5D Library › Dealer Gamma Exposure, Zero-Day Options and Volatility Harvesting Dynamics · 中文
Preface: Who Is on the Other Side
Note
This is not investment advice; please read this first. This book studies the mathematics of option hedging and of selling volatility. It does not recommend buying or selling any security, option or strategy, and nothing in it is a forecast of any market. Options can lose more than their premium; a short option position can lose many times the premium collected, and leverage can wipe out an account in a single day — Chapter 4 computes how close the growth-optimal size of a short-volatility strategy sits to that point. All the data in the book are synthetic, generated by the programs printed in it, unless a published study is cited. Rules for trading options, and who may trade them, differ between countries; readers are responsible for knowing their own.
Every option has two sides. A trader who buys a call has a counterparty who sold it, and if that counterparty is a dealer who does not want the direction, the dealer hedges — and the hedge trades the underlying. That simple fact has grown a large popular literature: charts of “dealer gamma exposure” that are said to predict the day’s tops and bottoms, and daily short-volatility strategies that are said to pay a steady income because they win most days.
This book writes the mechanism as mathematics, runs it, and tests the popular claims against it. Like the other books of 5D Library, it disproves first and then sets bounds, and it reports negative results as plainly as positive ones — including where the book’s own model supports the popular story.
What the book contains
Chapter 1 computes the size and shape of gamma near expiry. Chapter 2 writes the hedging feedback loop as one equation and derives dampening, momentum, instability and strike pinning from it. Chapter 3 asks whether a chart built on open interest can know the dealers’ sign, and what it predicts if it can. Chapter 4 prices daily short strangles and iron condors under jump diffusion and asks what a win rate, a track record and a growth-optimal size say. Chapter 5 adds tail protection, discrete hedging and trading costs. Chapter 6 treats variance swaps: replication, skew, jumps and convexity. Chapter 7 reads the Guanzi’s “gathering and scattering” and the Shiji’s Balanced Standard beside the long-gamma hedge, with a toy model of the stabiliser’s limit. Chapter 8 gives a four-step check and a pre-registered test.
The book neighbours several others in 5D Library, and where they overlap it only recalls them briefly: Book 19 treated non-ergodicity and the Kelly criterion (Chapter 4 uses its framework); Book 33 treated pre-registration, power and decision rules (Chapter 8 follows it); Book 38 treated implied-volatility skew and vertical spreads (this book does not repeat it — skew enters here only through variance replication and wing prices); Book 40 treated tail dependence and barbell allocation; Book 45 treated information asymmetry in the order book (Chapter 2’s price impact); Books 53 and 54 treated structural identifiability (Chapter 3 uses the idea).
Claims corrected after checking
- Near expiry, gamma is a spike, not a wall. At-the-money gamma with thirty minutes left is 19.76 times its 30-day value, but the strike half-width over which gamma stays above half its peak shrinks from +8.726 % to +0.413 % (p01). Whether expiry-day gamma matters depends on whether the price is inside a band a fraction of one percent wide.
- “Long gamma calms the market” is a statement about horizon. With lagged hedging, one-minute volatility rises for both long and short gamma (1.154 times at ); the dampening appears only at the daily horizon, 0.684 for long gamma against 1.958 for short (p03).
- Pinning is real in the model, and so is its mirror. With dealers long the strike, closes within 0.10 of it rise from 0.052 to 0.132; with dealers short, the strike repels (0.000 at ) and the price can gap near it (p04).
- The dealers’ sign is not in the data a GEX chart uses. Identical open interest is consistent with dealer gamma of +18.9 or −94.5, and 0DTE positions opened and closed intraday — invisible to open interest — turn +18.9 into −260.7 in the example (p05). This is a failure of identification, not of estimation.
- Even when the mechanism is true, a GEX estimate does not predict direction. Its correlation with the next day’s signed return is indistinguishable from zero for every share of correctly signed days; its correlation with the size of the move is −0.238 when the sign is always right and fades to −0.010 when the sign is a coin toss (p06).
- The win rate says nothing about the expectation. The same strangle wins 0.862 to 0.873 of days whether its expectation is negative, zero or positive; only its price relative to its true value decides (p07).
- “A few gaps eat all the premium” is also wrong. At the fair price, jump days repay 0.232 of the premium collected; most of the premium is repaid on ordinary days (p07). The gaps dominate the tail, not the total.
- The danger of a genuine edge is its size. The growth-optimal leverage, 12.40, sits next to the 13.6 at which the worst simulated day wipes the account out, and 13 of 25 ten-year records estimate an optimum beyond that cliff (p09).
- Tail protection helps by raising what can be carried, and costs can reverse it. Overpriced wings lower the mean but raise growth at optimal size from +0.404 to +0.781 (p10); at a cost of 0.005 per contract the same narrow condor has the lowest net Sharpe ratio of the four structures, −0.57 (p12).
- A volatility swap is not the square root of a variance swap, and a hedged variance seller loses in crashes. The convexity gap is about 1.0 vol points in the example (p14), and a −20 % jump pays the hedged strip only 0.9365 of the variance it adds (p13).
- An attribution. The Guanzi passage on gathering and scattering that this book’s outline placed under Qingzhong Jia is, on checking, in Guo Xu (Chapter 7).
How to read the boxes
Marked boxes separate published evidence, runnable Python calculations (16 programs, standard library only, fixed random seeds, each run at build time with its real output printed beneath it), classical sources, notes, and the author’s hypotheses, each with the result that would show it to be wrong.
Disclosure
The author uses the works of 5D Library, including this book, as the methodological basis of his community courses, the 5D Ascension Cards (Dimensional Shift Card Decks) and the Yafu AI knowledge base, and therefore has a commercial interest in the subject of this book. The author may hold or trade the financial instruments discussed in this book, including index options. The book promotes no course, card deck, software, broker, signal service or fund, provides no link to or recommendation of any trading platform, and accepts no sponsorship from any financial firm. It is not investment advice. The study in Chapter 8 uses public market data only, is to be run and analysed by researchers independent of the author’s business, and its code and protocol are published openly for anyone to check.
Fan, Y. (2026). Dealer Gamma Exposure, Zero-Day Options and Volatility Harvesting Dynamics: Delta-Hedging Feedback Loops, Strike Pinning, Iron Condor Tail Ruin and Variance Swap Replication [做市商伽马敞口、零日到期期权与波动率收割动力学] (Version 1.0.0). 5D Library. https://doi.org/10.5281/zenodo.24090160 · Licensed under Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0).