Play-to-Earn Hangover: After the Crypto Gold Rush Ended

The California gold rush made some prospectors rich, but it created greater riches for shovel sellers, saloon keepers, transporters, and land speculators. Favorable geometries were simple: the poorer the river-walker, the deeper the pockets of those selling spades. The crypto-native equivalent was pioneered by play-to-earn designers: guild owners, VCs, token insiders, exchanges, and market makers gathered at the bar. Players with sore backs, cheap androids, and empty pans received the lion’s share of the gold.

This is where the play-to-earn narrative needs to begin: not with the lie about the kid playing Axie until dawn, but with the con that underpinned it. Unemployment, pandemic-related monotony, cheap phones, and token rewards made “gaming” cool for the first time for hundreds of millions. For desperate Filipinos, Venezuelans, and would-be entrepreneurs, Axie Infinity’s “play-to-earn” pitch was an invitation to join a revolution. The chance to battle cute little monsters to earn SLP tokens and turn them into dollars was irresistible poverty alleviation. The industry’s fatal weakness was in calling what amounted to a token issuance in game fuel a “play-to-earn” opportunity; the value capture mechanics were tilted to the advantage of those selling.

Gold rush narratives are forever about the future. You are not digging your own pan; you have staked your claim to the mother lode. You are not unemployed; you are unbanked. You are not playing a game; you are building a nation. Axie Infinity’s token sale talked about “residents of Lunacia.” Guild managers referred to their scholars as “players.” VCs hyped up multi-billion market caps for a proto-economy and sweat equity. It was all much easier to get away with because everyone was lying to themselves most of the time. If someone was selling something to someone else, there is always a bigger margin for self-delusion on the other side of the transaction.

In fairness to the token promoters, it was not entirely their fault that scarcity pressures and user growth made Axie’s economy all gas and no go. The math was brutal, but it was the sort of math that makes sense to anyone who has seen the long tail of an ICO euphoria. SLP tokens were distributed to players, and the more players there were, the more SLPs there had to be, and the less each one was worth. The game’s economy was designed to prioritize volume over value capture for the majority. And so, the wider world’s relentless march toward tokenization collided with crunch time in the Philippines and a pandemic freeze on everything else.

The problem with play-to-earn games is that they do not reward people for playing them. Not really. The best Axie Infinity could manage was to provide a wage for hundreds of thousands of people for a brief period. It was exhausting, demoralizing, and tokenized poverty alleviation. It is not that the math was wrong; it was that the macroeconomic conditions were permanently favorable for those selling at the top of the pyramid and permanently unfavorable for everyone else. The long tail for Axie’s “players” was always going to slope downward because their issuance schedule and virality mechanism made it so.

None of this is to deny the human cost of Axie’s collapse or the immensity of the financial wreckage it caused for hundreds of thousands who placed their trust in its SLP.

Axie Infinity’s “scholarship” system meant that many workers were effectively employed by their managers — and the latter’s compensation, at 50 percent in many cases, was far steeper than minimum wage. The most visible representative of this system was Samerson Orias, a line cook at a Takoyaki bar in the Philippines who sold his food to pay rent and sign up with an Australian manager for Axie.

He had to play five to six hours to earn 120 SLP, half of which went to his provider. Friends had told him that he could make hundreds of dollars a month, and he could afford to try. Stories like Samerson’s were the ones that fueled the industry’s “jobs platform” narrative, the ones that suggested that the P2E sector was doing something responsible by enabling these scholarships.

That perspective does not survive deeper inspection. A job, by its very nature, should not be dictated by supply-demand dynamics or the whims of a third party. If Samerson’s income depends on his SLP yield, which in turn depends on how many other scholars are earning SLP, then his job is not as stable as it is advertised. By Naavik’s research from late 2021, Axie scholars were seeing yields drop below the threshold of the Philippines’ minimum wage — for everyone but the top 1 percent. The problem was not that play-to-earn made players lazy; it was that the token economics made the labor market self-regulating.

Rappler’s investigation offered a peek into the industry’s other side – the managers’ experience. McGlay, looking to invest in a team of Axie scholars, saw his millions of pesos evaporate. Christopher Cruz, a Filipino businessman and manager of around 200 scholars, stated he made as much as 600,000 pesos per day by taking 60 percent of his scholars’ earnings. They were all students or residents of the region’s rural areas, eking out a living for about 450 pesos per day – just enough for a minimum wage. Then the SLP dropped, 150 SLP turned into 50, and his scholars’ earnings failed to meet the threshold.

Meanwhile, the Ronin bridge heist represented a different kind of industry reckoning. The exploit, which had stolen 173,600 ETH and 25.5 million USDC from the Ronin Network smart contract, marked the equivalent of nearly 625 million USD in damages. Sky Mavis’ 150 million raise, led by Binance, has since reimbursed users, with the bridge being refilled to capacity using the proceeds and the company’s funds. The transaction, as well as the overall transparency, may have helped the firm retain much of its user base. Still, the Ronin heist demonstrated how much the industry had relied on the confidence of its users.

