Skip to content
VolSurface Desk
Directory

Dealer Gamma Vanna Charm Flows

Signed Dealer Gamma Reconstruction: Cboe Open-Close Data vs Naive OI GEX

Stop trusting naive "Call OI minus Put OI" GEX charts: learn how institutional desks reconstruct true signed dealer SPX gamma from Cboe Customer/Firm Open-Close transaction files and quote-rule signing.

By Dr. Alistair Finch, CQF, FRMReview not recorded · Updated
Signed Dealer Gamma Reconstruction: Cboe Open-Close Data vs Naive OI GEX — VolSurface Desk 2026 quantitative & supervisory data card
VolSurface Desk (0DTE Index Options, Dealer Gamma Pinning, Vanna & Charm Rehedging): Replacing broken "Calls minus Puts" heuristics with exact Cboe Open-Close inventory accounting.Source: VolSurface Desk Institutional Research DeskLicense: CC-BY-4.0

Open five free retail "GEX dashboard" websites at 09:45 ET and you will often see five wildly contradictory S&P 500 gamma numbers—one claiming dealers are +$9.2 Billion Long Gamma while another claims −$3.4 Billion Short Gamma on the exact same OCC open-interest tape! Why? Because raw OCC end-of-day open interest is unsigned: if 50,000 SPX 5,900 calls exist in open interest, OCC data alone never states whether customers bought those calls (leaving dealers short gamma) or sold those calls to overwrite equity portfolios (leaving dealers long gamma).

Within the 0DTE Index Options, Dealer Gamma Pinning, Vanna & Charm Rehedging silo and our 0DTE Dealer Gamma, Vanna & Charm Flows Guide, institutional desks solve this observability gap using Cboe Open-Close Volume Summary accounting (SPX 0DTE GEX Benchmark) combined with live OPRA complex-order-book signing.

Exact Inventory Accounting Law From Cboe Open-Close Files#

Instead of guessing signs via the broken retail heuristic (Dealer = +100% Call OI − 100% Put OI), Cboe DataShop's Open-Close Volume Summary partitions daily volume for every strike K and expiration T across non-Market-Maker origins o ∈ {CustomerSmall, CustomerMed, CustomerInst, Firm} into Buy-to-Open (BTO), Buy-to-Close (BTC), Sell-to-Open (STO), and Sell-to-Close (STC). Because market clearing is a zero-sum identity across all participants, exact **Market-Maker Net Signed Position NMM(K, T, t)** accumulates deterministically day by day:

Quantitative Pricing Model: Cboe Open-Close Signed Dealer Gamma vs Naive OI GEX#

Formula or model from the articleNot independently verified
N_MM(K, T, t) = N_MM(K, T, t−1) − Σ_{o ∈ {Cust, Firm}} [ (BTO_{o,t} − STC_{o,t}) − (STO_{o,t} − BTC_{o,t}) ]
  • Exact Signed Dollar GEX: $GEXtrue(S) = Σ{K,T} [ NMM(K, T, t) · ΓBSM(S, K, T, σ(K,T)) · S2 · 0.01 · $100 ].
  • Intraday Real-Time 0DTE Update (t + τ): Add signed OPRA single-leg & complex-book prints ΔN_MM^OPRA(τ) classified via NBBO quote-rule + auction responder tags!
  • Worked SPX 5,900 Call Audit (Total Gross OI = 80,000 Contracts at SPX = 5,850, 14-DTE, Γ_1% = $112k/lot):
  • Naive Heuristic assumes NMM = +80,000 => +$8.96B GEX! Cboe Open-Close reveals +42,000 Institutional Overwrite STO vs +31,000 Hedge-Fund BTO (and 7,000 Firm-to-Firm) => True NMM = +11,000 (+$1.23B GEX)—an 86% overestimate by naive models!

Once desks extract true signed contract counts N_MM(K, T), they plug those exact inventories into our Option Greeks & GEX Calculator to map accurate Zero-Gamma Flip Strikes.

