The short answer
Positive gamma and negative gamma describe how hedging reacts to a move, not which way the move goes. A positive-gamma book has to trade against price to stay hedged, which damps moves. A negative-gamma book has to trade with price, which extends them. Applied to the aggregate inventory of options dealers, that difference is what people mean by a positive or negative gamma regime.
The phrase gets confusing because it is used at two scales: your own position, and the dealer book on the other side of everyone's positions. The signs at those two scales are often opposite. This page covers both, then the part most explainers skip: what the published evidence says about each regime, and why neither one is a bullish or bearish signal.
What gamma measures
The Options Industry Council defines it plainly: gamma is how delta is expected to change given a $1 move in the underlying. Its worked example is a $50 call with a delta of .50 and a gamma of .06. If the stock rises to $51, delta becomes roughly .56.
Two rules from the same page do most of the work here. Long options, calls or puts, always carry positive gamma; short calls and short puts carry negative gamma. And gamma is highest for options that are at the money and close to expiration. Stock has no gamma at all, which is why hedging with shares or futures can neutralize delta but never gamma.
Long gamma vs short gamma in your own book
If you own options, you are long gamma. Your delta grows as the move goes your way and shrinks as it goes against you, so gains accelerate and losses decelerate. You pay for that convexity in time decay.
If you sold options (credit spreads, iron condors, short strangles, covered calls), you are short gamma. Losses accelerate as the underlying moves through your strikes, and gains are capped at the premium collected. That premium is what you are paid to carry the curvature.
Neither sign is a forecast. A long straddle wants a large move in either direction; a short strangle wants a small one. Both are bets on distance and speed. The negative gamma explainer works through the short side in detail.
Positive vs negative gamma in the dealer book
Options dealers take the other side of customer orders and hedge the resulting delta in the underlying, usually in shares or index futures. Nobody publishes their inventory, so every gamma exposure model has to assume which side they hold. The common convention treats dealers as long calls and short puts; Baltussen, Da, Lammers and Martens state the same assumption in their Journal of Financial Economics study, and note that their exchange data misses positions traded over the counter.
Using the OIC numbers as illustrative arithmetic, here is what each side forces a hedger to do. One equity option contract covers 100 shares.
Positive gamma (dealer long the call). Delta .50 means the call behaves like 50 shares, so the dealer hedges by shorting 50.
- Stock up $1: the call's delta rises to about .56, the position is now long 6 shares net, and the dealer sells 6 shares into the rally.
- Stock down $1: delta falls to about .44, the position is short 6 net, and the dealer buys 6 shares into the decline.
Negative gamma (dealer short a put). Put delta of -.50 makes the short put worth about +50 shares, hedged with a 50-share short.
- Stock up $1: put delta moves toward -.44, the hedge is now 6 shares too short, and the dealer buys 6 shares into the rally.
- Stock down $1: put delta moves toward -.56, the hedge is 6 shares too long, and the dealer sells 6 shares into the decline.
Scale that across every strike and expiry and you get the two regimes. When modeled dealer gamma is net positive, hedging sells strength and buys weakness, the pattern associated with tighter ranges and mean reversion. When it is net negative, hedging buys strength and sells weakness, the pattern associated with moves that extend. Gamma exposure (GEX) is the aggregate estimate of which side the book sits on, usually expressed as dollars of hedging per 1% move.
Is gamma directional?
No, and this is the most common misreading of the regime. A positive-gamma reading is not bullish, and a negative-gamma reading is not bearish. The sign describes how the market is likely to respond once a move starts, not which way it starts.
Two documented sessions make the point better than theory. On 15 August 2023, Cboe's own market-maker data showed dealers long about $2bn of SPX gamma at 3pm as a selloff began, in Mandy Xu's account of 0DTE activity. And on 5 August 2024, when the VIX rose 180% to almost 66 before the open, the BIS found that market makers had positive options gamma exposure at the start of the regular trading session. The same bulletin traces the spike to widening bid-ask spreads, with less liquid put options accounting for more than 85% of it, and concludes that market makers' option hedging does not seem to have contributed meaningfully. Its charts also show market-maker gamma turning negative later that day as they likely provided liquidity, a reminder that the sign is a snapshot, not a fixed state.
So a positive regime can coexist with a sharp decline, and a negative regime amplifies short-covering rallies exactly as readily as it amplifies selloffs.
What the research finds
The evidence for a regime effect is real, but it is narrower than most commentary implies.
- Intraday momentum, not daily direction. Baltussen and coauthors find that across more than 60 futures markets the return in the last 30 minutes is predicted by the return over the rest of the day, and link it to hedging short gamma, which requires trading in the direction of price. In S&P 500 options from 1996 to May 2020 they count 2,930 negative net gamma days and 3,158 positive ones, and report no significant intraday momentum on the positive days. The effect reverts over the following days.
- Liquidity does part of the work. Barbon and Buraschi's working paper Gamma Fragility links negative (positive) aggregate gamma imbalance to intraday momentum (reversal) through its interaction with illiquidity, with the effect strongest in the least liquid names.
- Volatility and tail moves. In the Review of Financial Studies, Ni, Pearson, Poteshman and White report a noninformational channel through which market-maker hedge rebalancing affects stock return volatility and the probability of large price moves.
- Size matters as much as sign. The same Cboe analysis put intraday dealer net gamma from 0DTE SPX contracts at $170mm to $670mm, or 0.04% to 0.17% of daily S&P futures liquidity, because customer flow was largely two-sided.
Read together: the regime conditions how a move behaves inside the session, and its force depends on how large the hedging requirement is against available liquidity.
Where one regime becomes the other
Dealer gamma depends on spot, because the strikes that dominate the book change as price moves through them. The level where modeled gamma crosses zero is the gamma flip, which SquawkFlow calls the Flip Line. Above it the model has dealers net long gamma; below it, net short. The GEX flip price explainer covers how the level is computed.
Three things move the boundary or change its force:
- Expiry. Because gamma concentrates at the money near expiration, the book's curvature builds into an expiry and then vanishes at the close. That is part of why the market pins on Fridays.
- Time and volatility. Charm and vanna shift dealer deltas even when price stands still; vanna and charm explained shows how.
- Stale positioning. Open interest settles once a day, so any open-interest-based regime is computed from the prior close. When spot genuinely crosses the Flip Line during the session, the side of the line spot is on is the more current read, and a morning net GEX total that still shows the old sign should not overrule it.
Reading it on SquawkFlow
The free SPX GEX page shows the modeled regime, the Flip Line, the call and put walls, and the open interest settlement date the model was built from. Treat the regime as context for what a move may feel like, built on an assumed dealer sign, never as a call on direction.
Educational content, not financial advice. See our risk disclosure.