Hand of Fate 2

Hand of Fate 2 is a game of wildly unequal parts, and the reason it works is that the good part is so good it drags the mediocre part along behind it like luggage.

Defiant Development — a small Brisbane studio that shipped the original Hand of Fate in 2015 — built a sequel that understands exactly what people loved about the first game and what they merely tolerated. It launched on PC, macOS and Linux on November 7, 2017, at $29.99, and later made the jump to Nintendo Switch, a platform its structure fits naturally.

The setup is theater. You sit across a table from the Dealer, a hooded card-master with a voice like aged whiskey and a century-old grudge. He lays out your journey as a board of face-down cards. You walk it. Cards flip. Sometimes you find a merchant. Sometimes you find bandits. Sometimes you find a moral choice that the game will remember and quietly bill you for four encounters later.

Where the first game was a proof of concept — clever, thin, repetitive — the sequel is the actual product. Twenty-two structured challenges replace the original’s endless-ladder monotony. Companions arrive with their own personalities, board perks and combat behaviors. Mini-games diversify from a single mashy quick-time event into pendulum timing, dice pushes and shell-game card shuffles. And critically, the deck-building layer now has teeth: you choose which encounter, equipment and companion cards enter the deck before every run, which turns preparation into a genuine strategic act rather than a menu you click through.

The market positioning is unusual and it’s why the game endures. This is not a deckbuilder in the Slay the Spire sense — you don’t play cards in combat. It’s a roguelite dungeon crawler that uses cards as its map, its inventory, its narrative engine and its random number generator simultaneously. Nothing else does that. Its critical reception — a Metacritic of 80 — reflects an audience that found something it couldn’t get elsewhere and forgave the game its sins accordingly.

• The Living Board: Each challenge is a physical layout of face-down cards you traverse token-by-token. Fuel and food deplete as you walk. The board is the map, the timer, and the tension all at once — a pacing device few roguelites attempt.
• Pre-Run Deck Construction: Before every challenge you seed the deck with earned encounter, equipment and companion cards. This is the game’s real strategy layer: you’re not optimizing a build, you’re curating the probability space of your own story.
• The Dealer: A narrator, antagonist, and dungeon master rolled into one, delivered with a performance that reads every one of your bad decisions back to you with audible delight. He is the reason people finish this game.
• Skill-Based Fate Resolution: Outcomes hinge on pendulum-timing swings, dice rolls and card-shuffle gambles rather than blind RNG. It’s a small design decision with an enormous psychological payoff — you lose by failing, not by being unlucky, even when the math says otherwise.
• Companions With Dual Utility: Recruitable allies grant both board-level perks and battlefield support, so party selection ripples across two completely different systems.
• Deep Unlock Economy: Hundreds of artifacts, items and encounter cards enter the pool over time, meaning late runs draw from a materially different universe than early ones.

The Gameplay Loop

The loop is a nesting doll, and that’s the design’s central achievement.

Outer layer: you pick a challenge and build a deck. Middle layer: you walk a board of face-down cards, managing food, gold and health as a resource economy. Inner layer: individual cards resolve into shops, dialogue trees, mini-games or combat. Each layer feeds the one above it. A mini-game flubbed at the middle layer costs you resources that constrain the outer layer’s next run.

What makes it sing is information asymmetry. You know what’s in the deck because you built it. You don’t know where. So every step across the board is a decision made under partial knowledge — you’re not gambling blind, you’re playing odds you personally set. That is a fundamentally different emotional experience from standard roguelite randomness, and it’s why losses here sting differently. You built the deck. You chose the risk. The Dealer will remind you.

The fate cards deserve specific praise. When an outcome is uncertain, the game shuffles a small set of cards face-down and asks you to pick the good one — or asks you to stop a swinging pendulum inside a narrow window. Statistically, this is a coin flip dressed in ceremony. Psychologically, it’s transformative. Interactive randomness feels earned in a way that a hidden dice roll never does. Defiant understood something about player agency that studios with fifty times the budget still get wrong.

Where It Breaks: Combat

And then a fight card flips, and the game you’re playing changes.

