45%: Most gutted (MTL). 7%: Least touched (BOS). 4.2: Top returning core (BOS). ceiling: How to read the new cores Expansion took forty players out of the eight established teams and handed them to Detroit, Hamilton, Las Vegas, and San Jose. The easy take is that everyone got a little worse and the new teams are bad. The truer story is that the damage was wildly uneven, and it scrambled the league's pecking order more than the standings will admit at first. Here is what each team has actually committed for 2026-27, measured as the projected value of its core. Read this as a floor, not a forecast. These are the players under control today. The established eight all have holes where departures were and will fill them through free agency and the draft, so their real 2026-27 teams will be better than the bars below; the construct understates them. The new four front-loaded their builds, so their numbers move less from here. We pull goaltending out of the bars, because it is the noisiest projection and the league's biggest swing, and show it separately. And we do not rank one through twelve, because the two groups are not the same kind of number (more on that at the new teams). The established eight: who got gutted Each bar is a team's returning skater core (its top nine projected skaters who did not leave). The faded extension is the value it shed to expansion; solid plus faded is what its core was a year ago. The dot is its best returning goalie's projection, on the same scale, kept separate on purpose. MTL is the quiet collapse. A year ago it had one of the best cores in the league (4.3); expansion took 45% of it (1.9 projected WAR of skaters), the most of anyone, and dropped it toward the bottom of the returning ranks. What keeps it from free-fall is the crease: a +0.83 goalie projection, among the best in the league, is now most of the reason to believe in MTL. A 45-percent-gutted skater group papered over by an elite goalie is exactly the kind of team whose record will swing on save percentage. BOS and BOS barely got touched (7% and 7% of their cores), and that is its own kind of win. They did not get better; the league got worse around them. When your rivals lose a third to a half of their cores and you lose almost nothing, you climb without lifting a finger. That rise-by-attrition is the single biggest mover in this projection and the part a standings table will hide. BOS (returning 4.2, was 4.5, shed 0.3 = 7% of its core; goalie +0.67). VAN (returning 4.0, was 4.4, shed 0.4 = 9% of its core; goalie -0.17). MIN (returning 3.6, was 5.2, shed 1.6 = 30% of its core; goalie -0.16). NY (returning 3.0, was 4.1, shed 1.1 = 27% of its core; goalie -0.31). TOR (returning 2.8, was 4.1, shed 1.3 = 32% of its core; goalie +0.28). OTT (returning 2.6, was 3.9, shed 1.3 = 34% of its core; goalie +0.82). MTL (returning 2.4, was 4.3, shed 1.9 = 45% of its core; goalie +0.83). SEA (returning 2.3, was 3.8, shed 1.5 = 40% of its core; goalie -0.10) The four new teams: a separate tier Now the part that needs a different scale. The established cores above are returning players in roughly their old roles. The new teams' cores are players moving to a weaker team, tougher minutes, and worse linemates, so their old projections over-state what they will actually be worth. That is why we show the new cores as a band, from a context-discounted value up to the raw signed sum, rather than a single bar you could rank against the eight. Detroit (signed core 3.4, about 2.9 after a new-context haircut; goalie +0.00). Las Vegas (signed core 3.4, about 2.9 after a new-context haircut; goalie -0.47). Hamilton (signed core 3.1, about 2.7 after a new-context haircut; goalie -0.64). San Jose (signed core 2.7, about 2.3 after a new-context haircut; goalie -0.06) Here is the uncomfortable part for the plundered teams. Detroit's signed skater core (3.4) already projects above what the most gutted established teams have left (2.3 at the bottom). Even after the new-context haircut (2.9), the best new core is in the same neighborhood as a couple of 2025-26 playoff teams' leftovers. Read that as a ceiling against a floor, not a settled fact, but the gap between the league's bottom and its newcomers is smaller than 'expansion teams are bad' suggests. What this read isn't Not a final-roster forecast, and not a 1-to-12. It is committed value today. The established eight will fill their holes and rise more than the new four, so the bars understate them unevenly. We deliberately do not merge the two groups into one precise ranking.. The new cores are ceilings. Their value was earned in better contexts and will not transfer cleanly. The band shows the haircut; reality is somewhere inside it, probably nearer the bottom.. Lost WAR is not the whole damage. Losing one star hurts more than losing the same value spread over six role players, because the market can backfill role players and cannot backfill a star, and the expansion-year free-agent pool is thinner. So a concentrated loss is worse than its number, and a team that lost a play-driver may see its returning teammates' projections dip too.. Goaltending is the least reliable line. The goalie dots carry the widest error of anything here. Trust the goalie-dependent rankings least, and note that a great returning goalie is the main hope for the most gutted teams. Methodology Core. The summed projected next-season WAR of a team's top 9 skaters; beyond that, value is near replacement and adds little. Goaltending is excluded from the core and shown as the best returning goalie's projection on the same axis.. Returning vs shed. For the established eight, we take last season's top-9 skater core and split it into players who stayed and players who left to expansion; the faded segment is the latter.. New-team band. Each new core's raw signed sum, and a 15 percent haircut for the weaker new-team context; the truth sits between, and the band keeps us from a false-precise comparison to the established cores.. The unit. Our "WAR" is a value-above-replacement number built from per-game Game Score, not literal standings wins; it is the value scale the rest of the site uses. Numbers regenerate on every build as rosters fill. This caps our expansion series: the protection simulator , the signings post-mortem , the new-team cores , and now the whole league. Re-run it as free agency fills the holes.