ETH daily report · 2026-10-21

Bottom line: ETH remains in a five-day TRANSITION regime. The raw regime-weighted multi-anchor score is about -0.18, mapping to hold/watch. Structural health is 0, incomplete, because exact L2-to-L1 capture and exact burn remain unavailable. Valuation and triggers are separate: the completed October 20 close remained below $2,500 and same-day ETH/BTC remained below 0.030, so the existing two-part reduction review trigger remains active.

1. Data health and fallbacks

Missing means unverified, not zero. USDT live prices and USD closes remain distinctly labeled. Returns are computed locally from consecutive completed closes.

2. Layer 0 regime, structure and sub-zone

Price is above SMA200, while its slope is below -5%. October 16–20 were five consecutive TRANSITION days. The sub-zone is Weak-Transition: price remains above trend support and RSI is near oversold, but ETH/BTC and fee yield are weak. A low MVRV ratio is not a clean buy signal.

Structural health starts at zero. The five-year ETH/BTC percentile is 16.26%, so the below-10% penalty does not apply. Exact L2 capture and exact burn trend are unavailable; the fee proxy's +11.3% YoY change does not substitute for burn. Final structural health is 0, incomplete, with no positive action multiplier.

3. Multi-anchor valuation

TRANSITION weights are MVRV 20%, real yield 25%, fee yield 10%, ETH/BTC 15%, burn/issuance 10%, L2 capture 10%, network activity 5%, macro 0%, and derivatives 5%.

4. Total score and action

The raw TRANSITION score is about -0.18, inside the -0.3 to +0.3 hold/watch band. The stale, non-Z MVRV proxy contributes zero, and missing anchors are neutral rather than filled. Structural health is zero but incomplete, so it does not improve the action.

Separately, the completed close below $2,500 and ETH/BTC below 0.030 keep the reduction-review trigger active. That trigger is not the valuation score, and near-oversold RSI is not an instruction to chase the decline.

5. Independent horizons

• Tactical, 1–4 weeks: RSI 29.73 raises rebound risk, but price and ratio remain below their trigger lines. Wait for a reclaim; avoid both panic selling and premature reversal calls.
• Cyclic, 1–6 months: price above SMA200 conflicts with a deeply negative slope. Weak fee yield and relative strength leave the cyclic view at hold/watch.
• Strategic, 6–24 months: positive real staking yield and strong activity help, while L1 value capture, exact burn, ETH/BTC and treasury-company dilution remain the decisive falsifiers. Retain only structurally sized exposure; low MVRV alone does not justify a heavy position.

6. Hard trigger lines

• Add review: fresh MVRV ≤1.05 while price holds dynamic SMA200, or a completed close above $2,700 with ETH/BTC >0.032. Neither is confirmed.
• Reduce review: completed close below $2,500 and ETH/BTC below 0.030. Both are currently met.
• Trend red line: a completed close below dynamic SMA200, about $2,152, while slope remains below -5%.
• Structural red lines: five-year ETH/BTC percentile <10%, negative real yield, L2 capture <2%, or a long-duration burn decline >70%. Exact data for the latter two are missing.

7. BMNR mapping and scenarios

BMNR closed October 20 at $22.82, +5.80% for one session and -2.85% over five sessions. SBET closed at $7.87, +5.50% and -6.42%, respectively. Returns use consecutive closes.

BMNR's October 19 disclosure, with holdings as of October 18, reports 6,041,481 ETH, including 5,067,309 staked, plus 215 BTC, $582m cash and marketable securities, a $180m Beast Industries stake, and $74m of ORBS. The denominator is FMP's 569,576,806 common shares outstanding, sourced to a May 31 filing. It is a stale denominator and does not fully capture ATM issuance, preferred stock, liabilities or potential dilution.

Using the completed ETHUSDT close, ETH-only NAV per share is about $25.98. Adding disclosed non-ETH gross assets gives gross-asset NAV per share of about $27.48. BMNR trades near 0.88× ETH-only NAV and 0.83× gross-asset NAV; neither is a precise liquidation discount.

Caveats: ATM/preferred dilution, stale share denominator, persistent treasury-stock discounts, ETH/BTC weakness and weak L1 fee capture.

8. Cross-asset and macro context

ETH/BTC sits at the 16.26th five-year percentile, low but not at a new five-year low. BTC dominance was not obtained from an authorized same-timestamp source and remains DATA_MISSING. Latest FRED observations are DFII10 2.80% and DGS10 5.15% on Oct 19; WALCL $6.756tn and TGA $885.8bn on Oct 14; RRP $0.227bn on Oct 20; and M2 $23.343tn in August, STALE. A reliable aligned Yahoo DXY series was not obtained in this run. Macro weight is zero in TRANSITION, so this backdrop does not alter the score.

9. Sources, caveats and conclusion

Sources: Binance ↗, FMP ↗, CoinMetrics Community API ↗, DeFiLlama ↗, FRED ↗, BMNR October 19 SEC 8-K ↗, and Exhibit 99.1 ↗.

This is an evidence-layered research brief, not investment advice. An MVRV ratio is not presented as MVRV-Z; L2 TVL is not L1 value capture; disclosed gross assets are not liquidation NAV. Final view: hold/watch on valuation, reduction review still active tactically, and strategic size capped by structural gaps and dilution risk.