The Soulknife Rogue has the following ability:

Psi-Bolstered Knack. When your non-psionic training fails you, your psionic power can help: if you fail an ability check using a skill or tool with which you have proficiency, you can roll one Psionic Energy die and add the number rolled to the check, potentially turning failure into success. You expend the die only if the roll succeeds.

However, some DMs I know (including myself) don't have a clear DC for skill checks. For example, on a knowledge checks, the amount of knowledge given depends on the exact result of the roll. A 6 gives no results, a 13 gives a bit of information (for example, that the writing system is from this particular ancient civilisation) and a 17 gives some more (for example, that it's a tablet with funerary records).

Or, for a task like lockpicking a difficult lock, a 15 would mean that the lock is picked but that it took a long time/was quite noisy and a 20 would mean that the lock is picked efficiently and silently.

Or, for some checks, rolling too low (like under 10) gives negatively consequences, rolling above 10 but still failing means there are no negative consequences.

Then there are checks with results directly proportional to the roll, for example, negotiating for a decrease in prices gets a discount in gold equal to the number rolled if it's above 15 (so save 17 gp by rolling a 17, save 22 gp by rolling a 22).

My question is: how should the Soulknife's ability be handled ? When should they be allowed to spend psionic dice, and when is the dice expended ?

Should the rogue be allowed to expend psionic dice whenever they haven't reached the best possible results (for example, getting information from a knowledge check, but wanting some more), or only when it's clearly a failure (no results at all or negative consequences) ?

If the rogue uses a die and turns a result with negative consequences into a failure but with less drastic negative consequences, is the die spent ? Since the roll is still a failure, but it's still a less punishing failure, so the die did have effects.

Personally, I think that the die is expended if it changed the result of the skill check, but I am not sure whether it should be allowed to be used anytime.

  • 1
    \$\begingroup\$ Would the DM be open to you declaring beforehand "For the purpose of PBN interacting with this check, 'failure' means I jam the lock or attract attention due to noise, and 'success' means anything else." or "'Failure' means I can't get the gist of what the writing is about, 'success' means I get the gist of it or even an exact translation"? \$\endgroup\$
    – Snowbody
    Commented Apr 2, 2022 at 15:14

3 Answers 3


Adapt the feature to work with sequential categories of adjudication.

As you said the problem is to adapt a feature that work in a binary adjudication system to an "sequential" adjudication system.

I will suggest using this adaptation:

Psi-Bolstered Knack (Homebrew). When your rely on your skill training, your psionic power can help: if you roll an ability check using a skill or tool with which you have proficiency, you can roll one Psionic Energy die and add the number rolled to the check, potentially changing the adjudicated result. You expend the die only if the roll change the adjudicated result. You may use this feature after roll but before outcome is determined

I think this change will be a correct adaptation, but be aware that some one could said that change the result in a sequential adjudication system (potentially between different failure categories, between top failure category and bottom succes category, and between succes categories) is not equivalent, or even powerful, than change it in a binary one (just between success and failure).


You've uncovered an edge case when making house rules

Per the rules of ability checks:

For every ability check, the DM decides which of the six abilities is relevant to the task at hand and the difficulty of the task, represented by a Difficulty Class. The more difficult a task, the higher its DC.
To make an ability check, roll a d20 and add the relevant ability modifier. As with other d20 rolls, apply bonuses and penalties, and compare the total to the DC. If the total equals or exceeds the DC, the ability check is a success — the creature overcomes the challenge at hand. Otherwise, it’s a failure, which means the character or monster makes no progress toward the objective or makes progress combined with a setback determined by the DM.

As you pointed out, it is designed as a binary system. I also brought this up in this other question, that making ability checks a sliding scale is very common. Unfortunately, that is a house rule so everything else needs to be adapted to handle that rule.

Here is an example of a house rule that caused unforeseen problems

I watched a video about how to make lock picking more interactive; you roll a number of d6 based on the DC. Based on the results, you either picked the lock, failed to pick the lock, or jammed the lock. It allowed anyone to try to pick the lock, but having proficiency with the tools helped. It also had a scale for adding time as a factor. It seemed to have a lot going for it. So I brought it up with my group...

  • It didn't take into account Dexterity
  • It didn't take into account guidance as a bonus
  • It didn't take into account hex as a penalty
  • If doesn't work with the Gloves of Thievery bonus to Dexterity
  • And the list went on

Basically, it replaced the entire skill check roll with rolling d6s. So in order to incorporate this house rule, it needed to adapt to factor in other bonuses and penalty.

You need to adapt the sliding scale to have at least one binary element

You still need to have a binary pass/fail element in place so that all the other features still work. Luckily, for most things, that's already built in with DCs.

So to keep your sliding scale working, you base the amount of success on how far above or below the roll is from the DC. If it just matches, then it's a minimal success. If the roll greatly exceeds the DC then the success has additional information/benefits. Similarly, if the roll is below the minimum, the DM can decide if nothing happens, or a penalty such as misinformation or catastrophic failure.

For things that don't normally have a DC, the DM can use the Task Difficulty table under the Ability Score heading. So in your example of a knowledge check, the DM needs to consider how prevalent the information is. Hieroglyphics are a "writing system from this particular ancient civilization", but most people can at least identify them so the recognition DC would be pretty low. However, if the PC's roll is exceptionally high, they might even recognize a symbol or two and realize it contains "funerary records".

Another option is to have set break points

With any given DC, if the roll matches or exceeds by 5, it is a success. If it is between 6-10 over, it's a success with favor. It it beats the DC by 11 or more, it is not only a success but the PC gains something extra. And it works for penalties as well.

