What is dealer gamma? The hedging mechanic that shapes intraday moves
Market makers hedge the options they sell, and that hedging buys and sells the underlying. Whether it buys strength or sells it depends on one thing: the sign of dealer gamma. A plain-language explanation of the mechanic, the formula, and the assumption everyone glosses over.
Dealer gamma is one of the few concepts in options analysis where the mechanism is genuinely mechanical. Not a pattern someone noticed, not a ratio that correlated once. It is a hedging obligation that produces buying and selling in the underlying, and you can derive it from first principles.
It is also widely misexplained. This post covers the mechanic properly, the formula, and the one assumption sitting underneath every gamma chart you have ever seen.
Start with who is on the other side
When you buy a call, someone sells it to you. Usually that someone is a market maker whose business is capturing the spread, not betting on direction.
A market maker who has just sold you a call is now short a bullish position. If the stock rallies, they lose. They do not want that exposure, so they neutralise it by buying shares. That is delta hedging: holding enough of the underlying to offset the directional risk of the option.
If the option's delta is 0.40 and they are short 100 contracts, they buy 4,000 shares. Now a small move in either direction leaves them roughly flat.
The problem: delta does not stay put
Delta changes as the stock moves. A 0.40 delta call becomes a 0.55 delta call if the stock rises, and a 0.25 delta call if it falls.
Gamma is the rate of that change. It is the second derivative: how fast delta moves when price moves.
Because delta keeps changing, the hedge keeps needing adjustment. And the direction of those adjustments is the entire story.
Two regimes, opposite behaviour
When dealers are net long gamma:
Price rises, their delta rises, they are now too long, so they sell shares into the rally. Price falls, their delta falls, they are now too short, so they buy the dip.
Their hedging pushes against the move. This dampens volatility. Ranges compress, moves get absorbed, price tends to chop around rather than trend.
When dealers are net short gamma:
The sign flips. Price rises and they must buy to stay neutral. Price falls and they must sell.
Their hedging now pushes with the move. This amplifies volatility. Small moves become larger ones, and a sell-off can feed itself because the hedging flow is selling into the decline.
That asymmetry is why the same piece of news can produce a quiet 0.3% drift on one day and a violent 2% flush on another. The market's shock absorber is either engaged or it is working in reverse.
The formula
Gamma exposure, usually written GEX, aggregates this across the whole option chain. Per contract:
`` GEX contribution = gamma x open interest x 100 x sign ``
The 100 is the contract multiplier (one option covers 100 shares). Sum every strike and you get net GEX for the name.
Scaling varies. The common convention multiplies by spot to express the result in dollars of hedging per 1 point of movement, or by spot squared x 0.01 to express it per 1% move. Both appear in the wild, which is why absolute GEX figures from two sources rarely match. The sign and the shape are what matter, not the headline number.
The assumption nobody mentions
Here is the part most explanations skip, and it is the most important thing in this post.
Open interest does not tell you who is long and who is short.
The exchange reports that 10,000 contracts of a strike are open. It does not report whether dealers are long or short them. The sign in that formula is not observed. It is assumed.
The most widely used convention assumes dealers are long calls and short puts, on the reasoning that customers systematically buy puts for protection and sell calls for income, leaving dealers with the opposite book. That is a reasonable premise for index products. It is a weaker premise for a single name in the middle of a takeover rumour, where customers may be buying calls aggressively and dealers are short them.
We treat the dealer model as an explicit configurable parameter rather than a hard-coded truth, precisely because it is a modelling choice. A legacy alternative assumed dealers were short everything, which makes net GEX permanently negative and the flip level meaningless. The convention you choose changes the answer.
So when a chart tells you dealers are long gamma, the honest reading is: under the standard assumption about who holds what, the aggregate position implies dampening hedging flow.
Three more limits worth knowing
Open interest is stale. It updates once daily after the close. Intraday, you are looking at yesterday's positioning plus today's guess.
Gamma comes from a model. It is a Black-Scholes output that depends on an implied volatility input. Different IV surfaces produce different gammas.
"The dealer" is not one entity. It is many firms with different books, some of which hedge on different schedules, in different instruments, or not continuously at all.
None of this makes GEX useless. It makes it a conditioning variable rather than a signal: a read on what kind of day this is likely to be, not a prediction of which way price goes.
What the mechanic does and does not tell you
The mechanic is directionally symmetric. Long dealer gamma dampens moves in both directions. Short dealer gamma amplifies moves in both directions.
Nothing in the hedging logic says whether the next move is up or down. It speaks to magnitude and character, not direction. Any framework that converts a gamma reading straight into a directional call has added something the mechanic does not contain.
That is the correct mental model: gamma tells you how the market is likely to move, not which way.
Next in this series: how to read call walls, put walls and the gamma flip level, and what the options tape can and cannot tell you.