Web3 gaming burned through up to $15 billion chasing a token-driven future that gamers never bought into.

Data from Caladan, a market-making and trading firm, shows roughly 93% of so-called GameFi projects are now effectively dead, with token values down about 95% from their 2022 peaks and funding to studios collapsing 93% by 2025.

Investors and studios poured billions into tokens and non-fungible tokens (NFTs) before building blockchain-based games containing tradable properties. Then capital shifted into AI, asset tokenization and infrastructure, and more than 300 games shut down, turning Web3 gaming into a cautionary tale about chasing speculation over product-market fit.

"Capital was destroyed at every layer simultaneously," the report states, pointing to venture capital, retail NFT buyers, gaming guilds and Telegram's 300-million-user tap-to-earn wave as parallel casualties. Hamster Kombat alone lost 96% of its users within six months of launch. YGG, the flagship gaming-guild token, trades 99.6% below its November 2021 peak.

Individual post-mortems are brutal. Pixelmon raised $70 million in a 2022 NFT mint and, four years on, still has no public game. Ember Sword burned through $18 million over seven years of development before shutting down last May with no refunds. Gala Games is embroiled in a lawsuit alleging its co-founder diverted $130 million in tokens. Square Enix quietly wound down its Symbiogenesis experiment last July.

Structural mismatch

The failure wasn’t just a bad cycle or weak execution. The data indicate it was a structural mismatch between a model built around financial incentives and an audience that consistently signaled it wanted entertainment instead.

At the heart of the boom was GameFi, the play-to-earn model that turned gameplay into a financial feedback loop.

Players bought tokens or NFTs, earned rewards in those same assets, and cashed in as long as newcomers kept piling in. Once the inflows slowed, the math broke down. Token prices slumped, rewards thinned out, and users walked away — dragging entire in-game economies down with them.

Axie Infinity, the sector's one-time flagship, watched daily active users crater from roughly 2.7 million at the peak to around 5,500 today, according to DappRadar data.

The demand side never caught up with the flood of capital. Even at the height of the mania, just 12% of gamers had tried a crypto game, according to a Coda Labs survey, cited by Caladan.

Capital allocation made the problem worse. Studios raised tens or hundreds of millions of dollars before shipping viable products, removing the pressure to build games that could retain players.

The most telling data point may be where the money went instead. Gaming commanded 62.5% of all Web3 venture investment in 2022; by 2025, its share had collapsed to single digits as AI, real-world-asset tokenization and layer-2 infrastructure absorbed the displaced capital.

Even Animoca Brands, the sector's most prolific backer, has cut gaming to roughly 25% of its portfolio and is pivoting to stablecoins, RWAs and AI.

At the same time, development timelines stretched three to five years, while tokens traded in real time and demanded constant momentum. By the time many projects were ready to launch, their associated tokens had already collapsed.

The result is a sector that expanded rapidly on speculative demand and contracted just as quickly when that demand faded. More than 300 blockchain games have shut down, according to DappRadar, and remaining investment has shifted away from titles toward infrastructure.

What was once pitched as the future of gaming now looks more like a cautionary example of what happens when financial engineering runs ahead of product market fit.