Automatic success means you can't ever leave a prisoner alone


Skills, Feats, Equipment & Spells


(I just posted a "general feedback" post based on my group's DDD experience, but wanted to go a little deeper with this scenario here)

Here's a scenario that played out in a pf1e group recently.

The party captured a major NPC, and took him to a secure location for questioning. Once they tied him up, I was surprised to find the NPC was actually at their mercy, (this is a published adventure) with no ranks in escape artist, and a bulky CMB to contend with. They were actually able to safely rest and then leave him at that place while they proceeded to the next dungeon, and even got more info out of him about the dungeon by explaining it was in his best interest for the group to return alive and let him out instead of him starving to death. Even with a 20, the NPC would be unable to break his rope bonds, and he really is stuck in a little nook of the city sewers until the party returns. I was pretty satisfied with how it turned out, and I'm looking forward to see how they deal with him later in the game.

I was curious to see how this would work in the playtest, and, as I feared, rolling a 20 is an automatic escape from your restraints. The DC is beyond him, but a natural 20 is still a regular success. (yeah, see page 292 if you think it's a crit. Finally learned that, at least.) It's only an action to attempt an escape, so maybe he'd at least be fatigued if it takes him 300 tries (10 minutes)?

Please correct me if there's some errata or other rule I'm missing here.

I've been irked by automatic success for skills since I first read the playtest CRB. As it stands, you can't physically restrain a person for any amount of time unless there's an actual lock involved (requiring thieves tools and multiple successes). And the players never need to fear restraint unless somebody gets them in an actual cell either. Since, even untrained, it's just a matter of rolling Acrobatics over and over to get out, statistically freeing themselves within 7 rounds (21 actions).

I also noticed that the Acrobatics: Escape entry talks about manacles and says the DC would be determined by their type, but the manacles entry only mentioned the Lock Pick DCs for different qualities. It's safe to assume acrobatics would be more difficult, but it still leaves no way to make it less than a 5% chance to break out. Really it's weird *any* prison is working with this.

Mostly just wanted to draw attention to what's hardly a fringe case in any group that slightly values NPC life. I guess my larger point here is "I don't like automatic success (or failure) for skills, and this is a good example of why". It's not a bad idea for story-telling, always having *some* risk or chance, but I personally believe that NPCs are playing the same game, even if they are built differently, so this creates a lot of problems in the world. This would be alright in a system like FATE or Numenera where the players are the only ones who roll, but Pathfinder has always been more consistent about how PCs and NPCs interact with the world.


4 people marked this as a favorite.

I'm pretty sure the rules say the GM can say when a task is just flat out impossible. You aren't going to jump to the moon with an athletics check, and a level 1 character will never break adamantine manacles.

Grand Lodge

Captain Morgan wrote:
I'm pretty sure the rules say the GM can say when a task is just flat out impossible. You aren't going to jump to the moon with an athletics check, and a level 1 character will never break adamantine manacles.

The rules actually say that jumping more than 8' is flat out impossible. On a critical success of a High Jump check - one of the few listed DCs that exist.


This is simply one of those many cases where the needs of the story and GM administration must direct how those rules are used. Technically, by the rules, every criminal could also be a high-level Rogue, or a dragon in disguise, or every treasure hoard could be full of ancient artifacts, but it's up to the group and the story's needs on whether or not to implement that stuff, correct?

Does it create dramatic tension that the prisoner has a chance of escaping? Okay, leave that open. Roll in secret at various intervals. Or determine in secret...one morning the town wakes up and this most feared prisoner is gone! What do we do now?! If it doesn't fulfill any storytelling purpose, then there's no chance, just like there's no chance an elder dragon will randomly drop out of the sky and lay the town to waste, even though it COULD happen. Or circumstances don't allow for a roll, particularly for your everyday cutpurses who have minimal skill.


in◆⃟ wrote:
Captain Morgan wrote:
I'm pretty sure the rules say the GM can say when a task is just flat out impossible. You aren't going to jump to the moon with an athletics check, and a level 1 character will never break adamantine manacles.
The rules actually say that jumping more than 8' is flat out impossible. On a critical success of a High Jump check - one of the few listed DCs that exist.

To be fair, that's 30 inches better than the theoretical maximum in real life...


1 person marked this as a favorite.
Jesikah Morning's Dew wrote:

This is simply one of those many cases where the needs of the story and GM administration must direct how those rules are used. Technically, by the rules, every criminal could also be a high-level Rogue, or a dragon in disguise, or every treasure hoard could be full of ancient artifacts, but it's up to the group and the story's needs on whether or not to implement that stuff, correct?

Does it create dramatic tension that the prisoner has a chance of escaping? Okay, leave that open. Roll in secret at various intervals. Or determine in secret...one morning the town wakes up and this most feared prisoner is gone! What do we do now?! If it doesn't fulfill any storytelling purpose, then there's no chance, just like there's no chance an elder dragon will randomly drop out of the sky and lay the town to waste, even though it COULD happen. Or circumstances don't allow for a roll, particularly for your everyday cutpurses who have minimal skill.

Unless it's a PC tied up, in which case, they'll often be unhappy with "I know the rules say, but there's more dramatic tension if you just stay tied up." They'll just start making Acrobatics checks and be out in a couple minutes.

Bad place for an auto success on 20 rule.


Draco18s wrote:
in◆⃟ wrote:
Captain Morgan wrote:
I'm pretty sure the rules say the GM can say when a task is just flat out impossible. You aren't going to jump to the moon with an athletics check, and a level 1 character will never break adamantine manacles.
The rules actually say that jumping more than 8' is flat out impossible. On a critical success of a High Jump check - one of the few listed DCs that exist.
To be fair, that's 30 inches better than the theoretical maximum in real life...

