Bringing Complex Skill Checks to 5e

[Calimehter]

DnD 5e leaves a lot of blanks to fill for the DM when it comes to resolving out-of-combat ability checks, and the task resolution process is considerably less spelled out for most any task except combat.

To help fill in the blanks a bit and make such tasks more interesting, one thought I'm considering for a campaign I'm working on is importing the idea of the Complex Skill Checks from the 3.5 Unearthed Arcana book.

Let me know your thoughts (and any example situations if you care to share :smallsmile: )

THE RULE:

Ability checks that take more than one round or require retries or group checks will be resolved by Complex Checks whenever appropriate. A Complex Check basically involves rolling multiple checks, where a number of successeful rolls are required to complete a task, and you must roll these successes before you roll THREE failures. Unless stated otherwise for particularly simple or particularly complex tasks, you will need to roll 3 successes before you roll 3 failures.

PROS OF THE RULE (i.e. reasons for bothering in the first place):

- Reduces some of the variability that allows poorly skilled individuals to beat highly skilled individuals on a single die roll. With single checks, there is a great degree of variability. The DMG relates the example of the STR8 character rolling high and forcing open a stuck door that the STR18 character could not just because of a single lucky die roll. Contrast that situation to the much more detailed set of rules and multiple die rolls involved in combat. In combat, a higher level swordsman will beat a lower level swordsman the vast majority of the time, due to the fact that combat isn’t a single d20 “Strength(Swordsmanship)” opposed ability check. They are instead rolling multiple attack dice over multiple combat rounds against their respective AC and HP values. Complex skill checks bring some of that level of detail and reduced chance of an “upset” result to ability checks.

- Allows ‘Degrees of Success and Failure’ and the possibility of retries to be more easily quantified. Many skill checks will just be ‘success’ or ‘failure’ with nothing in-between . . but Complex Checks allow the DM to set up some rules for what happens if you succeed with 2 failures as opposed to 0 failures, especially for skills where the amount of time a check takes and whether re-tries are allowed are not clearly defined in the basic rules. For instance, the allowance of retry attempts could be based off of the number of successes one had before failing.

- Allows for some drama and possible changes in the course of action over the course of a given skill check. Consider the classic case of disabling a trap. Rolling one die means that you either succeed, or you fail . . . with no option to make decisions during the course of action. Rolling multiple dice means you can ‘sweat’ a bit as you gradually work your way through the mechanisms of a trap being disabled, and get a good feel for ‘how it is going’. Having one or two failures right off the bat may result in a player deciding to abandon the attempt altogether (rather than risk triggering the trap) and/or decide to spend resources (spells or other limited-resource advantages or bonuses) they might not have otherwise spent if the check was being decided by a single die roll. Likewise, a player who is roleplaying a Charisma(Deception) check may change tactics altogether partway through a conversation if the first few rolls do not ‘look good’ (i.e. the mark is not buying the story) rather than risk an outright failure.

- Allows for multiple ability and proficiency scores to be used for a given task. While many tasks will only require a check against a single ability and proficiency, there are also many tasks where more than one attribute or proficiency would be helpful to have. Consider the ‘book’ example of using a Constitution(Athletics) check to swim a long distance in place of a Strength(Athletics) check . . . this is all fine as is, but shouldn’t a stronger character *still* have a better chance of succeeding at such a task, even if Constitution is also very relevant to resolving it? With single checks, you have to pick one and go, but with Complex checks you can include both. A Complex check for swimming a long distance could be resolved by alternating Strength(Athletics) and Constitution(Athletics).

CONS OF THE RULE (or when not to use it):

- Takes more game time. In theory, the DM shouldn’t even be calling for skill checks at all unless failure is important (i.e. dramatic), so the time spent on Complex Checks should not be considered ‘wasted’ if the rolls add some drama and granularity to a situation that deserves it . . . but not every situation deserves it. Over-use could bog down gameplay in much the same way that overly-drawn out combats can.

- Requires more DM work. Planning out a particular skill check and mapping out the possible outcomes does not require as much work as setting up a combat encounter (random or otherwise) . . . but it is still more involved than just calling for a single die roll with a binary result, and may be more work than the DM wants to put into a check, especially an unplanned one.

