Options-Confirmed BTC Shock Transmitted to Alts: when BTC's UTC-day move is >=1 sigma AND Deribit BTC DVOL reprices up >=1.5 sigma the same day, trade an equal-weight 12-alt OKX USDT-swap basket in BTC's direction for a fixed 7 days (OKX 1-DAY, long-short, 5% per leg, BTC is signal only)

Original Research Lead hypothesis. It extends the promoted DVOL-confirmation design (lesson L275: Deribit BTC/ETH linear perps, 69 trades, PF 1.89; ETH COIN-M, 80 trades, PF 1.59) in a new direction: the options market confirms the BTC shock, but the position is taken in the higher-beta ALT complex on a different venue (OKX), not in BTC itself. BTC is used ONLY as the signal leg; no BTC position is ever held. Mechanism: a BTC daily move that options traders actively reprice (DVOL jumps) is information, not noise; alt perps follow BTC with higher beta and slower, crowded positioning (leveraged alt longs unwind for days after a confirmed risk-off shock, and alt rotation lags a confirmed BTC risk-on day), so the continuation is larger in the alt basket than in BTC. SIGNAL (evaluated at each UTC daily close of BTCUSDT.OKX): z_move = BTC log return of the day / std of the prior 60 daily BTC log returns (excluding today); z_dvol = (DVOL_BTC at today's close - DVOL_BTC at yesterday's close) / std of the prior 60 daily DVOL changes (excluding today), read point-in-time with self.supp_as_of('DVOL_BTC') / supp_window. Long event: z_move >= +k_move AND z_dvol >= +k_dvol. Short event: z_move <= -k_move AND z_dvol >= +k_dvol (DVOL must RISE in both cases: the options market confirms the shock). TRADE: at that close, open the 12 alt legs equally in the event's direction (5% of equity notional per leg, 60% gross, 1x per leg), exit all legs on a fixed 7-day clock (7 daily bars after entry). While a basket is open, new events are ignored (no stacking, no flips). Execution: a market order at the signal bar's close; on this 24/7 market that is the faithful equivalent of a next-bar-open fill (capability next_bar_open_fill: use_equivalent). UNIVERSE: fixed pre-2021 OKX USDT swaps SOL, ADA, LINK, DOT, AVAX, ATOM, LTC, BCH, DOGE, XRP, FIL, ETC (no overlap needed with prior DVOL runs, which traded only BTC/ETH); a leg is traded only if listed_since >= 60 days at the event (universe_at/is_listed); missing legs are skipped, not reweighted. PRE-STUDY (Research Lead, OKX daily closes 2021-03..2026-11, 0.12% RT per leg cost incl. modelled spread, non-overlapping): 89 distinct qualifying trigger dates, 60 non-overlapping basket events (23 long, 37 short) at k_move=1.0, k_dvol=1.5; per-leg-average 7-day return after cost +2.08% (median +2.10%, win 57%, PF 1.83, t 1.82) vs +1.48% for BTC itself; long arm gross +4.6%, short arm gross +0.7% (the short arm is thin and is pre-registered separately); per year 2021 +3.1%, 2022 +1.9%, 2023 +0.9%, 2024 +3.7%, 2025 -0.4%, 2026 +6.9%. At 60% gross the event-level drawdown was 18.1% and the worst event -8.9% of equity. The stricter k_dvol=2.0 gave only 39 events (below lesson L275/L272's ~60 floor), so 1.5 is the baseline. The ETH-DVOL analogue was weak in the pre-study (PF 1.08) and is NOT part of this hypothesis. FEE ARITHMETIC: expected capture ~2.0% per leg per 7-day trade vs ~0.10% OKX round trip (+ ~7 days of 8h funding, typically +-0.1%): >10x cost. RISK CONTROL route (b): signal/time exit (fixed 7-day clock) with capped sizing, each leg 5% of equity, gross 60%, never above 100%; annualized in-market basket vol ~85% x 0.6 gross ~51%, time-in-market ~20%, so realized strategy vol ~25%, inside the conservative 60% band; no leg contributes >120% vol (5% x ~90-130% ≈ 5-7%); leverage 1x, liquidation buffer far beyond 8 daily moves. Fills over the 2021-03..now DVOL span only.