OTOH, there's plenty of other superhuman things that can be done, so I'm not really fond of such hard limits.


thejeff wrote:

Unless it's a PC tied up, in which case, they'll often be unhappy with "I know the rules say, but there's more dramatic tension if you just stay tied up." They'll just start making Acrobatics checks and be out in a couple minutes.

Bad place for an auto success on 20 rule.

Each roll doesn't have to represent a single attempt at breaking free. The GM is well within their rights to limit the amount of rolls based on time, fatigue, etc. One roll could account for an hour of trying. Just like you can't sit there and roll over and over again to lift bars. Or at least, the rules might not say you can't, but the GM and the narrative are under no obligation to allow endless attempts at it.


Pathfinder Lost Omens, Rulebook Subscriber
Jesikah Morning's Dew wrote:
thejeff wrote:

Unless it's a PC tied up, in which case, they'll often be unhappy with "I know the rules say, but there's more dramatic tension if you just stay tied up." They'll just start making Acrobatics checks and be out in a couple minutes.

Bad place for an auto success on 20 rule.
Each roll doesn't have to represent a single attempt at breaking free. The GM is well within their rights to limit the amount of rolls based on time, fatigue, etc. One roll could account for an hour of trying. Just like you can't sit there and roll over and over again to lift bars. Or at least, the rules might not say you can't, but the GM and the narrative are under no obligation to allow endless attempts at it.

Its also far more tense to have "player rolls while the guard is away can they break out before he gets back" than "you aren't guarded at all oh look you get away what a surprise." Like it seems odd to complain about poor narratives when you are setting up boring scenarios in the first place.


2 people marked this as a favorite.
Jesikah Morning's Dew wrote:
thejeff wrote:

Unless it's a PC tied up, in which case, they'll often be unhappy with "I know the rules say, but there's more dramatic tension if you just stay tied up." They'll just start making Acrobatics checks and be out in a couple minutes.

Bad place for an auto success on 20 rule.
Each roll doesn't have to represent a single attempt at breaking free. The GM is well within their rights to limit the amount of rolls based on time, fatigue, etc. One roll could account for an hour of trying. Just like you can't sit there and roll over and over again to lift bars. Or at least, the rules might not say you can't, but the GM and the narrative are under no obligation to allow endless attempts at it.

It is much, much easier to bend the rule 0 in the players' favor then against. Any PC attempting to do nearly anything that isn't 'jump to the moon" impossible but still well beyond their capabilities is going to expect the game to work consistently, and is going to view any GM fiat of that nature extremely unfavorably. "You can't do this thing the rules say you can do because I said so."

The system assumes that anything that is attempted has, at a minimum, a 5% chance minimum of either success or failure or that what's being attempted is preposterous for anyone to attempt, like jumping to the moon. It doesn't, RAW, take into account people who just lack the ability to ever succeed like someone else night, nor does it handle repeated attempts very well.

Struggling against ropes barely requires a minute of actions which isn't even fatiguing, so repeated attempts here result in a weird amount of success for characters with no investment in the skill. Disabling traps is another example, requiring a lot of repeated rolls where nothing happens, waiting for either a critical failure to trigger the trap or a success to finally disable it, and since the DC's are so high to make critical failure a real threat it means 10 results, or 50& of all rolls, result in nothing at all happening.

The old system in PF1 did at least give the GM the option of citing the character's own lousy stats for why the roll won't work. As a random, poorly thought out example of an improvement, it could be adjusted so that if a character would critically succeed on a 1 or critically fail on a 20, they do that automatically, ignoring the normal rules for 1's and 20's. You can still have the player roll their dice and tell them what happened, as things generally do happen in those cases that void the need for further rolls, but the GM no longer needs to rely on fiat for doable tasks that nonetheless require superhuman capabilities. And also keeps high level adventurers from ever failing DC 5 checks.

Dunno how to fix the trap problem other than to have a better way to handle repeated attempts so it's not just watching someone roll dice endlessly. Half of the possible rolls shouldn't be "nothing happens."


Helmic wrote:
Jesikah Morning's Dew wrote:
thejeff wrote:

Unless it's a PC tied up, in which case, they'll often be unhappy with "I know the rules say, but there's more dramatic tension if you just stay tied up." They'll just start making Acrobatics checks and be out in a couple minutes.

Bad place for an auto success on 20 rule.
Each roll doesn't have to represent a single attempt at breaking free. The GM is well within their rights to limit the amount of rolls based on time, fatigue, etc. One roll could account for an hour of trying. Just like you can't sit there and roll over and over again to lift bars. Or at least, the rules might not say you can't, but the GM and the narrative are under no obligation to allow endless attempts at it.
It is much, much easier to bend the rule 0 in the players' favor then against. Any PC attempting to do nearly anything that isn't 'jump to the moon" impossible but still well beyond their capabilities is going to expect the game to work consistently, and is going to view any GM fiat of that nature extremely unfavorably. "You can't do this thing the rules say you can do because I said so."

This is so true. I feel like auto success isn't taking that into account right now. The rules simply say that a nat 20 is a success and don't say "this only applies when you could realistically actually do the thing in question".

They would either have to adjust that, or add an extra rule about how if you can't possibly succeed except by the nat 20 success (ie, the DC is higher than your highest possible check result), the DM can disallow rolling entirely and the auto success doesn't apply.

