ETH ATM curve exits inversion as skew diverges

ETH’s 1W ATM volatility decline removed the front-end curve inversion, but the smile adjustment was uneven. Downside skew eased at 1W and from 1M outward while deepening at 2W; butterfly premiums narrowed at 1W but widened across the middle tenors. The result is a change in the location and shape of protection premiums, rather than uniform relief across ETH options.

Derivasys Research

Report date06 Nov 2026

Current snapshot06 Nov 2026 · 17:31 UTC

CoverageETH

Comparison snapshot05 Nov 2026 · 17:31 UTC

The reading

What the evidence says

ETH’s front-end ATM curve exited inversion because 1W volatility declined more than 1M volatility. Over the aligned crypto comparison, 1W at-the-money implied volatility changed by -0.70 vol points ^[1] to 52.30% ^[2], while 1M ATM IV changed by -0.06 vol points ^[3] to 52.38% ^[4]. The supplied 1M-minus-1W slope changed by +0.64 vol points ^[5], leaving 1M ATM IV above 1W ATM IV. The decline also reached 2W, where ATM IV changed by -0.68 vol points ^[6] to 52.07% ^[7]. At 2M, ATM IV changed by -0.09 vol points ^[8] to 52.53% ^[9]. This establishes an endpoint change in the front-end curve, without establishing a persistent normalization regime.

ETH’s skew relief extended beyond 1W, but the 2W exception prevents a surface-wide directional conclusion. RR25 measures call-versus-put IV asymmetry: negative levels retain a downside bias. At 1W, RR25 changed by +0.20 vol points ^[10] to -5.66 vol points ^[11]; at 1M it changed by +0.61 vol points ^[12] to -3.61 vol points ^[13]. Relief continued at 2M, with a change of +0.71 vol points ^[14] to -2.91 vol points ^[15], and at 3M, with +0.49 vol points ^[16] to -2.38 vol points ^[17]. At 6M, RR25 changed by +0.63 vol points ^[18] to -1.49 vol points ^[19]; its absolute move ranked at 100th percentile ^[20] within the rolling-window comparison, 30 observations (observed sampling; 07 Oct 2026–05 Nov 2026) ^[20]. At 9M, RR25 changed by +0.41 vol points ^[21] to -0.94 vol points ^[22]. These negative endpoints show reduced downside asymmetry at the named tenors, rather than a transition to call-favored pricing.

ETH’s butterfly adjustment also differed by tenor. BF25 is the average wing IV premium over ATM, so it describes smile curvature rather than directional bias or absolute wing IV. At 1W, BF25 changed by -0.28 vol points ^[23] to 1.96 vol points ^[24]. The 1W narrowing contrasted with 2W BF25, which changed by +0.04 vol points ^[25] to 1.92 vol points ^[26], and with 1M BF25, which changed by +0.15 vol points ^[27] to 1.82 vol points ^[28]. Middle-tenor widening was clearer at 2M, with a change of +0.33 vol points ^[29], and at 3M, with +0.19 vol points ^[30]. Further out, 6M BF25 changed by -0.03 vol points ^[31] to 1.59 vol points ^[32], while 9M BF25 changed by -0.11 vol points ^[33] to 1.45 vol points ^[34]. Thus, nearly unchanged 1M ATM volatility concealed changes in both 1M skew and the relative wing premium; options repricing was broader than the 1W ATM decline.

ETH’s 1W volatility decline accompanied a spot return of -1.08% ^[35] over the aligned crypto interval, which limits a straightforward interpretation of the curve change as improving price sentiment. Trailing realized volatility over the measured weekly window was 50.15% ^[36], recomputed from canonical minute returns with complete observed coverage. Forward 1W ATM IV exceeded that trailing measure by +2.15 vol points ^[37]. That comparison locates the 1W implied level relative to recent realized movement without establishing a volatility regime or trading edge. During the news window, Cointelegraph reported Bitcoin and Ether ETF redemptions relating to the preceding trading session. Such reporting supplies demand context, but cannot explain why ETH 2W downside asymmetry deepened while 1W and 1M skew eased.

Qualifications and limits

ETH 2W RR25 changed by -1.01 vol points ^[38] to -5.36 vol points ^[39]. That deepened an existing downside bias and directly qualifies the relief observed at 1W and from 1M outward.
The rolling-window rank describes eligible observed absolute changes at the native source horizon; missing dates and sample selection can affect its interpretation. It establishes neither population rarity nor joint unusualness across the smile.
Forward 1W implied volatility and trailing weekly realized volatility cover different periods. Their spread is descriptive; the realized measure uses squared minute log returns under the stated annualization convention, rather than a publisher’s reported volatility summary.

Further qualifications

News publication, the preceding ETF trading session and the crypto endpoint comparison have distinct clocks. Reported redemptions do not establish event-aligned causation, investor positioning or the reason for the ETH 2W skew exception.

