Lido & stETH Weekly News Review — Week of September 21–27, 2026

1. The week in one arc

This edition covers September 21 – September 27, 2026 — six dated stories, all sourced, all inside the window. Last week Lido announced the first consolidation of its CMv2 migration; this week the migration acquired its second dated data point, a detailed account of that first consolidation by the operator that executed it. Around it, Lido's governance moved twice in five days: the DAO put an execution-delegation and deposit-reserve package on-chain on September 21, and on September 25 its Curated Module Committee gained a new dial over the 1,500 ETH deposit reserve — with a published plan to set it to zero. The tape rallied underneath all of it: ETH closed at $2,709.26, up 5.1% on the week's close-to-close series; LDO posted its best week since this site began publishing at +22.3%; and US spot ETH ETFs took in $689.8 million across five consecutive positive sessions — the mirror image of last week's −$140.6M. The stETH/ETH ratio spent the week at or below par and closed at a 5bp discount.

Source: every item below carries its own source and date; price series from the DefiLlama coingecko feed (same-session prints for Sep 20 and Sep 27); ETF flows from the Farside daily table.

2. Sep 21: Bitwise's first CMv2 consolidation, documented in detail

On September 21, Crypto Briefing published a detailed account of the first validator consolidation executed on Lido Core — carried out by Bitwise, which acquired staking infrastructure provider Chorus One earlier in 2026 and operates as a node operator under Curated Module v2. The mechanics are unchanged from last week's announcement and now carry a second, independent account: legacy 0x01 validators are consolidated into the 0x02 format introduced by Pectra's EIP-7251, which permits an effective balance of up to 2,048 ETH against the old 32 ETH cap. Lido projects a roughly one-third reduction in its total validator count and an approximately 29% drop in attestation messages per epoch across a migration covering roughly 8.4 million staked ETH — about $22.8 billion at the Sep 27 ETH price of $2,709.26 (DERIVED). The DAO approved the framework in July 2026; the first consolidations executed in September; full migration is targeted for Q1 2027. CMv2 covers 34 curated operators, all of which now face ETH bonding requirements. Bitwise has not disclosed how many validators it consolidated or how much stake moved.

Source: Crypto Briefing (Sep 21, 2026); syndicated summary via Brief (Sep 21, 2026); migration scale per the Ebunker Ethereum staking weekly (Sep 28, 2026), citing Lido.

Take: one consolidation is an announcement; two accounts of it, a week apart, are the beginning of a record. What is still missing is the number that would make the migration measurable from outside — how much stake actually moved in step one. Until Bitwise publishes it, the 8.4M figure is a plan and the one-third validator reduction is a projection. Our own falsifiable marker stands: the network validator count, weekly, which we could not refresh this week (section 8 of the data report).

3. Sep 21: Aragon Vote #205 goes on-chain

On September 21, Lido DAO enacted Aragon Vote #205, a governance package with three parts: the Execution Delegation Framework (LIP-37), an upgrade of the DepositSecurityModule, and the establishment of a deposit reserve target with a ceiling of 9,600 ETH, overseen by the Curated Module Committee. The measures are live on-chain.

Source: Aragon Vote #205 execution (Sep 21, 2026), via Bitcoinist; package summary per CoinMarketCap research coverage.

Take: the headline items are the delegation framework and the security-module upgrade, but the number to hold onto is the 9,600 ETH ceiling. It is the constitutional limit on the dial that became operative four days later (section 4), and it tells you the DAO has already priced the worst case: a reserve ten times the current setting would still be smaller than one day of churn-limit throughput. Governance moved in the same direction as the engineering — fewer hands, clearer mandates, bounded discretion.

4. Sep 25: the committee gets the dial — and plans to zero it

On September 25, the Curated Module Committee's 5-of-9 multisig gained Easy Track authority to set the deposit reserve target — the slice of Lido's buffer ETH protected for new validator deposits ahead of the withdrawal queue. The target stood at 1,500 ETH as of Sep 27, and no motion had been filed. The committee's published plan is explicit: set it to zero until the 0x02 Community Staking Module launches — an October launch per the committee, Q4 per Lido's own documentation — then consider restoring 1,500 to 2,000 ETH once node operators show deposit-ready demand. The committee's sizing study, 500 simulations over 360 days of staking inflows and withdrawal requests, puts average withdrawal finalization at 2.3 days with a zero reserve against 2.6 days at 1,500 ETH in the normal case, widening to 6.3 versus 7.9 days under a stress scenario of a roughly 30-day network exit queue plus about five days of skimming and oracle processing; a 2,000 ETH target stretches the stressed mean to 8.5 days. The stated rationale: the reserve existed to seed Curated Module v2's migration keys, those keys are now seeded, and the existing CSM has few depositable keys before its 0x02 version ships.

Source: CryptoSlate (Sep 2026) and WhaleFactor analysis, both citing the committee's Sep 2 statement and the Easy Track activation of Sep 25, 2026; buffer mechanics per Lido's contract documentation.

Take: 1,500 ETH is a rounding error against a $26.5 billion protocol, which is exactly why the story is worth reading slowly. What changed is not the money — it is the authority, the ceiling, and the published plan. Zeroing the reserve trades a modeled 1.6 days of stressed-case withdrawal speed for deposit capacity the committee says it does not need until October. If 0x02 CSM slips and withdrawals stay heavy, the same ETH gets expensive fast — the committee's own stress table is the receipt.

