How GEX Is Estimated

The Ceiling, Floor and Flip Line all start as one number per strike. Gamma exposure estimates how many dollars of stock dealers might need to buy or sell for each $1 move if they keep their hedges neutral. Nobody outside a dealer can see its book, so this is a model. Knowing how it’s built tells you when to lean on it and when to hold it loosely.

The formula, one piece at a time

1. Gamma per share

   A pricing model gives each contract’s gamma: how much its delta changes per $1 move. In a standard model, a call and a put at the same strike and expiration have the same gamma.

2. × open interest × 100

   Multiply by the number of open contracts and by 100 shares per contract. That’s how many shares of delta those contracts gain or lose per $1 move.

3. × spot

   Multiply by the current price to turn shares into dollars: GEX = gamma × open interest × 100 × spot, in dollars of stock per $1 move.

4. Apply the sign and add

   Count calls as positive and puts as negative, then add them to get the strike’s net GEX. The model assumes dealers are net long calls and net short puts, because many investors sell calls against stock they own and buy puts for protection. That’s an assumption, not observed inventory.

From profile to levels

Line up net GEX for every strike and you have the gamma profile. The Ceiling is the strike with the largest positive net GEX. The Floor is the strike with the most negative. The textbook Flip Line comes from a running total: add net GEX strike by strike from the lowest strike up, mark the first place the total crosses zero, and interpolate between the two strikes on either side. The example and the builder below use that running total. The live chart uses a local version: the strike gap nearest price where net GEX turns from negative below to positive above; on the nearest expiry it is held between readings so it doesn’t hop on small moves. On a lopsided map the running total can sit at the edge of the strike window or never cross at all, while the local crossing stays near price. The Magnet comes from a separate, volume-based estimate of the largest-gamma strike from the data provider, so it doesn’t have to match any of the others.

Assumptions and limits

• The sign is assumed. Under the dealer-side sign assumption, calls add and puts subtract. When customers buy calls heavily or sell puts, real dealer gamma at those strikes can have the opposite sign, and the map can mislabel a level or the regime. That’s model risk, and it never fully goes away.
• The line is a shortcut. A fuller method would recompute every option’s gamma at many hypothetical prices and find where total dealer gamma changes sign. The running total is faster, but it depends on which strikes are in the window and can land somewhere else, or nowhere.
• Open interest is a day old. Open interest is tallied once a day, overnight, so positions opened today only show up in tomorrow’s count. Same-day contracts opened today never show up at all, because they expire first. Intraday changes come from price and time reshaping gamma.
• A profile is a snapshot. Each bar is computed at today’s price. Because gamma is highest near the money, a strike’s GEX tends to grow as price moves toward it and shrink as price moves away, so a big bar $15 away can matter less right now than a smaller bar at price.
• One expiration, about ±30 strikes. StratosHedge uses the nearest expiration, 0DTE on most SPY days, and about 30 strikes on each side of price. Positioning in later expirations is left out, and the map can change a lot when an expiration rolls off.
• Units differ between tools. Multiplying by spot turns shares into dollars of stock. Within one symbol it scales every bar equally, so it doesn’t change which strike is biggest. Across symbols, “per $1 move” still isn’t apples to apples: $1 is 0.2% of a $500 stock but 2% of a $50 one, which is one reason some tools scale per 1% move instead. Raw numbers from different tools aren’t comparable.