Using Ability Checks in B/X

One gripe I have always had with D&D ability scores is - what are they for? Regardless of edition, they are only ever used for deriving other numbers. It would be much clearer if it were simply a -3/+3 stat since that is how it actually gets used. And yet, no one wants that. We all love our 3-18 rolls that we end up never using. It annoys me that such a prominent feature of PCs mechanically means so little.

The exception of course, is the ability check in TSR D&D, introduced in B/X and which achieved peak infamy with 2e non-weapon proficiencies. The mechanic where you actually get to roll against your ability score.

To begin with, let's acknowledge that there are many good old school reasons not to roll for ability checks in B/X:

  • Most things that DMs in later editions require rolls for, shouldn't require a roll in the first place as long as the players can describe what they are doing properly - And the ability score should be factored in by the DM in those cases anyway.
  • It encourages a binary "Success or Failure" approach to task resolution. 
  • There are other mechanics already handling those events (mostly the X-in-6 roll).
  • Ability scores should have low relevance to a character's capability compared to level.
  • B/X assumes character competence and further down the road of ability checks lies the evil of looking at your character sheet to discover all the things you now can not do and sticking to what it says you can do. 
I basically agree with all these observations and would argue for them as reasons not to have a skill system. That notwithstanding, I want to have ability checks for for reasons of gaming aesthetics - It's one of the most prominent features of character creation and the character sheet. It should matter somehow, in and of itself and not just as a conversion to a different number.  There should be times where having rolled 18 Wisdom for my fighter actually matters mechanically.

Rolling against your ability score is one of the most obvious and intuitive rulings that the character sheet present to you. It belongs in the game somehow. Despite my support for all the observations in bullet points above, I believe there is a measured case for ability checks. Here's two ways I use them:

Make a CHA check to show the world that you are FABULOUS

Ability Check as Failover Mechanic

I call this the "Oh. Really?" roll. Use it in case of:
  • Player failing to describe what they are doing properly.
  • Players attempting poorly considered things, or failing to use the proper tools or preparation for the job.
  • Players attempting foolhardy things that a competent person couldn't normally feel assured of accomplishing.
Basically, whatever could prompt the DM to say "Oh. Really?" can be immediately followed with a "make an ability check then." Make players aware of this guideline too - Ability checks probably means you could have played this one better. 
I believe this is also a nice bridge between those hardliners who insist in player competence over character competence and vice versa,

Make a WIS check to realise you need to level up before seeking revenge against Bargle

The second way I use it is

Ability Check as Cost Determination Mechanic

When players are attempting something that you have already determined will succeed, it can happen that circumstances suggest there could be a cost to it. 
For example, climbing a very difficult surface. The cost could be lost time (always a precious resource when you track time with the random encounter die), broken tools, exhaustion, injury, etc. Or the inverse - even faster than expected, went up real quiet, etc. 

For that, you roll an ability check, not to determine success or failure, but to have a "margin of success or cost." You then use the margin of the die roll against the ability score to inform your judgement of cost or additional success.

In this way, ability scores don't go down the road of gatekeeping who can and can't do tasks or creating binary outcomes from something that doesn't need binary outcomes, but they still inform you how a character interacts with the world based on those scores.

Modifiers are simple - Grant advantage if you have especially favorable circumstances (for example, attempting something especially relevant to your class or secondary skill) and disadvantage for especially hard circumstances. Keep reading to find out why this is mathematically pleasing in relation to the regular X-in-6 roll. 

Walking on the ceiling is a DEX check, right?

Using ability checks in place of X-in-6 rolls

As a last exercise, let's see if there are not other scenarios where an ability check can be used: Converting the X-in-6 roll.  

I use this basic formula for conversion from X-in-6 to to a d20 roll: (20/6)*X. This tells me the "Target Number" I am trying to roll at or under on a d20. With rounding, it goes like this:

  • 1-in-6: 3 (3,33)
  • 2-in-6: 7 (6,66)
  • 3-in-6: 10 
  • 4-in-6: 13 (13,33)
  • 5-in-6: 17 (16,66)
This progression maps also very nicely to the bonuses given by ability scores, if we take 3-in-6 and 10 as our respective baselines. A 7 ability score gives a -1 ability modifier. 13 gives a +1 and 17 +2. 

Which means that ability score modifiers applied to an X-in-6 roll gives about the same benefit as the ability score itself when applied to a standard 1d20 roll under check. 

