Dealer Hedging Flow in Shares: Gamma, Vanna and Charm, Priced
Gamma exposure in dollars is an abstraction. The number a desk actually uses is the size of the hedge: how many shares change hands when spot, volatility or the clock moves.
The short version
A market maker who sells an option has to hedge it. The hedge changes as spot moves (gamma), as implied volatility moves (vanna) and as time passes (charm). Sum those sensitivities over every open contract, multiply by the contract size and the spot price, and you get the flow: the shares the dealer community must trade for a given change. We compute it from the public option chain with our own Black-Scholes model and show it in shares and dollars, per +1 % of spot, per +1 point of volatility and for the rest of the session.
The arithmetic, in full
For one contract with gamma Γ and open interest OI, a 1 % move in spot S changes the delta hedge by Γ × OI × 100 × S × 1 % shares, which is Γ × OI × 100 × S² × 0.01 dollars. That dollar figure is the gamma exposure everyone quotes. Divide it by spot and you are back to shares. Calls and puts are signed the usual way, calls positive and puts negative, which encodes the standard assumption that dealers are long the calls customers sold and short the puts customers bought.
Vanna is the change of delta per point of implied volatility. Multiply by open interest, contract size and spot and you have the shares the dealer must trade if implied volatility rises one point with spot unchanged. Charm is the change of delta per day; scale it by the fraction of the session still to run and you have the shares that must be traded by the close simply because time passed.
We solve implied volatility from the quoted mid of each contract rather than trusting a published figure, with the risk-free rate taken from the three-month bill and the dividend yield of the underlying. Deep in-the-money quotes with no solvable volatility fall back to the published value; contracts with no quote are dropped, never invented.
How to read the three lines
- Gamma per +1 %. Positive dollars: dealers sell into rallies and buy dips; the structure dampens moves. Negative dollars: dealers chase; the structure amplifies. The share count tells you how big the reaction is against the day's volume.
- Vanna per +1 vol point. On a day when implied volatility jumps, this is the delta dealers must re-hedge. Negative vanna flow with rising volatility means dealers buy on the spike, which is the classic reason why crashes stall.
- Charm to the close. Positive: delta builds as time passes and dealers sell to stay hedged into the close. Negative: delta decays and dealers buy back. On expiry days this is a large share of the afternoon flow.
The widget also lists the strikes that contribute most to each number, so the flow is not a black box: you can see that half the gamma sits at one strike and treat that strike accordingly.
What the number is not
It is not a forecast. It says what must happen if spot, volatility or time moves, not which of them will move. It assumes dealers hedge every contract, which is a simplification: proprietary books, spread positions and cross-asset hedges all reduce the real flow. It is computed on delayed public quotes, so on a fast day it is fifteen minutes behind. And it uses a single volatility surface for every contract, which is fine for direction and order of magnitude, not for pricing.
Where it lives
On the terminal in the Own model preset, on /structure under the dealer map, in the Chrome overlay's Map tab, and as a one-line summary in every levels response so the native platform ports can print it. The journal stores the read at the moment you log a trade, so your reports can later split your results by whether dealers were dampening or chasing when you entered.
Educational content. Nothing here is a signal, investment advice or a prediction. Every number is a description of option-market structure or of public official data at a point in time.