Ballionaire

There is a moment Ballionaire is built to produce — once a player stops reading the trigger descriptions and starts reading the board — when the ball drops and nothing about the outcome is predictable. Not because the game is random. Because the machine on the board has more interactions in it than anyone can hold in their head. A ball hits a chicken. The chicken lays an egg. The egg falls into an axe. The axe splits it. Both halves hit a multiplier placed twelve drops ago for entirely different reasons. The screen fills with numbers. The player did that, technically, but couldn’t have predicted it, and that gap between authorship and outcome is the entire product.

Ballionaire is the work of newobject, a solo-scale developer operating under the Raw Fury publishing banner — a label that has spent the better part of a decade betting on structurally weird games and being right more often than not. It launched December 10, 2024 on Steam, arriving into the post-Balatro gold rush when every publisher on earth was hunting for the next number-goes-up machine that hides its depth behind simple rules. Most of those hunts produced Balatro with a different noun. This one didn’t.

The positioning is clever and the marketing copy is, for once, accurate: this is a pachinko roguelike. A run starts with an empty board. A ball drops. It ricochets down through nothing and earns nothing. Then the game drafts one of three triggers — gadgets, essentially — for placement. Drop again. Draft again. Twenty drops later there’s a contraption. Three ELDERS demand escalating tribute along the way, and missing a quota kills the run outright. That’s the frame. Everything interesting happens inside it.

The reception tells the story of a game with a spectacular opening stretch and an argument about the long game: a warm initial reception that has since cooled, with recent Steam reviews sliding toward mixed — and that shift is the most honest signal about this game.

• The Autobonker Loop: Draft one of three triggers, place it, drop the ball, watch physics do the rest. The player is not aiming. The player is engineering. The distinction is the whole design thesis.
• A Deep Trigger and Boon Library: A combinatorial library deep enough that experienced players are still discovering interactions well past the point most games run dry — chickens laying eggs, axes splitting coins, boomerangs returning balls to the top of the board for another pass.
• Board Variety With Teeth: The Pyramid, the Danger Wheel, and a full pinball table each bend the ruleset rather than just reskinning it. A board that changes ball behavior changes which synergies are viable, which is real variety and not cosmetic churn.
• The Laballatory: A sandbox mode with no quotas, no ELDERS, and no death — build the absurd machine a real run couldn’t afford, then share it.
• Lua Custom Content API + Steam Workshop: Players can author their own triggers. Not a toy mod loader. An actual scripting surface with distribution attached.
• Steam Deck Verified: And, by design, arguably the platform the game was made for.

The Gameplay Loop

Most roguelike deckbuilders ask for a decision and then execute it. Ballionaire asks for a decision and then surrender. The trigger is placed. The ball drops. Physics takes it from there, and physics does not care about the plan.

This should be a disaster. Handing the resolution of a strategy game to a bouncing sphere is, on paper, an abdication. What saves it — what makes it sing — is that the randomness is spatial, not arbitrary. A card game’s RNG hands over a card nobody wanted. Ballionaire’s RNG bounces the ball three pegs to the left. The difference is that the second one is legible. The board shows why it happened, and more importantly, a player can build against it. Cluster triggers where the ball tends to fall. Use a boomerang to buy a second pass through an overinvested region. Place a splitter above a dense field because two balls in a good neighborhood beat one ball anywhere.

That’s the skill curve, and it’s a real one: the game is not about learning to predict the ball. It’s about learning to build machines that don’t need the ball to cooperate.

The draft-of-three cadence keeps this tense in a way a static builder wouldn’t. The game constantly offers pieces that are individually mediocre and contextually devastating, and the read — “is this good in the machine already on the board?” — is the actual game. The ELDER quota structure applies the pressure that makes those reads cost something. Without escalating tribute this would just be a toy. With it, the toy has to pay rent, and those two goals fight.

The Onboarding Problem

Ballionaire does not explain itself. This is the criticism that surfaces in nearly every negative review and it is entirely earned. Trigger descriptions tell what a trigger does in isolation. They do not tell how it interacts, what counts as a “bonk” for its purposes, whether it fires on spawned balls, whether it stacks, or whether the thing about to be built is a combo or a dead end. The game’s answer is trial and error — burn a run, learn a rule, repeat.

