D&D 5E (2014) Can you retry a failed skill check? How long?

[BookTenTiger]

This thread hasn’t been posted in for 11 years until now.

It's okay, there's nothing in the rulebook that says you can't retry a skill 11 years later!

 

[BookTenTiger]

On the topic, if a skill can be tried over and over again until the PC succeeds, then I usually switch up the success parameters.

I might have the check represent how much time it takes to succeed, or how quietly or skillfully they complete the task at. So if a thief is trying to pick a lock, and they can just keep rolling into infinity, I'll just have them roll once. A high roll might mean it was done quickly and quietly, and a low roll might mean it was a loud and dirty job.

 

[James Gasik]

I have never had a thief try twice in one round using their bonus action. They mostly like to attack and try to open a door or lock. Out of combat, I tend to rule they cannot use their bonus action/ cunning action.

I have not really tried the 2024 rules with getting advantage opening locks using tools and proficiency.

Ok, but a part of my point is that the ability to perform a task as a bonus action should imply that it takes a very short amount of time to make the check, even outside of combat, right? So if, as in the original post (so long, long ago), the DM wants to say "ok, the check takes 1 minute, a retry takes 5 minutes" and so on, what does that mean for a Thief?

Because it doesn't really make sense that a Thief can do something in combat along with a normal action and movement inside of 6 seconds, but out of combat, suddenly it takes 10x that amount of time or longer.

 

[Horwath]

I have never had a thief try twice in one round using their bonus action. They mostly like to attack and try to open a door or lock. Out of combat, I tend to rule they cannot use their bonus action/ cunning action.

I have not really tried the 2024 rules with getting advantage opening locks using tools and proficiency.

I would say that if ability gives you that you can do something as Bonus action instead of Action, I would say that in extended action, you can do it 2× or 3× faster.

I.E: you need 20 successes to break open a safe, someone can make 1× roll per turn, but a thief would make 2× or maybe 3× if I value 1 Action = 2 Bonus action.

 

[aco175]

Ok, but a part of my point is that the ability to perform a task as a bonus action should imply that it takes a very short amount of time to make the check, even outside of combat, right? So if, as in the original post (so long, long ago), the DM wants to say "ok, the check takes 1 minute, a retry takes 5 minutes" and so on, what does that mean for a Thief?

Because it doesn't really make sense that a Thief can do something in combat along with a normal action and movement inside of 6 seconds, but out of combat, suddenly it takes 10x that amount of time or longer.

I would say that if ability gives you that you can do something as Bonus action instead of Action, I would say that in extended action, you can do it 2× or 3× faster.

I.E: you need 20 successes to break open a safe, someone can make 1× roll per turn, but a thief would make 2× or maybe 3× if I value 1 Action = 2 Bonus action.

Some of all this might be a holdover from earlier editions where in combat time was different than out of combat time.

There is also some input from the failed check. If a PC misses the check by 5 on a standard lock I may say that it appears harder than first thought and it takes another minute to open it. They failed, but given a few more rolls, they would get it whether in 6,12, or 18 more seconds- or half that if they are a thief and can use a bonus action as well. Again, out of combat time does not matter as much- at least to me.

 

[Li Shenron]

I wonder if anything has changed over the last 11 years for anyone?

Nothing has changed over the last 51 years.

The problem I have with "one and one" is it doesn't make any sense.

There is hardly anything RAW that makes sense after a large enough amount of scrutiny and reality checks. A game rule is supposed to be usable, not perfect, and realistic only up to a certain level, but as players keep looking purposefully at counter-examples to prove their point, no level is realistic enough for them.

My front door warps a bit with the weather. This means sometimes I try to open it, and it sticks, forcing me to try a few times. Imagine if I lived in a "one and done" world! I'd be calling a locksmith every time I tried to open my door and failed!

What benefits to a game would rolling multiple times bring in such a case? Do you want a chance that you have to call the locksmith? If yes, then roll, once, with numbers chosen so that more or less the probability of failure suits your expectations. If no, don't roll, just narrate that this time you had to try many times.

I like the old 3E style approach in this regard. If you are not under duress and there are not dangerous consequences of note, you can "take 10", so the outcome of your roll is 10. If you have lots of time, and no consequences for failure, then you can "take 20", so the outcome of your roll is 20, but the cost is that you take 20x the time to do the action very carefully. In the case of say picking a lock, taking 20 may not be permissible if the lock could jam, or if your lockpicks could break. Up to the DM, if you can take 10 or 20 in a given situation.

I do not fully endorse "Take 10" or "passive skills" rules, they have the byproduct of making it not possible to do your best, and there are plenty of complications due to how different skills can work in different circumstances, so to me these rules don't improve the game but only complicate it with unnecessary discussions. The original 3e "Take 20" rule was more solid, as is something like "Reliable Talent" which sets a minimum default after rolling.

