I’m taking the plunge and going with three attributes: Might for all things based on physical prowess and melee; Finesse for all things delicate and ranged; Wits for all things mental and magical.
Having (barely) taught English in university and school classes, my aim is avoid some of the struggles I encountered. I wish to be able to explain the rules in the English language and yet keep it as far from the structure of the English language as possible. I don’t want to describe rules and then explain why those rules don’t apply in this or that situation.
And so as much as I like the simplicity of the roll-under-attribute system of The Black Hack, I hope to keep the theme running throughout the ruleset that ‘Higher Is Better’, which allows for the 1-2 = failure, 3+ = success mechanics of Usage Dice. If it is possible to avoid have a natural 1 on a d20 be a critical success but a failure on a d12, then I will do so.
So I will be sticking to adding a modifier to a d20 with a target number to beat, as is the case with most D&D-esque games. Hopefully that will also make converting a lot of old material relatively straightforward.
But what modifier? It has never seemed satisfactory to derive a modifier from a larger number representing an attribute – it seems an unnecessary bit of mental arithmetic to ask of a player (not usually too complicated on its own, but all these little references to different values and subsequent calculations – as well as double checking with the GM to make sure everything’s being done correctly soon add up), and it renders the original attributes as awkward bits of decoration taking up valuable room and attention on a character sheet.
Knave makes the attribute itself be the modifier, so you just add the value of the attribute to the d20 roll. I would like to follow suit. In terms of impact this might have on the game itself, I suspect the characters might advance a little quicker than other systems which do use modifiers, since an increased attribute will have bigger impact on resolving challenges instead of contributing to an eventual increase in modifier. This means starting off with much lower stats than a D&D-esque player might be used to, but will make much more sense to a newcomer to the hobby.
No more explaining why you need to roll these higher numbers because they lead to higher lower numbers.
What range of numbers are we looking at, then? If they start at 1 (or perhaps 0), what is the cap?
D&D 5e has Attribute Modifiers going up to +5, with an additional proficiency bonus up to +6. That’s a bit too high for my liking. I have no wish to accommodate superheroes with these rules – I want to aim for ordinary people doing extraordinary things, with those ordinary people eventually getting a little less ordinary at doing those extraordinary things. If one were to follow the old D&D houserule of capping levels at level 6, and apply it to 5e, the max modifier a character could get would be +8. An top attribute of 8 seems fair at this point, accounting for the fact that there won’t be a list of skills in this system to which one could apply additional modifiers.
I rather like The Black Hack’s character leveling system where you have to roll a d20 for each attribute and increase it by 1 if you roll higher. It’s another instance of rolling higher is better in some instances but not in others (like challenge resolution), but in a system where Higher Is Better it fits in nicely. It will balance out characters who start off with worse stats than others and yet keep some of the randomness of character creation going. An attribute range of 0 to 8 allows a player to just roll one die to try this, and encourages an impression of a character making reasonable strides in ability (going from 5 to 6 seems a greater leap than from 15 to 16).
Alright, so now I have a spine of a system which hopefully still accommodates smooth converting of OSR and D&D material. Now to ponder the ribs…