The One Rule

To play ForeSight you learn one rule. It's called resolution, and it runs every uncertain thing anyone attempts — a task — from picking a lock to winning a trial. Everything else in this book is that rule applied to a situation, or an example of applying it. Learn this page and you can run the game from memory.

The core

Every ability — an attribute (innate, like Strength or Perception) or a skill (trained, like Firearms or Stealth) — has a Score and a Base Ease Factor (BEF).

• Score is how good you are: 1 is hopeless, 10–15 is competent, 20+ is very good. Skills climb far higher than attributes.
• BEF is how forgiving the ability is by nature — usually around 5.

To resolve a task, judge how hard this particular attempt is as an ease-factor modifier, add it to the BEF to get the Ease Factor (EF), and multiply:

SC = Score × Ease Factor

Roll D100. Roll ≤ your Success Chance and you succeed.

*Amy fires a stun pistol at a suspect. Firearms BEF 4; the suspect is 12 m off (−2 range) and moving (−2), but she braced (+1), took a bead and got it (+2), and the pistol is well-balanced (+1). EF = 4 − 2 − 2 + 1 + 2 + 1 = 4. Her Firearms Score is 17, so SC = 4 × 17 = 68%. That is about as complicated as it ever gets.*

How ease factors work

Ease factors aren't a linear scale at the bottom — they run:

0 — ¼ — ½ — 1 — 2 — 3 — 4 — 5 — 6 — 7 — 8 — 9 — 10 — 11 — 12 — 13 — 14 — 15

Adding a modifier moves right along the ladder, subtracting moves left. Sum every modifier first, then apply it (6 + 10 − 4 is 6 + 6 = 12, not 15 − 4). EF can't fall below 0 (flatly impossible) or rise above 15 (as easy as things get). A Score over 20 counts as 20 but adds +1 EF per 3 points above — written 20+1 (23–25), 20+2 (26–28)… — which is how the truly gifted do what lesser mortals simply cannot.

One step is a big lever, and it scales with skill. In the sweet spot, +1 EF adds your whole Score to the SC — Score 15 going EF 5→6 jumps 75%→90%. Down at the ¼/½ rungs a step is worth about half your Score. Two consequences: modifiers stay small (a typical situational modifier is ±1 or ±2 — each is already potent), and the same good conditions help a skilled character more, which is exactly right.

Fractions round to nearest (½ or more rounds up): Score 13 at EF ¼ → 3.25 → roll ≤ 3.

Designer's Note — a lever you can feel. One EF step is a big lever, and a fixed-size one: because SC = Score × EF, a step moves your success chance by your whole Score in points wherever you sit on the range — 30%→45% and 60%→75% cost the same single step. Almost no other dice engine can say that. A 2D6 or 3D6 bell curve gives a lever that's huge near the middle and near-useless at the tails, so a +1 barely dents a long shot or a near-certainty. A flat D20 is the reverse — a modest lever in the middle that turns brutal at the 1/20 edges, and the change in odds is nastier still. A raw flat D100 is the cautionary tale: a flat +10 runs 10→20 (doubling your chances) in one place and 45→55 in another — the "same" modifier doing wildly different work, a giant lever with poverty traps at the bottom. (Savage Worlds' exploding dice are more erratic yet.) Score × EF is deliberately none of these: the same-sized step wherever you pull it.

Competence at a glance

What a Score buys you on an ordinary task (a typical BEF ≈ 5, no modifiers):

Challenge at a glance

What the situation is worth. This table is really all you need to remember — every other modifier in the book is an example of it.

Modifiers accumulate, and can run past either end: wounded and firing at a small, distant, dodging target with a badly-made gun can pile well below −7. So it should be.

Success comes in degrees — Quality Ratings

One roll tells you whether and how well. Read the Quality Rating (QR) off where your roll fell under the SC:

And succeeding is not the same as prevailing. A Success Chance of 100 makes the roll certain, never the outcome: most interesting tasks are opposed (his defence meets your attack, better QR taking it) and most consequences are then mitigated — by armour, by cover, by location, by what the world does about it. So the quantity that actually decides things is your net result relative to his, not your number on its own. This is why stacking several small ease-factor advantages beats holding out for one big one, and why an expert profits from bad conditions: a penalty applied to both of you comes out of his chance of succeeding at all, and merely out of your surplus.

Two hard limits: a roll of 100 is never better than QR7 (even a near-certainty can fail), and 99 is never better than QR4 (you can always succeed, but never brilliantly, on a 99). (The character sheet's Resolution Table precomputes the four thresholds so you never divide at the table.)

