Under/Over Rolls
D&D attributes are a curious beast insofar as they come in two parts; a smaller number that actually gets used most of the time, and the 3-18 rating from which it is derived. This would make sense if attributes routinely went up in value, making the large number a kind of XP value for the smaller one, but – at least in earlier editions – they don’t.
d20 Under Attribute
When writing a clone or heartbreaker there’s an incentive to jettison one or the other. SEACAT reduces attributes to modifiers. The Black Hack ditches modifiers and has players roll d20 under their attribute.
The latter always appealed to me; it’s such a simple mechanic, and has a clear probability, second only to a percentile roll. No bell curve, no arithmetic, just compare one number to another.
But this meant that monsters in the Black Hack have no variation in AC. Worse, if a monster is higher or lower level than you, it applies a penalty or bonus to the roll. So in a party of mixed-level PCs the clarity of “roll under your STR” is quickly lost.
These modifiers apply to everything, including sneaking, deception, intimidation – this kind of unvaried level-based difficulty makes it feel pointless coming up with alternative approaches to a situation, since the difficulty is always going to be about the same.
Under/Over
The first thing I did was to throw out the level difference modifier and introduce a difficulty mechanic. Since one point of AC is worth 5% on the hit roll, and 1 HD increases a monster’s accuracy by roughly the same amount, I simply made AC ascending 1-10 and said “to hit you must roll within your STR and over the target’s AC.”
Defense rolls, player-side and made on DEX in the Black Hack, roll directly against HD. Eventually I made player AC start at 9 and ascend based on armour.
This was easy to abstract into a general difficulty system. Thief skills such as ‘climb sheer walls’ have a base difficulty of 7, meaning a DEX 10 average PC will have a 20% chance of success (as well as having higher DEX on average, thieves get a discount.) A saving throw is a test against 2 (Poison) 6 (Spells) or 4 (anything else.) Monster saving throws are caster's INT vs monster's HD modified by the save type.
I kept 1s & 20s as critical failure and success, because rolling a 20 is just more fun than rolling a 1. This had some interesting consequences in terms of what the different numbers on the die represent.
| ⬛ | 🟥 | 🟥 | 🟥 | 🟩 | 🟩 | 🟩 | 🟩 | 🟩 | 🟩 | 🟩 | 🟩 | 🟥 | 🟥 | 🟥 | 🟥 | 🟥 | 🟥 | 🟥 | ⬜ |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 |
Here we have an attribute of 12 and a difficulty of 4. Odds of success are 40%.
What does it mean to roll a 3 here? If it's a hit roll, you know that you've failed specifically because you hit your target's armour. If it's a defense roll, you failed because of the ferocity of the monster. Conversely a 14 would mean your attack missed, or you were unable to evade your foe’s attack, and your armour wasn’t hard enough to deflect it.
The height of the roll might imply a degree of in-the-moment aggression and (over)confidence: Too timid and your efforts will be shrugged off, too bold and you'll exceed your own ability.
And the 20, being unaffected by skill, represents luck – perhaps you overextended – but you did so exactly when your opponent was off guard.
This information can be useful. For example, if you're firing into melee – a 3 is safe – your arrow has struck your foe's armour and isn't going anywhere. A roll of 15, however, is a shot that has gone wild, and might hit an ally. Since most such misses are going to be in the 9-19 range, which is the same range as PC AC, I simply compare the number directly to the AC of an ally to check for friendly fire, with no more rolls made.
Pros & Cons
It's slightly counter-intuitive. “Roll high/low” is easy to grasp; “aim for the middle” or “roll as high as you can without going over” not so much. But players are making so many of these rolls they soon get used to it, and the advantage is that, only having to compare three numbers, they're very quick.
These rolls handle penalties well enough – you can just pump up the difficulty until the roll is 20-or-nothing.
But they don’t like bonuses at all. At best they ruin the nice clean mathless roll. At worst, the bonus causes the attribute to exceed 19 and what do you do then? Wrap around and reduce difficulty instead?
So I've had to be creative in avoiding positive modifiers – relying heavily on the Advantage mechanic, or offering difficulty discounts instead.
Inspiration
This mechanic was inspired by, of all things, Mark Rein*Hagen’s lost Exile game, which featured percentile rolls with the degree of success being equal to the number on the tens die as long as you didn’t exceed your skill, and with particularly difficult tasks requiring a certain threshold of success to accomplish.
