Lido & stETH Weekly Data Report — Week of August 24–30, 2026

1. Executive summary

The second snapshot delivers what the founding issue promised: the first true week-over-week comparisons. Five findings, two of them about our own data:

• ETH re-based to ~$2,500. The $1,845 we carried on Aug 27 came from a stale price cache and is withdrawn. Two independent reads now agree — Bybit at $2,487 and OKX at $2,500–2,555 (both Aug 31) — and Ether's verified weekly gain on that corrected basis is +28% (CoinDesk tally).
• Lido TVL rose to $18.7B from $17.9B — +4.5% (DERIVED; the source does not state its snapshot date, so the delta carries a vintage caveat).
• LDO fell 12.0% on the week on a clean single-source daily series: $0.4007 (Aug 23 close) → $0.3527 (Aug 30 close).
• Two figures we could not reconcile, printed as they stand: stETH supply reads 9.63M ETH on one source and 8.86M on another, and our supply×price check against TVL no longer balances. Both are shown unresolved rather than averaged into a clean-looking number.
• The macro regime shifted: September hike odds 35% → 56–60% after Warsh's keynote — the first genuinely risk-negative macro print of the site's lifetime.

2. The WoW table — where deltas are real, and where they are not

Rows marked n/a refuse to manufacture a delta from mismatched vintages — that is the column's contract. Sources for every row are in the Sources section.

3. The yield table, re-baselined honestly

This week brings the first fee-honest cross-provider comparison from a single aggregator table (CoinLaw, citing stake.lido.fi and DefiLlama, read Aug 31):

One careful note: these APRs are the aggregator's current table and are not directly comparable to last week's 2.17% APY print (APY compounds, APR does not; and the sources differ). From next week we hold this table's source constant so true provider-level deltas begin. What the fee-adjusted view already shows: Lido's net yield (~2.16%) edges out cbETH's (~2.12%) despite the lower gross rate — the 25% fee on cbETH eats its headline advantage, which is exactly the kind of fact gross-only comparisons hide.

4. Competitive position: the concentration story gets sharper

The freshest aggregator table sharpens last week's custody thesis. Full holdings board, per the same source:

The dollar column is the source's own, and it prices ETH at roughly $2,105–2,111 — below the ~$2,500 we carry as this week's snapshot. Re-based at our price, Lido's 8.86M ETH would read ~$22.2B (DERIVED). We print the source column as published and flag the gap rather than silently re-price it: it is one more sign that the supply and TVL readings on this page come from different vintages.

Three reads, all computed from the rows above. One: the seven non-Lido protocols together hold 5.27M ETH worth $11.11B, and wBETH alone takes $7.80B of that — about 70%. Set against the $29.82B combined board, Lido is ~63% and Lido plus wBETH ~89%. "Lido dominance" critiques that ignore custodial wBETH are measuring half the board. Two: the decentralized tier is bleeding — rETH -22.75% and mETH -20.58% over 30 days while wBETH and cbETH grew — the field is consolidating into one protocol plus one exchange, not diversifying. Three: the "who is #2" question now has a clean answer (wBETH, decisively), which makes future share-of-stake comparisons testable rather than rhetorical.

5. Anomalies & data-quality notes — including one on ourselves

• An ETH price we have withdrawn. On Aug 27 we carried $1,845 for Ether. That figure came from a stale cache; two independent sources now put it near $2,500 (Bybit $2,487, OKX $2,500–2,555, both Aug 31). Every number on this site that depended on the old read has been recalculated, and the archived Aug 27 editions carry a note pointing here.
• stETH supply conflict: lido.fi's counter read 9,626,252 ETH (Aug 27) while an aggregator table shows 8,863,785 ETH (undated cache). We publish both, withhold supply WoW, and re-verify with same-day reads next issue.
• The TVL cross-check no longer balances. In our Aug 27 report, supply × price matched TVL within rounding — but both sides were priced off the stale quote we have since withdrawn. On the corrected basis: 9.63M × $2,500 = $24.1B against a TVL of $18.7B. Something in that triangle is stale, most likely the supply counter or the TVL cache. Until it reconciles we print the gap instead of a clean-looking number.
• Operator-count basis: lido.fi's scorecard shows 600+ node operators; a third-party table cites 900+. Different counting bases; we keep the protocol's own scorecard and label the discrepancy.
• LDO circulating supply: last week's tracker said 895.8M; this week two sources agree on 836.3M. Market cap re-derived to $295M with the basis change flagged.

6. The macro overlay that now dominates everything

The data column cannot ignore the regime change: September hike odds at 56–60%, the 2-year at 4.34%, and a Fed chair who has explicitly retired forward guidance. For staked ETH, the arithmetic that matters is now staking's ~2.2% net yield versus a 4.34% two-year Treasury — the widest the gap has been in the site's lifetime. Liquid staking's pitch cannot be yield; it is liquidity, composability and ETH-upside participation. Every scenario in our deep dive this week is built on that fact.

7. Four-week lineage — the honest trend line

With two snapshots on file, the trend table starts — narrow, because only these figures have same-basis reads:

One discrepancy worth naming: our Aug 27 report listed LDO at $0.336 from a market tracker, while the Coinbase daily series we now use as our standard puts the same day's close at $0.3665. Two sources, one date, a 9% spread — the same pattern as the stETH supply conflict above. We are standardising on the Coinbase daily series and will re-verify the tracker read next issue.

Small table, real deltas. Next issue the same-day discipline widens it: supply reads, share-of-stake and the yield series all join once their sources are re-verified on matched dates.

A word on how to read a table that contains corrections, conflicts and n/a rows alongside clean deltas. The instinct is to skip the flagged rows and quote the tidy ones; this column asks the opposite. The re-based ETH price changes every number that touches it, so it is printed with its old value visible rather than swapped in silently. The supply conflict and the failing cross-check are left visible because smoothing them would manufacture false precision — the reader deserves to know which cells are measurements and which are disputes. And the n/a rows are not gaps in effort: where a source has not printed a fresh value inside the window, the table says so instead of carrying a stale figure forward. A week-over-week table is only as honest as its worst row.

8. Conclusion

The second snapshot does its job in three ways. One: real WoW deltas exist now — LDO -12% on a clean series, TVL +4.5% with a caveat, wBETH +11% directionally — and the rows we refuse to compute are as informative as the ones we do. Two: three things did not reconcile this week: an ETH price we had to withdraw, a stETH supply that two sources report 9% apart, and a supply×price check that no longer balances against TVL. All three are printed rather than smoothed over. Three: the macro overlay changed the denominator of every yield comparison this site makes. Next issue: same-day supply reads to repair the cross-check, and provider-level yield deltas on a frozen source.

FAQ

Related reading

• Lido & stETH Weekly News Review — Week of August 24–30, 2026
• The Hawk at Jackson Hole: What Warsh's Debut Means for Staked ETH