Marginalia (MARG) — mainnet thesis

Claim: BUY

BUY MARG with a target of 200,000 IMD per MARG by 2026-10-22 23:59 UTC. This is a speculative, asymmetric claim, not a statement that the target is already supported by cash flow. At the pinned latest block, 26,191,217, the Uniswap v4 pool’s spot state implies approximately 112,837.52 IMD per MARG (token0 is MARG and token1 is IMD). The target is therefore about 1.77 times the observed spot, while remaining within the contract’s permitted call range.

The reason to buy is not a treasury yield stream today. It is the combination of a functioning, permissioned note-and-bond mechanism, a live pool with substantial concentrated liquidity, and a fixed-supply token whose marginal demand is coupled to posting notes. The critical qualification is that the treasury currently contributes nothing observable: its IMD balance is zero and no drip has occurred in the observed deployment-to-snapshot range. The thesis is consequently a valuation of future usage and fee/pot activity, not a claim on existing treasury reserves.

Evidence

1. PoolManager state, read through the MarginaliaHook pool ID: at block 26,191,217, poolId() returned 0x436763bf49aa9ea6db1d8d4d80ac8c7b3c36ad58fa6281a06ea39c92d69e2f20. The v4 StateLibrary storage word gave liquidity() = 1,347,914,627,805,428,131,718,690, sqrtPriceX96 = 26,613,766,687,065,974,167,194,241,530,122, tick 116,342, protocol fee 0, and LP fee 12,500 (1.25%). These are direct PoolManager reads, not values copied from a frontend. The liquidity is evidence that the market is not an empty deployment; it is not evidence that the liquidity is permanent or broad across price ranges.

2. MarginaliaHook.noteCount(): at block 26,191,217, noteCount() returned 14. noteOf(bytes32(1)) through noteOf(bytes32(14)) showed every record still in an open thesis/call state at the snapshot (the stored status values were 6 for thesis-open or 0 for open), with no observed settled forfeiture. This is modest adoption, but it is real on-chain use. The hook’s code fixes MARG_BUY = 0.25 IMD for every note and BOND_MIN = 0.25 IMD; its _bondOf(uint256) function defines the bond as imdIn - 0.1 IMD - 0.25 IMD, capped at 10 IMD. Fourteen notes therefore imply 3.50 IMD of required MARG-buy notional if all fourteen use the ordinary buy path. That is a mechanism-level demand floor, not a forecast of volume.

3. MarginaliaHook.potMarg() and forfeiture events: at block 26,191,217, potMarg() returned 0. A direct log read for Settled and ThesisForfeited from the deployed hook through block 26,191,217 found no such forfeiture event, so the total forfeited bond observed is 0 IMD. This is important negative evidence: the pot is not currently being enlarged by failed calls. It also limits what can be inferred from the 14 notes; open bonds are not revenue and may later be returned.

4. MarginaliaTreasury.balanceOf and Dripped history: at block 26,191,217, IMD.balanceOf(Treasury) and MARG.balanceOf(Treasury) both returned 0. The Treasury clone’s drip() code reads its IMD balance, computes balance * 200 / 10,000, and transfers that amount to the hook once per UTC day. Directly reading the IMD transfer/event history for the deployed Treasury and the hook’s Dripped(uint32,uint256) event produced zero Treasury inflows, zero Treasury outflows, and zero drips in the observed range. Therefore the current 2%-of-balance claim is exactly 0 IMD, not a positive yield.

5. Pool price and scarcity: the fixed MARG contract at 0xfa0e...759c is the pool’s token0, and the direct slot word at block 26,191,217 encodes a nonzero price and nonzero liquidity. The price calculation is (sqrtPriceX96² / 2¹⁹²), yielding 112,837.5175569702 IMD per MARG. The inference is that the market has established a high nominal IMD price despite only 14 notes; the uncertainty is whether this price survives a larger sell order, because a single slot read does not reveal depth across ticks.

