BtcDeribitMonthlyAtmLongCallRollRegime1D

BTC Deribit Monthly ATM Long-Call Roll with 200-SMA Bull Regime Filter

A long-only single-instrument LONG-CALL OPTIONS strategy on BTC perpetual options on the DERIBIT exchange (deepest crypto options venue with 1486+ active BTC/ETH contracts) using daily bars on the underlying BTC for signal generation and monthly ATM call options for execution. This proposal fills the LARGEST structural gap in the portfolio: zero options exposure across ALL 11 in-pipeline strategies. Every existing strategy uses linear-payoff instruments (spot, futures, perps); NONE provide convex (gamma-positive) exposure. Options fundamentally change portfolio risk dynamics: when BTC moves +20%, an ATM call delivers +200-400%+ depending on time-to-expiry and IV; when BTC moves -10%, loss is CAPPED at 100% of premium. This asymmetric payoff structure improves Sharpe, Sortino, and max-drawdown metrics in ways that linear strategies cannot — convexity is genuinely orthogonal to every other strategy class. The trigger mechanism is intentionally MINIMAL: BTC > 200-day SMA (bull regime) → maintain a 30-day-to-expiry BTC ATM call position. Close existing call when 7 DTE remaining, immediately open a new 30-DTE ATM call. Close all positions if BTC closes below 200-SMA. This is mechanistically distinct from every portfolio strategy: it's NOT a directional spot/futures play (those have linear payoff); NOT a trend-cross entry (the SMA gates the position; there's no entry/exit on each cross — we hold continuously while in regime); the rolling structure is deterministic (every 23 days, close + open). Deribit is chosen because (a) Deribit has by far the deepest crypto options liquidity (~85% of crypto options volume), (b) BTC is the most liquid options underlying, (c) Deribit fees are competitive (~0.06% RT on options), (d) Deribit has 4-year+ options data history sufficient for multi-window walk-forward. CRITICAL: this proposal is conceptual — implementation complexity is non-trivial (strike selection, expiry rolling, premium tracking). The developer agent will need to use crypto_options data type and select instruments dynamically. Only 3 explicit parameters (regime SMA period, target DTE on entry, roll DTE on exit) — minimum-parameter options strategy.

Fills the largest structural portfolio gap: zero options exposure across all in-pipeline strategies, every one of which uses linear-payoff instruments. A long ATM call provides convex, gamma-positive exposure genuinely orthogonal to every linear strategy class: a +20% BTC move can pay +200-400% while a -10% move caps loss at the premium, improving Sharpe/Sortino/max-drawdown in ways linear strategies cannot. The trigger is intentionally minimal and non-directional-cross: BTC > 200-SMA gates a continuously-held monthly ATM call (no entry/exit on each cross), rolled deterministically (~every 23 days, close at 7 DTE then open a new 30-DTE call), with an immediate close on a bearish regime flip and a premium-loss catastrophe stop. Deribit is chosen for the deepest crypto options liquidity (~85% of volume), BTC as the most liquid underlying, ~0.06% RT fees, and 4y+ history for multi-window walk-forward. Genuine options exposure (real theta/gamma/premium decay), NOT a synthetic-futures proxy: the rolling expander injects real monthly call legs and the options_no_option_trades gate ensures the analyzed backtest places real option-leg trades; the smoke path fires ONLY in verification where no schedule exists. Minimal 3-parameter surface resists overfitting. leverage=1.0 (options venue cap), equity-relative sizing, no unused-leverage gate.

Not worth the 2 hours: negative risk-adjusted return at the pre-opt gate plus a twin that already failed optimization. The mechanism is implemented correctly (30 long ATM calls, vega +1923, theta -1734, convex), but the +1337% total_return is a COMPOUNDING MIRAGE: Sharpe is -0.28 and Sortino -0.78 (the arithmetic mean daily return is negative from monthly theta bleed at a 37% win rate; the headline is carried entirely by two trend months, 2024-02 +994% and 2024-11 +1035%). The equity path is a leverage-blown lottery — avg_position_pct 57% of equity in a single long call produces annualized_volatility ~5800%, max-drawdown in multiples of equity, downside_deviation 21, and a 2023 of -383% — and it is statistically indistinguishable from noise (probabilistic_sharpe 0.46, sharpe CI -10.07 to +4.67, only 30 trades). NOT iterate: the obvious code issue (reckless sizing) is scale-invariant to Sharpe — resizing the 57% position would tame the volatility but cannot turn a negative risk-adjusted edge positive — and the underlying edge is the same regime-gated monthly ATM long-call roll that JUST failed optimization on the ETH twin (8e053554: out-of-sample Sharpe [0.0, 0.0, 0.0] across all three walk-forward windows, is_overfitted=TRUE, degenerate holdout, zero-trade optimized config), where it even started from a POSITIVE in-sample Sharpe. Optimizing the BTC instance would burn two hours re-confirming that zero-OOS failure. FAILURE PATTERN: regime-gated (200-SMA) monthly ATM long-call roll is regime-dependent long-asset convexity whose 'edge' is a handful of bull-month lottery payoffs; it produces a flashy compounded total_return but negative Sharpe/Sortino with reckless option sizing, and the family does not survive walk-forward on either BTC or ETH — convexity is not generalizable alpha here.