But saying "a nat 20 is a success" in the rules without any further clarification, and then expecting a DM to tell their players "I know what the rules say, but you're not allowed to do that anyway" is... not great. This isn't some weird edge case like trying to shoot a dragon in the eye with a ballista at 400'.

Quote:
The system assumes that anything that is attempted has, at a minimum, a 5% chance minimum of either success or failure or that what's being attempted is preposterous for anyone to attempt, like jumping to the moon. It doesn't, RAW, take into account people who just lack the ability to ever succeed like someone else night, nor does it handle repeated attempts very well.

Yep.

Quote:
Struggling against ropes barely requires a minute of actions which isn't even fatiguing, so repeated attempts here result in a weird amount of success for characters with no investment in the skill. Disabling traps is another example, requiring a lot of repeated rolls where nothing happens, waiting for either a critical failure to trigger the trap or a success to finally disable it, and since the DC's are so high to make critical failure a real threat it means 10 results, or 50& of all rolls, result in nothing at all happening.

Disable device seems like a case where +/- 10 should actually be +/- 5 instead, just to limit how many rerolls are going on. Narrow the window for success or failure and let the game move on. Nothing is gained by making us sit there rolling nine times in a row until we get a result that actually does something.

Quote:
The old system in PF1 did at least give the GM the option of citing the character's own lousy stats for why the roll won't work. As a random, poorly thought out example of an improvement, it could be adjusted so that if a character would critically succeed on a 1 or critically fail on a 20, they do that automatically, ignoring the normal rules for 1's and 20's. You can still have the player roll their dice and tell them what happened, as things generally do happen in those cases that void the need for further rolls, but the GM no longer needs to rely on fiat for doable tasks that nonetheless require superhuman capabilities. And also keeps high level adventurers from ever failing DC 5 checks.

This makes sense, but it's only going to apply when you start getting to higher levels. Until someone has a +14 total in a skill, they won't be able to reach the necessary 15 on a nat 1 to meet the criteria.

Similarly, with a 0 in a skill, the lowest one that a 20 roll would still be a crit failure is a DC 30, and you wont' be seeing many of those at low level.

I'm not sure what the solution is, but I agree that the status quo is a problem. There also shouldn't be infinite retries. That was what take 20 was for (avoiding wasting everyone's time by rolling over and over again to go 20 fishing), and if you couldn't succeed on that, then there was no need to roll at all and you just couldn't do it. So maybe that's an option: if it's something that allows retries and you could succeed with a 20 on the dice (excluding the auto success), you can simply take 20 and do so given enough time. If you couldn't succeed with that, then the auto success doesn't apply and it's impossible for you to do it.

Quote:
Dunno how to fix the trap problem other than to have a better way to handle repeated attempts so it's not just watching someone roll dice endlessly. Half of the possible rolls shouldn't be "nothing happens."

Well, one option there is what I mentioned above to lower the crit success/failure thresholds from 10 to 5, causing those to be much more frequent and the range of "does nothing" results to shrink. Another option is to lower the DCs across the board and make a normal failure do something.

But, what we don't want is what I've seen with people trying to recover from dying, where they get dying 2, succeed to get dying 1, fail to get dying 2, succeed to get dying 1, fail to get dying 2... Someone did that for five rounds in a row in the last playtest game I was in, it was just silly.


Lock picking mechanics are an issue right now, specifically with the crit failure and success models. Currently you can succeed on a 20 as long as you're trained and have some tools. The only penalty for a critical failure is breaking your tools, which has them treated as poor quality. But this doesn't stop you from using them; it only gives you -2 item penalty and removes any potential item bonus they gave you.

This means that when you can just use some poor quality lock picks to eventually open any lock. It may take a stupid amount of rolls but statistically you can get to 3 nat 20s in a row eventually. We are basically back to an unwritten Take 20, except you can go above 20 as written.

I think the conflict is that auto succeed on a 20, auto fail on a 1 can be a fun mechanic where failure has consequences, and it makes logical sense. Even the smoothest talker can say one wrong thing, and even the least skilled person might land a lucky hit or say the right thing at the right time. But it doesn't work great in a context with no time or failure constraints, like lockpicking or escaping bonds.

One elegant solution might be to make locks and restraints proficiency gated like many hazards are. Right now higher quality locks just have higher DCs, but that doesn't prevent anyone trained from just used poor tools to get through them eventually. But if you need to be an expert to pick an expert quality lock, it creates an elegant way for your low level rogue to get past it while also explaining why some random schmuck could not. And once you start being able to afford Master locks, you've stopped anyone below 7th level from being able to pick it, ever.

We currently have no DCs for breaking or escaping bonds, but I think similar principles apply. The hero or villain managing to escape their bonds is too iconic to remove entirely from the game, but even setting aside the auto-success on a 20 rule it seems like +level means every creature gets increasingly tough to restrain if you merely rely on DCs.

But if instead we have:

Rope: Trained in Acrobatics, or Expert in Athletics-- You can't lock-pick a rope, but a sufficiently motivated person with the physical training is going to get out eventually. I used expert for athletics because I don't know that the average farmer should be busting out of bonds with two minutes of work, but an expert adventurer? Sure. I feel like acrobatics training should be ubiquitous, so I left it low.

Manacles: Trained in Thievery, Expert in Acrobatics, or Master in Athletics. It shouldn't be that hard to pick basic manacles *IF* you have picks. I figured it is probably harder to wriggle free from irons than rope. And 7th level masters seem like a good point to where you can just start popping chains on pure muscle.