Trung Nguyen, CEO of Sky Mavis and co-founder of Axie Infinity, was no longer the star attraction. The play-to-earn ecosystem’s most visible representative was now talking about tokenomics, community management, and the rebirth of Axie as part of the broader Ronin network. The same went for Aleksander Larsen, the co-founder and long-time business lead, who stepped down from his executive roles to focus on the security space. His comments about the Ronin exploit were the most telling: a major play-to-earn studio had just suffered a state-sponsored attack, and one of its founders was now in the cybersecurity space.

Jeffrey Zirlin, also known as Jihoz, was among the few public figures in the space talking honestly about the challenges facing the P2E space. By announcing that he and his team would halt SLP emissions in Axie Origins from January 7, 2026, Zirlin seemed to understand the limitations of bot farms. The move was largely motivated by fears of rampant botting and wash trading, as stated in the developer’s public letter. In 2025, SLP has enjoyed a net burn of 250 million, and Zirlin believes that a healthier SLP supply will benefit the whole ecosystem.

The token that had propelled the P2E concept to prominence was set to become a “deflationary force,” and the company was preparing to enter the next phase of its development as a game chain and a metaverse infrastructure provider. This was not the end of Axie Infinity, but its transformation from a nation-state simulation to a more conventional game design with better tokenomics. If the play-to-earn market’s lesson was anything, it was that the token distribution model is frequently far more important than the game itself. The more users, the more supply, but the more supply, the more competition for value.

Guild Spread and Greater Fool

Yield Guild Games was one of the first P2E organizations to formalize the scholarship system, and it has since become the labor platform’s preferred form of worker exploitation. Gabby Dizon, a Filipino game developer, recognized the need to organize the scholarship during the pandemic and launched the blockchain-based game guild. YGG’s community managers trained the scholars while also collecting a cut of their earnings.

The venture capital firms and founders thought it was a fantastic idea. Yield Guild Games had more than 4,700 scholars and generated over 8.6 million in profits for its members, with disbursements reaching more than a million each week, according to Andreessen Horowitz. The a16z partners highlighted YGG’s capacity to provide much-needed jobs in emerging economies throughout their blog posting. The industry’s most successful “gaming” organization, therefore, did not simply promote itself as a traditional employer. While most guilds only rented NFTs to players and collected a sizeable royalty, YGG appeared to provide a more accessible entry point to the P2E world. Everyone could join and try their luck, but few would succeed beyond the initial scholarship.

Everything worked out for the first time, and everyone was eager to acquire more shares. As the YGG token rose, so did the number of guilds and the value of their NFT collections. People who purchased the token were greater fools for believing that YGG’s P2E ecosystem would generate reliable earnings for its growing number of scholars. The entire system was constructed on a solid consensus about the value of the underlying technology. VCs were correct to believe that P2E could disrupt the traditional gaming industry and offer a new source of income for emerging-market residents. Even if the scholars eventually graduated and moved on to other opportunities, a steady stream of young individuals eager to earn would always be available to fill depleted ranks. The agreement was, in many ways, ideal for all parties.

That is, till it wasn’t. The “greater fool” theory of value has a tendency to fail when the fool stops being greater. For the majority of the P2E industry’s players, the transition from a speculative asset to a yield-generating tool was anything but smooth. If we accept that game design is not the primary source of value capture for play-to-earn studios, we have to accept that the “earn” component of the proposition was the key. The more scholars, the more supply, and the less valuable each SLP token would become. It was a self-regulating system, but one that placed too much emphasis on scarcity and competition.

This is how STEPN’s token rush was revealed to be a bubble. The “move-to-earn” platform’s token, GMT, peaked well above $4 but failed to defend its spot and eventually dropped below $0.007 by mid-2026 — a nearly 99.8 percent decrease. Find Satoshi Lab’s co-founder and CEO continued to release new NFT-based products to his followers, such as the STEPN GO running app and the GMT Pay payment system, but the magic was already gone. When the company’s representatives admitted that they would have to cut earnings in half to combat inflation, the “move-to-earn” industry’s viability was questioned. As a “move-to-earn” token, GMT was no longer a viable option for advertising.

YGG’s story had a comparable conclusion. YGG’s token, which had seen a more than 99 percent drop from its peak, eventually failed to attract a large number of scholars. With the shift toward the metaverse, yield farming, and Web3’s broader creator economy, the YGG scholarship system had limited appeal to emerging-market workers. The token’s value proposition no longer made sense in the context of the company’s transformation. On the other hand, the shift from NFT lending to game publishing was relatively seamless. The YGG representatives continued to highlight the revenue from LOL Land, as well as the value of the organization’s treasury, which still contained tens of millions of dollars. The game studio’s survival was not in question, but its ability to draw in the same number of scholars was.

Ronin’s current condition is far more intriguing than the previous two examples. In addition to its network’s security issues, Pixels, a farming and social massively multiplayer online game transitioning to Ronin, became one of Web3’s most popular games by user counts, with particular traffic sources indicating that it regularly hits more than a million daily active users.