So for those 3-in-6 rolls where ability score modifiers apply, you can basically just make an ability check and get roughly the same outcome, albeit with more granularity. For the hard ones starting at 2-in-6, apply-3 to the ability check and -7 for 1-in-6 rolls. And vice versa, +3 for 4-in-6 and +7 for 5-in-6

However, there is another way to get much closer to those unrounded probabilities of the d6, using a modern mechanic instead of those static modifiers to the roll - Advantage. 
For 2-in-6, apply Disadvantage to the roll.
[dis-]advantage skews the average of about 3.4 so is actually closer to the unrounded d6 progression of  3.33. 
For those rare 1-in-6 roll, apply a further -3 to the d20 roll (the probabilities of advantage interact with the static modifier, hence hewing to -3 rather than -4).
For 4-in-6, grant Advantage and 5-in-6, an additional +3 to the roll.

I like advantage over static bonus here, because it means high ability scores retain a (very) small chance of failure that gets lost with static modifiers and overall it is more pleasing to roll differently for different circumstances than adding and subtracting, especially when the probability math of advantage is so close to the original X-in-6.

Looking over the X-in-6 rolls in B/X, let's see which ones allow modifiers from ability scores and can therefore be translated into an ability check:
  • Climbing (the rules even recommend an ability check instead)
  • Opening Stuck Doors. 2-in-6 -> STR check with disadvantage.
Ok, that is actually not a lot, although I do believe stuff like boarding vessels and hiding could receive bonuses from ability score modifiers. This exercise is perhaps more about examining the maths and demonstrating some equivalence between the ubiquitous X-in-6 roll and the ability check, showing that there can be a place for it in B/X.


  1. Yeah, I often think about that too. There is no single solution to this. Basically, when facing a task, you have to ask yourself:

    A - Should chances improve with a high ability score?
    B - Should chances improve with a high level?

    There are very few cases in B/X when BOTH apply*.

    (E.g., attack rolls, spell saving throws).

    But for most modern games (and also to my gut feeling, TBH) this is often the case.

    So you usually have to make a choice between A and B.

    B/X does contain one solution (that I do not like): ability scores make you reach high levels a bit faster (due to XP bonus).

    A better solution IMO is just letting you improve your ability scores as you level up - even if very slightly.

    I also agree that the most important part (besides A and B, above) is finding out how hard you want the task to be.

    1-in-6 seems to be common in B/X. You could adapt this to a d20 using a "Target 18" (e.g., roll 1d20 plus ability mod) or even Target 30 and adding your entire ability.

    But it is hard to reduce B/X to a single mechanic, which is why we end up in this dilemma... roll high, roll low, roll ability scores, use modifiers instead, improve with level, should anyone be able to try, etc.

    1. OK Eric, let me start out by saying that it drives me mad that you asterisked something and then never explained it.
      A lot of different topics to unpack here, so I am going to make several replies to your post to keep it organised.

      Rabindrath over on dragonsfoot made an excellent system for "Action Throws" that I think combines level and ability scores very well.

      If I were looking for a combo of level and ability scores, I think this is as good as it gets. Easy to recalibrate too. When I am not embracing it, it's because:

      A/ The complexity is high enough that I'd want it pre-calculated as entries on the character sheet to roll against
      B/ I am not looking for a mechanic that encourages *more* ability checks and if its on the sheet, one should expect to use it.
      C/ I still want a way to use the original ability scores as more than a number to derive other numbers from, damn it.

    2. Lately, I've come to profoundly dislike ability score improvements, and at this point consider it to be one of the biggest turn offs of WotC D&D.

      This is the road of bland homogeneity, where all 6th level fighters have their STR maxed out, all 8th clerics have WIS maxed out and so forth. The higher the level, the more they all begin to look alike.

      It makes ability scores meaningless as character defining attributes, something I think that, at heart, is the central point of ability scores in the first place.
      I think this could be a grousing blog post of its own in the near future. :D

    3. I think I lean towards class/level being relevant for class related features and ability scores for everything else. I like the idea of higher level characters still having more mundane challenges to resolve outside their class speciality, without falling into the 5e fallacy of 20th lvl characters facing +0 saves against DC19 spells.

      One thing I surprised myself with in my thinking with all this is actually to lean away from "finding out how hard it is". Another topic for a grousing blog post perhaps, but I am beginning to suspect DCs are a straight up bad mechanic.

      My first draft for the ability check was actually a TARGET20 roll rather than roll under, since I figured that makes it easy to adjust the difficulty to T25, etc (and I prefer T20 for combat rolls anyhow).

      But then I realised that is not really the right mindset for this type of roll. It leads down a road of imputing DCs on the world around the PCs.
      I don't want a game where the PCs interact with a world of game mechanics, but one where they interact with the fiction and the game engine helps resolve that and then disappears again.

      I think simply calling for an unmodified ability roll already sets a baseline of "here is a challenging situation for your ability to be tested against."