Expert Manacles: Expert in thievery, Expert in Acrobatics, or Master in Athletics. Lock picking scales just like a normal lock does. I don't think the athletics or acrobatics should feel much harder though.

Master Manacles: Master in thievery or acrobatics, Legendary athletics. Now we have upped the ante, and the physical constraints bump up a step. This will restrain even freakishly strong people, but a master acrobat can still dislocate their way out.

I'm not 100% satisfied with that progression. I'd like it if TEML lined up more directly with the item quality, but that just doesn't seem reasonable that a trained person could bust iron manacles. But I think the idea has legs.


Pathfinder Lost Omens, Rulebook Subscriber
Tridus wrote:


This is so true. I feel like auto success isn't taking that into account right now. The rules simply say that a nat 20 is a success and don't say "this only applies when you could realistically actually do the thing in question".

Yes there is. Pg 292 "If you lack the proficiency for a task in the first place, or it’s impossible, you might still fail on a natural 20."


Malk_Content wrote:
Tridus wrote:


This is so true. I feel like auto success isn't taking that into account right now. The rules simply say that a nat 20 is a success and don't say "this only applies when you could realistically actually do the thing in question".

Yes there is. Pg 292 "If you lack the proficiency for a task in the first place, or it’s impossible, you might still fail on a natural 20."

Oh hey, so it is. Good to know, thanks. :)


1 person marked this as a favorite.
Helmic wrote:
It is much, much easier to bend the rule 0 in the players' favor then against. Any PC attempting to do nearly anything that isn't 'jump to the moon" impossible but still well beyond their capabilities is going to expect the game to work consistently, and is going to view any GM fiat of that nature extremely unfavorably. "You can't do this thing the rules say you can do because I said so."

Rule 0 doesn't say "You can do or attempt to do anything all the time without any consequences or consideration for what is actually happening," though. As a GM you are under no obligation to let a PC throw a thousand-ton boulder just because they're gonna roll Strength over and over until they get a 20.

Helmic wrote:
Struggling against ropes barely requires a minute of actions which isn't even fatiguing, so repeated attempts here result in a weird amount of success for characters with no investment in the skill. Disabling traps is another example, requiring a lot of repeated rolls where nothing happens, waiting for either a critical failure to trigger the trap or a success to finally disable it, and since the DC's are so high to make critical failure a real threat it means 10 results, or 50& of all rolls, result in...

Have you ever struggled against ropes or wrestling holds or the like? It is very much fatiguing. And sure, these are fantasy heroes, so not likely to wear out as easily as us mortal folk, but they still can't go on and on forever. It is neither unreasonable, nor even explicitly against the rules to say "Okay, your third attempt has moved to now taking half an hour for each roll, unless you can find a way to change the circumstances" or something like that. This is a case where game rules are not able to simulate narratives or any sense of "in-world physics" or the like, and require a social contract to favor the in-game events over out-of-game unrealistic endless attempts.

There's nothing wrong with rolling over and over until you get what you want, but there's nothing wrong with saying that it's not feasible, either. Presumably the PCs have other things they can do, like find that sharp stone hidden in the straw on the cell floor to try to cut their bindings or the like, rather than just saying "Nope, tenth time failed, gonna do it again."


1 person marked this as a favorite.
Jesikah Morning's Dew wrote:
Rule 0 doesn't say "You can do or attempt to do anything all the time without any consequences or consideration for what is actually happening," though. As a GM you are under no obligation to let a PC throw a thousand-ton boulder just because they're gonna roll Strength over and over until they get a 20.

We're not attempting something obviously impossible for all adventurers, though. We're trying to break restraints. A level 20 Barbarian with a 22 in STR and Legendary Athletics may be capable of breaking chains bound around their body, but a level 5 Wizard with a 10 in STR and only Trained in Athletics should have no hope of ever doing that. Hell, even regular rope should be beyond their capabilities.

And since breaking rope isn't intuitively a trained thing you do — you just flex really really hard — it's also an awful candidate for proficiency gating. It's something whose raw DC should make impossible, but only for some characters.

Helmic wrote:
Struggling against ropes barely requires a minute of actions which isn't even fatiguing, so repeated attempts here result in a weird amount of success for characters with no investment in the skill. Disabling traps is another example, requiring a lot of repeated rolls where nothing happens, waiting for either a critical failure to trigger the trap or a success to finally disable it, and since the DC's are so high to make critical failure a real threat it means 10 results, or 50& of all rolls, result in...
Jesikah Morning's Dew wrote:
Have you ever struggled against ropes or wrestling holds or the like? It is very much fatiguing. And sure, these are fantasy heroes, so not likely to wear out as easily as us mortal folk, but they still can't go on and on forever. It is neither unreasonable, nor even explicitly against the rules to say "Okay, your third attempt has moved to now taking half an hour for each roll, unless you can find a way to change the circumstances" or something like that. This is a case where game rules are not able to simulate narratives or any sense of "in-world physics" or the like, and require a social contract to favor the in-game events over out-of-game unrealistic endless attempts.

All physical activity can be fatiguing, but that's not what the game term is referring to. The game term refers to an extreme kind of tired that can't just be cured by catching your breath for a second. Fighting for even just a minute is, IRL, going to be far more tiring than most physical activities. Struggling against rope for a few minutes is not going to intuitively be that sort of tiring. Sure, struggling for an extended period of time greater than ten minutes might need you to take a break, but a couple minutes of struggling might not even require you to catch your breath.

The issue is that you're relying entirely on GM fiat to screw over a player for trying something the rules say they can do very quickly (it's literally just an action, one that's far less tiring than trying to attack with a weapon while keeping your guard up). That makes it ripe fodder for arguments and feelings that the GM is being unfair and arbitrary, or is going out of their way to punish someone.

