pages you already read. Free, no account, no tracking.

Jet Greaves Is the NHL’s Most Underrated Goalie

Columbus' Jet Greaves was a top-7 NHL goalie in 2025-26 by shot-quality-adjusted GSAx — all on an $812,500 cap hit. The case the box score hides.

Related: All Posts Methodology Glossary