Your job at the table
If you're the Game Master, your job isn't to beat the players or to recite rules from memory. It's to describe a world that feels real, play the people and creatures in it honestly, and decide what happens when the dice leave things in doubt. You're a narrator, a referee, and the players' biggest fan all at once.
You don't need to know every rule. You need to know how the core loop works — the loop from Chapter 2 — and you need to be willing to make a call and keep things moving. A confident ruling now beats a perfect one five minutes from now.
Setting a DC
When a player tries something uncertain, you set the Difficulty Class — the number to beat. You have two reliable ways to land on one.
For a task whose difficulty comes from skill, use a simple DC keyed to the proficiency it would take:
- Untrained — 10
- Trained — 15
- Expert — 20
- Master — 30
- Legendary — 40
For a task whose difficulty comes from the situation — picking this lock, climbing this cliff — use a level-based DC that rises with the challenge's level. A few anchors: level 1 → DC 15, level 5 → 20, level 10 → 27, level 15 → 34, level 20 → 40, filling in a point or so per level between. Either way, you can nudge the DC by ±2 for a clever approach, a hard circumstance, or a lucky break.
The four degrees in play
Chapter 2 laid out the four outcomes. Running them well is mostly about narration. A success moves the story forward. A failure should too — it rarely means "nothing happens." The lock holds, but now you hear footsteps coming. The jump falls short, but you catch the ledge.
Save the big swings for the criticals. A critical success is a moment to reward — the players will remember it. A critical failure is a chance for a complication, not a punishment: the rope snaps, the guard turns, the spell fizzles into something worse. Keep failure interesting and the table stays leaning forward.
When to call for a roll
The most common new-GM habit is asking for too many rolls. A roll is for a moment that's both uncertain and interesting — where either outcome would change the story. If a character wants to do something they'd obviously manage, let them. If failure would just stall the game with no interesting consequence, let them.
Reserve the dice for the fork in the road. When the answer is "obviously yes," say yes. When it's "obviously no," say no and offer another path. Save the roll for "maybe" — that's where the tension lives, and that's where the loop from Chapter 2 does its best work.
Building an encounter
When you want a fight to land at a chosen difficulty, build it on an XP budget. For a standard party of four, the pool is sized to the fight you want:
- Trivial — 40 XP
- Low — 60 XP
- Moderate — 80 XP
- Severe — 120 XP
- Extreme — 160 XP
You spend that pool on creatures, and each one costs by how its level compares to the party's: a foe 2 levels below costs 20, 1 below costs 30, an equal-level creature costs 40, +1 costs 60, and +2 costs 80. So a Moderate fight might be two equal-level enemies, or one a level up plus a couple of weaklings. (Add about 10–20 XP per extra player beyond four.)
The budget is a starting point, not a straitjacket. Terrain, numbers, the party's condition, and a little surprise all tilt a fight. Build to the budget, then trust your read of the table.