What makes this seven worse is that, because it's in the rules, players are going to plan around it, and that's perfectly reasonable. It's why we're playing Pathfinder instead of FATE, the rules are generally consistent enough that you can plan ahead. It means that if one player starts rolling to hope to auto-succeed on a 20 (they probably don't even know the DC is that high, they don't know the rope was magically strengthened), it's going to feel awfully arbitrary if their three rolls took an entire hour and fatigued them while someone else did it in a matter of rounds, not because of the rolls but because the GM decided to make someone else's attempt take longer. It is just bad practice, I've seen GM's do it before and it's never gone over well. It just feels like the GM is cheating to be a jerk, even if that's not what they're trying to do.

The game rules explicitly lay out how long attempts are, and it balances stuff around the assumption that each attempt takes the same amount of time unless stated otherwise. Traps, for example, assume repeated attempts. If each attempt after the second took an hour, that would be extremely unreasonable. The party often has to go through ten or more rolls to do that.

Jesikah Morning's Dew wrote:
There's nothing wrong with rolling over and over until you get what you want, but...

There clearly is, though. I think needing to roll repeatedly and wait for GM input was one of the strongest complaints about the playtest. It's an OOC time waster.

Jesikah Morning's Dew wrote:
there's nothing wrong with saying that it's not feasible, either.

Yes, there is, as I've explained. If the rules say one thing and you day another and the rules would work in the player's favor, it's going to be an extremely unfun situation. The better solution isn't to just "out-GM" the problem, it's to fix the rules so players aren't being led to believe they should be breaking rope bindings with no training at level 1 within a couple minutes. If the rules didn't directly imply that, we wouldn't have an issue to begin with.


1 person marked this as a favorite.
Helmic wrote:

We're not attempting something obviously impossible for all adventurers, though. We're trying to break restraints. A level 20 Barbarian with a 22 in STR and Legendary Athletics may be capable of breaking chains bound around their body, but a level 5 Wizard with a 10 in STR and only Trained in Athletics should have no hope of ever doing that. Hell, even regular rope should be beyond their capabilities.

And since breaking rope isn't intuitively a trained thing you do — you just flex really really hard — it's also an awful candidate for proficiency gating. It's something whose raw DC should make impossible, but only for some characters.

I'm not really talking about proficiency-gating. I'm just talking about it's not realistic or possible for someone to continuously try to break ropes or lift bars or something for hours on end without a break until they succeed or get a critical failure. That is where pure rules must take a backseat to any kind of narrative sense, unless that's not what you're after as a game experience. Which, fine if you are, but I'm not going to let characters sit and roll and roll and roll to lift a gate or something if it makes no dramatic sense.

If there's no drama or storytelling need for them to fail, they can just succeed. If it's a combat or high-pressure situation, you can let them attempt a few times while events progress around them—a prison break or a fight breaks out, and they're desperately trying to break free before the opportunity is gone or something. Okay, sure. If they're just supposed to be trapped for hours with nothing else going on, awaiting execution in the morning? Roll once or twice to represent the initial attempts (which may take longer than a single action), then challenge them to alter the circumstances (such as finding said sharp stone).

Helmic wrote:
The issue is that you're relying entirely on GM fiat to screw over a player for trying something the rules say they can do very quickly (it's literally just an action, one that's far less tiring than trying to attack with a weapon while keeping your guard up). That makes it ripe fodder for arguments and feelings that the GM is being unfair and arbitrary, or is going out of their way to punish someone.

If that's the case, then the entire game relies on GM fiat. I could just as easily say an elder dragon drops down in the middle of the town and attacks. I don't feel it's punishing someone to state that, "No, you can't roll a hundred times in a row and take all night to try to lift those bars. If you can't do it yet, something has to change for you to try again." I've never had this be an issue, I'll admit. Just blindly rolling until you win feels a little like the sort of thing that happens if you try to just view the rules in a vacuum, or take a more wargaming aspect to it, ignoring narrative and logical consequences.

Now, don't get me wrong—I don't say this in a judgmental fashion, or casting aspersions on a playstyle. Certainly that's not "wrong" or not fun or anything like that. Just not my preferred style of play.

Helmic wrote:

What makes this seven worse is that, because it's in the rules, players are going to plan around it, and that's perfectly reasonable. It's why we're playing Pathfinder instead of FATE, the rules are generally consistent enough that you can plan ahead. It means that if one player starts rolling to hope to auto-succeed on a 20 (they probably don't even know the DC is that high, they don't know the rope was magically strengthened), it's going to feel awfully arbitrary if their three rolls took an entire hour and fatigued them while someone else did it in a matter of rounds, not because of the rolls but because the GM decided to make someone else's attempt take longer. It is just bad practice, I've seen GM's do it before and it's never gone over well. It just feels like the GM is cheating to be a jerk, even if that's not what they're trying to do.

The game rules explicitly lay out how long attempts are, and it balances stuff around the assumption that each attempt takes the same amount of time unless stated otherwise. Traps, for example, assume repeated attempts. If each attempt after the second took an hour, that would be extremely unreasonable. The party often has to go through ten or more rolls to do that.

It does NOT say in the rules that you are allowed to attempt something without in-game consequence. You can try to pickpocket the guard, but you're not guaranteed success there any more than you are here, or unlimited attempts, or freedom from the consequences.

