RPG systems that are actually fun as games.

[Pedantic]

I think anything where the results are based on DM improvisation is doomed to feel like there is no agency. See Bearworld.

That being said, "Yes, and..." is the rule of improv, and RPGs are a form of improv. That is why old-school DnD had so many tables. Tables were the original procedural generators, and rolling on those is a lot more fun than "you fail, game stops."

Yeah, having variable failure states is the interesting part of "fail forward" design. There's no reason that couldn't be combined with a design that respects player agency, it's just the opposite of the rules-light design those games have usually embraced. You'd want to spell out the effects that can happen on failure preemptively, and probably even give players some ability to choose their preferred failure state or mitigate failure states they particularly don't like. You need more detailed skill descriptions, with more outcomes specified preemptively. There's even space for some DM improvisation there for outcomes outside of the usual failure results, as long as it's called out before the check and exists as a known possible outcome or is something players can opt into.

The other thing that really should be considered more clearly is the role time plays as a resource. Take 10/20 or equivalent rules mean most of the binary "checks to see if the adventure continues" situations actually amount to a time penalty. That's only interesting if time is a meaningfully limited resource, which mostly comes down to scenario design right now. More of that could be offloaded to resource systems (fatigue, ability durations, etc.).

There's also the problem attitude that rolling dice is in and of itself interesting, which prompts GMs to call for rolls for situation that don't actually merit them. I think your take 10/20 rules need to be less a thing a player calls for and more a result of the game state. You should have to declare that there is a time pressure, or a stressful situation or whatever and otherwise have the default result be the highest possible player outcome.

Checks for "does a character know something?" are the biggest binary "does the adventure continue" check, and need to be rooted out or otherwise accounted for. You can do it in scenario design, by just telling GMs never to lock progression behind those checks, but it's always a risk to offload that work on to GM shoulders. At the very least, you'd want to give players the ability to spend a meta currency for success on those checks.

This sort of sounds like PCs going from random table to random table, though. Maybe buttressed by an encounter with a statblock. I really enjoy random tables, but this sounds excessive. Maybe I'm misunderstanding.

The key is the player point of interaction. If you know the possible outcomes of trying to pick lock vs. casting knock vs. bashing the door down, you can make a decision about which results have the greatest upside and which risks are least bad for you. Once you've decided on a risk to take, having a randomized result for how that risk plays out makes the outcome interesting, and leads to emergent situations that are fun for the GM and players to work through.

[K]

Imagine failing a roll to unlock a lock, but rather than just failing, you roll on a failed lockpicking chart and get the result "Pick breaks loudly, alerting anyone inside that the door was tampered with. They investigate."

So the PCs lose the chance for surprise, but the game continues because the guards come boiling out of the room. It's a randomizing event, but at no point on the chart is "Bears attack from nearby bushes."

The problem I have with this is that some players will feel like the failure is not the way their characters would have handled things. For example if you have a "you break all your picks trying to open that lock" result, some players will complain that their character would have stopped before being out of picks.

So something that I like to do to have this kind of mechanism without robbing players of their agency is to let the player decide: either the attempt is failed or he can choose (either from the table or something he makes up) what to trade for the success or he can decide to take his chance and roll a die with a 50% chance of critical failure and a 50% chance of success.

The thing is, players don't feel like they lost agency when they get hit by a monster who beat the character's AC. Bad things can happen to PCs that they don't control and its OK. It's actually the preferred state, because the verisimilitude of an RPG breaks down the instant you go "and now I don't accept that this event happens." The rules of "Yes, and..." is that both sides are saying "yes" all the time.

The trick is to not make the bad thing a "no." Telling the PC that they broke all their picks is a way of saying "no more lock-opening for you." The player is correct that they would keep spares or in some way account for the lock-pick-breaking problem ("I make extra picks from the bones of golems we kill!")

Manipulating the results after they have been determined with rerolls or PCs choosing outcomes... that's too much PC agency and equally destructive because you aren't improvving any more. You are saying "no" to the narrative.

