Recently, I've seen quite a few discussions about the Olympus DAO project, and it’s indeed quite interesting. Coming from the DeFi 1.0 era of liquidity mining, these days projects are starting to play new tricks.

Speaking of Olympus DAO, many people's first reaction is the high APY and the (3,3) meme. Indeed, since its launch in 2021, this project has attracted a lot of attention with its highly creative economic model. What it aims to do is quite ambitious — to create a decentralized stable asset, rather than relying entirely on centralized institutions for backing like USDT or USDC.

The core mechanism is actually based on using DAI as an anchor, allowing OHM to float freely based on its intrinsic value relative to 1 DAI. Sounds good, but in practice, it’s much more complex. Olympus DAO introduced a bond mechanism, allowing users to buy discounted OHM with assets like wETH and DAI, while also earning yields through staking. This design is indeed innovative and has attracted many.

But the game theory logic inside is the key. The (3,3) concept borrows from the Prisoner’s Dilemma model; in theory, if everyone participates in staking, it can lead to a win-win situation. The problem is, this model relies entirely on rational decision-making and collective consensus from participants. Once someone chooses to exit, the entire system begins to destabilize.

Looking at historical data, you can see it clearly. Olympus DAO’s TVL once surged to $860 million, with APYs exceeding 1000%, seeming invincible. But with market sell-offs, TVL plummeted to $260 million, a 70% drop. The whale sell-off of over 80k OHM in early 2022 even directly triggered a price crash.

In essence, the price of OHM is still maintained by new buyers entering the market. The project itself has no real application scenario or cash flow support; users buy OHM purely to earn that high APY. But high APY inevitably means high inflation, and high inflation will eventually lead to a price decline. This is a classic death spiral.

Comparing it to staking protocols like Lido reveals the difference. Lido users stake ETH and receive stETH 1:1, which has actual use cases. But Olympus DAO’s staking rewards are in OHM itself, and the value of OHM is entirely influenced by market sentiment. The risk factors are on completely different levels.

My view is that projects like Olympus DAO are still in the early experimental stage. They require all participants to maintain the (3,3) spirit; once a traitor appears, it collapses. In the fast-changing crypto world, maintaining this kind of dynamic balance is extremely difficult. Users seeking high returns through Olympus DAO must be prepared to bear the corresponding risks.