This is an easy sort of thing to work out ahead of time with your group. Mine tends to be the type to not reduce things to "rolling endlessly to auto-succeed on a 20" because that's just not fun for us. If there is no pressure on whether or not they succeed, then they can just succeed, or if failure is more interesting, we explore those consequences. If there IS pressure, then they can make some rolls as the tension in the scene mounts and some looming consequence adds to the drama.

Disarming a trap: Okay, so the roll part is important, but there's no other pressure? Fine, just sit there and try it, because it's interesting, and makes sense. You screw up badly and you trigger the trap. So keep trying to pick the lock. That's different, though.

Context matters. The scene helps determine how. I'm not sure how else to explain it to your liking.

I'm okay with there not being an auto-success rule at all, honestly, and an auto-failure one. I'm not a big fan of absolutes in gaming systems, and I prefer other ways to make critical successes/failures a thing. So I'd be okay with this being removed entirely, or making it a house rule—but any play groups would be keenly aware of that at the outset.


1 person marked this as a favorite.

Repeatedly trying to escape from ropes seems like the sort of thing that would occur in gameplay often enough that it would be good if the rules could handle it. Something along the lines of "natural 20 counts as a +5/+10 bonus rather than auto-success, a certain number of failures / critical failures in a given hour / day fatigues you".


2 people marked this as a favorite.
Matthew Downie wrote:
Repeatedly trying to escape from ropes seems like the sort of thing that would occur in gameplay often enough that it would be good if the rules could handle it. Something along the lines of "natural 20 counts as a +5/+10 bonus rather than auto-success, a certain number of failures / critical failures in a given hour / day fatigues you".

This would make more sense in general. Instead of auto-success, treat it as +10. If that's enough to make it a success, then it is. If not, then you simply can't do it. That'd also push anything that was already a success into crit success territory, so it's working on that front too.

That's fairly elegant, and provides an easily understandable situation of "no, even a nat 20 doesn't let you succeed on a DC 50 check with your +3 modifier" without any DM fiat required.


The Escape action is poorly written for people tied up with rope.

[[A]] ESCAPE
When trying to escape from restraints or a creature’s Grapple, attempt an Acrobatics check. If you’re tied up by a creature, the escape DC is that creature’s Thievery DC. If you’re restrained by manacles or some other device, the DC is based on the type of restraint. Against a Grapple, the escape DC is the Grappling creature’s Athletics DC.
Success You escape from the restraints or the Grapple, losing the grabbed or restrained condition.
Critical Success Per a success, and you can move up to 5 feet. If you move, this action gains the move trait.
Critical Failure You don’t escape, and you take a –2 circumstance penalty to Acrobatics checks to Escape until the end of your current turn.

Restrained
You’re tied up so you can barely move or a creature has you pinned. You have the immobile and flat-footed conditions, and you can’t do anything with the attack or manipulate traits except Break Grapple or Escape. The restrained condition overrides grabbed.

Let's put this into a context. My 3rd-level bard detective is tracking down a criminal gang. While investigating a warehouse that he suspects contains goods stolen by the gang, he is ambushed. Rather than killing the bard, they take his weapons, beat him almost unconscious, and tie him to a chair to interrogate him about who hired him. The gang's boss does not do the tying himself, because that would be beneath his dignity. He assigns a 1st-level minion to it, with Dex 18 and trained in Thievery, so passive Thievery +7, DC 17. My bard has Dex 16 and is trained in Acrobatics, so with +8 he has a 60% chance of a success or better.

Once bard bluffs in response to the interrogation, the boss departs and leaves only two minions to guard the bard. When his guards get comfortable, sheathing their weapons, he can spend one action for a 55% chance of escaping his rope bonds and then spend his next two actions casting Soothe heightened to 2nd level for 3d6 healing. If he fails, the guards notice, but he could calm their worries with deception saying that he was just uncomfortable and was trying to squirm to a more comfortable position. He trie again once the guards sit down again. On the second failure an angry guard will knock him unconscious. That gives him two 60% chances of success, for a combined 84% chance of escaping in the first 5 minutes.

That situation is too manageable. Only the guards are a challenge. If the GM wants to convert this scenario into a situation where the bard's companions have to rescue him, it will take a lot of GM fiat.

Paizo need to correct the problem that the bard can escape on a roll of 9 as they fix the problem that he can escape on a roll of 20.


Pathfinder Lost Omens, Rulebook Subscriber

How does it take a lot of GM fiat? The GM fiat needed is "they have manacles" or "the guards rotate occassionally to keep them alert." That isn't bloody huge.


GM fiat isn't really "big" or "small" in this context. The issue is that there are very explicit and hard rules that interact with other parts of the system, including a feat or two IIRC. Which means if you bring in GM fiat at all so that ropes and manacles work as they do IRL (they're a pretty reliable means of keeping someone restrained) then you're directly contradicting the RAW and RAI actions that say it works a certain way and they're actually very easy to circumvent. And whenever you, as a GM, have to tell a player that the extensive rules they were planning on using to escape actually don't work at all because you said so, it's going to be a bad time even if it doesn't result in an argument. You're going to be accused of railroading, hard.

Dramatically speaking, I'm OK with characters like that bard being able to escape their restraints. They'rec trained, they've got levels over their m the person who tied them up, it's a PC, is thematically appropriate for that to happen. Breaking out of ropes at all, ever, isn't my issue. Having to houserule it so NPCs work differently and don't even attempt the check is really annoying, but sm it's a lot easier to do that without creating conflict at the table.

