Dealer positioning

What the exchange publishes and what it does not

OPRA and the listing exchanges publish quotes, trades and once-daily open interest. Every open contract has a long and a short side, and the report says nothing about who sits on which. A model that wants to know whether a hedging book is long or short gamma at a strike therefore has to assume a holder. Public GEX dashboards generally assign positive exposure to calls and negative exposure to puts, or the reverse, and present the result as if it were measured. It is not.

The usual convention and its logic

• Clients are assumed to buy puts for protection and sell calls for income, so the dealer on the other side ends up short the puts and long the calls.
• Under that reading, call open interest contributes positive dealer gamma and put open interest negative dealer gamma: the calls-positive, puts-negative scheme most public dashboards print. Models that assume dealers are short every option their clients hold, or that classify the options tape trade by trade, give different signs for the same chain.
• Mirroring the assumption negates the net gamma of every strike. The zero-gamma strike does not move, because a negated cumulative profile crosses zero at the same place, but the regime on each side of it swaps, and the walls can change, since the strikes that dominate above and below spot are not the same once the signs flip.
• The convention is a reading of the street, not a law. Netting, spreads and other hedges inside a real book can diverge from it on any given day.

Why it decides the hedge direction

A delta-neutral hedger with a long-gamma book sells after a rise and buys after a fall; with a short-gamma book the adjustments reverse. Whether a strike is long or short gamma for the dealer is exactly what the positioning assumption sets. This is why a call wall does not, on its own, tell you whether dealers sell into a rally there; the gamma-walls guide on this site works through an isolated short-call example where the hedge buys more after a rise.

In Senzoukria

The GEX module states this assumption instead of hiding it. The Calculation assumptions panel has a Dealer positioning setting with two options, Dealers short gamma (clients own the options) and Dealers long gamma (clients sold the options), and its hint recalls that OPRA publishes open interest, never who holds it. In arithmetic terms, the default, which reproduces the backend's levels, counts call gamma as positive and put gamma as negative; the other option negates the net gamma of every strike. The How firm the key levels are section shows how far zero gamma, the call wall and the put wall move when the 0DTE treatment varies and, once a setting is changed, how far each level sits from its default value.

Common mistakes

• Calling a dealer GEX number a measurement of dealer books.
• Comparing two providers' flips without checking whether they use the same positioning convention.
• Attributing a print near a gamma level to dealer hedging because the model predicted it.