- Reduced chance of an “upset” result or succeeding against the odds. Using complex checks, the odds of success go down much faster than usual for a difficult task than they do for a single check. A character who needs an ’11’ or higher has a 50:50 chance with both kinds of checks. However, a character who needs to roll a ’16’ to succeed will make it 25% of the time on a single check, but will only make it 11% of the time on a complex check! This is a feature really, not a bug, but it is not always appropriate to each situation, and the smaller chance of success on difficult task could actually reduce dramatic tension if the outcome seems nearly pre-determined. If luck should play a big role in something or it is something that is a ‘snap’ ability check (playing a yahtzee-style dice game against someone, or grasping at a waving rope before it gets out of reach), a complex check is probably not appropriate. Likewise, if the DM wants to give the “needs a 20” character a shot at something, it is a lot more likely to happen on a single roll.

AN EXAMPLE:

A Complex check for breaking free of the regular PHB manacles is a DC 20 Strength check (no proficiency applies), with 3 successes needed before 3 failures are rolled. If in combat, keep rolling dice on your turn until you have all 3 successes or get a single failure.
- Each Failure will add one round to the amount of time that the attempt takes. In combat, you will stop rolling and can continue the check (if desired) on your next turn.
- Getting 3 failures means that the complex check has failed, and will need to be restarted (i.e. resetting the number of successes and failures back to zero)
- Rolling 3 failures without a success means that a retry cannot be attempted, you are simply not going to break those manacles without a significant change in the situation.

Note that when compared with using a single check, this very much reduces the odds of a character actually breaking the manacles, especially in a timely fashion. A STR 18 character has a 1 in 4 chance of breaking them in a single round using a single check, but the complex check reduces that chance of that single round break (i.e. 3 straight successes w/o a failure of any kind) to 1 in 64! Their odds of succeeding at all over time are only 11% for any given check, and their odds of being denied a retry (3 straight failures) are high at 42%. A STR 10 character’s chances would go from 5% (and really 100% if the DM is allowing unlimited check attempts) down to virtually nothing.

Using this rule makes DC20 manacles very hard to escape from, rather than a temporary hinderance that can nearly always be overcome, which is a bit more like the performance of the real-life equivalents of these devices. This could be good or bad for the PCs depending on who the manacles have been applied to! :smallbiggrin: It is also worth noting that failure to break the manacles does not eliminate the option to try other methods (the Dex option to escape them - though this has tough odds too - or having a teammate with free hands try to destroy them with damage or pick the lock).

A gaming group that wanted to keep manacles as more of a 'marginal' item like they are with single checks by keeping the odds closer to the single-check scenario . . . can just use single checks if desired. They can also reduce the number of successes needed to, say, 2 or even 1 (it is a fairly non-complex task after all) or even swap the DC down to something like 18.

[Calimehter]

I guess I nailed it in one. :smallcool:

[strangebloke]

Its a totally fine way to do things some of the time, and it should probably be in the DMG as a variant rule.

Really, I use this or something similar every time a task takes more than a few minutes, and its usually worked pretty well. A difference between what I do and what you have written out here is that I usually will require multiple different checks.

EX: Tracking an enemy across rough terrain.
-INT(Investigation): find the enemy's last known location.
-WIS(Survival)x3

If the party fails the first investigation check, they'll get the info eventually due to other NPC action. However, the total number of successes out of the total here will lead to a different result. 0-1 success means that the enemy's general location is known, but that the enemy has the chance to get the drop on them. 2 successes means that the enemy's exact location is known. 3 successes mean that the party has a chance to get the drop on the enemy, and can make a stealth check to sneak up on him. 4 successes mean that they make the stealth check with advantage.

[EggKookoo]

THE RULE:

Ability checks that take more than one round or require retries or group checks will be resolved by Complex Checks whenever appropriate. A Complex Check basically involves rolling multiple checks, where a number of successeful rolls are required to complete a task, and you must roll these successes before you roll THREE failures. Unless stated otherwise for particularly simple or particularly complex tasks, you will need to roll 3 successes before you roll 3 failures.