Long-only BTC Deribit MONTHLY ATM long-call roll gated by a 200-day SMA bull regime. The primary instrument is BTC-PERPETUAL.DERIBIT on 1-DAY bars (the roll clock and the SMA(200) underlying source); the tradeable structure is a SEQUENCE of monthly near-ATM BTC call contracts resolved by the backtest rolling-options expander from parameters[rolling_options] (underlying BTC, roll monthly, dte_target 30, ATM offset 0.0, 1-DAY legs, 4y lookback) into concrete legs + a per-period roll schedule. Each monthly expiry the strategy opens ONE ATM call ~entry_target_dte_days (30) before expiry and rolls it out ~exit_roll_dte_days (7) before expiry, BUT only while BTC closes above its 200-SMA (confirmed bull regime). It closes the open call immediately if the regime flips bearish (BTC <= SMA) or on a premium-loss catastrophe stop (stop_loss_pos_pct). Sizing is equity-relative (risk_frac) with a 1-contract floor. The base RollingOptionsStrategy owns schedule parsing/entry/roll; this subclass adds the regime gate and early exit. It subclasses BOTH RollingOptionsStrategy and FactoryStrategy and re-declares the four abstract hooks so Layer-1 AST recognizes it while the MRO routes roll logic through the rolling base. A verification-only long-perp smoke trade on the Deribit primary proves order plumbing in the Layer-3 sandbox (no chain expansion there) and is gated OFF whenever a real schedule exists, so the analyzed backtest always trades real option legs. 3 economic parameters (sma_period, entry_target_dte_days, exit_roll_dte_days); pure OHLCV signal, zero supplementary data.

Mechanism correctly implemented: 30 long ATM calls (moneyness 1.0087), net_entry_vega +1923 (long vol), net_entry_theta -1734 (pays decay) — genuine convex long-call exposure matching the hypothesis; fills the Deribit options gap.

Headline compounded return is large (+1337%) driven by real convex winners in trend months (2024-02, 2024-11).

NEGATIVE risk-adjusted return: Sharpe -0.28 AND Sortino -0.78 despite the positive compounded total — the arithmetic mean daily return is negative (monthly theta bleed, win_rate 37%); the +1337% is a compounding mirage carried by just two months (2024-02 +994%, 2024-11 +1035%).

Reckless position sizing: avg_position_pct 57% of equity in a single long call drives annualized_volatility ~5800%, max-drawdown in multiples of equity, downside_deviation 21, and a 2023 of -383% — the equity path is a leverage-blown lottery.

Not statistically distinguishable from noise: probabilistic_sharpe 0.46 (coin flip), sharpe CI -10.07 to +4.67, only 30 trades.

Known-failure-pattern match: the identical regime-gated monthly ATM long-call roll just failed optimization on the ETH twin (zero OOS Sharpe across all 3 windows, is_overfitted=true, degenerate holdout, zero-trade optimized config) — and that version had a POSITIVE in-sample Sharpe; this BTC one is negative.

57% equity/call, vol ~5800%

sane sizing

BtcDeribitMonthlyAtmLongCallRollRegime1D bought monthly ATM BTC calls while BTC held above its 200-SMA, the first options strategy meant to give the all-linear portfolio convex, capped-loss upside. The greeks were correct and the compounded return eye-popping (+1,337%), but it was a mirage carried by two bull months: Sharpe -0.28, Sortino -0.78, with 57%-of-equity single-call sizing producing ~5,800% vol and a -383% 2023, and only 30 trades at a noise-level PSR. The analyst ruled it negative risk-adjusted at the pre-opt gate and a direct match to the ETH twin that had already failed optimization with zero out-of-sample Sharpe, so resizing or tuning could not rescue it. It ended after one iteration as abandoned, never advancing to optimization or risk review.

Regime-gated monthly ATM long-call roll is regime-dependent long-asset convexity whose 'edge' is a handful of bull-month lottery payoffs: a +1,337% compounded headline masks a negative Sharpe/Sortino from theta bleed, and the family does not survive walk-forward on either BTC or ETH — convexity is not generalizable alpha here, and a flattering compounded total must be checked against arithmetic risk-adjusted metrics and sane position sizing.

It was abandoned at the pre-optimization backtest-review gate (verdict: abandon) because the risk-adjusted return is negative despite the flashy compounded total — monthly theta bleed at a 37% win rate makes the mean daily return negative — with reckless sizing, a noise-level PSR, and a direct match to its ETH twin that had just failed optimization (zero OOS Sharpe in all 3 windows, degenerate holdout, zero-trade optimized config); resizing cannot turn a negative edge positive, so optimization and later stages were never reached.

A long-only single-instrument convex options strategy on BTC Deribit — holding one near-ATM monthly call (opened ~30 DTE, rolled ~7 DTE) only while BTC closed above its 200-day SMA (bull regime), closing on a regime flip — the first options strategy proposed to fill the portfolio's zero-options gap with gamma-positive, asymmetric (capped-loss) exposure, using a minimal 3-parameter pure-OHLCV signal.

Across 30 real ATM call trades (correctly convex: net vega +1,923, theta -1,734, moneyness 1.009, skew +1.68) the headline compounded return was +1,337% but risk-adjusted it was negative — Sharpe -0.28 (CI [-10.07, +4.67]), Sortino -0.78, profit factor 1.39, win rate 37% — a compounding mirage carried by two trend months (2024-02 +994%, 2024-11 +1035%), with reckless sizing (avg_position_pct 57% of equity, annualized vol ~5,800%, 2023 -383%, downside deviation 21) and PSR 0.46 (indistinguishable from noise) on just 30 trades.