Lido & stETH Weekly News Review — Week of September 7–13, 2026
1. The week in one arc
This edition covers September 7 – September 13, 2026 — six dated stories, all sourced, all inside the window. Last week's edition ended with a paradox: a record share of ETH staked, an entry queue 36 days long, and an exit queue holding exactly zero. This week the paradox resolved itself in the least dramatic way possible — through maintenance. The staking provider Kiln began an orderly exit of its entire validator fleet after a hack at one of its clients, and by Sep 12 the exit queue held 2.656M ETH with a 46-day wait while the entry queue had collapsed to 725,000 ETH and 12.6 days. Around that pivot: a whale redeeming 10,000 ETH in two instalments, a corporate treasury staking enough ether to sit behind an eighth of the network's validators, Lido's first V3 vault products from a new operator, and an ETF tape that broke its inflow streak on Monday and answered with the biggest single-day inflow of the month on Thursday. ETH itself did almost nothing: $2,507.25 to $2,522.46 across the window on our series, +0.6%.
2. Sep 8: Justin Sun redeems another 5,000 ETH — ten thousand since Aug 26
On September 8, on-chain analyst Ai Yi (@ai_9684xtpa) recorded a second 5,000 ETH redemption request against Lido from Justin Sun's address — bringing cumulative redemptions since August 26 to 10,000 ETH. Of the latest batch, 3,500 ETH was routed to a Poloniex hot wallet and 1,500 ETH went into Morpho. The same tracker shows the address still holding 238,000 stETH, which the source valued at roughly $594 million.
Source: on-chain analyst Ai Yi (@ai_9684xtpa), via GetChainNews (Sep 8, 09:22) and ChainCatcher (Sep 8).
Take: ten thousand ETH against 43.1M staked is 0.02% of the network — a rounding error in the flow tables. What makes it a story is the mechanics: one redemption request, however large, waits in the same queue as everyone else's, and at this week's exit-wait of 46 days the second instalment alone is still inside the machinery. The redeem-to-Morpho leg is the detail to watch: staking redeemed into a lending venue is not staking leaving the system, it is staking changing shape.
3. Sep 8: BitMine's treasury hits 5.93M ETH — and most of it is staked
On September 8, BitMine Immersion Technologies reported its treasury at 5.93 million ETH — about $15.7 billion at its own print — after adding 28,086 ETH that week. The company disclosed that 5.067 million ETH of the position is already staked.
Source: BitMine Immersion Technologies disclosure (Sep 8), as reported by Coinliva and Fixygen (Sep 11).
Take: two divisions put that in context. Against the 43.1M ETH now staked network-wide (Sep 12 read), BitMine's staked position is 11.8% — DERIVED, our division. Ethereum's security budget assumes validators are many and independent; one corporate balance sheet behind an eighth of the staked supply is the sharpest test of that assumption the network has produced. Note what is not claimed here: 11.8% of staked ETH is not 11.8% of validators — BitMine's stake is delegated across operator infrastructure, which spreads but does not eliminate the concentration question.
4. Sep 9: Stakely ships two vault products on Lido V3 — public and institutional
On September 9, node operator Stakely launched two Ethereum staking products built on Lido V3's stVaults infrastructure: a public vault for individual users that combines staking with Lido's EarnETH strategy (the staked position is mintable into stETH through a DeFi wrapper), and non-custodial institutional vaults where each institution configures custody, permissions, fees and liquidity while Stakely runs the validators. No deposit size, client names or adoption targets were disclosed.
Source: Lido/Stakely official announcement (Sep 9), via CoinOKC and CryptoNEWS; also catalogued in P2P.org's September DeFi Dispatch.
Take: V3's operator-led product race is producing ships, not roadmaps — two more SKUs on an infrastructure layer that did not exist a year ago. But the disclosure is honest about its own limits: no flows were announced. Treat this as capability news, not demand news — Lido's ecosystem is diversifying into differentiated vault products while the core pooled protocol keeps doing what it has always done.
5. Sep 10: Kiln begins exiting its entire validator fleet — 1.6M ETH in motion
On September 10, staking provider Kiln announced it had begun an orderly exit of all its Ethereum validator nodes — roughly 1.6 million ETH of delegated stake — following the hack of SwissBorg, to which Kiln's infrastructure was linked, in order to ensure the integrity of the staked assets it operates. Kiln stated that the exiting ETH would "probably be re-staked with new validator keys, meaning they will not be sold."
Source: Kiln statement (Sep 10), reported by BlockBeats via WEEX (Sep 12).
Take: the honest framing is churn, not flight — but the word "probably" is doing a lot of load-bearing work in Kiln's sentence. A full fleet exit is also the largest live demonstration yet of a point this site keeps making: custodial staking concentrates operational risk. When something breaks at a provider with 1.6M ETH under management, the network's queue machinery absorbs the shock — which is exactly what it is for, and exactly what we watch next.