Certainly, it would be very tricky to write up the tables to always say "yes" to the narrative while still feeling like a failure, but I think it's possible.

[Blade]

The thing is, players don't feel like they lost agency when they get hit by a monster who beat the character's AC. Bad things can happen to PCs that they don't control and its OK.

I'm fine if my character gets hit by a monster who beat his AC, but I'm not fine if it becomes: "You take a step back to avoid the monster's sword and fall down the hole behind you." (unless there's a critical success/failure involved) because the character can't decide if he gets hit or not, but he can decide if he'd rather risk getting hit by the sword or risk falling down the hole (unless there's a critical success/failure involved).

Let's return to the lock-picking example, let's consider that we have a table with the following fail-forward options:
- "You break one of your lock-picks, but you manage to open the door"
- "You make a lot of noise, but you manage to open the door"
- "You take longer than usual, but you manage to open the door"

None of them are saying "no" (except possibly the breaking lock-pick if it's the character's last pick) but depending on the situations, these options are not equal. And the character should be able to steer towards one or another: if we're chased down, I don't care if I make noise, but I don't want to take longer than usual. There can still be unwanted consequences, but they should be less likely than "accepted" bad consequences.

[violence in the media]

The thing is, players don't feel like they lost agency when they get hit by a monster who beat the character's AC. Bad things can happen to PCs that they don't control and its OK.

I'm fine if my character gets hit by a monster who beat his AC, but I'm not fine if it becomes: "You take a step back to avoid the monster's sword and fall down the hole behind you." (unless there's a critical success/failure involved) because the character can't decide if he gets hit or not, but he can decide if he'd rather risk getting hit by the sword or risk falling down the hole (unless there's a critical success/failure involved).

Let's return to the lock-picking example, let's consider that we have a table with the following fail-forward options:
- "You break one of your lock-picks, but you manage to open the door"
- "You make a lot of noise, but you manage to open the door"
- "You take longer than usual, but you manage to open the door"

None of them are saying "no" (except possibly the breaking lock-pick if it's the character's last pick) but depending on the situations, these options are not equal. And the character should be able to steer towards one or another: if we're chased down, I don't care if I make noise, but I don't want to take longer than usual. There can still be unwanted consequences, but they should be less likely than "accepted" bad consequences.

Are we actually presuming that people are tracking individual lock-picks in this hypothetical game? Are they tracking arrows, pitons, spell components, rations, and liters of water as well? I thought the "you break a pick" was just part of the flavor text for "it makes a lot of noise and alerts whoever is in the room." Also, every one of your examples has them opening the door on a failed roll--what is the point of even making the skill check? Just build a simple success right into the lock-picking table.

As an aside, I've broken several screws and drill bits without having any intention of doing so. There often isn't a moment where you get to think, "I should stop, this is gonna break."

[shinimasu]

Even rules light games sometimes track resources, they just tend to use pools of "resource points" instead of listing out each individual thing the player has. With small items of no huge significance not counting towards the pool.

So you get something like "You have 10 supply, your tool breaks, you have 9 supply but the door is open" in which case the failure still cost something but the something wasn't the story's forward momentum.

[K]

This sort of sounds like PCs going from random table to random table, though. Maybe buttressed by an encounter with a statblock. I really enjoy random tables, but this sounds excessive. Maybe I'm misunderstanding.

It should be auto-success if you have enough skill. I've lost too many Rogues in DnD from falling out of trees because I botched a climbing roll. It's certainly "funny" and creates tension to roll dice all the time, but it's not good design.

There is a sweet spot for randomizing events, especially if they are going to affect overall narrative. Too many and it's all chart-look-up and mad-lib narrative forever. Too few, and you start to get risk averse to the situations where they are featured (I can't tell you the number of random corruption tables I've avoided rolling on by refusing to get into that situation).