Counter-argument: the 2% drip decays geometrically

The bear case is correct in its mechanical premise. If a Treasury starts with balance B, receives no inflows, and drips 2% once per day, the balance after n successful drips is B × 0.98ⁿ; the daily amount also declines geometrically. More importantly, this live deployment has no starting reserve at the snapshot: balanceOf(IMD, Treasury) = 0 at block 26,191,217, so the next permitted drip would calculate to zero and revert with NothingToDrip. The direct history also shows no observed Treasury inflow, outflow, or Dripped event. It would be misleading to describe the treasury as an active source of recurring support.

The answer is therefore conditional, not rhetorical. The hook has other accrual paths: its source code sends stamps, swap fees, and forfeitures through _accrue, splitting 70% to the weekly thesis pool and 30% to the daily writer-refund pool. The same block-26,191,217 read shows 767 Accrued logs in the deployed hook’s observed history and two PotFunded logs, but those logs do not prove that the treasury funded them, nor do they prove profitable demand. The current state shows 14 open notes, zero forfeited bonds, and zero MARG held as pot. Thus the non-treasury activity is evidence of mechanism use, not evidence of distributable earnings.

This makes the BUY thesis depend on adoption: more notes create more mandatory 0.25 IMD MARG purchases, while stamps and later failed calls can fund pots. If notes remain at 14 and no outside funding arrives, the thesis fails; there is no geometric reserve to rescue it. A further unanswered question is the pool’s executable depth at 200,000 IMD/MARG, which requires a trade simulation or tick-by-tick liquidity inventory rather than the single slot read used here.

Conclusion

MARG is under-valued only under the explicit adoption scenario in which note volume and non-treasury accruals grow; at block 26,191,217 the evidence supports a BUY with a 200,000 IMD target, but does not support treating the empty treasury or zero forfeitures as present income.

Posting spec

• Side: BUY.
• Target: 200,000 IMD per MARG, represented as a 17,720 bps target against the observed 112,837.52 IMD/MARG spot (the on-chain contract requires at least a 1,000 bps deviation; this target is above spot and should be encoded with the exact venue baseline at posting).
• Deadline: 2026-10-22 23:59 UTC.
• Bond: 0.25 IMD. This is the minimum permitted bond, appropriate because the claim is deliberately sized as a falsifiable test rather than a high-conviction capital allocation. The ordinary buy requires 0.10 IMD stamp + 0.25 IMD MARG buy + 0.25 IMD bond = 0.60 IMD imdIn before slippage.
• Router signature: note((address,(uint8,bytes32,address,bool),uint8,uint32,uint16,bytes32,uint8,string),(address,address,uint24,int24,address),uint256,uint256,uint256) returns (uint256).
• Encoded arguments: p = (0xfa0EfEa559bdE229031c2B0A219472316647759c, (1, 0x436763bf49aa9ea6db1d8d4d80ac8c7b3c36ad58fa6281a06ea39c92d69e2f20, 0xD34a99Bc0f67aE1bbd63c660e6d0b0dd03E263B7, true), 0, 17720, 7, 0x0000000000000000000000000000000000000000000000000000000000000000, 1, "MARG BUY thesis: adoption must sustain the 200000 IMD target"); v4Key = (0xfa0EfEa559bdE229031c2B0A219472316647759c, 0xD34a99Bc0f67aE1bbd63c660e6d0b0dd03E263B7, 12500, 60, 0xf0457dd0cf526686190e4f057ba19db14d34a0cc); imdIn = 600000000000000000; minMargOut = 0; deadline = 1792713540. In Solidity notation this is abi.encodeCall(MarginaliaRouter.note, (p, v4Key, 600000000000000000, 0, 1792713540)); minMargOut must be replaced with a caller-chosen slippage floor before any real submission.

No transaction was submitted and no key was requested. The report uses direct reads from the supplied addresses and the verified source of the Hook, Router, Treasury, and shared types: Hook, Router source, Treasury source, and types.