My issue is more that, because of how repeated rolls and automatic successes on a 20 interact, anything that can be rapidly attempted using three or fewer actions will likely result in success in a couple minutes, no fatigue, regardless of skill. The rules don't have anything for making it impossible for a DC that's really high, just those obviously impossible for all characters like jumping to the moon. It's an inherent problem when trying to represent repeated attempts where a single success on a D20 accomplishes the entire task, it can't represent odds smaller than 5% and that 5% is only going to swell as you keep rapidly repeating the check.

It's an underlying math problem that I imagine might pop up in multiple undesired situations. Automatic successes on a 20 are fun and help mitigate that "might as well do nothing" feeling, especially during combat (where rapidly attempting and succeeding once a minute is very far from an issue).


Malk_Content wrote:
How does it take a lot of GM fiat? The GM fiat needed is "they have manacles" or "the guards rotate occassionally to keep them alert." That isn't bloody huge.

Okay, I wrote a scenario where the bad guys are likely to own a pair of manacles (cost 30 sp), but the original post by Starcatcher was about the good guys keeping a prisoner. They don't own manacles, because they never captured a prisoner before. They would have rope, because it comes with the climbing kit.

I looked up the price and DC of manacles, pages 185 and 186.
Standard manacles, 30 sp, Pick the Lock DC 22 and 3 successes, Escape DC unknown
Expert level 2 manacles, 150 sp, Pick the Lock DC 27 and 4 successes, Escape DC unknown
Master level 7 manacles, 2,250 sp, Pick the Lock DC 32 and 5 successes, Escape DC unknown

The prisoner picking the lock on manacles would be require thieves' tools, perhaps improvised thieves' tools that give a -2 penalty, and would be a manipulate action for which manacles impose a DC 5 flat check. Keep any pieces of wire away from the prisoner and he won't pick the lock.

The Escape action (page 144), however, says it works on manacles and says nothing about requiring tools or multiple successes. The GM can set the DC high, such as DC 37 for expert manacles, but that sitll gives a success on a natural 20. Which was Starcatcher's original point. The prisoner can free himself in 20 attempts on average, If he attempts one Escape action per turn, he won't fatigue himself. Thus, if he is left alone for 2 minutes, he will slip out of the manacles. And an adventuring party is too small to leave a party member behind to guard the prisoner all day. One hireling could be easily overpowered by an escaped prisoner. We have to go up to two hirelings. If the party just wants to keep the fellow bound for two hours while they explore the rest of the dungeon, that is too much trouble.


1 person marked this as a favorite.
Tridus wrote:
Matthew Downie wrote:
Repeatedly trying to escape from ropes seems like the sort of thing that would occur in gameplay often enough that it would be good if the rules could handle it. Something along the lines of "natural 20 counts as a +5/+10 bonus rather than auto-success, a certain number of failures / critical failures in a given hour / day fatigues you".

This would make more sense in general. Instead of auto-success, treat it as +10. If that's enough to make it a success, then it is. If not, then you simply can't do it. That'd also push anything that was already a success into crit success territory, so it's working on that front too.

That's fairly elegant, and provides an easily understandable situation of "no, even a nat 20 doesn't let you succeed on a DC 50 check with your +3 modifier" without any DM fiat required.

One advantage of making a nat 20 a +10 instead of an auto-success or turning success into crit is that it is easy to phrase as a rule: "Rolling a 20 on a d20 attack, check, or saving throw grants an automatic +10 bonus to the roll. Rolling a 1 on a d20 attack, check, or saving throw grants an automatic -10 penalty to the roll." The rule does not have to parse out success, failure, critical success, or critical failure. It just adds numbers to numbers.


Pathfinder Lost Omens, Rulebook Subscriber
Mathmuse wrote:
Malk_Content wrote:
How does it take a lot of GM fiat? The GM fiat needed is "they have manacles" or "the guards rotate occassionally to keep them alert." That isn't bloody huge.

Okay, I wrote a scenario where the bad guys are likely to own a pair of manacles (cost 30 sp), but the original post by Starcatcher was about the good guys keeping a prisoner. They don't own manacles, because they never captured a prisoner before. They would have rope, because it comes with the climbing kit.

Well people unprepared for taking prisoners should have to deal with not being able to take prisoners very well in my opinion.


2 people marked this as a favorite.
Malk_Content wrote:
Mathmuse wrote:
Malk_Content wrote:
How does it take a lot of GM fiat? The GM fiat needed is "they have manacles" or "the guards rotate occassionally to keep them alert." That isn't bloody huge.

Okay, I wrote a scenario where the bad guys are likely to own a pair of manacles (cost 30 sp), but the original post by Starcatcher was about the good guys keeping a prisoner. They don't own manacles, because they never captured a prisoner before. They would have rope, because it comes with the climbing kit.

Well people unprepared for taking prisoners should have to deal with not being able to take prisoners very well in my opinion.

So your argument is that tying people up just shouldn't work and that's fine. No one in world should expect people bound with ropes to be held for more than a couple minutes.

Okay.


1 person marked this as a favorite.
Malk_Content wrote:
Mathmuse wrote:
Malk_Content wrote:
How does it take a lot of GM fiat? The GM fiat needed is "they have manacles" or "the guards rotate occassionally to keep them alert." That isn't bloody huge.

Okay, I wrote a scenario where the bad guys are likely to own a pair of manacles (cost 30 sp), but the original post by Starcatcher was about the good guys keeping a prisoner. They don't own manacles, because they never captured a prisoner before. They would have rope, because it comes with the climbing kit.

Well people unprepared for taking prisoners should have to deal with not being able to take prisoners very well in my opinion.