It basically has to be baked into the background of everything where you'd be tempted to roll or when you are deciding something minor that was not previously decided in the adventure design. "Is there a piece of furniture in this castle hallway that I can throw down the stairs?" is great for a chart, but wandering monster tables are not because the first enhances the narrative and the second replaces it.

[Trill]

Also, every one of your examples has them opening the door on a failed roll--what is the point of even making the skill check? Just build a simple success right into the lock-picking table.

Violence, that's the point of fail-forward.
Yes, no matter whether you fail or not you open the door. But success means you manage the door without negative consequences.
It's the question: "Did you manage to open the door?". Succeeding means you answer with "Yes". Failure means you answer with "Yes, but..."
Because the other option is to have them not open the door, which either means that they sit there trying again and again (which is quite boring), or just leave.

To give another example: Suppose you try to jump on top of a wall.
Success means you jump up and are now on top of it.
Failure instead can mean that you got up there, but took a lot of time
or that you scraped up your knees
or that you spilled your stuff
or whatever

You still want to succeed, because those consequences can harm your plans.

[Stubbazubba]

Because the other option is to have them not open the door, which either means that they sit there trying again and again (which is quite boring), or just leave.

If there was no urgency, no consequence to picking the lock slowly, why did you have them roll in the first place? Just narrate the Take 20 outcome and move on.

If there's a consequence to failing to pick the lock quickly or stealthily or whatever else, then if they fail that consequence happens. They may not necessarily be aware of the consequence, e.g. if it's a pretty small room and whatever's inside notices the attempt to pick the lock. In which case they fail once and you narrate that the lock was noisier than they expected, but it clicks open (and out burst the monsters, if they're of that persuasion).

If they simply want to try to beat their own speed record, and they fail, then you narrate that consequence; it takes a lot longer than they hoped but the lock eventually opens. Move on.

The check isn't to see if you succeed, because given enough time and focus, you'll eventually succeed. The check is to see if you perform well enough to avoid whatever consequences accompany failure. Once you've triggered those consequences, you don't keep rolling because the consequences are already triggered, there's nothing the check can tell you (unless there are more consequences that can be triggered but haven't yet been).

I am opposed to the idea of a random table to determine just what consequence it brings, I think the context should make it clear, and 99% of the time the player should be aware what the consequence for failure is: either you know there's a guard patrol making its rounds, or you're on a stealth mission and silence is key, or whatever else, and the MC should be explicit about it when calling for the check, "OK, the lock looks pretty tough, if you're not on your game the guards might loop back around before you get it, roll for it," or "The handle is locked, you can pick it but there's a chance they'll notice as you do, roll for it." Those two might not even be the same skill; the latter might be stealth/sleight of hand or something. But also, in the former you may not end up picking the lock, if the guard patrol shows up and you combat instead. Or you leave your allies to fend off the guards while you finish picking the lock. The point is the context provides all of those possible outcomes, not a random table.

And when there's not a direct consequence from the context, yeah, you should just auto-succeed. There's no need to have a check just to have one.

You still want to succeed, because those consequences can harm your plans.

You have to be careful with how you phrase the question the check is answering, otherwise no, those consequences will never actually harm your plans. At some point, the answer to the question is "No." If I want to Persuade the cultists to stop the ritual to sacrifice the virgin, that can't be a guaranteed success and I'm just rolling to determine ancillary consequences: failure sometimes means no.

[Lord Charlemagne]

Another issue with fail-forward mechanics is that they can often struggle to play nice with other mechanics of the game that aren't fail-forward as well. If a system's skill system is entirely fail-forward while the combat system isn't, then there is a real incentive to just focus solely on combat as that is the mechanic where you can lose. You can avoid this by making it so that knowing how to pick locks is something anyone can do with zero or near zero investment, but then you create the fact that the skill of lock-picking is incidental and then the skill check is just a roll to modify the next encounter.

