QQQ gamma exposure: what does a negative GEX reading mean?

A negative QQQ GEX reading means a model estimates dealers were net short gamma, so their hedging could follow moves. Worked on the Oct 26, 2026 snapshot.

A negative GEX reading on QQQ means a gamma exposure model estimates that options dealers were net short gamma. If those dealers delta hedge the way the model assumes, their hedging would buy as QQQ rises and sell as it falls, which can add to moves instead of damping them. QQQ gamma exposure is a model estimate that rests on an assumption about who holds the options, and the data cannot confirm that assumption. As of the 2026-10-26 session, CaymanBot's delayed QQQ snapshot showed total GEX of -$4.05B per 1% move, with no gamma flip in range.

This post covers only what the sign of that number means, why it came out negative for QQQ on that day, and what it leaves unanswered.

What gamma exposure measures

Gamma is how much an option's delta is expected to change for a $1 move in the underlying. Long calls and long puts have positive gamma; short calls and short puts have negative gamma. At-the-money options usually have the most sensitive delta, so they carry the most gamma.1

Market makers adjust their positions to stay delta-neutral as the market moves, buying and selling the underlying to offset directional risk, with no view on direction behind the trades.2 A gamma exposure (GEX) model asks what that hedging might look like across a whole options chain.

The model needs one assumption that the data cannot supply. Open interest counts the contracts still open, and the OCC updates it once a day,3 but it does not show who holds each contract. GEX assumes dealers are long the calls and short the puts that customers trade, and that they delta hedge.4 Under that convention, calls add positive gamma and puts add negative gamma. CaymanBot multiplies each strike's gamma by its open interest, by 100 shares and by the price squared, then shows the sum per 1% move. Our post on gamma exposure (GEX) covers the full method and its assumptions.

Negative and positive regimes

Cboe describes the two signs this way. When market makers are long gamma, they would hedge against the market's move (sell into a rally, buy into a decline), with the potential to dampen moves. When they are short gamma, they would hedge in the same direction as the move, with the potential to exacerbate it.5

So a positive GEX total is read as modeled dealer hedging that leans against moves, and a negative total as modeled short gamma hedging that leans with them. Both readings depend on the word "potential". The same Cboe post shows a session where market makers turned short gamma and the index stabilized anyway, the opposite of what the mechanism alone would suggest.5

Worked example: QQQ as of 3:45 pm ET on 2026-10-26

Every figure below comes from CaymanBot's public QQQ gamma exposure page and the file behind it, read on 2026-10-29.6 The page states: "On Oct 26, 2026, QQQ's total gamma exposure was -$4.05B per 1% move, a negative gamma regime, with no gamma flip in range and the stock at $713.46."

That page is an end-of-session snapshot, delayed at least two days and updated once a day. By the time you open it, it will likely show a later session than 2026-10-26.

StrikeCall OIPut OICall GEX (millions)Put GEX (millions)Net GEX (millions)
68028,088304,63967-804-737
69026,549293,97272-1,113-1,041
70065,929300,306216-1,389-1,173
71054,278134,268448-830-382
71557,089109,636734-70925
72099,022136,668588-834-246
73092,91860,727464-294170
740112,89343,411389-175214
750122,59252,117350-175174
760149,22518,850341-50292
770111,2136,727197-17180

GEX figures are in millions of dollars per 1% move. Each column is rounded on its own, so net can differ from call plus put by $1M.

What -$4.05B per 1% move means

A 1% move from the snapshot price is:

1% move=price100=713.46100=7.13\text{1\% move} = \frac{\text{price}}{100} = \frac{713.46}{100} = 7.13

That is about $7.13 per QQQ share. Under the model, the dealer hedge would change by about $4.05B of QQQ for a move that size, in the same direction as the move. It is a modeled quantity; nobody observed those trades.

Why the total was negative

The puts below the price did most of the work. Strikes 680, 690 and 700 each held about 294,000 to 305,000 put contracts, and they carried the largest negative net GEX in the table. At 700, for example:

net=call GEX+put GEX=216−1,389=−1,173\text{net} = \text{call GEX} + \text{put GEX} = 216 - 1{,}389 = -1{,}173

Strike 700 sat close to the price, which is why its puts carried so much gamma:

distance=1−strikeprice=1−700713.46=1.89%\text{distance} = 1 - \frac{\text{strike}}{\text{price}} = 1 - \frac{700}{713.46} = 1.89\%

Above the price, most strikes were small positives, from 170 to 292 million per 1% move. Among the listed strikes (643 to 784), net GEX below $713.46 summed to about -$6.3B per 1% move and above it to about +$2.3B. The page lists only strikes within 10% of the price while the model sums the whole chain, so those two figures do not add up to the page total. They only show that most of the negative gamma sat below the price.