There’s a defense of this. Discovery is the pleasure here, and a game that front-loads a synergy encyclopedia has spoiled its own best moments. Balatro withheld plenty and was celebrated for it. But Balatro’s atoms were playing cards — a rule set every adult on earth already carries. Ballionaire’s atoms are bespoke gadgets with bespoke verbs, and expecting players to reverse-engineer the interaction matrix through run death is not elegant minimalism. It’s onboarding friction dressed up as design philosophy. The distance between “I don’t know what will happen” and “I don’t know what the words mean” is enormous, and the game keeps stumbling across it.

The players who push through find a system with real depth. The ones who bounce off aren’t wrong. They were failed by a tooltip.

The Ceiling

The second complaint — the source of that recent-review slide — is content exhaustion. Past the point where most games would still be introducing ideas, the seams show. The trigger pool has been seen. The strong openings are known. The boards stop surprising. The draft-of-three starts feeling less like a decision and more like a lookup.

The shape of the criticism should reassure newobject and worry them in equal measure: almost nobody is criticizing the core design. They’re asking for more of it. More boards, deeper meta-progression, a reason to keep going after the machine is solved. That’s not a game that failed. That’s a game whose players fell in love and ran out of road — and the Lua API and Workshop are clearly the intended answer. Whether a community-authored trigger economy can carry a game past its own content ceiling is the open question, and it’s the one that will decide whether Ballionaire is remembered as a great stretch of hours or a permanent fixture.

The art direction is loud, chunky, and cheerfully unserious — bright gadgets, thick outlines, a board built to read instantly at a glance even when it’s carpeted in machinery. That last part is not a small achievement. Readability is the hardest problem this game had to solve, and the design mostly solves it: the board is engineered so that even with dozens of triggers firing and balls in flight, what the machine is doing stays parseable.

Mostly. In a truly runaway cascade — the screen-filling kind the game is built to produce — legibility collapses into confetti. Numbers stack on numbers. The win is obvious; the why is not. That’s arguably the intended payoff and it’s still a genuine information loss, especially for a player trying to learn an interaction the tooltips wouldn’t teach.

The physics simulation is the technical star by design: crisp, consistent, and built so it never reads as cheating. That matters enormously in a game where the loss condition is a ball going somewhere it wasn’t wanted. And as a 2D game of this scope, it is light on hardware — the best moments involve spawning a hundred balls, and the engine is built to absorb that.

On Steam, mouse and keyboard is exactly right — placing a trigger is a point-and-click act, and there’s no aiming skill for a controller to fumble. But the more interesting fact is that Ballionaire is Steam Deck Verified, and the Deck is arguably its native habitat.

This is a game of short decisions and long spectacles. One choice, then fifteen seconds of watching. That rhythm is torture on a machine where a browser tab is one alt-tab away and heaven on a handheld held on a couch. The Deck turns Ballionaire’s spectator beats into the point rather than the pause. We recommend it there.

The Steam Workshop integration is the platform’s other real contribution — it’s the distribution layer that makes the Lua API more than a curiosity. Custom triggers are only interesting if they can find a player, and Workshop is how they do.

Ballionaire earned the word “autobonker” — a stupid word for a smart game. It found real space between Peggle and Balatro that nobody had noticed was empty, and it built something there that neither of those games could have made. The moment a machine outruns its builder’s understanding of it is the best idea in the roguelike genre right now.

But great ideas don’t excuse negligent teaching, and a game that hides its own rules behind dozens of failed runs is picking a fight with its audience for no reason. The tooltip problem is fixable. It should have been fixed before launch. The content ceiling is a harder problem, and newobject has bet the game’s long-term life on the Workshop community solving it — a bet that’s honest, ambitious, and unproven.

We rate it a Good: recommend it to the right audience, with eyes open. Expect a superb stretch of hours and a slightly wistful ending. That’s a better deal than most games at this price offer, and it’s a lot less than this one nearly was.