According to Woofun AI, after a 21% surge in August, Bitcoin slipped into a two-month trading range, yet the DeFi sector staged an independent rally.
From August 22 to October 20, Bitcoin's daily closing price oscillated between $75,600 and $86,600, delivering an overall gain of just 5.6%, while Ethereum actually fell 2.7% against the trend.
However, among 19 leading DeFi tokens screened for having a token-holder dividend mechanism, as many as 14 managed to beat Bitcoin.
Among them, Raydium topped the list with a stunning 178% gain. The median gain for this group of tokens reached 12.9%, more than double Bitcoin's advance, signaling a strong preference among investors for projects with genuine profit-sharing capability.
Why the market is focusing on "real payouts to holders"
The deeper reason is that looking only at a protocol's total revenue can easily be misleading. Many protocols post rich headline income, but the profits are not returned to token holders; instead they are retained by the team or used for other non-dividend purposes.
Therefore, this study adopts DeFiLlama's data methodology to specifically measure "actual yield," which includes token buybacks and burns, fee burns, and earnings paid directly to token holders.
The screening criteria were extremely strict: only core DeFi projects were included, covering DEXs, perpetual contracts, lending, collateralized debt positions, liquid staking, yield tools, and yield aggregators, while public blockchains, oracles, and launchpads (such as Pons) were explicitly excluded.
In addition, qualifying tokens had to have a market cap above $100 million on August 22 and an actual dividend record within the past year. Price data came from CoinGecko, the statistical window was locked to August 22 through October 20, and 19 samples were ultimately selected.
The "market cap/annualized holder yield" metric mentioned in the article refers to dividing total market capitalization by all payouts to holders over one year at the current yield level. The lower this figure, the less investors are paying for each dollar of dividends, and the stronger the theoretical valuation appeal.
Woofun AI compiled data showing that when tokens were grouped and compared by dividend growth, the results differed markedly: the dividend-growth group posted a median gain of 18.3%, while the other group managed only 6.9%. The top-ranked projects all had clear logic for earnings growth.
Leading projects and their catalysts
Raydium saw a key turning point on September 5, when tokenized stock issuance platform StonkFun connected to its LaunchLab business, driving RAY token buybacks from $360,000 in July to $3.94 million in September, an 11-fold increase.
Uniswap benefited from massive trading volume brought by Robinhood Chain (HOOD.US), which contributed nearly half of the protocol's revenue in September, and combined with its existing fee-sharing mechanism, its performance was strong.
After merging with Velodrome, Aerodrome token holders can not only capture trading fees but also share in the fees projects pay to buy liquidity. This portion of revenue reached $14.9 million in September, exceeding the $10.5 million paid to liquidity providers for the first time.
Conversely, edgeX saw the actual yield paid to holders fall 15% and its token price plunge 22%, making it the worst performer in the sample and confirming that declining yields drag down token prices.
Valuation traps and outlier samples
Valuation traps and anomalous samples reveal the complexity of the market: a low valuation does not necessarily equal a high-return opportunity.
AERO had the lowest valuation multiple at just 8.7 times, and its token price soared 77%, while edgeX at 15 times was the second-lowest in the sample yet finished at the bottom. This shows that relying solely on high or low valuations cannot predict price moves.
Some tokens defied conventional logic: Hyperliquid (HYPE) saw the yield paid to holders jump 79%, but its token price rose only 6%. Measured by fully diluted valuation (FDV), its valuation reached as high as 85 times, and roughly $800 million of HYPE unlocked and flowed out each month, while monthly buybacks totaled only $75 million, with the massive selling pressure capping the token price.
During these 59 days, Hyperliquid distributed $146.9 million in yield to holders, accounting for 59% of the sample's total yield, while the team unlocked about 9.9 million HYPE per month.
Sky (SKY) passed a governance vote on August 13 to use 55% of each period's yield for token buybacks and distribute 45% to stakers. Holder yield quadrupled, but the token price rose only 11%, and its valuation multiple fell from 98 times to 28.5 times.
Aave, affected by the rsETH security incident, suspended buybacks and paid no dividends during the statistical period, yet its token price still rose 24%. Lido's buybacks must be executed uniformly through a DAO vote, so a low single-month yield does not mean dividends have stopped. Meteora's dividend function was only recently launched, so there is no historical data for comparison. THORChain's RUNE burn ratio was cut from 5% to 1% on September 12, and its token price movement was unrelated to dividend growth logic.
How to approach DeFi tokens if Bitcoin keeps ranging
Facing a possible continued sideways move in Bitcoin, the investment strategy for DeFi tokens needs to return to fundamentals. An effective stock-picking method must satisfy three conditions at once: first, the project's actual yield paid to token holders is trending upward; second, there is a clear positive catalyst, such as a new business launch, a surge in trading volume from fee-sharing, or a protocol merger; and third, token unlocks and fully diluted valuation must be strictly reviewed. Hyperliquid's case warns us that large unlock-driven selling pressure can severely cap price gains.
It should be noted that this study includes only 19 token samples over a 59-day statistical window, so the conclusions are not universally applicable. If the observation period were extended, the rankings of mid-tier tokens could change substantially. This method is only an observational indicator and does not mean that rising tokens could have been precisely selected on August 22.
The valuation multiples in the article are calculated based on circulating market cap. Many projects have large locked positions, which makes valuations appear cheap. For example, Hyperliquid's circulating market cap valuation is 19.6 times (circulating market cap of $18.1 billion), but its FDV valuation is as high as 85 times (FDV of $77.8 billion). Although DeFiLlama data is already the industry's highest-quality source, it is not 100% accurate, and investors need to combine multiple dimensions of data for comprehensive judgment.
Comments