Read one column: fast, good, or much

A QR1 is quick or superb or plentiful — never all three. Decide before you roll what was at stake, and read only that column:

• Yield — how much? QR1 100%, QR2 80%, QR3 50%, QR4 25% of the maximum. Damage, intel gathered, ground covered, gold panned, rope salvaged. (Damage = Yield × Damage Class ÷ 2.)
• Time — how fast? QR4 takes roughly twice a competent estimate; QR1 is done in a flash. This is what answers "did you pick it before the guard rounded the corner?"
• Product (PM) — how good, for later? When a task makes something used in a later task, carry it forward as PM = 4 − QR: QR1 → +3 … QR4 → +0, QR7 → −3. (That's why the scale is numbered 1-2-3-4-7-10.)

Never penalize the same choice twice. If you took a penalty to do it a certain way, success already includes that thing. Pick the lock fast under a haste penalty and a plain success means you were fast — no separate Time roll on top. Call the head-shot at a penalty and a hit lands where you aimed. Pay once, up front.

Half of all successes are QR4 — twice the time, a quarter of the take — and that's just life: the plumbing rarely gets fixed well. When a result is hidden (was the alarm well-installed?), a failure yields and takes time exactly as a QR4 would; the dice stay quiet and the GM knows.

Speed answers "why can't I just try again?"

No special rule needed. A failure already spent the time — a QR7 eats ~200% of the estimate, same as a QR4 success — so retrying isn't free, and when time matters the guard is coming. A QR10 shuts the door: you've fouled the approach (snapped the pick, convinced yourself the safe turns the wrong way) — find another way. The worst-to-best conga line never forms.

Opposed resolution — oppose or build

Tasks rarely stand alone. Two moves cover every chain:

• Oppose — both roll, compare QRs; the better QR wins. A big gap is a rout, a small gap a narrow win, and equal QRs cancel — the blow just parried, the tail almost spotted. That's a feature: nothing decisive happens, the tension holds, you keep going, and a string of ties is a long taut struggle. If someone must act now, the higher Score edges it.
• Build — a prior result feeds the next as a modifier: +(5 − QR). QR1 → +4 … QR4 → +1. The follow-up still uses its own ability and BEF, nudged by what came before. Opposing a standing result instead (hacking a system someone built well)? Take the same value as a penalty.

Worked examples

• Picking a lock — Lockpicking (or Sleight of Hand) vs a difficulty for the lock's quality, minus haste, plus good picks. Time is usually the real stake: take the haste penalty and success is beating the guard.
• Fixing an engine — a chain. First diagnose it: a Knowledge task (the relevant field, with Perception or Intelligence to read the symptoms). The diagnosis is the product that feeds the fix (PM = 4 − QR) — a spot-on read (QR1) is +3 to the repair; a botched one (QR7) sends you chasing the wrong fault at −3. The repair may still be hard for its own reasons (access, a seized bolt), and its Yield is how much you restore — capped at as good as before. So: +3 to do the repair from a perfect diagnosis, never +3 to the thing repaired. A repair only restores.
• Hot-rodding it — the same plan-then-execute chain (plan the modification as a Knowledge task; the plan feeds the execution), but where a repair only restores, hot-rodding improves a chosen quality at a tradeoff — and it's the book's clearest picture of what a QR actually means:
  • QR1–2 — the gain, with the tradeoff shrunk or gone; maybe a little extra on top.
  • QR4 — the gain and the full tradeoff: quicker off the line, worse through the bends (the normal deal).
  • QR7 — all downside, no upside: you paid the price and got nothing.
  • QR10 — you broke it completely.
• Persuading the Duke — Charm (or Confidence) opposed by his resolve (WP, or Confidence if he's arguing back). Compare QRs; the gap is how convinced. A tie: he wavers but won't commit — yet.
• Picking a pocket — Sleight of Hand opposed by the mark's Perception. Reaching for a guarded inside pocket is a penalty and a called outcome: succeed and you got that pocket.
• Sneaking past — Stealth opposed by their Perception (or Search, if they're actively looking). A tie means they sense something and can't place it.

Random QR

When you need a QR out of nowhere — area damage catching a bystander, how well an unseen NPC did their job — roll D10: 1→QR1, 2→QR2, 3–5→QR3, 6–10→QR4.

Designer's Note: resolution is adapted from Victory Games' James Bond 007 — the finest published RPG of its era — generalized, then beaten into shape by four decades of play. It is not negotiable: a cleverer-looking dice engine was built to replace it, looked fine on paper, and failed at the table. Score × EF works because the value is a percentage, and people understand percentages and multiplication in their bones. The QR scale is the secret engine — combat damage, a chase's gained ground, an interrogation's leaked truth, a forged passport's quality are all "read the QR in the column that matters." Learn these two things and the rest of the book is examples.