A few hundred missiles taking out large fixed critical infra like refineries is existential. Advanced rocketry tech is eroding fortress American with will greatly limit US expeditionary posture that relies on unmolested homefront. Think precise long range strikes against ships in port or strategic bombers in harden bunkers, space infra stations or production facilities. Everything changes when CONUS becomes vunerable to foreign power projection - something that constrains nearly every US adversary who has to consider homeland strikes.

E: over post limit

IMO RU has been relatively restrained in attacking power infra, mostly lower level nodes (substations) on the grid that's repairable vs what happened to Baghdad during first gulf war, worsened by western sanctions that prevented repair. UKR resilience mostly largely due to having access to spare parts from global producers. It hasn't been a "hands off" situation. Replacing transformers is very different from rebuilding power plants. Severely grading infra is closer to "call it quits' ' total war which war hasn't yet devolve to.

> it couldn't have won the war anyway

This is more or less the point. Critical infra vulnerability on CONUS opens up the “let's not have unwinnable war condition” - it's essentially an escalation rung that US hasn't had to deal with, and will circumscribe the ability for the US to respond. It's function more useful in context deterrence via conventional MAD in peer scenarios (i.e. US vs PRC over TW). There's a reason Biden specifically communicated to PRC and RU that even cyber attacks on US critical infra will be interpreted no different than kinetic war. The value of global precision strike hypersonics is predominantly to shape/limit US behaviour in the same way US carrier groups or strategic bombers projected off adversaries' shores shape/limit theirs. And so far the US has near unilateral monopoly over such deterrence/coercion/persuasion.