How is GEX calculated, and is it accurate?
By Faizan Ahmad. Updated .
GEX is calculated by taking each option's gamma, multiplying it by the contract's open interest and by 100 shares, converting it to dollars of stock per 1% move (gamma × open interest × 100 × price² × 0.01), counting calls as positive and puts as negative on the assumption that dealers are long calls and short puts, then adding up every strike and expiry. It is accurate as a rough map of where hedging pressure sits, not as a precise dollar figure. It rests on a once-a-day open interest count and on guesses about which side dealers are on, so trust the sign and the biggest strikes more than the exact total.
The GEX formula, step by step
- Get each option's gamma. Gamma comes from an options pricing model using the stock price, strike, time to expiry and the option's implied volatility. It's the change in delta for a $1 move in the stock.
- Scale by open interest and contract size. Gamma × open interest × 100 gives the shares dealers would trade for a $1 move at that strike.
- Convert to a 1% move in dollars. A 1% move is price × 0.01 dollars, and each share is worth the price, so multiply by price twice and by 0.01.
- Apply the dealer-side sign. Calls count as positive and puts as negative under the standard assumption.
- Add everything up. Sum across strikes for the by-strike chart and across all strikes and expiries for the total.
GEX per strike = gamma × open interest × 100 × price × price × 0.01 (calls positive, puts negative)
SqueezeMetrics' original paper, which named the measure, wrote it in shares (gamma × open interest × 100, negative for puts) and noted it switched to dollars for SPX. The dollars-per-1% version is what most tools show today, because it compares across stocks with very different prices.
Working it backward with SPY on October 30, 2026
SPY's total dealer gamma after the close was -$335.55M. Divide by the $757.12 close and you get about 443,000 shares: the estimated amount dealers would sell on a 1% drop (about $7.57) and buy on a 1% rise.
The strike detail shows why the total alone can mislead. The $758 strike carried -$364.81M by itself, more than the whole total. Negative strikes at $750 (-$251.12M), $770 (-$188.25M) and $760 (-$177.86M) added to it, while positive strikes such as $755 ($147.58M) offset part of it. Two stocks with the same total can have very different setups: one with gamma spread evenly, one with it piled onto a single strike next to the price.
The assumptions behind every GEX number
| Assumption | Why it's made | Where it breaks |
|---|---|---|
| Dealers are long calls and short puts | Investors tend to sell calls against stock and buy puts for protection, so dealers end up on the other side | Stocks where customers pile into calls (dealers end up short calls, flipping the sign), and any day the flow runs against the pattern |
| Every contract has a hedging dealer on one side | Market makers take the other side of much of the volume | Customer-to-customer trades and funds that don't hedge |
| Dealers hedge exactly to delta | It's the only thing that can be computed | Real desks hedge in bands, net across books, and use futures or other options |
| Open interest is current | It's the only public count of positions | The OCC tallies it from each day's trades, so today's opening and closing trades, including most 0DTE activity, aren't in it |
| One implied volatility per option | Needed to compute gamma | When volatility jumps, gamma spreads out and the strike picture changes |
| This ticker's options tell the whole story | Data is per symbol | SPY's hedging sits inside a larger S&P 500 complex that includes SPX index options and futures |
SqueezeMetrics spelled out the first three openly in its paper, including that real market makers use "hedging bands" rather than hedging every tick. They are reasonable working assumptions. They are not facts about any dealer's book.
So is GEX accurate?
As a direction-of-hedging map, it's useful. As a dollar count, it's an estimate with wide error bars. Three data points make the case:
- Vendors disagree, a lot. Cboe wrote in 2023 that published estimates of market maker positioning in SPX ranged from "record short" to long $50bn of gamma. Same market, opposite readings.
- The people with real data see different things. Cboe, measuring market maker positions in SPX 0DTE options, reported that market makers were long about $2bn of gamma at 3 PM on August 15, 2023, when that day's selloff began.
- Researchers had to go beyond open interest. The 2025 study by Amaya, Garcia-Ares, Pearson and Vasquez used proprietary trade data to determine the aggregate market maker position, because public open interest doesn't say who is long and who is short.
The fair summary: the sign is most trustworthy for big index products, where the dealer-side assumption fits best; the largest strikes are usually right, because big open interest is hard to miss; and the exact total is the least reliable part. Skeptics who say GEX is stale by the open have a point about timing. Defenders have a point too: a map drawn from yesterday's positions still shows where the big strikes are. Whether that's enough to trade on is the question in does gamma exposure actually help you trade.
How to sanity-check a GEX reading
- Check the flow side. If customers have been buying calls at the ask, dealers are likely short those calls, the opposite of the assumption. How to tell if options were bought or sold covers the side chips.
- Check where the gamma sits. A wall inside the expected move matters more than one far outside it.
- Check the expiry. Gamma that expires this Friday disappears after Friday.
- Check the size against volume. A few hundred million dollars per 1% is small in SPY and huge in a mid-cap.
How to see it on Tradytics
Tradytics computes dealer gamma from open interest by strike and shows it per strike with the call wall, put wall and gamma flip, with total gamma in dollars of stock dealers trade per 1% move. Find it under Dealer levels on any options research page, such as SPY options, or ask the AI analyst (Chat button on any page) for a ticker's gamma by strike. It's on every plan.
Common questions
Why do two GEX tools show different numbers?
Different dealer-side assumptions, expiries, volatility inputs, strike windows and update times. Even the definition of the walls varies.
Does GEX use volume?
The standard method uses open interest, not volume. Some tools blend in the day's flow to adjust the dealer-side guess, which helps on days with heavy one-way buying.
Why multiply by price twice?
Once to turn a 1% move into dollars of price change, and once to turn shares of hedging into dollars of hedging.
Does GEX include 0DTE options?
Only positions still open at yesterday's close. Contracts opened and closed in the same session never reach the open interest count.
Sources
- SqueezeMetrics, Gamma Exposure (GEX), revised December 2017: the formula and its four stated assumptions.
- Options Industry Council, Open Interest: Why It Matters: the OCC calculates open interest by consolidating each day's opening and closing trades.
- Cboe, Evaluating the Market Impact of SPX 0DTE Options (September 8, 2023): estimates ranging from "record short" to long $50bn; market makers long about $2bn on August 15, 2023.
- Amaya, Garcia-Ares, Pearson and Vasquez (January 25, 2025): measuring actual market maker gamma from proprietary trade data.
- SPY options traded $331.78M in premium on 2026-11-02: $193.21M in calls and $138.57M in puts, a net bullish lean of $31.95M.
- Put/call ratio 0.72.
- Options price a move of about ±$6.07 by 2026-11-02.
- Max pain for 2026-11-02 is $765.0.
Updated 2026-11-02 16:02 UTC