This is of course highly system dependent. If the quest is about hacking the mega-corporations top computer system from their base of operations, you need to have it so the party is capable of actually doing that and the plot needs to be able to advance forward. This can be with a fail-forward system, the existence of multiple ways to solve the problem with certain skills making it easier, or through the plot being able to advance forward with the party being unsuccessful.

[JigokuBosatsu]

Maybe it's not important, but does the whole "bag of disposable lockpicks that break when you fail to pick" trope have any TTRPG provenance, or is it just a vidya game thing? I mean... I've never broken a lockpick even when I've jammed so hard that the retaining clip on the barrel broke. I suppose it doesn't break immersion for people who've never tried it, but I'm still curious.

[Foxwarrior]

Random lockpicking thread from the internet

Turns out that it's vaguely based in reality, some people do break their lockpicks every once in a long while. I've never seen it as a mechanic in a TTRPG though.

[Zaranthan]

I've only seen it in games where a computer is counting your lockpicks and you have a stack of them like ammunition. TT games that mention picks breaking never seem to offer a cost to replace them, but I've seen several that treat it like "your pick breaks, which makes a loud snap, and one of the wandering encounters comes to investigate".

[JigokuBosatsu]

Ahh, I see people are wearing them out. That makes sense, and I suppose if fantasyland has shitty steel that would happen faster. But... ehh. I don't know that it ultimately matters. I suppose in a fail forward scenario if you fuck up your lockpicking roll then a previously nonexistent beartrap inside the lock could snap your pick.

[Meikle641]

I think Tasslehoff Burrfoot broke one of his in a novel, but that's about it...

In the Night Angel trilogy, Azoth lusts after a special metal set of picks, because it would never break from metal fatigue and was flexible, but was a bitch to make and hence expensive.

[violence in the media]

Also, every one of your examples has them opening the door on a failed roll--what is the point of even making the skill check? Just build a simple success right into the lock-picking table.

Violence, that's the point of fail-forward.
Yes, no matter whether you fail or not you open the door. But success means you manage the door without negative consequences.
It's the question: "Did you manage to open the door?". Succeeding means you answer with "Yes". Failure means you answer with "Yes, but..."
Because the other option is to have them not open the door, which either means that they sit there trying again and again (which is quite boring), or just leave.

To give another example: Suppose you try to jump on top of a wall.
Success means you jump up and are now on top of it.
Failure instead can mean that you got up there, but took a lot of time
or that you scraped up your knees
or that you spilled your stuff
or whatever

You still want to succeed, because those consequences can harm your plans.

I get what you're saying about fail forward, I just don't understand where it should or shouldn't force you to try another approach. The previous example had the lock-picking fail and the party had to resort to kicking the door in. Is fail-forward supposed to be saying "yes" to every single thing the players attempt, or is it supposed to be looking at a larger picture. In the locked door scenario, the goal is to get into the room. Whether that happens by picking the lock or not is ultimately irrelevant.

[deaddmwalking]

In the [url=hhttps://playclassic.games/games/role-playing-dos-games-online/play-eye-of-the-beholder-online/]Eye of the Beholder Video Game[/url] you had to have lockpicks in order to try to unlock a door; a thief character would start with a set and you could find some in various places. Those lockpicks could break with a bad enough roll.

[Blade]

Is fail-forward supposed to be saying "yes" to every single thing the players attempt, or is it supposed to be looking at a larger picture. In the locked door scenario, the goal is to get into the room. Whether that happens by picking the lock or not is ultimately irrelevant.

I have two problems with this approach. The first is that it leads to vague narratives: "You somehow enter the building". It's either that or you arbitrarily take a decision for the characters and that decision can have impacts later on.

Which is my second problem: the way something is done can have impacts on the way the story goes. Kicking the door will be more likely to draw attention than picking the lock.
This you can solve by having a "modifier" mechanism: "the goal is to get into the room without raising an alarm".