I've actually been doing this for a while now. I think I got the idea from how death saves work. I love it, and my players like it too. It works very well to create escalating tension, and mitigate the feeling that a single bad roll will ruin your day.

[Hadoken]

My DM uses this rule in our game, and I like it a lot. I like that it encourages creative application of skills to particular situations. 5e really needs to dedicate some dev time to enhancing the gameplay around skills.

[Calimehter]

Good to hear others are doing it too, and thank you for the tracking example!

For those who have been running it already: Do you see a lot of these checks being made with advantage? For the tracking example, the odds of success go way up on an INT check if anyone can help the "primary' tracker?

[Helldin87]

I have used a variant of this at my table. The rules are put in place for complicated non-combat encounters. Players roll initiative and perform ability checks to influence the encounter against a DC ice predetermined. They must succeed in 3 before failing 3 and each player must roll a check different than the player before. This means not every player can roll deception but someone could opt to use deception followed by intimidation and then deception again.

It’s added to the RP as well since I ask them to explain their check and role play the actions.

Seems to result in fun!

[strangebloke]

Good to hear others are doing it too, and thank you for the tracking example!

For those who have been running it already: Do you see a lot of these checks being made with advantage? For the tracking example, the odds of success go way up on an INT check if anyone can help the "primary' tracker?

I often used 'proficiency as permission' in a sense.

A history/religion check related to a thing you never would have studied (because you have no proficiency and it's not something you were raised around.) Would be impossible or made with disadvantage. I also rule that if you can't make the check without disadvantage, you can't help with the check.

Then there are checks where 'help' makes no sense. Stealth, persuasion, insight (though you can roll seperately) and Perception all come to mind.

But yeah, generally PC's do have advantage. That's fine; the DC's are pretty hard to consistently hit otherwise.

In an example like the one above, the DC for tracking the enemy is based off of his Dex/Wis survival/stealth ability, but without Avatar because there's no way to help with covering tracks; two people are inherently more trackable.

[Pelle]

AN EXAMPLE:

A Complex check for breaking free of the regular PHB manacles is a DC 20 Strength check (no proficiency applies), with 3 successes needed before 3 failures are rolled. If in combat, keep rolling dice on your turn until you have all 3 successes or get a single failure.
- Each Failure will add one round to the amount of time that the attempt takes. In combat, you will stop rolling and can continue the check (if desired) on your next turn.
- Getting 3 failures means that the complex check has failed, and will need to be restarted (i.e. resetting the number of successes and failures back to zero)
- Rolling 3 failures without a success means that a retry cannot be attempted, you are simply not going to break those manacles without a significant change in the situation.


This is just pointless rolling. For "skill challenges" to make sense, the situation needs to change sufficiently so that the player can keep making new decisions. Here it it just continuing doing the same for 5 rolls. If it is one decision, just resolve it with one roll, and set the DC according to the probability of success you want.

I like the idea of skill challenges and similar, they have a lot in common with chase rules. However, if they are suited for the task they usually happen organically in play anyways. Trying to impose the mechanics feels weird, and you need to buy-in to using an abstract mechanic instead of keeping your agency to affect the situation. In the above example, what does 1 failure mean? Why can you keep trying after 1 failure, but not 3? In effect, the characters actions need to be fitted to the mechanics, instead of the opposite.

So my opinion is that these types of checks are cool if the situation is suited for it, like if multiple steps are required to achive something and there is limited time. But if this is the case though, this type of check will happen organically anyways, so it is not really needed. The biggest benefit I see with it, is that the players are explicitly made aware of the required successes and time/failures available, making the resolution of the complex task more dramatic since they know the stakes ooc.

[EggKookoo]

This is just pointless rolling. For "skill challenges" to make sense, the situation needs to change sufficiently so that the player can keep making new decisions. Here it it just continuing doing the same for 5 rolls. If it is one decision, just resolve it with one roll, and set the DC according to the probability of success you want.

And yet my players like them. Next session I'll tell them to stop having fun.