But we are talking about Pathfinder 2nd Edition, which has stronger emphasis on good skills that are not primary skills. Being minimally prepared should make them ready for minimal challenges.

As a GM I like to throw new challenges at my players. Therefore, I give them an easy prisoner, a 0-level goblin warrior, Skills –2; Acrobatics +3, Athletics +3, Stealth +5. The 4th-level party has rope and Thievery skills for using rope. By apparent PF2 standards, they are prepared for an easy prisoner. By Rules Update 1.6, the Dex 18 rogue has passive Thievery DC 19 (+5 expert proficiency, +4 Dex). Oops, the goblin can escape on a roll of 16 or better.

If the party had brought standard manacles, the goblin can escape on a natural 20.

Malk_Content's threshold was "take prisoners very well," but without a guard or a cage, the party cannot hold prisoners for longer than 2 minutes.

Perhaps auto-success is a fundamental limitation in Pathfinder 2nd Edition. Or perhaps it is simply a flaw in the Escape action. What if Escape said that escaping from rope bonds takes the amount of time the captor took to tie the bonds (maximum ten minutes) instead of one action? (A Quick Escape feat could shorten that.) Then the goblin would need 40 minutes on average to escape. That would be enough time to check the next room in the dungeon.


Nice idea, Mathmuse. I like it.


I think it would be easier to make it just a ten minute activity, to match up with trap disabling. Much simpler, GM doesn't need to think hard about it, and it doesn't make sense that a knot needs to take ten minutes to tie just to hold someone in place for an equal amount of time. I'd laugh at anyone taking an entire minutev to tie such a knot, it should barely take more than a round.


Pathfinder Lost Omens, Rulebook Subscriber

I think that is perfectly reasonable. If we go by movie/tv time that seems about right for the amount screen time that happens for someone to break out of rope bonds if left uninterrupted. I just don't see the problem with "you've only tied them up (and not even bothered to use aid another to push the number up) so someone might need to watch them." Tying up at that point is still useful, it lets a single watch man control many prisoners for example (with a ranged weapon pretty much anyone within its optimal increment.) That is pretty great.


Malk_Content wrote:
I think that is perfectly reasonable. If we go by movie/tv time that seems about right for the amount screen time that happens for someone to break out of rope bonds if left uninterrupted. I just don't see the problem with "you've only tied them up (and not even bothered to use aid another to push the number up) so someone might need to watch them." Tying up at that point is still useful, it lets a single watch man control many prisoners for example (with a ranged weapon pretty much anyone within its optimal increment.) That is pretty great.

For the cinematic argument, are we taking about the time for the hero to escape (a few minutes), for the person the hero rescues (untied by hero after many hours), or for the innocent guard the hero had to subdue (many hours)?

Does Aid work on a passive DC? Let me check page 307.

[[R]] AID
Trigger An ally is about to use an action, activity, free action, or reaction that requires a skill check.
Requirements The ally is willing to accept ...

Nope, passive Thievery DCs don't provide a proper trigger for Aid.

On the other hand, the Aid suggestion inspires another solution. What if Paizo puts a Tie Bonds skill under Thievery that makes a secret-roll Thievery check to tie up the prisoner? Success against the prisoner's active Acrobatics DC would give the minumum time before they could try Escape. Failure would be any time, with critical failure a guarantee of successful escape. Success would be 1 hour with critical success 1 day.


3 people marked this as a favorite.
Mathmuse wrote:
Malk_Content wrote:
I think that is perfectly reasonable. If we go by movie/tv time that seems about right for the amount screen time that happens for someone to break out of rope bonds if left uninterrupted. I just don't see the problem with "you've only tied them up (and not even bothered to use aid another to push the number up) so someone might need to watch them." Tying up at that point is still useful, it lets a single watch man control many prisoners for example (with a ranged weapon pretty much anyone within its optimal increment.) That is pretty great.

For the cinematic argument, are we taking about the time for the hero to escape (a few minutes), for the person the hero rescues (untied by hero after many hours), or for the innocent guard the hero had to subdue (many hours)?

Does Aid work on a passive DC? Let me check page 307.

[[R]] AID
Trigger An ally is about to use an action, activity, free action, or reaction that requires a skill check.
Requirements The ally is willing to accept ...

Nope, passive Thievery DCs don't provide a proper trigger for Aid.

On the other hand, the Aid suggestion inspires another solution. What if Paizo puts a Tie Bonds skill under Thievery that makes a secret-roll Thievery check to tie up the prisoner? Success against the prisoner's active Acrobatics DC would give the minumum time before they could try Escape. Failure would be any time, with critical failure a guarantee of successful escape. Success would be 1 hour with critical success 1 day.

Except I don't want that either. I want the skilled escape artist to be able to break out easily and quickly with a successful check. I just don't want it to be automatic on the 20 you'll eventually get.

There's all types of hacks people are proposing to get around the basic problem while leaving the autosuccess rule intact. That rule is the problem.


1 person marked this as a favorite.
thejeff wrote:

Except I don't want that either.

There's all types of hacks people are proposing to get around the basic problem while leaving the autosuccess rule intact. That rule is the problem.

As I said over in a blog thread, the design decision to make all checks have all four degrees of success possible at all times for all characters, regardless of proficiency was one that was interesting and done to show that it is possible, but not fun.

Community / Forums / Archive / Pathfinder / Playtests & Prerelease Discussions / Pathfinder Playtest / Player Rules / Skills, Feats, Equipment & Spells / Automatic success means you can't ever leave a prisoner alone All Messageboards

Want to post a reply? Sign in.
Recent threads in Skills, Feats, Equipment & Spells