Verified market snapshot

Current levels and changes over each recorded native interval. Only receipt-matched quotes are shown; an em dash marks a missing matching quote.
Asset · tenor	ATM IV	ATM IV change	RR25	RR25 change	BF25 change
ETH · 1W	52.30 %^[2]	-0.70 vol pts^[1]	-5.66 vol pts^[11]	0.20 vol pts^[10]	-0.28 vol pts^[23]
ETH · 2W	52.07 %^[7]	-0.68 vol pts^[6]	-5.36 vol pts^[39]	-1.01 vol pts^[38]	0.04 vol pts^[25]
ETH · 1M	52.38 %^[4]	-0.06 vol pts^[3]	-3.61 vol pts^[13]	0.61 vol pts^[12]	0.15 vol pts^[27]
ETH · 2M	52.53 %^[9]	-0.09 vol pts^[8]	-2.91 vol pts^[15]	0.71 vol pts^[14]	0.33 vol pts^[29]
ETH · 3M	52.53 %^[40]	0.03 vol pts^[41]	-2.38 vol pts^[17]	0.49 vol pts^[16]	0.19 vol pts^[30]
ETH · 6M	53.10 %^[42]	0.10 vol pts^[43]	-1.49 vol pts^[19]	0.63 vol pts^[18]	-0.03 vol pts^[31]
ETH · 9M	53.49 %^[44]	0.15 vol pts^[45]	-0.94 vol pts^[22]	0.41 vol pts^[21]	-0.11 vol pts^[33]

Weekly versus monthly curve

Matched native endpoints; inverted means monthly ATM IV is below weekly ATM IV. Regimes compare verified levels; no numerical slope is inferred.
Asset	Prior 1W	Prior 1M	Current 1W	Current 1M	Curve regime
ETH	53.00 % ^[46]	52.44 % ^[47]	52.30 % ^[2]	52.38 % ^[4]	Inverted → Upward sloping

ETH BF25 curve

Current · 06 Nov 2026 · 17:31 UTC · Comparison · 05 Nov 2026 · 17:31 UTC. Tenors are equally spaced categories, not elapsed time. Gaps remain unconnected.

Figure sources: ^[55] ^[56] ^[57] ^[58] ^[59] ^[60] ^[61]

ETH RR25 curve

Current · 06 Nov 2026 · 17:31 UTC · Comparison · 05 Nov 2026 · 17:31 UTC. Tenors are equally spaced categories, not elapsed time. Gaps remain unconnected.

Figure sources: ^[62] ^[63] ^[64] ^[65] ^[66] ^[67] ^[68]

BF25 change by tenor

ETH · 05 Nov 2026 · 17:31 UTC → 06 Nov 2026 · 17:31 UTC · 24h. Own-run receipt-matched changes; missing or ambiguous points remain gaps.

Figure evidence: ^[23] ^[25] ^[27]

What to watch next

Each monitor describes a verified component or central counterexample of the selected story; observation tests, not a joint forecast.
Story component	Next snapshot	Observed condition	Comparison period	Evidence
ETH · 1W ATM implied volatility	07 Nov 2026 · 17:31 UTC	Current level: at or below 52.30 % (current level). Between current and prior: strictly above 52.30 % and strictly below 53.00 %. Prior level: at or above 53.00 % (prior level).	05 Nov 2026 · 17:31 UTC → 06 Nov 2026 · 17:31 UTC	^[2] ^[46] ^[1]
ETH · 2W ATM implied volatility	07 Nov 2026 · 17:31 UTC	Current level: at or below 52.07 % (current level). Between current and prior: strictly above 52.07 % and strictly below 52.75 %. Prior level: at or above 52.75 % (prior level).	05 Nov 2026 · 17:31 UTC → 06 Nov 2026 · 17:31 UTC	^[7] ^[71] ^[6]
ETH · 1M ATM implied volatility	07 Nov 2026 · 17:31 UTC	Current level: at or below 52.38 % (current level). Between current and prior: strictly above 52.38 % and strictly below 52.44 %. Prior level: at or above 52.44 % (prior level).	05 Nov 2026 · 17:31 UTC → 06 Nov 2026 · 17:31 UTC	^[4] ^[47] ^[3]

Historical sampling methodology

Each rank describes its supplied population. Exact consecutive daily endpoints use fixed UTC anchors and 24 elapsed hours, including across daylight-saving changes. Missing pairs are omitted without interpolation; their absence does not establish why the provider omitted them. Full captured history is bounded by this run. Different captures and windows can have different denominators even when their ranks agree.

Measure	Population	Observed pairs or levels	Requested window	Excluded pair or level endpoints
ETH 6M.rr25.absolute_move^[20]	rolling 30 calendar days (UTC)	30	2026-10-06T17:31:00+00:00 to 2026-11-05T17:31:00+00:00	None

Source reporting

Bitcoin, Ether ETFs post $530M outflows as election gains evaporate · Published 2026-11-06T08:19:18+00:00

BF25 = (25-delta call IV + 25-delta put IV) / 2 − ATM IV. Levels and changes are measured in volatility points.

Retrospective report: native market dates are distinct from later retrieval and run clocks. Later capture does not establish availability at the original cutoff.

Source notes