Play drops into a real-time action-RPG arena — dual daggers, two-handed hammers, parries, a combo meter feeding special attacks against soldiers, mutants and raiders. It’s an unmistakable Batman: Arkham derivative: rhythmic counter-timing, crowd management, prompt-driven parries.

It’s fine. It is aggressively, persistently fine.

The animations lack weight. Hit feedback is soft. Enemy variety is real on paper — mutants and raiders behave distinctly enough — but the encounter design rarely exploits it, and once the parry window becomes muscle memory, the melee stops asking questions. The combat isn’t broken; it’s underwritten. In a game where every other system is layered, considered and surprising, dropping into a flat brawler feels like the Dealer paused mid-story to make you do chores.

The structural problem is worse than the mechanical one. Combat consumes a disproportionate share of runtime relative to how much it contributes. Every minute spent in a mediocre arena is a minute not spent at the table, where the game is at its strongest. Defiant clearly knew this — the sequel’s fights are meaningfully better animated and more varied than the original’s — but improving a weak pillar isn’t the same as fixing the load-bearing math.

Difficulty and the Luck Problem

The challenge structure introduces spikes. Some challenges impose conditions — protect an NPC, survive on limited food, complete under a token limit — that interact badly with deck randomness. A run can be functionally lost in its first few cards, and the game will make you play it out anyway.

Defenders will say roguelites are supposed to do that. They’re half right. The issue isn’t that runs fail; it’s that failure here can be legible before it’s final. You can see the run is dead and still have a long stretch of walking ahead of you. That’s a pacing bug wearing a design-philosophy costume.

Interface and Onboarding

The UI is one of the game’s quiet triumphs. Card layouts read clearly at a glance. Resource counters sit where your eye already is. The transition animations between board and combat — cards physically swelling into 3D arenas — are the connective tissue that makes two disparate systems feel like one object.

Onboarding friction is low. The Dealer teaches through narration rather than tooltips, which is both good design and good writing. Newcomers do not need the first game; the century-later setting is a genuine fresh start.

The art direction is the point. Skeuomorphism done with conviction — this is a game about a table, and it commits to the table. Cards have thickness and grain. Tokens have weight. The Dealer’s hands move like hands, not like animation rigs. The whole presentation sells the fiction that you’re sitting across from a person who is enjoying your suffering.

The 3D combat arenas are the weaker half of the aesthetic. Environments are functional and slightly generic — competent fantasy dressing that doesn’t match the tactile specificity of the tabletop layer. Texture work is serviceable. Lighting is flat.

As a 2017 release with modest requirements, the game is light on hardware, and the macOS and Linux builds are real ports rather than afterthoughts.

Steam (PC/Mac/Linux): The definitive version. Mouse-and-keyboard is excellent for the board layer — precise card selection, fast navigation, no cursor drift. Combat, however, wants a controller, and the game supports one fully. The optimal setup is a hybrid: mouse for the table, gamepad for the fights, which is awkward and which nobody will actually do. Pick the controller and accept slightly slower menus.

Nintendo Switch: The port makes an argument for itself that the PC version can’t. This is structurally ideal handheld design — a run is a discrete session, encounters resolve in minutes, and the card-flip cadence suits interrupted play. It’s the version we recommend to anyone who plays in twenty-minute increments.

The compromises are real, though. A 3D game of this scope ported to Switch gives up resolution and stability in busy combat arenas, and the longer waits between board and battle blunt the rhythm the board layer works so hard to establish. Text is legible undocked, which is not a given for a card-heavy game. Verdict: buy it on Switch if you’ll play it portable, on Steam if you won’t.

Hand of Fate 2 is a game worth handing to a designer as a teaching text. Its structural ideas — interactive randomness, player-authored probability, a narrator who functions as antagonist and UI simultaneously — are more inventive than most of what shipped with a hundred-million-dollar budget that year. Nearly a decade on, nothing has replicated the specific thing it does.

It’s also a game with a hole in the middle of it. The combat that occupies a large share of playtime is competent filler attached to a brilliant frame, and every minute spent in it is a minute of wanting to get back to the table. Defiant improved it over the original and still couldn’t make it interesting, which suggests the problem was never execution — it was the decision to build a brawler at all.

Buy it for the right reasons. The Dealer is worth it, and the table is worth more.