But then there are other consequences that you cannot always be aware of, especially longer-term consequences. You can deal with them with a similar mechanism: "The goal is to get into the room without raising an alarm nor leaving traces", this is more or less the approach of "On Mighty Thews" (or at least the version I've played, I'm not sure how close to the original rules we were).
But then the game turns into a game of "find the fail-proof combination of modifiers" vs "find the weakness in the modifiers" and you get into the "it's fine if you play with the right people" situation.

[Orca]

The only tabletop RPG which has a serious, listed chance of breaking lockpicks that I've seen is PF2. I expect that Rolemaster has a fumble chart with a result like that somewhere, but I don't remember seeing such.

[K]

I get what you're saying about fail forward, I just don't understand where it should or shouldn't force you to try another approach. The previous example had the lock-picking fail and the party had to resort to kicking the door in. Is fail-forward supposed to be saying "yes" to every single thing the players attempt, or is it supposed to be looking at a larger picture. In the locked door scenario, the goal is to get into the room. Whether that happens by picking the lock or not is ultimately irrelevant.

Fail Forward is a term of art from rules-lite games, and I don't know if it's the same as what I'm talking about. It seems to be a lot of "your checks all succeed, and the result is how much I fuck with the you."

What I am talking about is that good design would say that failure should not end the narrative, but it should add to the narrative while not stopping it.

For lock-picking, failing the check should not mean that there is no other way into the room. In a lot of bad RPGs, both tabletop and cRPGs, failing that roll means that whatever is in the room is now inaccessible. Dungeon Crawl Classics doesn't have rules for breaking objects to get into chests and past doors, but 3e DnD does. Baldur's Gate won't let you break down a door or chest, but Divinity: Original Sin 2 does. The former games have to carefully design adventures to keep no locks between the main narrative because it might cause an total adventure fail on a single die roll, while the latter games are chocked full of locked doors and chests and a lot more fun to play.

Take another example: Spellcraft to learn a spell from a scroll in 3e DnD. On a failure, you can't try to learn the spell again until you level. What if that is an important spell for the narrative (Plane Shift, for example, can easily be the difference between playing an adventure and not, because it gets you to the adventure.)?

Well, what if you fail that check? Unlike a door, we can't just say that the beefy fighter throws a shoulder against it. So what if we had a table of random failures where there are partial fails like needing an expensive focus to cast the spell, or a side effect on the spell, or letting you learn the spell after buying some reference materials, but also full failures that force you to go to a back-up system like letting the casters do long rituals to cast any spell they don't know?

The point is that the RPG doesn't have to end on a failure screen because that's only the easiest option and not the best option. Failure should change the adventure and maybe even the goal of adventure, and it should not end players' ability to keep playing.

[OgreBattle]

I'm guessing in D:OS2 the reason to pick locks and not smash everything is stealth related? Like then the guards come, or your assassination target prepares.

Dungeon Crawlers, even ones where you should avoid fights, usually have an emotional high point of fighting or at least killing. So I can see "fail forward" meaning that if you succeed you go into the fight with more advantages, and "fail forward" means the fight begins with less advantages or more penalties.

In Darkest Dungeon, when your path is blocked and you have no shovels to clear it, you take stress and hp damage from clearing with hands.

[MGuy]

I think the consequence for breaking things is that it rapidly degrades your weapon and in the case of chests I think breaks what's inside or something.

[...You Lost Me]

Picking locks is more convenient.

If you hit a door with a melee weapon, it can wear down and break. You can repair weapons easily with a repair hammer. Or just shoot fire at a door until it breaks. Or you can shoot a ranged weapon at it.

Sometimes, breaking down a door will trigger a fight, but you have a top-down view and you can easily break anything without people noticing.

Breaking a chest can cause you to lose stuff inside it. But not quest items, and most of the stuff you lose isn't that valuable. You will be rolling in money 90% of the time.

The real problem is just convenience. Objects tend to have a lot of HP. You're probably already stealing things with one of your characters, so you might as well get some lockpicks (or play an undead) and open doors / chests quickly.