Mostly no. A play-to-earn game is sustainable only if what players spend inside it (crafting, upgrades, fees, burns) pays for the tokens it hands out, and most of the 2021 to 2022 cohort never got there.

A report from trading firm Caladan, cited by CoinDesk in April 2026, says roughly 93% of GameFi projects are effectively defunct. It says more than 300 blockchain games have shut down and that gaming tokens are down about 95% from their 2022 peaks. Exceptions are possible in principle and rare in practice. So the useful question is not whether the genre is sustainable. It is whether the one game you are about to fund with an NFT or token purchase is, and that can be checked from public data.

How the math runs

Every reward token has a tap and a drain.

  • The tap is emissions. These are tokens minted per match, per quest or per day, plus team and investor allocations that unlock on a schedule.
  • The drain is sinks. These are tokens spent on breeding, crafting, upgrades, repairs and fees, some of them burned outright.

When the tap runs faster than the drain, circulating supply grows. If buyers do not grow with it, the price falls. A session that pays the same number of tokens then pays fewer dollars. The players who came for the dollars leave, which shrinks the drain, which widens the gap.

The variable that decides the outcome is where rewards are funded from. There are three possible sources:

  1. Newly minted tokens.
  2. Money from newer players buying in.
  3. Revenue from people who spend because they enjoy the game.

Only the third survives a slowdown in new signups. Caladan's conclusion is that the model assumed emissions and speculation could substitute for player demand, and that this was a structural flaw rather than a bad cycle.

The tap can run hotter than most players realize. Ronin, the chain built for Axie Infinity, ran RON issuance above 20% a year until May 2026, per AMBCrypto.

Running the check on one game

  • Emissions and unlocks

    From the token's own docs or unlock tracker, write down four things: max supply, circulating supply, daily reward emission and the vesting calendar. The gap between circulating and max is supply that has not hit the market yet.

  • Sinks

    List every place the token gets spent. Mark which sinks you must pay to keep playing and which are optional. Also note whether a sink burns the token or routes it to a treasury, where it can come back out. A cosmetic shop nobody uses is not a sink.

  • Who pays your reward

    If the answer is the emission schedule or the treasury, you are on a timer. If it is player spending, find the revenue figure.

  • Direction of activity

    DappRadar [shut down on 17 November 2025](https://www.blockchaingamer.biz/news/40900/dappradar-shutting-down/), so use the studio's own published dashboard if it has one, and a chain-level page such as DefiLlama for the game's chain. Look at the trend in daily active addresses over several months. The trend matters more than the level. For a baseline, sector-wide daily active wallets fell 6% quarter over quarter to 5.8 million in 2025, per an SNS Insider report.

  • Studio runway

    Check whether the team has shipped before and what it has in the bank. In a Blockchain Gaming Alliance survey, 32.6% of developers named cash shortages as their top challenge.

The weighting is close to the one used in scoring DeFi protocols for risk. Supply schedule and revenue matter more than anything on the roadmap.

Three that failed it

Axie Infinity. This is the textbook case. Daily active users fell from 2.7 million to roughly 5,500, according to the Caladan report. None of the sources here carry dated SLP emission and burn figures, so the collapse can be shown through its player count but not reconstructed token by token. That gap is worth noting in itself, because the numbers that would have warned players were never easy to find.

Pett.ai. This is the clearer recent example. The game managed emissions with an "AI Central Bank" that adjusted rewards for inflation. According to EGamers.io:

  • AIP peaked at $0.04323 on January 22, 2025.
  • It sat at $0.0001584 by the shutdown announcement on August 22, 2026, a 99.6% decline.
  • About 349.51 million of a 500 million max supply was circulating, with a market cap near $55,300.

The servers went dark September 7, two weeks after the announcement. An algorithm that tunes the tap cannot create demand for the drain.

Fishing Frenzy. This one fails a different way. It had 9 million installs, a peak of 25,000 daily active users and more than $1 million in revenue. Its studio still shut the servers on June 25, 2026, saying it could not find product-market-business fit. Revenue passes the check only if it is large relative to what the game costs to run.

The counterexample, such as it is

The brief called for a game whose sinks demonstrably held, with dated figures. The sources here do not contain one, and inventing one would defeat the point of the article.

The closest case is Ronin's own fix. On May 12, 2026, alongside its move to an Ethereum OP Stack Layer 2, Ronin made two changes:

  • It cut RON inflation from above 20% to below 1%. The team calls this a "20x reduction in new token emissions."
  • It raised the marketplace fee going to the treasury from 0.5% to 1.25%.

That is the right shape: less tap, and more of the treasury funded by fees. It is also a chain token rather than a game's reward token, and it is only a few months old. Whether fees keep pace with what the ecosystem pays out is not yet known.

What it is not

Often mistaken for sustainability Why it is not the same
Revenue YGG Play peaked at $3 million a month in October 2025 and was still sunset by August 1, 2026.
Big-name backing Proof of Play raised $33 million led by a16z and shut down on August 4, 2026. PIRATE's market value was about $803,000 at the time.
Heavy onchain usage Proof of Play's networks were among the two largest gas consumers on Ethereum rollups at peak. Pirate Nation's daily users fell from 100,000+ to about 10,000.

Before you buy in

Ask one question and do not accept a roadmap as the answer: who pays my reward? If it is emissions or the next wave of players, treat the entry cost as money spent on a game, not money invested. Size it so a two-week shutdown notice like Pett.ai's costs you nothing you would miss.

If the answer is revenue, ask the studio for the figure and see whether it holds up. DappRadar, which used to show game revenue and active wallets, shut down on 17 November 2025, and we found no free replacement for its per-game numbers. A chain-level page such as DefiLlama's Ronin page shows active addresses, fees and revenue for a whole chain, which tells you whether the chain a game runs on is growing, not whether the game is. Then compare it against the games that pass or fail this test in our rankings.