Loop Hero

What it is. Your hero auto-walks a looping track and auto-fights whatever’s on it. You don’t control the hero — you place terrain cards (meadows, rocks, vampire mansions) around the loop from a hand you drew, each of which spawns loot, resources, and stronger enemies. Build up, cash out at camp, go again.

Player fantasy / why it’s fun. I author the world; the hero lives in it. The pull is a genuinely novel loop stitched from three familiar genres — placement, auto-battler, and deckbuilder — where your only lever is the map itself, and every tile is a double-edged bet: more reward always means more danger.

Design DNA

Fuse three solved cores into one: the auto-battler’s watch-don’t-touch combat, the deckbuilder’s draft-and-play hand, and a placement puzzle where the board is a fixed loop. Then bind them with one rule — everything you build both helps and threatens you. A rock gives HP but a mountain spawns harpies; a road-side spider farm is loot and a death trap. The player’s whole agency is indirect: you don’t fight, you cultivate the conditions of the fight, then decide when to flee to camp with your spoils.

The lesson Loop Hero teaches about composition is the most valuable thing to extract: fusing genres is not stapling them together — it’s finding the one rule that makes them the same game. Three loops that would each be complete on their own are welded by the single principle that every placement raises reward and danger together. That binding rule is what stops the fusion from feeling like three minigames in a trench coat; it makes the deckbuilder’s draft, the placement puzzle, and the auto-battle all about the same decision. When you fuse genres (see process-composition), your real design work is discovering that rule.

The novelty isn’t a new verb — it’s the combination, and the discipline of satisfying all three parent genres at once. Indirect control is the delivery vehicle: because you shape conditions rather than act, the placement, the draft, and the fight are naturally one continuous authored consequence.

Load-bearing structures

Structure	Why it works
Indirect control	You never move the hero; you shape the world and watch. Agency lives in placement + when-to-retreat, not in reflexes. `system-emergent-systems`.
Every tile is double-edged	Each card raises reward and danger together — pure `pattern-risk-reward` baked into placement; there are no free builds.
Fixed loop as the board	The track never changes shape, so the puzzle is what you add to it — a bounded, legible placement space.
Deck-drawn placement hand	You play from a drawn hand of tiles, so runs vary and building is a draft. `genre-deckbuilder` / `system-build-diversity`.
Auto-combat resolves the bets	Watch-don’t-touch fights turn your placement decisions into visible outcomes — the auto-battler’s prep-then-watch. `genre-auto-battler`.
Push-your-luck retreat	Leave the loop to keep resources; die and you lose most of them — the tension of when to cash out. `pattern-risk-reward` / `system-session-structure`.
Two economies (run vs base)	Run resources feed a persistent base that unlocks new cards — meta-progression across loops. `system-meta-progression` / `system-resource-loops`.

What to steal

Fuse genres by binding them with one rule. The parts (place / draft / auto-fight) are borrowed; the invention is “everything you build both helps and threatens you.” A single binding rule turns three genres into one game. See process-composition.
Indirect control. Let the player author conditions and watch consequences — agency through the world, not the avatar. Distinct, low-input, deeply strategic.
Double-edged placement. Make every build-choice raise risk and reward together, so there are no dominant plays — only bets. pattern-risk-reward.
Push-your-luck cash-out. A “retreat to keep your spoils vs push for more” decision gives every run a self-authored climax.
Two economies (run/base) so loops feed a persistent unlock tree — system-meta-progression. Run resources are volatile (lost on death unless you retreat); base resources are permanent — the same ephemeral/persistent split anchor-hades uses, here gating the card pool itself.
Bound the board so placement is legible. A fixed loop (rather than an open map) keeps the placement space small enough that every tile’s effect is readable and every bet is comprehensible — the constraint is what makes indirect control strategic rather than fiddly.

What’s just theme (drop it)

The pixel-fantasy / “the world was erased” fiction. Cosmetic framing.
The specific tiles. Terrain cards that spawn paired reward+threat is structural; “vampire mansion vs oblivion” is flavour — system-build-diversity.
The hero classes. Loadout variants; the loop is class-agnostic.
The retro aesthetic. Aesthetic, not structure.

Composes into

genre-auto-battler — supplies the prep-then-watch combat.
genre-deckbuilder — supplies the draft-and-play hand.
system-resource-loops — the gather-during-loop, spend-at-base cycle.
pattern-risk-reward — double-edged tiles + retreat timing.
process-composition — Loop Hero is the exemplar of genre-fusion-with-a-binding-rule.

Twist seams

Loop Hero but the loop is a real map you traverse (structure) — swap the fixed track for an open path; placement becomes route-authoring. Moves toward genre-exploration.
Loop Hero but two players share one loop — one places, one fights (perspective) — split indirect and direct control across a coop pair; the placer’s bets are the fighter’s problem. Feeds system-coop-and-competition.
Loop Hero but tonal — you’re gardening, not adventuring (tonal) — the double-edged tiles become plants that yield harvest and pests; auto-combat becomes an auto-tending sim. Pairs with genre-farming-sim.
Loop Hero but the binding rule is scarcity, not danger (constraint) — every tile you place consumes a shared budget that also powers the hero, so building up literally starves your walker; the double-edge becomes economic rather than combat. Sharpens the “everything helps and threatens” rule along a new axis. Pairs with system-resource-loops.