Crucible

Autopsy Report

Crucible holds a unique distinction in the graveyard: it’s the game that was so bad its publisher un-launched it. Developed by Relentless Studios and published by Amazon Games as the tech giant’s first major foray into game publishing, Crucible launched on May 19, 2020, collapsed within weeks, was pulled back into closed beta in July, and was permanently shut down by November. A 174-day lifespan from launch to death — the fastest AAA game lifecycle in modern gaming.

The 43% positive review score from 10,669 reviewers tells you everything. When more than half of all reviewers actively dislike a game from one of the richest companies on Earth, the problems aren’t surface-level bugs or balance issues — they’re fundamental design failures. Crucible tried to be three games simultaneously: a hero shooter (like Overwatch), a MOBA (like League of Legends), and a PvE experience (like Destiny). The result was a game that couldn’t answer the most basic question a player asks: “What am I supposed to be doing?”

Matches were confusing. Characters had MOBA-style ability kits but fought in shooter-style encounters. AI creatures roamed the map to be farmed for experience points, but their presence diluted the PvP tension rather than enhancing it. The store description — “Pick an alien, human, or robot hunter to fight in 4v4 matches” — is so generic it could describe any of a hundred team shooters. Even the marketing couldn’t articulate what made Crucible special, because nothing did.

The collapse was immediate. Crucible reportedly peaked at around 25,000 concurrent players on launch day — pathetic for a AAA F2P game backed by Amazon’s resources (Apex Legends hit 1 million in its first 8 hours). Within two weeks, concurrent players had dropped below 200. Amazon removed two of three game modes because there weren’t enough players to fill matches. Then came the un-launch: in July 2020, Amazon pulled Crucible back into closed beta, an unprecedented admission that the game should never have been released.

The 2-5 million estimated owners with a 187:1 owners-to-review ratio paints a damning picture of the F2P experience. Millions downloaded the free game, played one or two matches, and left without caring enough to even leave a negative review. At 149.2 reviews per month — a fraction of what comparable F2P launches generate — the game failed to inspire any reaction stronger than indifference.

The broader context makes it worse. Crucible launched one month after Valorant’s closed beta began dominating competitive shooter conversation. Fortnite was at its cultural peak. Call of Duty: Warzone had just launched. COVID-19 lockdowns meant more people were gaming than ever before — and all of them were playing established titles rather than giving Amazon’s confused experiment a chance.

Amazon eventually learned from the wreckage. New World launched in 2021 with a clearer identity (MMO), and Lost Ark publishing (2022) proved Amazon could succeed in gaming — just not by designing games by committee. Relentless Studios was shut down. Crucible was fully removed from Steam. And somewhere in the gaming history books, there’s a footnote about the time the world’s largest e-commerce company spent years and hundreds of millions of dollars making a game that lasted less time than a free trial.

Key Failure Factors

• Design-by-Committee Confusion: Hero Shooter + MOBA + PvE = nothing. The game’s tag soup (Hero Shooter, PvP, PvE, MOBA, Team-Based) reflects a design process where every stakeholder’s vision was included and nothing was edited down. The 43% review score — with 6,079 negative reviews — confirms players couldn’t find a coherent experience.

• Disastrous Launch Metrics: ~25,000 concurrent players at peak, dropping below 200 within two weeks. Two of three game modes removed for insufficient players. For a AAA F2P game with Amazon’s marketing resources, these numbers represent a total failure to acquire and retain players.

• The Un-Launch Heard Round the Industry: Pulling a live game back into closed beta six weeks after launch was an unprecedented move that confirmed the game was broken. It generated a news cycle of mockery and ensured no player would return. The un-launch was more memorable than the launch.

• Invisible in a Crowded Market: Launching one month after Valorant’s closed beta, during Fortnite’s peak, alongside Warzone’s explosive growth — Crucible had zero competitive positioning against established titles that all had clearer identities and deeper content.

Lessons for Developers

1. Money cannot substitute for design vision. Amazon reportedly invested hundreds of millions in Crucible and produced a 43% review score. No budget can fix a fundamental lack of creative clarity. The game tried to be a hero shooter, MOBA, and PvE game simultaneously and failed at all three because nobody committed to one vision.

2. An un-launch is worse than a bad launch. Pulling Crucible back to closed beta was intended as damage control but became the defining narrative. A bad game can be patched; an un-launched game is a punchline. The 174-day lifecycle from launch to permanent shutdown set a record nobody wants.

3. Tech company culture can be hostile to game development. Amazon’s engineering culture — data-driven, committee-based — produced a game designed by consensus rather than creative leadership. The tag confusion (Hero Shooter + MOBA + PvE) is the product fingerprint of a project where every feature request was approved and nothing was cut.

Related Deaths

• LawBreakers — Another expensive multiplayer game that launched into a saturated market with an identity problem and died rapidly, though at least it committed to being a shooter.
• Artifact — Valve’s card game that launched to strong expectations and immediately collapsed, proving that even gaming’s biggest names can’t guarantee audience adoption.
• Hyper Scape — Ubisoft’s battle royale entry that similarly launched without a clear identity and died within a year, confirming that publisher size doesn’t protect against design confusion.