5. Sep 21: Bitmine's treasury reaches 5.98M ETH

On September 21, Bitmine Immersion Technologies disclosed that its Ethereum holdings reached 5.98 million ETH, with combined crypto and cash holdings of $17.1 billion. On the circulating supply we carry (122.02M ETH, DERIVED) that leaves Bitmine roughly 120,000 ETH short of a clean 5% — the threshold its chairman calls the “alchemy of 5%.” The disclosure states more than 5.06 million ETH is staked — about 85% of the treasury — which at the company's own 7-day yield print of 2.62% implies a projected $359 million in annualized staking revenue (DERIVED). Against the 43.68M ETH now recorded as staked network-wide, Bitmine's treasury alone sits behind roughly 13.7% of all staked ETH (DERIVED).

Source: Bitmine Immersion Technologies disclosure (Sep 21, 2026), via Cryptopolitan, PR Newswire and KuCoin research; network staked total per the Ebunker weekly (Sep 28, 2026).

Take: the most rate-insensitive staker on the board added to its position in the same week the network's staking APR printed its largest drop in our records. Bitmine is not buying a yield; it is buying scarce supply with a yield attached. That distinction is the whole debate about treasury-company staking in one sentence, and this week handed us both halves of it: 5.98M ETH accumulating into a 2.18% network print.

6. Sep 21–25: the ETF bid returns — five straight days of inflows

US spot ETH ETFs posted their first full positive week since August. The Farside daily table reads +$270.0M (Sep 21), +$162.2M (Sep 22), +$104.5M (Sep 23), +$66.1M (Sep 24) and +$87.0M (Sep 25) — a window net of +$689.8M (DERIVED), against −$140.6M the week before: an $830.4M swing in seven days. The tape underneath: ETH printed $2,644.73 (Sep 21), a week high of $2,775.17 (Sep 22), a pullback to $2,684.34 (Sep 24), and a close of $2,709.26 (Sep 27) — +5.1% close-to-close. BTC closed at $84,876.06, +5.5% on the same series.

Source: Farside Investors Ethereum ETF daily flow table; prices from the DefiLlama coingecko feed, same-session prints.

Take: last week this column read the outflows as institutional de-risking into the Fed's hike, and predicted the money was parked, not gone, because Friday's $143.7M re-entry said so. Five green days later that read has held — but note what the inflows are not buying: the network staking yield fell to 2.18% in the same window. ETF demand is a bid for ETH the asset, not for ETH the yield instrument. The two legs of this market decoupled before; they rarely do it this visibly.

7. The window's close: LDO's round trip, and a week of discounts

Two market facts closed the week that belong in the record. First, LDO's round trip: after gaining through mid-week, it dropped 11.0% in a single session to $0.38510 on Sep 24, then recovered 27.0% off that low into the Sep 27 close of $0.48893 — +22.3% on the week, its best since this site began publishing, on a market capitalization of $408M (DERIVED on the resolved 834.151M supply basis). No Lido-specific disclosure inside the window explains the move; the governance package of Sep 21 and the committee's Sep 25 authority are the nearest dated events, and we describe the sequence without assigning cause. Second, the peg spent the week at or below par: the stETH/ETH ratio printed a low of 0.99866 (Sep 22, a 13.4bp discount), briefly touched a +2bp premium on Sep 25, and closed at 0.99947 — a 5bp discount. It is the most persistent discount stretch in the daily series this site holds.

Source: LDO and stETH/ETH daily series, DefiLlama coingecko feed, same-session prints; LDO supply basis per Analytics Insight market data (Sep 17, 2026), resolved in edition five.

Take: a wrapper that spends a rising market at a small discount is not broken — redemptions are open and the discount is the exit door's price. What would concern us is the same print with redemptions impaired or the spread widening past the mid-round-numbers. Neither happened: the worst print of the week, 13.4bp, is a fifth of the round trip LDO made in the same sessions. Watch the discount, not the narrative.

8. Numbers of the week

Seven figures carry the edition; all seven are sourced and dated. ETH closed the window at $2,709.26 (+5.1% close-to-close on the re-read basis). LDO closed at $0.48893 (+22.3%), its best week in our records, on a market capitalization of $408M. Lido TVL reached $26.47B (+2.9%); the staking ratio printed 36.19% (+0.42pp against the prior print); the network staking APR stepped to 2.18% (−39bp, the largest single-week move in our records). US spot ETH ETFs took in +$689.8M across five positive sessions, and the stETH/ETH ratio closed at 0.99947 (−5bp). Read together, the set describes a market that paid more for Ethereum in the same week Ethereum paid less to stake it — the thread this edition's deep dive pulls on.

Source: every figure above appears with its own source and snapshot date in sections 1–7 and the weekly data report.

9. Left out, and why

Three items a reader might expect this column to carry, and the reason each stays out. No “Lido protocol exploit” item. No such disclosure landed inside the window; manufacturing one to fill a perceived gap would violate the same window discipline the rest of the edition obeys. No fresh validator-count number. Both aggregator endpoints that publish an active-validator count returned access errors this week, so the 906,000 figure is carried from the last obtainable read — printed as carried, never dressed up as a trend. No competitor product launch. No rival liquid-staking or restaking protocol shipped a market-moving release we can date inside Sep 21–27; the board's movements this week are price and flow, not new supply. The rule is the same as the window rule: inside the dated week, or not in the edition.

Source: negative findings from the same source set; the validator-count access failure is disclosed in the data report's data-quality ledger (section 6).

FAQ

Related reading

• Lido & stETH Weekly Data Report — Week of September 21–27, 2026
• The 39-Basis-Point Week: What Just Happened to Staking Yield
• Lido & stETH Weekly News Review — Week of September 14–20, 2026