Naive VS CBOE GEX - Gamma Exposure Trading Course Ep. 4

Finance-focused summary (Options / Gamma Exposure Trading Course)

Key topics

• Gamma Exposure (GEX) & options flow: The session explains how to construct and interpret gamma maps for index options, with emphasis on the difference between:
  • “Convention-based” GEX: industry standard; uses assumptions/averages.
  • “Deep Gamma CBOE”: exchange-data based; built using licensed CBOE data about who opens/closes positions.
• Why index options matter (risk transfer):
  • Options on the S&P 500 (SPX) are positioned as a central venue where equity market risk is managed through hedging by market makers.

Instruments / tickers mentioned

• Index: S&P 500 Index (SPX)
• Futures: ES (S&P 500 futures)
• Nasdaq index: NDX
• ETF: Invesco QQQ (QQQ)
• Non-finance markets mentioned: FX / GBP-USD — explicitly noted that there isn’t enough option data for meaningful coverage.
• No specific equities were discussed beyond the above index/ETF instruments.

---

Market microstructure / data framework (step-by-step methodology)

The presenter describes how to build a more accurate gamma map using exchange-level position-side data:

• Build gamma maps from a “position by gamma, strike by strike” formula
  • The Greeks drive the arithmetic; the formula remains constant.
• Determine the missing “side” (who is long vs. short)
  • Open interest (OI) is public but doesn’t reveal which side (long/short) holds the positions.
  • The tape shows trades, but not the lasting “home” of positions or whether positions are truly new vs. carried over.
• Use CBOE exchange files (auction opening/closing & quotation files)
  • These files provide a market maker “side” field reported by the exchange (not inferred).
• Reconstruct “position memory” from the first listing day of each contract
  • Emphasized for 0DTE, because yesterday’s OI is not the same population that trades today.
• Apply controls / consistency checks
  • Examples:
    • Counter resets at evening close; avoid mixing similar contracts (e.g., SPX vs SPXW).
    • “House rule” cross-check: client-book vs intermediary-position must match with reported correlation ~1,000 (described as internal consistency, not claimed as absolute truth).
• Validate by sample days
  • Verify strike-by-strike to decimal precision before publishing to the platform.
• Heat map adjustments by time-of-day
  • Same magnitude can behave differently at open vs near close; heat map reflects session timing.

---

Explicit finance numbers & metrics cited

S&P 500 options volume & 0DTE mix

• ~4.5 million S&P 500 option trades/contracts per day
• ~60% traded on the same day
• > $3 trillion notional per day
• ~86% of 0DTE client-side volume comes from orders of < 10 contracts

0DTE open interest staleness (core caution)

• Measured: ~6 out of 10 near-the-price contracts are bought that same morning
• Implication: last night’s OI is missing much of today’s relevant positioning

Gamma magnitude thresholds (practical guidance levels)

• “Approximate” high/low guidance:
  • High gamma: examples imply levels around ~5 million
  • Low gamma: mentioned as below ~500–600 million (time-of-day dependent)
  • Mentions ~300–400 million as a lower-magnitude regime where price may not linger
• Note: Some numeric levels appear without clear universal units and may be chart-specific.

Scalping payoff reference

• When ES moves through a level, it “makes about 10–20 points”
• Framed as potentially favorable relative to operators who take “two or three points”

Start-of-day example

• During the first ~5 minutes after open:
  • $11 million worth of gamma at one level
  • $5 million worth of gamma at another level
• Combined with “call wall / seller absorption” and a coverage profile indicating upside amplification.

---

Key explanations / trading implications (risk management lens)

• Market makers hedge dynamically; directional bets are not the goal
  • Remaining risk in the options book is hedged via futures (ES).
  • Gamma determines how hedging pressure changes as price moves, requiring ongoing hedge adjustment.
• Why exchange-data GEX differs from conventional GEX
  • Conventional method effectively assumes/infers side (example pattern: long calls / short puts).
  • Exchange-based method can show material discrepancies (e.g., market maker short gamma may differ from estimates).
• Interpretation via profiles
  • Heat map: identify key gamma levels where price may react.
  • Gamma magnitude (“inside the numbers”): strength of positioning/absorption.
  • Coverage profile: models how market makers hedge under changing implied-volatility/Greek scenarios:
    • Blue profile: described as implying slowing moves (less amplification)
    • Pink profile: described as implying amplifying moves
• Probabilistic cautions about level behavior
  • Higher magnitude long-gamma zones (“long jacks”) purportedly correlate with longer time to cross.
  • Lower gamma levels (e.g., ~300–400m) purportedly lead to faster breaks or quick bounces.
  • Breakout vs rejection behavior varies depending on whether the profile suggests “free to move” or resistance.

---

Recommendations / product claims made in the session

• Use Deep Gamma CBOE (exchange-based) instead of convention/inferred GEX when possible to reduce side-estimation error.
• For day trading / 0DTE:
  • Use today’s reconstructed positioning, not only yesterday’s open interest.
  • Set DTE = 0 for day-trading (explicitly confirmed).
• Build/validate custom strategies using:
  • Deep Pattern Builder with overlays including:
    • JAXX heat map
    • coverage profile
    • jacks profile
    • real-time open interest profile
• Backtesting export limitation/disclosure
  • “You can’t export options data for backtesting,” but you can export derived CSV from the matrix system.

---

Disclosures / disclaimers mentioned

• No explicit “not financial advice” statement appears in the provided subtitles.
• Transparency claims:
  • The reconstruction is presented as “based on reported facts,” contrasted with inference/assumption methods.
• Internal consistency checks:
  • Described as method consistency, not absolute truth.

---

Presenters / sources (end of session)

• Rodolfo de Rovigo — creator of Deep Gamma suite; lead speaker
• Andrea Timitilli — host / collaborator mentioned during the session
• Fabio Valentini — mentioned/thanked
• Antonio de Rovigo — developer mentioned; manages technological aspects
• Marcus Sartorello — manages B2B area
• Platform/company source referenced: Volumetrica
  • Includes “Deep Charts/Deep Gamma/CBOE data licensed directly from the exchange”
• Exchange data source explicitly: CBOE (Chicago Board Options Exchange)