Dealer gamma positioning, explained simply

"Dealer gamma" has become a fashionable phrase, usually attached to charts of "gamma exposure" with confident claims about where the market will pin or explode. The underlying mechanism is real and worth understanding. The confident claims deserve more scepticism than they get. This lesson gives you the mechanism and the caveats in equal measure.

The mechanism in plain language

When you buy an option, a market maker (dealer) usually sells it to you. Dealers do not want directional risk, so they hedge: if they sold you a call, they buy some of the underlying; if they sold you a put, they sell some. The amount they hold is set by the option's delta.

Delta changes as the underlying moves. The rate of that change is gamma. So as price moves, the dealer's hedge is no longer the right size and they must adjust it. Whether that adjustment pushes with the market or against it depends on whether the dealer is net long or net short gamma.

Dealer long gamma (dealers have, on net, bought options from the crowd, or the crowd has sold them calls):

Price rises → dealer's delta rises → they sell underlying to rebalance.
Price falls → dealer's delta falls → they buy underlying to rebalance.
Dealers sell strength and buy weakness. Their hedging dampens moves. Realised volatility tends to be low; the index tends to mean-revert intraday.

Dealer short gamma (dealers have, on net, sold options to the crowd, especially puts):

Price rises → dealers must buy underlying to keep up.
Price falls → dealers must sell underlying to keep up.
Dealers buy strength and sell weakness. Their hedging amplifies moves. Realised volatility tends to be high; moves extend.

Long gamma regime:                   Short gamma regime:

price ─╱╲─╱╲─╱╲─╱╲─  (pinned, choppy)   price ╱╱╱╲╲╲╲╱╱╱╱╱ (trending, fast)
dealers: sell rallies, buy dips         dealers: chase both directions

That is the whole idea. Everything else is estimation.

Key idea: Dealers hedge the options they sell. When they are net long gamma their hedging dampens moves; when net short it amplifies them. The regime is real. Estimates of which regime you are in are approximate.

How exposure is estimated

Public "GEX" (gamma exposure) charts are built from open interest by strike, with an assumption about which side the dealer is on, most often "dealers are short every put and long every call". Multiply each contract's gamma by its open interest and by that sign, sum across strikes, and you get a curve of estimated dealer gamma against price.

The level where the sum crosses zero is called the gamma flip. Above it, dealers are estimated to be long gamma (dampening); below it, short gamma (amplifying). Strikes with very large positive gamma are often labelled "call walls" and expected to act as resistance because dealers sell into them; large negative strikes are "put walls".

Estimated regime	Expected tape	Typical when
Long gamma, well above flip	Low realised vol, intraday reversals, grinding	Bull market, options sellers active
Near the flip level	Unstable; small moves can change the regime	Corrections beginning or ending
Short gamma, below flip	High realised vol, trending, gap risk	Selloffs, after sharp declines

The caveats, which are not optional

The sign assumption is a guess. Open interest tells you a contract exists, not who is long it. The "dealers short puts, long calls" rule is a rough average over the crowd's habits; it is wrong for large segments of flow, especially institutional overwriting programs (which sell calls, making dealers long calls) and systematic put selling (which makes dealers long puts). Different vendors use different assumptions and produce different flip levels for the same day.

Open interest is stale. It updates once a day. Intraday, especially in the zero-days-to-expiry era, enormous positions are opened and closed before they ever appear in open interest. The gamma that matters at 2pm may not be in the morning's data at all.

Dealers are not the only hedgers. Volatility-targeting funds, CTAs, and leveraged ETF rebalancing all produce flows that push with or against the market, and they do not show up in options data.

It is not a directional forecast. Long gamma says "moves get damped", not "market goes up". Short gamma says "moves get amplified", not "market goes down". Traders who read a negative GEX as bearish are making an error the mechanism does not support.

The published record is thin. Vendors show the winning examples. Independent tests of gamma levels as support/resistance find modest effects at best, strongest around large monthly expirations and weakest in trending markets.

What is actually usable

Given all that, three uses hold up reasonably well:

Regime context for realised volatility. If a reliable estimate says dealers are deeply long gamma, expect a choppier, mean-reverting tape and size intraday fades accordingly; if deeply short, expect follow-through and give trend trades more room. This is the same day-type question from Module 2, answered from a different data source, and Module 2's internals should agree if the estimate is right.
Attention around large strikes near expiry. In the last day or two before a big monthly expiration, price does show a tendency to gravitate toward strikes with heavy open interest. That is the pinning effect, covered in the next lesson.
Recognising unstable zones. When price sits close to an estimated flip level, the regime can change with a small move. That is a reason for smaller size and wider attention, whichever way it goes.

What does not hold up: trading levels off a GEX chart as if they were support and resistance drawn by a market participant, or reading the chart as a directional call.

A sanity check you can run

If a gamma estimate says "long gamma, dampened tape" and the day is printing ADD +2,000 with TICK one-sided and a 3% index move, the estimate is wrong for today, whatever its methodology. Internals are observed; gamma exposure is inferred. When they disagree, the observation wins.

Try it: Find a free gamma-exposure chart (several vendors publish a daily version). For ten sessions, write down the stated regime (long or short gamma) and the flip level in the morning. At the close, write the day's range as a percentage and whether the day was trend or chop by the Module 2 fingerprint. After ten days you will have your own small test of whether the estimate described the tape. Expect it to be right more than half the time and wrong often enough to matter.

Recap

Dealers hedge options they sell; hedging adjustments dampen moves when dealers are net long gamma and amplify them when net short.
Public gamma exposure estimates come from open interest with an assumed dealer side; the flip level is where the estimated sum crosses zero.
The sign assumption is a guess, open interest is stale, other hedgers exist, and the effect is about volatility, not direction.
Usable: regime context for realised vol, attention near big strikes at expiry, awareness of unstable zones near the flip.
When gamma estimates disagree with observed internals, trust the internals.