So if the DC for reading the tablet is 15:

  • 1 - 4: Catastrophic failure; in an effort to examine the tablet, they drop it and cracks it making future reading more difficult
  • 5 - 9: Bad failure; they recognize some of the symbols, but incorrectly translate
  • 10 - 14: Normal failure; doesn't ring any bells
  • 15 - 20: Normal success; they recognize the language
  • 21 - 25: Better success; they see a couple of symbols they remember and know it's basic purpose
  • 26+: Critical success; not only do they recognize the symbols, they get an extra 1d4 on their next check with the same language

Some things don't have degrees

Also remember that not everything needs a sliding scale; or a scale on both sides. For instance, hitting a baseball. How well you succeed is how far you hit it, but any failure is still a failure and you didn't hit it at all.

  • 1
    \$\begingroup\$ As I commented on Pepijn's answer, your suggestion to still set a DC means a psionic die can help you avoid critical failure, but isn't expended if it doesn't push the total past the set DC? And can't be used to boost from success to bonus, only from failure to possible bonus? I guess consistent with the phrasing "When your nonpsionic training fails you", although that's based on the binary assumption. Still failing to get bonus stuff. \$\endgroup\$ Commented Apr 2, 2022 at 7:51
  • \$\begingroup\$ @PeterCordes, for option one, there is just a single DC--a very simple pass/fail scenario. If your roll means you failed, you can roll the psionic die and try to reach success. For the multiple breakpoints, it's still succeed/failure, but with graduation within each. So if the roll lands in any of the "something" failure, you can roll the die to try to get into the success ranges. But if you already succeeded, then there is no option to roll to make it a better. success. The feature only works if you failed, not if you non-optimally succeed. And you only expend the die if you reach a success. \$\endgroup\$
    – MivaScott
    Commented Apr 2, 2022 at 22:28

Set a 'minimum' DC alongside the range of DCs

It is a bit of a copout and obvious answer but as a DM I am also a fan of various skill checks/saving throws representing a range of outcomes rather than a binary success or failure. Unfortunately for us there are features that somewhat rely on the binary nature of skill checks/saving throws. Here you have noted the rogue's Psi-Bolstered Knack but another example is the Magical Guidance optional feature from sorcerer.

To accommodate these features I set a minimum DC first, which is the DC the player has to meet to do exactly what they set out to do but nothing less and nothing more. From there I determine things like "if they succeed by more than 5 they get this bonus success" and "if they fail by more than 5 something negative happens". This approach is very reminiscent of how some saving throws are handled, for example the Shortbow attack from the Sprite mentions (emphasis mine):

the target must succeed on a DC 10 Constitution saving throw or become poisoned for 1 minute. If its saving throw result is 5 or lower, the poisoned target falls unconscious for the same duration, or until it takes damage or another creature takes an action to shake it awake.

Although the line between success and failure can be blurry it is important to remember that skill checks are generally initiated by the player wanting to do something. A success must then obviously mean they do what they set out to do without short term negative consequences that could have reasonably been avoided.

In cases where the player will always succeed or fail and the roll is only made to see how much they succeed or fail, I think it is reasonable for these abilities to just not come into play and to be upfront with the player to avoid them spending time or resources on something that won't change. Alternatively in these cases you could decide to still let them use the ability if they roll below some sort of average DC.

  • \$\begingroup\$ Alongside the Sprite here are some other monsters that implement additional effects based on rolling below the normal DC: Pseudodragon's sting, Ghost's horrifying visage, and a Bodak's death gaze \$\endgroup\$
    – anon
    Commented Apr 1, 2022 at 10:45
  • \$\begingroup\$ So if the roll is already above the minimum DC, but not reaching the "bonus", they can't use a psionic die (or magical guidance) to boost it? Is that rule liked by players, or tolerated as a necessary compromise between rules. \$\endgroup\$ Commented Apr 2, 2022 at 7:42
  • \$\begingroup\$ Or if they bring the result up from failure to success-with-a-cost, they don't have to expend the psionic die because it didn't cross the minimum DC you set for complete success with no downsides (but no bonuses)? Or you don't do success-with-a-cost, you just do failure vs. failure with extra downsides? Your 2nd-last paragraph mentions success without downsides, implying that success with downsides was possible. It'd be great if you could expand on your answer to specify exactly what you're saying should happen in those corner cases. \$\endgroup\$ Commented Apr 2, 2022 at 7:42
  • \$\begingroup\$ @PeterCordes My approach relies heavily on players being detailed with what they are trying to do. For example instead of "I investigate the room" I will ask them to elaborate on what they are looking for. This allows me as the DM to set a single DC that better corresponds with what the player is trying to accomplish. From there I decide whether rolling 5/10/whatever above or below that DC does anything extra. I don't really view these as extra DCs, there is only one true DC that corresponds to what the player wants to do. This is definitely a style that won't work at every table. \$\endgroup\$
    – anon
    Commented Apr 2, 2022 at 12:27
  • \$\begingroup\$ For cases where the player might not yet know what the outcomes might be, like with some 'knowledge' checks, I set the initial DC based on the outcome that provides enough info to keep the story going but leaves out details. I know this is rather vague but I handle these on a case-by-case basis. I have not had players complain nor do I see why they would have any reason to. A success on the main DC means they did what they wanted to do or got enough info to keep going forward with the story, I don't think anyone would consider those outcomes a failure that would have to trigger their features. \$\endgroup\$
    – anon
    Commented Apr 2, 2022 at 12:37

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .