Incremental / Idle
What it is. An economy that grows itself. You buy producers, producers make currency, currency buys more producers; periodically a prestige reset trades progress for a permanent multiplier. The number goes up — the question is only how fast, and toward what next unlock.
Player fantasy / why it’s fun. “I keep crossing thresholds I couldn’t afford a minute ago.” The pull is a horizon that keeps arriving: every purchase is close, the next tier is visible, and a reset makes the whole climb feel faster. Fun is the pacing curve with no deserts — the moment you’re bored, the game has failed.
Pillars
- The curve is the game. First-buy times and the gaps between eras are the design surface. You are authoring a rhythm of “soon”, not a set of numbers.
- Every tier pays back on schedule. Cost-over-production per tier stays roughly flat (a ~15–25s payback). A rising payback silently strangles the late game.
- Intent is an action, never a mutation. A click is
input.press, not a direct state write — or the replay lies and the balance-sim can’t trust the arc.
The loop stack
| Layer | The beat |
|---|---|
| Moment | Afford the next thing → buy it → watch the rate tick up. |
| Encounter | Reach a tier / unlock / milestone that changes what you optimise. |
| Session | Climb toward a prestige; decide when to reset. |
| Meta | Prestige multipliers, permanent upgrades, new mechanic layers per reset. |
Essential systems
| System | Why this genre needs it |
|---|---|
system-economy | Faucets, sinks, and currencies are the entire object; inflation control keeps late tiers meaningful. |
system-progression | The power curve is the content; pacing power gain is pacing the game. |
pattern-feedback-loops | Production→currency→production is a runaway positive loop; prestige is the negative-loop reset that re-paces it. |
system-reward-schedules | Milestone unlocks and offline-earnings payouts are the variable, anticipated rewards. |
system-meta-progression | Prestige currency + permanent buffs are cross-run meta; power vs. new options is the tuning axis. |
system-onboarding | The first three buys must land in under a minute or the ramp never catches. |
Content & difficulty model
- Balance-sim the whole arc. Run the economy headless from zero to endgame; assert the pacing windows (first-buy times, era-to-era gaps), monotone production, and no unlock deserts — no stretch where nothing new is affordable for too long.
- Flat payback per tier. State the inequality:
cost_n / production_n ≈ const(~15–25s) across tiers. A tier whose payback balloons is a wall; assert it against the actual config, don’t vibe it. - No click-softlock. Prove the player can always progress without a required manual click storm — an idle game must idle.
- Difficulty = slope, not spikes. You never “make it harder”; you shape how long each threshold takes. A desert is the only real failure state.
- New layer per prestige. Each reset should unlock a mechanic, not just a bigger multiplier, or the late loop flattens into staring.
Signature-mechanic seeds
- Incremental but the producers are a factory graph — outputs feed inputs, and
the bottleneck moves (mechanic-swap toward
anchor-factorio; the curve becomes a logistics puzzle). - Incremental but every prestige is a poker deck — reset trades score for jokers
that reshape the next run’s multipliers (mechanic-swap toward
anchor-balatro). - Incremental but it’s a tended garden — production is crops maturing on a
calendar; the tonal bend swaps grind-anxiety for
genre-farming-simcalm. - Incremental but one hard constraint: one currency, one screen — no menus, the whole ramp legible at once (constraint; forces ruthless curve design).
- Incremental but the number going up is a story meter — thresholds unlock
narrative, not just power (tonal; pairs with
genre-narrative-decisions).
Common pitfalls
- Rising payback ratios. The classic strangler: late tiers cost far more than they return, the ramp stalls, the player quits. The flat-payback inequality exists to catch exactly this.
- Unlock deserts. Any dead stretch waiting on a wall is the genre’s cardinal sin; the balance-sim asserts against them.
- Direct-mutation UI. Buttons that write state instead of pressing an input make replay and the balance-sim untrustworthy — the arc can’t be proven.
- Prestige-as-only-content. Resets that add a multiplier and nothing else flatten the meta; each should open a new mechanical layer.
- Punishing idle. If afk means you fall behind, it isn’t idle. Offline earnings and a click-free path are load-bearing.
Anchors
anchor-balatro— the “number goes up” surge and the reset-for-multipliers loop; the direct reference for the prestige-as-deck seed.anchor-factorio— production feeding production, complexity scaling with the curve; the reference for the factory-graph bend.
Verify
Balance-sim the arc; assert pacing windows, monotone production, no unlock deserts, no click-softlock → design/FUN.md §14 · Incremental/idle. Author the curve here; prove its windows there.
Composes with
system-economy— the faucet/sink model the whole ramp rides on.system-progression— the power curve you are literally pacing.pattern-feedback-loops— the runaway production loop and the prestige brake.
See also
- design/FUN.md §14 — mechanical truth + verify recipe.
sandboxes/— the procgen/economy lab to prototype a payback curve headless before wiring UI.