Quickleaf
Legend
What's more interesting to me about your question are the implicit rules that influence it. And you're calling one of those out here explicitly: Lockpicking having almost zero guidance for how the GM can adjudicate it.But D&D doesn't even really have rules for lock picking in the first place. This is part of the problem for me. D&D has vaguely defined skills that say things like "You can attempt to do X" but it says nothing about how long it takes or anything relevant. D&D doesn't attempt to simulate this aspect at all, since it has no rules for it.
Also note again that my example rogue has infinite uses of the basic lock picking skill, but has a few limited uses of "success with no risk of failure, instantly".
It's hard to say that my system is less realistic than whatever is already in D&D 5E (since there is basically nothing to compare my system to in the first place)
In my own approach to Lockpicking, I try to handle this in 3 ways.
1 - More Lock Detail
Because I am interested in encouraging creativity, I try to think about and describe locks with a bit more detail (this is often with a broad brush "the entry door is X lock, all other locked doors in the dungeon are Y locks"). For instance:
- Barred Doors – technically not a lock, (5e RAW allows Knock to be used on a barred portcullis, for example, but I prefer the older AD&D interpretation, which is itself a long story because there were contradictions in the text, but I believe "can't open bars" is the intent), and while you cannot lockpick it, you can devise creative methods of opening it
- Knot & Wax Seal – technically not a lock, but a deterrent, easy to cut past, very hard to cover your tracks
- Door Latch – technically not a lock, but also trivially easy for anyone with a bit of skill to open, an automatic success / speedbump for a rogue with proficiency in thieves' tools
- Pin Lock – push key in, lift up against a few little pins, pull back and pull the bolt lock out
- Warded Lock – twist key, but there are obstructions that a key must be shaped to slide around in order to be twisted
- Push-Key Padlock (aka Spring Lock or Fetterlock) – these have a spring mechanism inside which can trap a lockpick / the wrong key or get jammed
- Combination Locks
I've been assuming the typical "thieves' tools" includes 10 lockpicks/skeleton keys – these can be resupplied by dealing with thieves' goods or the criminal underworld, but often they're illegal. Warded Locks are the most vulnerable to lockpicking, so expending a lockpick automatically opens a Warded Lock unless it's of masterwork quality or otherwise exceptional.
Similarly, when dealing with a Pin Lock, Warded Lock, or Push-Key Padlock there's the chance your lockpick might break. Could be if the check fails by 5+ (what I usually do), could be natural 1, could be some other determination the GM makes.
3. Knock's Limitations (and Uses) in My Games
Against a Push-Key Padlock on a chest or vault door? Knock spell is your weapon of choice! Against a Door Latch? Sure you could cast Knock, but that's overkill when the rogue can pop it open automatically. Against a Barred Door? Knock won't do crap, and it's up to the clever thief to tap out the hinges or whatever.
I like to have Knock leave some sort of an effect on a Pin Lock (such as yanking the bolt out of the locking mechanism entirely) or a Warded Lock (such as damaging the obstructions in the lock to prevent it from being re-locked). So it's not just the loud sound but also inspection will reveal tampering. There's a tradeoff consideration to be made.
4. Rules Ecosystem Around Lockpicking
Destroying a door in 5e RAW is trivially easy and has very little cost – it's a choice of "do we mind spending 1-2 rounds hacking/blasting it down and leaving obvious evidence?" Not always, but more than half the time I try to attach some consequence to destroying the door – e.g. giving monsters in the room beyond automatic surprise against the PCs.
Breaking a door in 5e (Strength check) suffers the same issue as many skills in 5e – no guidelines on repeatability & no guidelines on consequence. So I have rules that mitigate that:
- Let the Roll Stand – If the barbarian couldn't break the door down, their check stands. The barbarian can't simply try again & the gnome wizard doesn't stand a chance (and cannot make that roll), unless something shifts dramatically in the narrative.
- Failed Checks Have Consequences – The result is almost never "nothing happens." Instead, the PC ramming their shoulder into the wall takes some bludgeoning damage, can't use that arm to wield things for 10 minutes as it goes numb, they break something fragile on their person, etc.