I like the idea of skill challenges and similar, they have a lot in common with chase rules. However, if they are suited for the task they usually happen organically in play anyways. Trying to impose the mechanics feels weird, and you need to buy-in to using an abstract mechanic instead of keeping your agency to affect the situation. In the above example, what does 1 failure mean? Why can you keep trying after 1 failure, but not 3? In effect, the characters actions need to be fitted to the mechanics, instead of the opposite.

Arbitrary limits and goals are built into the game all over the place. Three is most often used because it's the smallest number that can reasonably be considered a sequence. And having an odd numbered sequence helps remove ambiguity. I've also done things where you need three successes eventually and the failures don't count, but usually when there's another constraint like you're in combat and potentially taking damage the whole time. You can mix and match.

Usually I phrase it so there's some time limit, or that something will happen if they don't get the successful checks they need. That provides a context for why they can't just keep trying indefinitely.


So my opinion is that these types of checks are cool if the situation is suited for it, like if multiple steps are required to achive something and there is limited time. But if this is the case though, this type of check will happen organically anyways, so it is not really needed. The biggest benefit I see with it, is that the players are explicitly made aware of the required successes and time/failures available, making the resolution of the complex task more dramatic since they know the stakes ooc.

They're good for time-based actions. If the room is flooding, you can use this to create tension in getting the door lock picked in time. It's also really good for social challenges (think rap battle). Any time the PCs need to do something that takes time and could be interpreted as having degrees of progress, or could have a possible two-steps-forward-one-step-back progression, this is a good way to do it.

I don't use it all the time (very rarely in comparison to how often the players roll overall) but it's always seen as an interesting minigame. Like, oh, he's bringing out the extended checks, something important is happening. It also does break up the mechanics a bit so not everything is just a single roll. It creates the feel of a task that is a little more drawn out and considered, rather than a quick action like you might see in combat. The players all get invested and can react as the successes or failures pile up. Depending on the situation, I let different players make different rolls in the sequence, also suggesting a team effort in a way that feels more involving than just using the Help action to provide advantage.

[Pelle]

And yet my players like them. Next session I'll tell them to stop having fun.


If it's fun, use it. But why is it more fun to roll 5 times instead of 1 to break the manacles above? It doesn't really add anything in that situation. Here I would challenge you to try make it more fun with just one roll. For example, I think you can get the same drama with letting the players know the DC instead, with 5 times less fiddling.



Arbitrary limits and goals are built into the game all over the place. Three is most often used because it's the smallest number that can reasonably be considered a sequence. And having an odd numbered sequence helps remove ambiguity. I've also done things where you need three successes eventually and the failures don't count, but usually when there's another constraint like you're in combat and potentially taking damage the whole time. You can mix and match.

Usually I phrase it so there's some time limit, or that something will happen if they don't get the successful checks they need. That provides a context for why they can't just keep trying indefinitely.


Sounds like you are using it kind of organically, which was what I wanted to encourage. If you announce "now we use the complex skill rules", then every action taken by the player and every failure needs to be adapted to work within the 3 success/failure rule. That needlessy constrains the actions players can take, or make the decisions made meaningless. Instead of establishing the situation and use the mechanichs to resolve it, you establish the mechanic and need to form the situation around the results. This is essentially a more storygamey type of approach, and it depends on taste if that is good or not.



They're good for time-based actions. If the room is flooding, you can use this to create tension in getting the door lock picked in time. It's also really good for social challenges (think rap battle). Any time the PCs need to do something that takes time and could be interpreted as having degrees of progress, or could have a possible two-steps-forward-one-step-back progression, this is a good way to do it.


Agreed about the situations, but in these cases, the normal skill checks works out the same anyways, you don't need a special complex skill rule. "the room will be flooded in 6 rounds", that's enough pressure. Or "the room will be flooded in 6 rounds, you must first climb up to that ledge, then pick that lock, and then push away the big rock". Same effect, but happens organically and feels much better.



