One-Roll, Player-Facing Resolution
Locked in. This is settled architecture, not an open question.
| Rule | Statement |
|---|---|
| One roll | One roll resolves the whole situation being resolved. Not one roll per step |
| Player-facing | The player rolls. The GM does not roll against them |
| No opposed rolls | Everything is resolved against a calculated Target Number |
Hitting a target is a Target Number. Winning an arm-wrestling match is a Target Number. The procedure does not change with the fiction.
What follows necessarily
The defender never rolls
Every defensive quality — Evasion, electronic defense, signature, cover, a rival's arm-wrestling strength — becomes a static number folded into the attacker's Target Number. Nothing about a defender is resolved by a die.
The brief already anticipated this: Target Number = Skill + Attribute + Equipment + Situation - Difficulty. Difficulty is where the opposition lives. That term now carries the entire defensive side of the game and needs a real derivation, since every defensive stat in the Unit Record has to reduce to it.
Margin carries everything downstream
If one roll resolves the situation, then hit-or-miss, quality of hit, location control, penetration, and damage magnitude all derive from one number: the Margin.
The brief was already halfway here. Edges are Margin-derived and already spend on location control, penetration, extra damage, suppression, and system targeting. This lock-in completes that logic rather than fighting it.
Two rolls in the current brief do not survive
| Roll | Status | Consequence |
|---|---|---|
| Damage roll | Cannot exist as a separate roll | Damage derives from Margin |
Component Check (2d10 + Breach Severity) | Cannot exist as a separate roll | Must derive from Margin, penetration depth, or become deterministic |
This closes a cluster of open questions rather than answering them:
- OQ-19 (die-size damage classes) is dead, superseded when the system went d6-only. Weapon identity has to come from Penetration, damage scaling against Margin, range, arcs, and properties instead of from die size.
- OQ-04 (Component Checks too lethal) changes shape. The 36% disable-or-worse rate was a property of a
2d10roll that no longer happens. The severity question survives; the distribution question does not.
Correction. An earlier version of this document also declared exploding dice dead here, on the grounds that an explosion is an extra sequential roll. That was wrong and is withdrawn. An explosion continues the same roll toward one result rather than asking a second question — there is no second decision, no waiting on another party, and one Margin comes out the end. Exploding dice are now the engine's core (OQ-09 resolved). See
04-dice.md.
NPC attacks need a player-facing form
If the GM never rolls, an attack on a player character has to become something the player rolls. This is not yet specified and is the largest gap the lock-in opens. See OQ-21.
This reverses the dice recommendation
Earlier analysis favoured a flat distribution, because on 2d10 a +1 is worth between 4 and 10 points depending on skill, which makes tactical tradeoffs hard to evaluate. That argument still stands on its own terms.
But under one-roll resolution, the Margin distribution is the damage distribution, and its shape now matters more than modifier legibility.
Margin mass by quartile at ~75% success (generated):
| Scheme | Low quarter | Mid-low | Mid-high | Top quarter |
|---|---|---|---|---|
2d10 | 38% | 33% | 21% | 8% |
d20 | 27% | 20% | 27% | 27% |
d100 | 25% | 24% | 25% | 25% |
2d10 produces a decaying damage curve: a typical result, with big results rare. That is exactly the feel multi-die damage (2d6, 3d6) would have given — the thing the one-set dice constraint otherwise makes impossible.
The flat schemes produce uniform damage. A maximum-damage result becomes as likely as an average one, which reads as arbitrary rather than dramatic.
So the bell curve, which looked like a liability when it only governed hit-or-miss, becomes an asset once it also governs damage. It is the only way to recover a damage curve under the dice constraint. This materially strengthens 2d10 and OQ-20 should be re-evaluated on these grounds.
The modifier-legibility objection does not vanish — it becomes a cost knowingly paid, and one the PA partially mitigates by displaying the delta.
What this fixes for free
| Benefit | Detail |
|---|---|
| Speed | The 12-step attack procedure collapses to one roll plus PA computation. The 45-second target stops being in doubt |
| Computability (OQ-17) | No opposed rolls means every Target Number is a pure function of game state — exactly what the PA needs |
| Player engagement | Nobody waits while the GM rolls. Every die thrown at the table belongs to a player |
| GM load | The GM stops rolling entirely and runs fiction and opposition instead |
| Dice budget | The PA spec's "2-3 physical rolls per turn" becomes one per resolution, comfortably met |
Open consequences
- OQ-21 — what a player rolls when an NPC attacks them.
- OQ-22 — whether "one roll" means one result or one throw. A simultaneous throw of d20 + d10 + damage die is one physical action but three readings. Under the strict reading it is out; under the permissive reading it is the natural implementation.
- Difficulty derivation — every defensive statistic in the Unit Record must reduce to a single number. This is now a blocking part of scale numerology (OQ-03).
- Whether Component Checks become Margin-derived, penetration-derived, or deterministic.