Institutional Hybrid Signed-GEX Reconstruction Pipeline#

Hybrid T+1 Cboe Open-Close and live intraday OPRA quote-signing pipeline for institutional GEX
Hybrid T+1 Cboe Open-Close and live intraday OPRA quote-signing pipeline for institutional GEXVerified Exhibit

Four Systematic Biases Where Naive Unsigned GEX Fails Catastrophically#

Quantitative Distribution · Resulting Error in Naive GEX Map
Resulting Error in Naive GEX Map
01Retail Buying 0DTE OTM Calls200% Sign Inver…02JPM HEQT Quarterly Collar (…Misses the 80%-…03Customer 10-Pt 0DTE Iron Co…Overstates stic…
Real Institutional / Retail TradeNaive Unsigned OI AssumptionTrue Dealer (MM) Gamma SignResulting Error in Naive GEX Map
Retail Buying 0DTE OTM CallsAssumes dealers are Long Calls (+Γ)!Dealers are Short Calls (−Γ)!200% Sign Inversion! Naive shows resistance where dealers actually chase upside!
Systematic Put-Write ETFs (PUTW/XYLD)Assumes dealers are Short Puts (−Γ)!Dealers are Long Puts (+Γ)!Misclassifies stabilizing put support as crash acceleration
JPM HEQT Quarterly Collar (+95P/−80P/−C)Treats 80% Put as dealer Short (−Γ)!Dealers are Long 80% Put (+Γ) & Short 95% PutMisses the 80%-put positive gamma floor completely
Customer 10-Pt 0DTE Iron CondorCounts massive gross Call + Put barsNet dealer gamma cancels within ±15 pts (Speed dipole)Overstates sticky gamma by 5x–8x

Naive Unsigned OI GEX vs True Cboe Open-Close Signed GEX ($B / 1%)#

Comparison of unsigned gross GEX estimate versus true Cboe Open-Close signed GEX across four SPX zones
Comparison of unsigned gross GEX estimate versus true Cboe Open-Close signed GEX across four SPX zonesVerified Exhibit

Empirical validation published by the Bank for International Settlements (BIS) and Cboe confirms that true signed net 0DTE dealer gamma averages under $500M–$900M on normal days, spiking past ±$2.5B only during unilateral macro sweeps—explaining why combining T+1 Cboe Open-Close files with intraday OPRA quote-signing under SEC market-data rules is mandatory for institutional accuracy.

Frequently asked questions

What exact assumption makes free public GEX websites structurally inaccurate?

Free websites only download unsigned OCC end-of-day Open Interest, which never reveals whether customers bought or sold to open! To force a sign, naive scripts assume Market Makers are +100% long every Call OI and -100% short every Put OI—completely mis-signing retail long calls and put-write ETFs!

What are the four trade-action buckets inside Cboe Open-Close SPX datasets?

For every option series and participant origin (Customer <100 lots, Customer 100–199, Customer ≥200 institutional, and Firm), Cboe reports exact contract counts for Buy to Open, Buy to Close, Sell to Open, and Sell to Close—allowing exact accounting of cumulative signed net inventory!

How do desks estimate intraday 0DTE signed dealer gamma in real time before T+1 Open-Close files?

Real-time OPRA tick engines classify every non-spread execution against prevailing NBBO bid/ask quotes (Lee-Ready / quote-mid signing plus Cboe complex-order-book auction flags), discounting vertical/condor multi-leg prints whose opposing gammas cancel within 10–20 SPX points.

How much of gross 0DTE SPX volume cancels out inside multi-leg spreads?

Cboe microstructure studies show over 55% to 65% of customer 0DTE SPX volume executes as complex multi-leg orders (verticals, iron condors, butterflies). Treating both legs as unsigned directional bets overstates net dealer gamma by 400% to 700%!

Publication record

Published
November 13, 2026
Updated
November 14, 2026
Author
Dr. Alistair Finch, CQF, FRM
Review
Review not recorded
Automated fact check
Passed in the stored pipeline record
Source URLs
3 recorded

AI assistance recorded: anthropic/claude-opus-4.6. This field does not establish human review or source verification.

Academic & Institutional Attribution

Cite This Research & Embed Live Data Card

Open Attribution License (CC BY 4.0)

Standard academic & institutional research paper citation format.

Dr. Alistair Finch, CQF, FRM. (2026, November 14). Signed Dealer Gamma Reconstruction: Cboe Open-Close Data vs Naive OI GEX. VolSurface Desk. https://volsurfacedesk.com/desk-notes/cboe-open-close-signed-dealer-gamma-positioning
Explore Dealer Gamma Vanna Charm Flows Hub (20) →

VolSurface Desk Data, Benchmarks & Contributor Records

Cboe Open-Close Signed Dealer Gamma vs Naive OI GEX | VolSurface Desk