Strategy · Options / GEX

GEX Regime

Dealer gamma decides whether a level holds or breaks. This page shows the regime live, and warns you as price walks into a wall. Our own testing said selling was the only side of the options trade worth taking; the ranked ladder that acts on that is still in build.

Start for $8
Powers the Dealer’s Hand strategy

The regime, right now

Net dealer gamma per name, in dollars. The sign is the whole story: positive means dealers trade against the move and the range holds; negative means they trade with it and levels give way. Index and ETF majors first, then the largest gamma names.
Scope: our study measured QQQ only. The regime reading is live for every name shown; the hit rates further down were not measured on any of them except QQQ.

Loading the live gamma feed…

Wall proximity warning

The trigger the strategy actually waits for. As price walks toward the call wall or the put wall, this escalates — and it stands itself down the moment gamma turns negative, because that is exactly when walls stop holding.

Waiting on the live feed…
The bands. Inside 0.25% of a wall is the trigger zone. Inside 1.0% is a heads-up. Wider than that is mid-range and there is nothing to do. Every band is a setting you can move on the desk — these are the defaults, not a law.

The 0DTE credit-spread ladder

When a wall goes hot in positive gamma, the desk ranks the defined-risk verticals you could sell against it — short strike at or just beyond the wall, wings bought for protection, priced off the live bid/ask mid.

Loading…
Defined risk only. Every rung is a vertical with a bought wing — no naked short premium, ever. Max loss is the width of the spread minus the credit, and it is printed on the row before you do anything with it.

How often it actually fires

A wall only matters if price gets there. Measured on QQQ, this is how often each wall was reached, and when in the session it happened.

WallTouched, all daysTouched, negative gammaTouches in the first 30 min
Call wall52%60%73%
Put wall32%53%52%
Both walls held and price pinned between them on 30.6% of positive-gamma days, versus 23.1% of negative-gamma ones.
Read the second column carefully. Both walls are reached more often when gamma is negative — and negative gamma is exactly when this strategy refuses to sell. So the sell-side opportunity set is thinner than the headline rates suggest, and it is thinner on the put side than the call side. The call wall is the busier of the two by a wide margin. Most touches land in the first half hour; after 11am the wall trade is largely gone for the day. Measured on QQQ over the study window — not a promise about any other name or any future week. One caveat we cannot yet close: wall touches are detected from session highs and lows, and our own data notes flag raw high/low prints as unreliable before a later correction. We have not confirmed these rates were recomputed afterwards, so treat them as indicative, not settled.

How the strategy plays it

Dealer’s Hand is a regime filter first and a trade second. The regime decides whether you are allowed to sell at all.

1. Positive gamma — sell the wall

Dealers dampen the move, the range holds, and the wall is a fence. This is the only regime the strategy sells in. The measured result behind it is a put credit spread; placing the short strike at the wall, and the call-side version, are our construction and are not separately tested.

2. Negative gamma — stand down

Dealers amplify the move and walls stop holding. The strategy does not sell here. It says so on the panel and shows no ladder at all rather than a tempting one.

3. Wait for the wall, not the clock

No entry until price is actually in the trigger band. Approaching is a heads-up, not a signal. Most days never trigger, and that is the point.

4. Defined risk, priced at the mid

Both legs, one order, limit priced off the live bid/ask mid rather than lifting the offer. Max loss is known before entry and capped by the wing.

5. Every entry carries an exit

A profit target and a stop go on with the trade, not after it. Nothing is left to be watched by hand.

6. Flat before the close

0DTE positions are closed out in the last half hour. The strategy does not carry gamma risk into the bell hoping for a pin.

7. Losing streak pauses it

More consecutive losses than the strategy's own average means it stops and waits. Edges are cyclical — ride it while it works, pause when it stops, let it come back.

8. Everything is a dial

Proximity bands, spread width, minimum credit, position cap, session windows. Defaults ship sensible; nothing is welded shut.

What our own testing found

Not a textbook description of gamma exposure — the measured results from our study, including the parts that argue against trading it.

The walls reject far more than they break

Intraday, price bounced off the near-term wall 67–93% of the times it was tested. On positive-gamma days the put wall held 93–95%.

The sign sets the size of the day

Strongly positive gamma (over +$2bn) gave a next-day move around 0.6%. Negative gamma gave about 1.0%, and twice the rate of moves past 1.5% (23% against 47%). Comparing only the extremes — beyond −$4bn against beyond +$4bn — that gap widens to about .

Buying the bounce lost money

The bounce is real but tiny, roughly 0.15–0.33%. That does not clear the spread on a bought option. Every long-option variant we tested came out negative after real costs.

Selling it was the side that survived

Defined-risk put credit spreads on positive-gamma days: +12%, 95% win. That is the only structure with a measured number — the call-side version is untested, and so is placing the short strike at the wall.

Read this before you weigh those numbers. The study ran on QQQ only — 189 trading days of real options data, September 2025 to June 2026. That is one macro regime, bull into correction; a sustained bear or a dead-calm year is untested, and the figures were not measured on the other names shown above. Fills were modelled on 5-minute bars, not tick data. The credit-spread leg has not been crash-tested — its worst case is a gap straight through both strikes, and the June 5 crash sits outside the sample. The +12% / 95% figures were measured on the study's own spread construction, not specifically at 0DTE. Past results are not a forecast, and nothing on this page is financial advice.

Rent the strategy, not the guesswork

Dealer’s Hand runs on the Northtape desk: the live regime and the wall warning, kept current for you. The ranked 0DTE ladder and the streak pause are in build and are not live yet. Like every Northtape strategy the results are public and the method stays closed — you rent the running edge, not a file to maintain.

See pricing →

The results are public — the recipes are not. Every strategy name is a Northtape brand; the underlying method stays proprietary and closed-source. See the full track record → and the Dealer’s Hand certificate →. Hypothetical, not a guarantee of future results, and not financial advice.

Four ways in
Rent the desk, or own the edge.
Rent
$8/mo
The whole desk for six months, then twenty a month. Cancel anytime.
Rent the desk
Own
$99once
Buy one named strategy outright and keep it. Nothing recurring.
Buy a strategy
Bundle
Both
The named strategies and the algorithms together, in one package.
See the bundle
Algorithms
$199/mo
The algorithms, and we choose the stocks they run on each session.
See algorithms
One desk instead of six subscriptions.
Cancel anytime · secure checkout by Stripe
Monthly
$8/mo
founding rate · cancel anytime
3+ Stock Picks every trading day
Options Flow with OI, Greeks and GEX/DEX
Momentum & Volume Scanners and Dealer Walls
Top 7 Hedge Fund Portfolios to clone
Real-Time Quotes at no extra fee
Go monthly
Best Value
Annual
$20/mo
$240 billed yearly · one payment
Same as Monthly, only cheaper — one easy payment for the year and you’re done. Every new engine Jarvis learns ships straight to your desk, no upcharge.
Go annual
Add-On
Jarvis Fully Automated
$26.99/mo
first 3 months, then $49/mo · on top of any plan
The quantitative-reasoning engine goes hands-off: backtested entries, sizing, trailing exits and a hard kill switch, executed for you inside the same risk gates the desk shows you.
Add Full Auto