In general, I would be inclined to be lenient and allow characters to retry if they fail, but there could be consequences of intermediate failures.

Maybe a usable compromise could be, that retries are possible but there will be a cost for failure. But this has to be a real cost, not a potential cost that maybe it won't matter at all. This is why I am skeptic of house rules on adding more time spent to each attempts, because sometimes times matters and some other times it really doesn't.

Ok, but a part of my point is that the ability to perform a task as a bonus action should imply that it takes a very short amount of time to make the check, even outside of combat, right? So if, as in the original post (so long, long ago), the DM wants to say "ok, the check takes 1 minute, a retry takes 5 minutes" and so on, what does that mean for a Thief?

Because it doesn't really make sense that a Thief can do something in combat along with a normal action and movement inside of 6 seconds, but out of combat, suddenly it takes 10x that amount of time or longer.

The fault is allowing the Thief to do something unreasonable in combat in the first place. But apparently we want the game to be both realistic and unrealistic in the same time, because out of combat we want the "lord of the rings" feeling of taking ages to do everything, and in combat we want the "mission impossible" feeling of doing many over the top stuff at once. The idea that a PC can disarm a trap or pick a lock in one turn during combat is because players want a WOW! factor, but then some interprete this as a sort of "law of nature" that to disarm a trap always takes 6 seconds by default in the world.

If you want more consistency, you can make different decisions depending on how complex the challenge is:

a) the trap/lock can be handled with a combat action and a dice roll (because under pressure), while out of combat success is automatic
b) the trap/lock cannot be handled under pressure (automatic failure), and required a dice roll out of combat

 

[BJK Games]

I generally go for 1 attempt and the party can "help" with whatever that game's mechanics are. If they fail, it has to be a completely different kind of attempt on the skill check. Using Locks as an example, if they all fail to pick it, I'm not gonna allow it to be picked, but if they want to smash the chest and chance hurting whats in side or breaking the lock and chancing being heard by the monsters next door or whatever, I let that happen.

 

[Emerikol]

Also just want to say something as often as I can because it drives me nuts. There is no crit fail system in D&D, not this edition, not any edition, except as optional rules in one DMG sidebar and a very old Dragon Magazine article. In fact, in 5e like most editions a '1' on the d20 is only an auto miss in combat, with a skill check if the DC is low enough and your modifier is high enough a '1' can succeed but again in that case why bother asking for a check, just tell the expert he does his thing and move the story forward.

I think I started playing with crits in 1e and used them through 4e. I haven't played 5e to date. I dislike the swingyness of the d20 so the crit table might have a large section that just says: nothing further happens.

 

[AlViking]

Depends on the check and why I'm asking for it. If I think it will happen given enough time, if I call for a roll it will just be how long it takes. For something like sleight of hand to open a lock, I did a bit of searching and found that locksmiths will say that sometimes it's super easy to open a lock (although likely not the 6 seconds it takes in D&D) but other times it can take up to 20 minutes. So if the initial check misses by 10 or less they can try again. However, this second attempt means that they're familiar enough with this type of lock and they have the skill but it's going to take a while. So my house rule is you can try again but roll 2 D20 and that's the number of minutes it's going to take.

 

[James Gasik]

There is hardly anything RAW that makes sense after a large enough amount of scrutiny and reality checks. A game rule is supposed to be usable, not perfect, and realistic only up to a certain level, but as players keep looking purposefully at counter-examples to prove their point, no level is realistic enough for them.

Well sure, I get that, but a lot of posts in this thread are made by people who want to make rulings based on "realism" (or verisimilitude, at the very least). But when you produce counterpoints like lockpicking videos on YouTube or real life examples of how "one and done" checks are kind of ridiculous, then suddenly "it's a game, it has to have stakes!".

When your counterpoints are based in the rules, they reject the rules for being unrealistic.

When your counterpoints are based in reality, well "it's a game, not real life".

Kind of a double standard.

-

I mean, I once, long ago, had an argument with a DM about falling damage (especially how he felt that "terminal velocity" shouldn't exist...). The 2e PHB was actually kind enough to mention that people can survive falls from insane heights, and a little research on the internet put names to these events, from crippling (Vesna Vulovic's 6.31 mile drop) to Davide Capello falling 100 feet in a bridge collapse nearly unscathed (save for some occasional back and shoulder pain).

We went round and round about this for a bit, until he finally admitted he didn't care what was possible or probably in real life- he wanted falling to be scary in his games.

Ironically, a few weeks later, we managed to immobilize his big dragon boss while it was mid-air chasing our skyship. "Well, scratch one dragon." I said, referring to his falling damage rules.

"What? He's got enough hit points to survive that easily!" He replied.

"Oh yeah? How much damage would I take if I jumped off our ship right now?"

He sputtered for a bit then lamely replied, "That's different. You're a player!"