How Crypto Protocol Can Make Millions While Its Token Falls

DWF Labs’ November 9 report puts the gap down to delivery rather than earnings. The market maker and investor found that 13 of 15 tokens with an active or announced value-accrual mechanism beat Bitcoin this year, and it singles out SYRUP and LDO as tokens whose protocols shrank their buybacks or made them conditional.

Revenue has to clear four checkpoints before it reaches a token: the revenue figure a rule actually uses, the rule that turns it into a budget, whether that budget gets spent and where the bought tokens end up. Maple and Lido stall at different checkpoints.

Maple’s Business Grew While SYRUP Lost Nearly Half

Maple’s lending business expanded through a year in which DeFi lending shrank. The October Maple Memo reports $4.8 billion in assets under management and $1.67 billion in loans outstanding, up 12% from a year earlier, while the company says the wider DeFi lending market contracted 29%. Its financials page shows $21.37 million of protocol revenue over the past 12 months. DWF puts SYRUP down about 45% year-to-date.

Very little of that revenue reached the token for much of the year. Under MIP-019, the Syrup Strategic Fund received 25% of monthly revenue, but chairman Joe Flanagan explained in a July post that the allocation covered both buybacks and reserves, with no fixed monthly purchase schedule. The fund bought 8 million SYRUP in 2025. After that, Maple’s SYRUP dashboard shows no purchases between November 2025 and June 2026, a pause Flanagan attributed to a decision to build reserves for a prolonged downturn.

Maple’s response was MIP-021, which replaced the discretionary allocation with a fixed formula: 10% of monthly revenue below $1.5 million, 20% between $1.5 million and $2 million, and 30% above $2 million. DWF notes that the formula initially cut the share heading to SYRUP from 25% to 10%. Unlike the old allocation, though, it has to be spent, and purchases followed in July, August and September.

A 7% Revenue Gain Doubled SYRUP’s Buyback

The dashboard records $147,098 spent on 676,294 SYRUP for August and $314,116 spent on 1,371,166 SYRUP for September, a 113.5% increase in dollars spent. The October memo ties the jump to September revenue crossing $1.5 million. Because the MIP-021 rate applies to the whole month, crossing that line doubled the share taken from every September dollar, including the first.

Working backward from the purchases gives the revenue behind them: about $1.47 million in August and $1.57 million in September, a rise of close to 7%. If July’s $136,768 purchase also reflects the 10% rate, July revenue came to about $1.37 million and the quarter to roughly $4.41 million, which matches the quarterly total of just over $4.4 million cited by DWF.

The step works in reverse just as sharply. Had September revenue come in about $75,000 lower, under the line, the rule would have funded less than half of the purchase Maple made. DWF describes Maple’s revenue as sitting right at the threshold and expects buybacks to swing between tiers from month to month.

The $21.37 million figure is already the narrower measure. The same financials page lists $122.42 million of gross revenue for the period, so a reader sizing SYRUP’s buyback from the headline number would overstate its base nearly sixfold. It is the same kind of gap Coindoo examined in who benefits from the $5.9 billion users paid in fees.

The commitment is also time-limited. MIP-021 went to a SYRUP holder vote from July 13 to 17 and applies for six months, with purchases made after each month-end once revenue is finalised.

Lido Earns Millions, but NEST Has Bought No LDO

Lido’s core business kept earning as well. The DAO’s first-half 2026 report shows $15.71 million in net staking revenue from January to June and a $6.7 million positive result for the staking unit. DWF still counts LDO among this year’s decliners and describes both of Lido’s buyback programs as slow and conditional.

Buybacks are only part of the explanation. The same report shows Lido’s share of staked ETH falling from 23.93% to 21.18% in the first half, as new capital went to custodians, exchanges and institutional stakers. A token can fall because the business is losing ground, and no buyback design offsets that on its own.

NEST, short for Network Economic Support Tokenomics, went live in mid-August. Each day it compares the DAO’s share of staking revenue with a baseline of $40 million a year, about $109,000 a day, and adds half of the difference to a running balance. On a short day, half of the shortfall is subtracted. NEST buys only while that balance is above zero, within a $50,000 daily cap and a $10 million rolling annual ceiling, according to Lido’s launch overview and the LIP-36 specification.

Lido’s revenue is running under that baseline. The H1 staking figure works out to roughly $87,000 a day. At that pace, the running balance would shrink by about $11,000 every day, and the whole deficit would have to be rebuilt before NEST could make its first purchase.

Lido’s own assessment is blunt. The report estimates that, at current staking levels and market share, revenue reaches the daily baseline only with ETH near $2,730, and it states that “buybacks are not expected soon.” DWF reports that NEST has yet to execute a single buyback.

