Gamma range-signal EM-normalized — is GEX-01's range already priced by the straddle?

Adding the exact 09:35 net-GEX state to a strictly prior, expected-move-aware forecast improves untouched prediction of the remaining-session realized range beyond the same forecast without GEX; otherwise GEX-01 is a volatility-state description already priced by the option market.

Pre-registered trigger

At 09:35 ET, classify the exact point-in-time all-expiry net-GEX value by its strictly trailing 60-session distribution. The outcome starts with the first complete minute after the decision and is the remaining-session high-low range normalized by the exact D-1-close straddle expected move. No price after 09:35 may enter the trigger or its controls.

Outcomes and horizons

Outcomes: range_vs_em. Horizons: close.

Matched baseline

Primary comparison is paired out-of-sample loss for a baseline forecast versus the identical forecast plus continuous signed/magnitude GEX and prior-only GEX tails. Baseline features are D-1 expected move/ATM IV, 09:30-09:34 opening range and RV, overnight return/volatility, strictly prior HAR/range memory, day of week, macro flag, and available liquidity controls.

Matching fields: day_of_week · expected_move · opening_realized_vol · overnight_vol · recent_range.

Exclusions

{"tier1_macro_blackout_min":30}

GEX-01 (validated) establishes that net-GEX sign predicts realized range in absolute terms; VRP-01 establishes that the 0DTE straddle EM already prices most of that range. The open question that decides monetizability: does the straddle SCALE for the gamma regime (range ÷ EM ≈ 1 on both signs — the signal is real for vol but not an options edge) or does negative-gamma range OUTRUN the implied move (range ÷ EM > 1 — a tradeable surprise)? The load-bearing separation is the atm_iv control: EM is a monotone transform of atm_iv, so a real gamma effect must survive residualizing range ÷ EM on atm_iv, distinguishing "gamma predicts range beyond IV" from "gamma IS IV".

Prior evidence

(b) mechanically plausible, unproven. GEX-01 is validated in raw bps; VRP-01 is thin (native premium +6.3 bps, p=0.078). No published test of whether the dealer-gamma range premium survives EM-normalization at our granularity — this study is the graph bridge between the two mechanisms.

Required governed inputs

chain_snapshots · em_bands · intraday_snapshots · macro_calendar

At least one required data source is not yet captured; the publication remains data-blocked.

Current research notes

Published report: research/gamma_em_surprise/REPORT.md. Version 2 supersedes, but does not execute, research/gex01_em. No direction, support/resistance, magnet, straddle-P&L, or trading right is tested. Before outcomes open, an outcome-blind metadata audit must prove the exact 09:35 GEX and D-1 expected-move fields are point-in-time authenticated; otherwise publication is inconclusive with no numerical claim. The 2026-08-14 metadata-only audit found 682 candidate 09:35 GEX sessions and 280 expected-move sessions, but zero canonical temporal identities, zero content-bound GEX sessions, zero generation-bound expected moves, zero active Gold generations, and zero lineage-bound paired sessions. No outcomes were opened.

Full-distribution reporting requires n ≥ 30; smaller samples at or above n = 20 are limited to the pre-declared reduced summary.

Statistics

hit_rate_vs_baseline · median_excess · bootstrap_ci95_block_by_session · split_half_stability

Multiple testing

registry wide bh fdr 10.

Out-of-sample protocol

Expanding walk-forward only. Every prediction is fit on strictly earlier sessions; feature transforms, imputation, tail cutoffs, and model selection are learned inside that prior window. Baseline and challenger share identical paired evaluation sessions. Calendar-week blocks are used for inference; era and year stability are reported. This historical corpus can produce at most candidate evidence and requires a later prospective replication for validation.

Refute incremental GEX information if the untouched paired forecast-skill lower bound is not positive after the declared family correction, if either time-shifted or matched-random GEX placebo performs as well, or if the sign is not stable across powered eras. Mark inconclusive, never refuted, when authenticated point-in-time inputs or paired support are inadequate. This leaves GEX-01's absolute same-day range classification untouched.

The lifecycle label is not a confidence score. It describes where this exact version sits in the pre-registered validation process.

No structured reference note is currently linked. The prior-evidence section remains the declared literature context.

Study identity

GEX-37 v2

Definition hash

d59983902da4657be4cdbe4bf3da19055fa6c4d357dc2d34d926108aa4180fce

Result data window

2016-05-12 — 2026-08-13

Last research run

2026-08-14T05:26:00Z

Result identity

d59983902da4657be4cdbe4bf3da19055fa6c4d357dc2d34d926108aa4180fce

Supersession

No superseded result is being served.

Related studies

GEX-01, GEX-06, VRP-01