When Tordek is leaning back in his tavern chair, drinking a tankard of mead at 1 HP, he looks just like he does at 20 HP. So the White Mage comes over and casts Cure on him. Yet Tordek looks and acts exactly the same.
Why would the White Mage make the decision to cast the Cure spell on Tordek, if Tordek shows no discernible signs of needing it? Why would the White Mage, who can only cast Cure twice per day and needs to be picky about it, choose to cast the spell on Tordek but not Mialee, when both look equally uninjured?
The decision to cast a Cure spell can only rationally be made if the effects of its resolution are knowable to the spellcaster. Maybe you could do that in such a way that priests have "luck sense" which lets them see who is in danger of being doomed, but the easiest and most consistent way is to just say that HP correspond to something which is easily observable to everyone - such as state of physical injury. As long as we're talking about scratches and blunt trauma, rather than impalement, the
only issue is that we need to handwave the loss of physical capacity as a hero gets beaten up... in which case we already have the convenient explanation that this a
hero. Fighting through the pain is just basic hero stuff.
Stranger still: does the cure spell remove the dents from Tordek's armor, the nicks from his sword, and replace the boards of his shield?
Indestructible equipment is
close enough to real-world physics that we can probably ignore it. A chain shirt can take a lot of hits before it's deformed enough to no longer function, so as long as we have a passing mention of the fighter maintaining his gear during downtime, we probably don't need to track that. It's just one of those simplifying assumptions which saves us a lot of bookkeeping as long as we remember that we
are making an assumption, and as long as we remember to stop making that assumption when the pre-requisite conditions no longer hold.
Likewise, we
could make a simplifying assumption that archers don't need to track their arrows, under the conditions that an archer will always recover as many spent arrows as possible and replace the rest during downtime. Thus, we can save a lot of bookkeeping, as long as arrows aren't being shot off of a cliff and there are plenty of trees/vendors in the vicinity to allow for resupply. Once your archer is fighting fire elementals in the desert, those conditions no longer apply, so the simplifying assumption becomes unreasonable and it makes more sense to actually track your ammo.