What delta exposure measures
Delta exposure — DEX — is the aggregate delta of an options book, expressed as a quantity of the underlying. Gamma exposure asks how much hedging a move will force. Delta exposure asks how much underlying the book is already equivalent to, right now, before anything moves.
The building block is single-contract delta. The Options Industry Council defines it as "a theoretical estimate of how much an option's premium may change given a $1 move in the underlying," and states the range plainly: "for purchased options owned by an investor, Delta is between 0 and 1.00 for calls and 0 and -1.00 for puts." The reading that matters for DEX is the share-equivalent one. A call with a 0.40 delta behaves, for small moves, like 40 shares of stock, because the contract multiplier is 100 and 0.40 × 100 = 40.
Sum that across every open contract in the chain and you have a directional position measured in shares. That aggregation is the entire idea. Everything difficult about DEX is in the two steps that follow: what sign to attach, and what the resulting number entitles you to conclude. If delta itself is unfamiliar ground, the core Greeks come first.
The formula, and where the spot factor goes
Per contract, the expression is short:
share delta = delta × open interest × 100
Multiply by spot and share delta becomes dollar delta — the notional of underlying the position is equivalent to. FlashAlpha's DEX reference writes the aggregate dollar form as Σ (δᵢ × OIᵢ × 100 × S) × sign — the same expression summed over the chain with one factor of spot applied, plus a trailing sign term that carries the whole positioning convention and is the subject of the next two sections.
That single factor of spot is where DEX and GEX part company. The conventional GEX figure carries spot² × 0.01: one factor of spot to convert share gamma into dollar gamma, a second plus the 0.01 to quote the result per 1% move rather than per $1. DEX needs no per-move normalisation at all, because delta is not a rate of change of a hedge — it is the hedge. There is nothing to express "per move." So DEX takes one factor of spot and stops. Two conventions survive in the wild — share delta and dollar delta — and they differ by a factor of spot, which on a 7,700-point index is a difference of nearly four orders of magnitude. An unlabelled DEX axis is uninterpretable for exactly the reason an unlabelled GEX axis is, and the GEX formula walkthrough covers the same trap from the gamma side.
Three different numbers are all called DEX
The arithmetic is uncontroversial. The sign is not — three distinct quantities circulate under the one name.
Chain delta signs contracts as they exist: call deltas positive, put deltas negative, no assumption about who holds what. It is a property of the open interest, nothing more.
Dealer delta applies the standard positioning assumption — that customers initiated the open interest and dealers took the other side of it — and reports the dealer's resulting position. Because the dealer is on the opposite side of every contract, this number is the exact negative of chain delta. GEXmetrix describes the output in those terms, defining positive net DEX as "dealers are net long the underlying" and negative as "dealers are net short the underlying (put exposure dominates) and hold the underlying as a hedge," and names the crossing point the Zero Delta Level: "the price where total dealer delta crosses zero." MenthorQ's DEX guide uses the same dealer frame.
Trade DEX is a third thing entirely: a per-print figure, delta times size, signed by the aggressor. TradingFlow's glossary entry defines it as the "share-equivalent directional weight of an options trade," and warns against "equating high DEX with 'bullish' without reading call/put and aggressor side." It measures flow, not inventory.
Chain delta and dealer delta are the same book with the sign flipped. A trader reading "DEX is deeply positive" concludes customers are leaning long under one convention and dealers are long and will sell into strength under the other. Those are opposite trades.
How to tell which convention a chart uses
You do not need the vendor's documentation, which frequently does not say. You need one lopsided book and a calculator.
Take an index chain, where put open interest is typically heavy and put deltas are negative. Illustrative numbers on a $500 underlying: calls carrying an average 0.45 delta across 20,000 contracts contribute 0.45 × 20,000 × 100 = +900,000 shares of delta. Puts averaging −0.55 delta across 30,000 contracts contribute −0.55 × 30,000 × 100 = −1,650,000 shares. Chain delta is 900,000 − 1,650,000 = −750,000 shares, or −$375 million of dollar delta at spot.
Dealer delta on the same book is +750,000 shares, +$375 million. Same chain, same arithmetic, opposite sign — not a rounding difference or a units difference, a reversal.
So: compute the raw chain figure yourself from any free chain, then look at the sign your chart prints. Matching sign means you are reading chain delta and the dealer inference is still yours to make. Opposite sign means the dealer assumption has already been applied for you. Do it once per data source; every later reading depends on it.