I don't use it all the time (very rarely in comparison to how often the players roll overall) but it's always seen as an interesting minigame. Like, oh, he's bringing out the extended checks, something important is happening. It also does break up the mechanics a bit so not everything is just a single roll. It creates the feel of a task that is a little more drawn out and considered, rather than a quick action like you might see in combat. The players all get invested and can react as the successes or failures pile up. Depending on the situation, I let different players make different rolls in the sequence, also suggesting a team effort in a way that feels more involving than just using the Help action to provide advantage.

I do think skill crawl game structures are interesting, it's just that they feel very gamey and artificially constraining if the situation doesn't fit well. When it does fit, being explicit about the structure is what is the added benefit of it IMO.

[EggKookoo]

If it's fun, use it. But why is it more fun to roll 5 times instead of 1 to break the manacles above? It doesn't really add anything in that situation. Here I would challenge you to try make it more fun with just one roll. For example, I think you can get the same drama with letting the players know the DC instead, with 5 times less fiddling.

It depends. I would only use it if there was some way of visualizing progress. So if I felt like you needed to use it to break manacles, I would probably describe the manacles as having some gradual failure structure. Not easy to do with manacles I admit -- perhaps they're connected with multiple chains and you're snapping them one at a time. Maybe more easy to visualize with a conventional rope, where you strain it and the braids snap one after the other.

I would say five rolls is probably too much. You can convey the same thing with three, and if that feels too easy the up the DC. In fact you can have rising or falling DCs as part of this system as well.


Sounds like you are using it kind of organically, which was what I wanted to encourage. If you announce "now we use the complex skill rules", then every action taken by the player and every failure needs to be adapted to work within the 3 success/failure rule. That needlessy constrains the actions players can take, or make the decisions made meaningless. Instead of establishing the situation and use the mechanichs to resolve it, you establish the mechanic and need to form the situation around the results. This is essentially a more storygamey type of approach, and it depends on taste if that is good or not.

Yes, I agree that you should only be using this when it feel naturally appropriate to the task being attempted. It's just another tool in your toolbox.


Agreed about the situations, but in these cases, the normal skill checks works out the same anyways, you don't need a special complex skill rule. "the room will be flooded in 6 rounds", that's enough pressure. Or "the room will be flooded in 6 rounds, you must first climb up to that ledge, then pick that lock, and then push away the big rock". Same effect, but happens organically and feels much better.

Yeah, the flooded room might not be the best example, as it's going to flood pretty much at a set rate which you can describe as rounds. A better example might be something where the constraint itself can vary as well. For example, I ran some new-ish players through Lost Mine of Phandelver a while back. I used this mechanic to handle their negotiation with Venomfang. The players weren't up for acting out the conversation but they formed the logic of their argument. I had them pick a "spokesman" and roll Persuasion every time they wanted to make a point or statement, and said they'd convince the dragon if they got three success before they got three failures. It worked out great, they even got to 2-and-2 and panic started to set in. I reminded them that another character can Help the spokesman, which gave them advantage on their last roll, which succeeded.

These were new players to D&D (and TTRPGs in general) and I honestly think that experience more than any other sold them on the unique sense of satisfaction you get from the game.

[Pelle]

It depends. I would only use it if there was some way of visualizing progress. So if I felt like you needed to use it to break manacles, I would probably describe the manacles as having some gradual failure structure. Not easy to do with manacles I admit -- perhaps they're connected with multiple chains and you're snapping them one at a time. Maybe more easy to visualize with a conventional rope, where you strain it and the braids snap one after the other.

I would say five rolls is probably too much. You can convey the same thing with three, and if that feels too easy the up the DC. In fact you can have rising or falling DCs as part of this system as well.


5 times is because you keep rolling until you have 3 successer or 3 failures. So it's minimum 3, maximum 5. The thing is, if you can visualize it, like three links that need to be broken, taking an action each, you don't need the complex skill rules at all, you can just use conventional checks one at a time.



These were new players to D&D (and TTRPGs in general) and I honestly think that experience more than any other sold them on the unique sense of satisfaction you get from the game.

I think this type of storygaming can be very fun, but it's nice to be aware that you are doing it.

[EggKookoo]

5 times is because you keep rolling until you have 3 successer or 3 failures. So it's minimum 3, maximum 5. The thing is, if you can visualize it, like three links that need to be broken, taking an action each, you don't need the complex skill rules at all, you can just use conventional checks one at a time.