Put open interest also outweighed call open interest across the chain:

put/call OI ratio=put OIcall OI=3,780,2132,686,037=1.41\text{put/call OI ratio} = \frac{\text{put OI}}{\text{call OI}} = \frac{3{,}780{,}213}{2{,}686{,}037} = 1.41

Reading net GEX strike by strike, by sign, is covered in our post on reading a GEX heatmap, and the GEX heatmap page shows gamma exposure by strike across tickers.

What "no gamma flip in range" means

A gamma flip is the strike where the running total of net GEX, added from the lowest strike up, crosses zero. On 2026-10-26 that running total never crossed zero, so there was no flip strike to compare the price with. When that happens, CaymanBot labels the regime by the sign of the total, which was negative. Our post on the gamma flip level explains how the flip is found.

What a negative reading does not tell you

Four limits keep the reading honest:

  1. The sign rests on the dealer assumption. If customers had sold many of those puts to dealers, the real dealer position would differ from the model's.4
  2. Open interest updates once a day. The snapshot reflects positions as of the prior OCC update, and expirations can change it sharply.3
  3. It is one end-of-session snapshot. This post describes 2026-10-26 after the fact and makes no claim about what QQQ did afterward.
  4. It says nothing about direction. A negative reading describes how modeled hedging may react to a move up or down. It is no forecast of which way QQQ goes, and Cboe's own example shows a short gamma session where the index stabilized.5

How to practice this

Open the QQQ gamma exposure page, note the total, the regime and whether a flip is in range, then pick one strike below the price and one above and check the sign of each net figure. Compare them with the 2026-10-26 table above; the page will show a later session.

CaymanBot's $100K paper account lets you place practice trades without real money (open QQQ in the app). In the first two Learning Mode stages, each trade on the feed comes with a plain-English sentence you can check against the strike, expiry and premium. A free account shows options data delayed 16 minutes (60 minutes without an account). New accounts start with a 14-day Premium trial.

Educational only, not investment advice.

Notes

  1. Options Industry Council, "Gamma." https://www.optionseducation.org/advancedconcepts/gamma ↩︎

  2. Options Industry Council, "Options Market Makers: The Hidden Machinery Behind Every Trade." https://www.optionseducation.org/news/options-market-makers-the-hidden-machinery-behind-every-trade ↩︎

  3. Options Industry Council, "Open Interest: Why It Matters." https://www.optionseducation.org/news/open-interest-why-it-matters ↩︎ ↩︎

  4. SqueezeMetrics, "Gamma Exposure (GEX)" white paper, 2017. https://squeezemetrics.com/download/white_paper.pdf ↩︎ ↩︎

  5. Cboe, "Volatility Insights: Evaluating the Market Impact of SPX 0DTE Options." https://www.cboe.com/insights/posts/volatility-insights-evaluating-the-market-impact-of-spx-0-dte-options/ ↩︎ ↩︎ ↩︎

  6. CaymanBot public QQQ gamma exposure page, end-of-session snapshot of 2026-10-26 (3:45 pm ET), read 2026-10-29. CaymanBot delayed options dataset (OPRA data delayed at least two days). https://caymanbot.com/qqq-gamma-exposure ↩︎

  7. Options data from OPRA, delayed at least 60 minutes.

Frequently asked questions

What does negative GEX mean for QQQ?

It means a gamma exposure model estimates that options dealers were net short gamma on QQQ. If they delta hedge as the model assumes, their hedging would buy as QQQ rises and sell as it falls, which can add to moves.

What was QQQ's gamma exposure on Oct 26, 2026?

CaymanBot's delayed end-of-session snapshot showed total QQQ GEX of -$4.05B per 1% move, a negative regime, with no gamma flip in range and QQQ at $713.46.

What does 'no gamma flip in range' mean?

It means the running total of net GEX, added from the lowest strike up, never crossed zero. With no flip to compare the price with, CaymanBot labels the regime by the sign of the total, which was negative.

Is negative GEX bearish?

No. Negative GEX describes how modeled dealer hedging may react to a move in either direction. It says nothing about which direction QQQ moves next.

Why is GEX a model estimate and not a fact?

Open interest shows how many contracts are open at each strike, but not who holds them. GEX assumes dealers are long the calls and short the puts that customers trade, so its sign rests on that assumption.

Put the research in one place

Start free. Free account: options data delayed 16 minutes. No account: 60 minutes. Premium adds alerts and Strategy Lab, and real-time options data for users who complete the OPRA non-professional attestation.

Start free