The caps are ceilings rather than targets. Hitting the $50,000 daily maximum every day would exhaust the $10 million annual limit in 200 days. Lido’s backtest on 2024 and 2025 revenue produced $7.09 million of purchases over two years, or roughly a third of the annual ceiling per year.

NEST launched in treasury-only mode, which sends every purchased LDO to the DAO treasury. A second mode would split each allocation between LDO and wstETH and deposit both as DAO-owned liquidity in a Curve pool, but switching requires another onchain vote.

Lido’s rationale for the strict design is a fair counterweight. Its launch overview argues that a simple percentage of revenue can trigger buying when the treasury has no real surplus. The H1 figures show the risk: an operating surplus of $1.6 million before one-off items turned into a $4.5 million loss after costs tied to the Kelp DAO incident. A Maple-style percentage would have kept spending treasury funds on LDO through a half-year in which the DAO lost money.

Lido’s LDO Buying Runs Through a Committee

LDO purchases have continued by another route. On April 13, the DAO approved a committee-run accumulation program that authorizes up to 10,000 stETH from the treasury for LDO, spent in batches. The H1 report says the first two batches swapped 1,591 stETH for about 10.03 million LDO at an average near $0.30.

A September 25 execution report covers a third batch: 680 stETH exchanged for 4,583,559 LDO between August 31 and September 23 at a volume-weighted $0.37 before fees, about $1.7 million. The onchain transfer confirms the LDO returning to the DAO treasury. After that batch, 7,000 stETH of the authorization was still available.

The result inverts the design debate. The discretionary, human-run route that Lido rejected as NEST’s model is the one currently moving LDO, while the automated mechanism waits for its accounting to recover.

READ MORE:

Crypto Hack Recovery Has a Problem Bigger Than Stolen Keys

Hyperliquid Sends Fees Straight to a Burn

Hyperliquid removes both the threshold and the human decision. Its fee documentation splits trading fees among HLP (the exchange’s community liquidity vault), the Assistance Fund and market deployers, and spot and HIP-3 deployers can keep up to half of the fees their markets generate. The Assistance Fund converts its share into HYPE automatically as part of the chain’s own execution. The HYPE it holds is burned, which removes it from both circulating and total supply.

DWF estimates that Hyperliquid routes 95% to 99% of its revenue into this programmatic buying, and it sets HYPE’s gain against a Bitcoin decline of about 8.5% over the same period. It also says Hyperliquid and Pump.fun together accounted for nearly 90% of a record $638 million spent on buybacks between January and late August.

DWF raises the obvious objection to its own finding: part of that performance may reflect a crowded trade into the handful of large tokens with buybacks, rather than a lasting change in how the market values tokens. The report argues the shift is structural but does not measure how much of the move each explanation accounts for.

Bought Tokens Don’t Always Leave the Market

Flanagan wrote that SYRUP entering Maple’s treasury comes out of circulation, and the dashboard counts 126.14 million SYRUP held by Maple outside its 1.14 billion circulating supply. MIP-021, however, places buyback SYRUP inside the Strategic Fund, whose listed jobs include market making, DEX liquidity positions and strategic deals such as acquisitions. Those tokens have left circulation by classification, and the fund can still put them back to work in the market.

Lido counts treasury LDO the same way. Its H1 report presents the 10 million LDO from the first two committee batches as a circulating-supply reduction of roughly 1% of total supply. In NEST’s design, Lido rejected burning outright, calling it “a one-time signal, not a durable mechanism,” and all NEST purchases remain DAO-owned.

Hyperliquid’s burn is the only one of the three that shrinks total supply permanently. It also gives up the option Maple and Lido keep: burned HYPE can never fund contributors, liquidity incentives or a future acquisition.

In each model, the dollars go to whoever sells. Holders who keep their tokens receive no payment; they gain only indirectly, through added demand or a smaller float. DWF draws the line against equity here: a shareholder whose dividend is cut still owns a claim on the company’s profits and assets, while a token holder whose buyback stops is left with no economic link to the protocol. Purchases can also be outrun by new issuance, as Coindoo found when Aave and JUP buybacks fell behind new supply.

Signals That Revenue Is Reaching the Token

Maple’s October purchase had not appeared on its dashboard as of November 9. Its size against October revenue will show which tier applied, and whether the month-to-month swing DWF anticipates has started. MIP-021’s six-month term also puts a renewal, amendment or expiry decision in front of SYRUP holders around the turn of the year.

For Lido, the signal is NEST’s first purchase. ETH near $2,730 marks the level at which Lido estimates revenue reaches the baseline, but crossing it would only begin repaying the accumulated deficit; buying starts once the balance returns above zero. Until then, the committee has 7,000 stETH of its authorization left to deploy.

Hyperliquid has no threshold to watch. The variable there is fee volume, since the only claims standing between a trading fee and a burned HYPE token are the shares taken by HLP and market deployers.