6. Sep 10→12: the queue flips — 2.66M ETH now waiting to LEAVE
The validatorqueue.com reads tell the whole week in two snapshots. On Sep 10: entry queue 1.895M ETH, ~33 days; exit queue 25,300 ETH, ~10.5 hours. On Sep 12: entry queue 725,000 ETH, 12d14h; exit queue 2.656M ETH, 46d3h. At our series price of $2,513.56 (Sep 12), the exiting pile is worth ~$6.68B — DERIVED, our arithmetic on the ETH counts.
One conflict to disclose, because this site prints conflicts rather than smoothing them: a same-day Sep 12 counter-read (a staking-costs survey citing validatorqueue) reported the opposite state — exit queue zero, entry 1,858,000 ETH. We carry the BlockBeats read as the one consistent with the Sep 10 Kiln event it explains, and note the counter-read in the data-quality section of this week's report. We also rejected the BlockBeats article's own dollar conversions: its $9.03B figure for 2.656M ETH implies $3,400/ETH, which no venue quoted in the window — the ETH counts and wait times stand, the dollars did not survive our cross-check.
Source: validatorqueue.com reads as of Sep 10 (via Fixygen, Sep 11, and Coinliva) and Sep 12 (via BlockBeats/WEEX); counter-read via CryptoNewsToday (Sep 12).
Take: last week's deep dive was titled "The Empty Exit Queue" and this week the queue holds 2.66M ETH — more exiting stake than at any point since the September 2025 panic topped 2.6M. The difference is the word "orderly": nobody is fleeing; a provider is rotating keys. The queue is doing its job, visibly, at machine speed. Whether that stays true depends entirely on what Kiln does at the end of the 46 days.
7. Sep 11: the inflow streak breaks — and then a $216M Thursday
US spot ETH ETF flows, from the Farside daily table: Sep 8: −$24.3M (the first outflow after eleven consecutive inflow sessions), Sep 9: +$34.7M, Sep 10: −$29.9M, Sep 11: +$216.4M — the largest single day of the month. Net across the window's four trading days: +$196.9M — DERIVED, our sum of the four prints.
Source: Farside Investors, Ethereum ETF daily flow table (accessed Sep 15).
Take: the streak breaking on the window's first trading day was noise; what matters is the shape of the reply. Two shallow outflow days followed by a $216M day is what a bid that is consolidating looks like, not one that is leaving — and it landed on the same Thursday the queue story peaked, which is the kind of coincidence that makes the "supply compression" framing of this cycle look less like narrative and more like plumbing.
Sources
- Justin Sun redemptions (on-chain) — Ai Yi (@ai_9684xtpa) via GetChainNews (Sep 8) + ChainCatcher
- BitMine treasury & staked position — Coinliva (BitMine disclosure Sep 8)
- BitMine mirror — Fixygen (Sep 11)
- Stakely vaults on Lido V3 — CoinOKC (Sep 9, official announcement)
- Stakely mirror — CryptoNEWS (Sep 9)
- Kiln exit + Sep 12 queue — BlockBeats via WEEX (Sep 12; dollar figures rejected, see text)
- Sep 10 queue read — Fixygen citing validatorqueue.com (Sep 10)
- Sep 12 counter-read — CryptoNewsToday citing validatorqueue.com (Sep 12)
- ETF daily flows — Farside Investors (accessed Sep 15)
- ETH/LDO daily series — DefiLlama coingecko feed (fixed UTC-hour print)
- Lido TVL — DefiLlama API (direct read, Sep 15)
- Glamsterdam / L2 context — P2P.org DeFi Dispatch (Sep 2026)
FAQ
Is the 2.66M ETH exit queue a crash signal?
Not on the current evidence. It is dominated by one provider's announced, orderly key-rotation (Kiln, ~1.6M ETH), and Kiln has stated the ETH will be re-staked rather than sold. The risk case is a provider that says 're-stake' and then sells; that is why the queue's drain rate is on this week's watchlist, not the headline itself.
Why do you reject BlockBeats' dollar figures but keep its ETH counts?
Its own numbers do not reconcile: $9.03B for 2.656M ETH implies $3,400 per ETH, and no venue quoted a price near that in the window. The ETH quantities and queue wait times come from validatorqueue.com and are consistent across sources; the dollar conversions failed our cross-check, so we recomputed them on our own dated series.
Does Justin Sun's redemption mean whales are leaving stETH?
The tracked flows are 10,000 ETH redeemed since Aug 26 against holdings still at 238,000 stETH — a reduction of about 4% of one address's position over three weeks, with part of the redeemed ETH deposited into Morpho. That is repositioning, not exit. We describe the flows; we do not guess intent.
Is this financial advice?
No. Informational analysis only.