      Disadvantage is a fine enough metric to say "it's *really* challenging" or "you are going about this in a very non-ideal way" and advantage a fine way of saying "but you have favorable circumstances due to...".

      I am almost tempted to say that if it ends up with advantage, they can just do it without a roll, but I suppose B/X too has a place for those 4-in-6 rolls. And of course those cost-determination rolls.


    4. Hahahaha I'm sorry about the asterisk, my bad! It should just read:

      "There are very few cases in B/X when BOTH apply. (E.g., attack rolls, spell saving throws)."

      "Action Throws" - nice idea but I agree with you, especially your "A" and "C" points.

      Ability score improvements - I see your point (especially share you concern with homogeneity) but I like ASIs! Will read your new post and try to respond with my objections.

      (I can say that homogeneity could be avoided using RANDOM ASIs, for example).

      I love Target20. I'm unsure about DCs too - but "4-in-6" is practically a DC. The good thing about DCs is that the mirror AC (if you're using ascending).

      Roll under ability checks is a great mechanic... my only issue is the "roll high and roll low" confusion.

    5. I must add about "Action Throws": it would be 10 times easier to say: "roll 1d20, add (ability+level)/2, Target 20".

      Not the exact same chances (but very similar) and works perfectly well for B/X IMO.

      Maybe do the exact same for thieves' skills - but thieves add their FULL level: "roll 1d20, add ability/2 + level, Target 20".

      E.g., Dex 14, level 8 means +15 to hide if you're a thief, +11 if you are not.

      +11 seems a lot for a non-thief, but in the unlikely event you have a Dex 14 MU, it gives the sneaky bastard lots of personality!

    6. That's not bad for a different take on action throws.

  2. In reading over that Dragonsfoot post, is there a chance the Action Throw table is similar to Swords & Wizardry and White Box FMAG's approach to the single Saving Throw? I ask strictly because of the single number aspect, not because of any statistical analysis or the like which I have not considered in the least.

    1. And as a follow-up, does that Antonio fellow have the full Action Throw table available anywhere else now that the original link is dead?

  3. Fascinating analysis! Question: For 1-in-6, can we roll 3d6 and apply "double Disadvantage" (two of three d20s must be rolled under to succeed)? And a similar principle for 5-in-6. Would that work? What's the math?

    1. I am not entirely sure what you are looking to emulate with this - What should it compare to?

    2. I mean, instead of doing this: "For those rare 1-in-6 roll, [apply Disadvantage to the roll and] apply a further -3 to the d20 roll (the probabilities of advantage interact with the static modifier, hence hewing to -3 rather than -4)." Use this instead: "roll 3d20 and aim to get at least 2 dice under ability score to succeed". Basically, it is a way of having a roll that doesn't require that the player or GM do subtraction; instead, simply roll more dice. My question is whether adding a third dice and requiring two dice under the ability roll is equivalent to your formula, from a math perspective. And if it is not, whether my goal could be achieved by using a different third dice (e.g., a 1d16... yeah, a DCC one). Hope that makes sense.

    3. I'm sorry, I made a mistake. I meant to say "roll 3d20 and aim to get ALL THREE dice under your ability score". Then the question is: would that be the equivalent (math-wise) to rolling 2d20 w Disadvantage (where you need TWO dice under ability schore) and applying a further -3 to each d20 roll (i.e., lowering the Ability score by 3)?

    4. The "problem" with 'double [dis-]advantage' is that it skews the probabilities dramatically in favour of those with higher ability scores.
      It basically becomes -6/-7 for someone with a score of 10-16, but only a -3 for someone with 18.
      I say "problem" because maybe a probability that is brutally punishing for anyone but the very most exceptional (who somehow just pull it off more often than not) is exactly what you for "nearly impossible".
      But I consider it problematic because you end up with a resolution mechanic where the probabilities become very opaque and it is unclear that the dice favours the 17+ stats whilst heavily punishing everyone else.

      But yeah, it would be nice to continue the progression with a different dice throw. My alternate suggestion is to just add 1d6 to the highest d20, although that makes it swingy in a way I don't really like for "nearly impossible".
      Alternatively, the highest 2 of 3d12 should give you nice equivalent numbers for that ~ -3, but the shift from 2d20, take highest, to 3d12 take highest 2 is not a very intuitive one for me.

    5. Thank you for the detailed analysis!


Post a Comment

Popular posts from this blog

Fantasy Map Review IV: Forgotten Realms

Fantasy Map Review II: Greyhawk

Fantasy Map Review III: Dragonlance

Review: Five Torches Deep

Comparison: Five Torches Deep vs Into the Unknown