Why the sign is a harder assumption for DEX than for GEX
Public GEX rests on an assumption most explainers flag: open interest tells you a contract exists, not which side of it a dealer is on. DEX rests on that assumption plus a second one, for a reason rooted in the Greeks themselves.
Gamma is positive for a long call and a long put. Its sign therefore depends only on whether the holder is long or short — one binary. Delta splits: the OIC notes that "long calls have positive Delta; conversely short calls have negative Delta. Long puts have negative Delta; short puts have positive Delta." The sign depends on long-versus-short and on call-versus-put.
The practical consequence is that a strike where the positioning guess is wrong costs you more in DEX than in GEX. In GEX it flips one contribution. In DEX it flips a contribution whose magnitude may be several times larger, for the reason in the next section. The dealer positioning guide covers where the customer-buys-options assumption holds: a reasonable prior for index products, an unreliable one for single names around events, buy-writes and collars.
DEX and GEX weight opposite ends of the chain
The two metrics run over identical open interest and end up dominated by completely different contracts, because delta saturates and gamma does not.
Illustrative, on a $500 underlying with 5,000 contracts open at each of two strikes. An at-the-money call with 0.50 delta and 0.008 gamma carries 0.50 × 5,000 × 100 = 250,000 shares of delta and 0.008 × 5,000 × 100 × 500 = $2,000,000 of hedging per $1 move. A deep in-the-money $400 call with 0.97 delta and 0.001 gamma carries 485,000 shares of delta — nearly double — and $250,000 of dollar gamma, one eighth as much.
Delta climbs monotonically toward 1.00 and flattens; gamma peaks near the money and decays away from it in both directions. So DEX is dominated by deep in-the-money and long-dated contracts that barely register in GEX, while GEX is dominated by near-the-money, near-dated strikes that are a modest share of DEX. This is why DEX reads as a slow, structural measure and GEX as an intraday one. They are not two views of one thing. They are two different subsets of the chain.
The level is inventory; the flow is in the change
Here is the step that separates a usable DEX reading from a misleading one. A desk running delta-neutral has already traded the underlying against its option delta. The DEX level therefore describes hedges that have already been placed, not hedges that are coming. Read as a forecast, it is backwards.
What generates flow is the change in that number. Deltas drift as the clock runs and as implied volatility moves, and the hedge has to be adjusted to keep pace — vanna and charm are the Greeks that describe the drift, and expiration removes the delta and its hedge together. A DEX level that is large and static produces nothing. A DEX level that is large and decaying produces a persistent, opinion-free bid or offer.
The neutrality assumption is also softer than it looks. Yunjiang Dong's 2025 study of 2020 trade-level OPRA and TAQ data for S&P 500 stocks, Cross Market Price Discovery and Selective Delta Hedging in the Option Market, finds that "an option trade moves the underlying price by 0.56 basis points over five minutes, about one-sixth the effect of a stock trade," that the impact "increases monotonically with moneyness and absolute delta," and that it "concentrates in single-leg and limit-order-book trades, while auction and multi-leg trades have negligible effects" — evidence consistent with market makers hedging selectively rather than mechanically. Spread legs sit in open interest and carry delta into your DEX sum whether or not anyone ever hedged them in the underlying. How market makers hedge covers what the desk is actually optimising.
Reading DEX without fooling yourself
Four checks, in order. Confirm the units — share delta or dollar delta, a factor of spot apart. Confirm the convention with the lopsided-book test above, so you know whether positive means customers or dealers. Confirm the expirations and strike window included, since a long-dated tail moves DEX far more than it moves GEX. Then read the trajectory rather than the level, because the level is a record of hedging already done.
SquawkFlow publishes gamma levels rather than delta exposure — the call wall, put wall and flip on our free SPX GEX page — and the asymmetry in this article is the reason. Both metrics are model outputs built on the same unobserved positioning guess, but DEX carries that guess through two sign decisions instead of one and concentrates its weight in exactly the contracts most likely to be spread legs rather than hedged directional inventory. That is a defensible number to build; it is not a number to publish without the caveats attached. Gamma exposure is the companion metric. Neither is a signal on its own: both describe where the mechanical flows sit, and both are only as good as the positioning assumption underneath them.
Educational content, not financial advice. See our risk disclosure.