Command Palette

Search for a command to run...

Overview

A data-driven strategy for selling short-term options (straddles or calendar spreads) before earnings announcements to harvest two edges: the rapid post-announcement drop in implied volatility (“IV crush”) and stocks' tendency to move less than the options market has priced in. Backtested across 72,500 earnings events on 4,500 stocks (2007-2024), the key finding is stark: blindly trading every earnings event yields near 0% mean return — the edge exists only when filtering for high-probability setups.

Key Concepts

  • IV crush — implied volatility priced into options ahead of earnings is historically higher than the realized volatility of the actual move, because of an uncertainty premium that disappears once results are announced.
  • Who overpays for options — hedgers (institutions buying protection regardless of cost) and retail speculators (chasing lottery-like payouts on short-dated calls) inflate pre-earnings option prices, creating the exploitable edge.
  • Screening criteria (all three required) — term structure backwardation (front-month IV significantly above back-month IV), a high IV/RV ratio (ideally >1.5), and sufficient liquidity (minimizes slippage on entry/exit).

Trade Structures

Short StraddleLong Calendar Spread
MechanismSell ATM call + ATM put, same expirationSell front-month, buy back-month at same strike
RiskHigh — unlimited loss potentialMedium — limited to net debit paid
Mean Return9.0% per trade (filtered)7.3% per trade (filtered)
Win Rate64%66%
Position Sizing≤2% of capital≤6% of capital

Execution Rules

  • Entry: open the position 15 minutes before market close on earnings announcement day.
  • Exit: close the position 15 minutes after market open the following trading day.
  • Position sizing: never use full Kelly sizing — even with a genuine statistical edge, improper sizing leads to ruin. Straddles capped at ≤2% of capital, calendars at ≤6%.

The Four Pillars of Risk Management

  1. Never trade full Kelly — theoretically optimal growth, practically unacceptable volatility.
  2. Straddles ≤2% of capital — protects against tail-risk events with unlimited downside.
  3. Calendars ≤6% of capital — balances meaningful returns against capital preservation.
  4. Avoid low liquidity — wide bid-ask spreads can completely erase the strategy's edge.

At 10% Kelly sizing (6% per calendar trade), the simulation shows 10,000growingtoamean 10,000 growing to a mean ~6M over 10 years with 0% simulated bankruptcy risk — a ~90% CAGR.

Case Study: AMZN Earnings Trade

A Feb 7/Mar 7 call calendar (100 contracts, 3.33/spread,3.33/spread, 33,300 total risk) flagged “RECOMMEND” by the screening criteria. Actual move was +2.5% vs. an expected ~7% — a textbook IV crush, netting +$9,300. A straddle on the same event would have earned more, but the calendar's defined-risk structure is the explicit trade-off for protection against large unexpected moves.

Building a Screening Tool

  • Data needed: earnings calendar, options chains (IV, Greeks), historical prices, volume data.
  • Term structure slope: front_month_iv − back_month_iv — negative values (backwardation) are favorable.
  • IV/RV ratio: rv = stdev(log_returns_30d) × sqrt(252), then ratio = thirty_day_iv / rv — values >1.5 indicate overpriced volatility.
  • Volume filter: 30-day average volume, typically ≥500k shares/day minimum.
  • Recommendation tiers: RECOMMEND (all criteria met), CONSIDER (partial), AVOID (poor setup).

Key Takeaways

  • The single most important finding is that this is not a "sell every earnings straddle" strategy — unfiltered trading returns ~0%, meaning the entire edge lives in the screening criteria (term structure, IV/RV ratio, liquidity), not in the options structures themselves.
  • Straddles and calendars aren't competing choices but a risk-tolerance dial on the same underlying edge — straddles capture more of the IV crush but with unlimited tail risk, calendars sacrifice some upside for a hard-capped loss, reflected directly in their different position-sizing limits (2% vs. 6%).
  • The AMZN case study is deliberately chosen to show a "perfect" IV crush outcome, but the accompanying trade-off analysis is the more important lesson: it explicitly quantifies what defined-risk costs you in upside, rather than presenting the calendar spread as strictly superior.

Related Reading

Back to article