The most successful batch of projects in this cycle actually reversed things at a critical point Using automatic repurchase, the price curve and users were confused together A typical case is the early cycle darlings HYPE, ENA, JUP At a completely unfair level of valuation multiples, using extremely exaggerated prices to directly invest tens of millions of funds in repurchases is equivalent to using a treasury to lift a sedan chair for the market at the top As soon as the price rises, the narrative is ignited, driven by price Retail investors see the trend and narrative strengthening synchronously, FOMO rushes in to take over the market, and is finally harvested at the top This is not an individual judgment error, but a collective misjudgment created by the mechanism Worse still, many founders have been fed too much chicken soup by this self reinforcement, starting to believe that those multiples are reasonable because the price temporarily proves them The short-term feedback from the market is regarded as a certification of long-term value, which is the starting point of all subsequent contradictions Subsequently, the price entered a downward trend for several months, unable to return to its previous high, and there is no clear path for it to return to its previous high So a second, equally erroneous attribution emerged Some people are starting to blame the mechanism, saying that the price keeps correcting from the previous high level, repurchasing is useless, and buying back is meaningless This statement is also incorrect It has forgotten a financial common sense that has been repeatedly verified for hundreds of years: Repurchase is not an anchor that keeps prices at the wrong level forever, it is just a tool for capital allocation You try hard to buy when it's expensive, only to end up wasting your funds in the worst possible way Enlarge the foam at the top, and then appear powerless in mean reversion So the discussion should not stop at whether repurchases are useful, but return to a harder constraint condition If the project's income cannot even support development, then of course, don't spend limited funds on tokens, focus on supporting development and products first But once the project is successful and has sustained and stable income, as the holder, I only ask one question: If the token has neither dividends nor repurchases, and at least does not have a very clear financial purpose, what is the purpose of the token's existence If a token has no cash flow path and no clear financial utility, it can only prove itself by higher prices, which is not value, but self hypnosis Therefore, I give a more detailed solution to the debate on whether to repurchase: repurchase must be done, but the intensity and price must be bound, making repurchase more aggressive when it is cheap and more restrained when it is frenzied The repurchase amount varies with the price, which is a correct target When the price is cheap, it is advisable to repurchase as much as possible, because the same amount of money can take away a larger proportion of the supply, and the effect of supply contraction is exponentially enhanced Slow down when the market is too hot, avoid chasing high with the treasury, and avoid blowing the narrative out of control Some founders are more accustomed to the traditional way of companies, where the CEO or management makes decisions to repurchase at the last minute, and makes ad hoc management actions based on the timing of operation and valuation. This is completely reasonable because companies in the real world have always been like this Ad hoc makes decisions based solely on the situation But if the protocol is more decentralized, or transparency, predictability, and even legal risks are prioritized, a programmatic buyback framework is needed to lock in human nature at the regulatory level while giving the market predictable boundaries Here is a simple programmatic approach: using the price to earnings ratio as a valve, allowing the repurchase ratio to be determined by the valuation range Each protocol can design details according to its own situation, and a feasible example is as follows Perform exponential moving average on income, with half-life determined by project selection, and then annualize this EMA income as earnings Then, the income generated per day or per block is divided into PE tiers based on the token price that can be reached at the time of repurchase When the PE corresponding to the repurchase is less than 4, repurchase 100% PE repurchase 75% from 4 to 6 PE repurchase 50% from 6 to 8 PE repurchase 25% from 8 to 10 PE higher than 10, no repurchase The core of this approach is to transform buybacks from emotional behavior to value constrained behavior, striking hard when cheap and stopping when expensive At the same time, all income balances that have not been used for current repurchases enter the reserve pool for repurchasing in the event of a real downturn The repurchase rhythm of the reserve pool only depends on the price EMA, for example, using the past 90 day price EMA as a reference, the lower the price is below this moving average, the faster the repurchase speed This is equivalent to a structured downside protection and also a discipline of buying and falling Yes, this plan is more complex and requires more sophisticated financial engineering than an all or nothing automatic repurchase But after the successive failures of narratives such as Web3 gaming, Web3 socializing, and metaverse, it is now clear that the core of Crypto is finance, and finance is also being Cryptoized When you are working on a serious project and still use childish capital allocation logic to handle repurchases, you are essentially managing mature assets in an early entrepreneurial manner You may not have financial experts in the team, but at least you should hire top external consultants or professional institutions to get the mechanism right