Right, it works best if the opposing factor is also variable or at least unpredictable. Other examples might be running across a field while an opposing army is raining flaming arrows down on you (you can't withstand a single shot). Or you've infiltrated the king's banquet and you need to investigate the guests but avoid being exposed. Or you're sneaking through a crowd while trying to avoid the constables, who aren't necessarily searching for you but are on the lookout for "dangerous adventurer types." Maybe even withstanding a debilitating noxious gas bomb and trying to outlast its impact until the breeze clears it from the area.

Things that have fuzzy chaotic components that can incrementally increase or decrease your chances of success, but with a clear indicator of success itself.

[Man_Over_Game]

I use a similar system, utilizing a sum vs. DC rolling.

Utilizing a sum of skill rolls means it's easier for allies to contribute, better shows the contrast between skills vs. unskilled players, and allows me to modify things on the fly.

------------

The Bard is manacled and wants to break free. He has his friendly Barbarian with him. The Bard tries to use brute strength (Strength: Athletics), but he doesn't have the leverage he needs, so his roll has disadvantage. He rolled a 6.

Total: 6

The Barbarian uses a crowbar and briefly rages to pump his strength to ridiculous values. Crowbar adds +5 to his Athletic roll, and he has advantage from Rage. Barbarian rolls 30.

Total: 36

The Bard feels a weakpoint in the manacles with his high Investigation skill. Rolls 19.

Total: 54

Halfway into the second turn, a guard shows up to deal with the escaped prisoners. The barbarian fights him off while the Bard improvises a solution for the remaining 6 points left to break the manacles.

------------

What really helps is this formula:

Average Roll per turn = 10 + (Proficiency x 2)

With this, you can estimate what the average roll will be for a character who's making a skill check they're talented in. Use that formula x the number of actions you want the players to take, so that you can comfortably come up with various different scenarios for the players to deal with. In my prisoner example, the Bard was supposed to be manacled for multiple turns, but quick thinking by the Barbarian allowed the manacles to almost nearly be dismantled in a short time, and so the upcoming encounters (balanced to be against a barbarian and a crippled caster) should be a breeze.

In social challenges, skills used by the same person after the first are halved, but NPCs can contribute (if they're willing and have something to contribute to the challenge). For physical challenges, the same person using the same skill is halved after the first attempt.

Applicable tools or clues add a straight +5 to the final result of a player's roll. Some sort of breakthrough (like translating the language on a book) resets all penalties for a new challenge.

If the players have a lot of time on their hands, double the goal requirement.

----------

Your example is quite similar, but the main versatility and benefit of yours is to create scaling rewards to separate talented vs. untalented characters, but doesn't provide much for reacting on the players' part. It's adding additional rolls and mechanics, without adding much game. Consider allowing (or even rewarding) players to use other skill rolls, and punishing repeated skill rolls (so you can't just spam Athletics to break manacles effectively) to allow more interaction and decision making from your players.

[Calimehter]

Good points about making sure the die rolls have a point to them. "Boggy" complex checks could get just as bad as "boggy" drawn out combats for slowing the game down.

My goals with the manacles example were to make them less "random" to break at all and to set a barrier to endless retries (to avoid the STR 10 character from just auto succeeding), and also to provide a method for determining how long it would take to do in a time sensitive situation.

I agree, though, that just trying to burst them by brute force straining is not exactly a "complex" action, and I could see swapping that out for a single STR check, and just adding a bit where failing by 5 or more would result in no retries allowed (sometimes no amount of straining will succeed) to keep the STR 10 crowd from auto-success.

The "complexity" of the check could come back in organically if the players decide to attempt different methods, such as the DC 20 DEX escape (same rules as STR single check). If a teammate with free limbs is available, they could get in on this with a DC 15 lock picking attempt (That method might get a complex check even if the others were single checks) or the "smash em" method of dealing damage via the attacking object rules (which might get around any checks at all since it will eventually succeed - assuming no time crunch or need to avoid being noisy). Or even just help the manacled player with his/her checks to give advantage.