Hermēs DocumentationBasis System
The same "gamma level" sits at different strikes under different bases. This chapter lays out the three axes HermesGEX makes explicit — perspective (customer vs dealer), measure (customer gamma / net-flow gamma / net GEX) and expiry (0DTE / next-day / 90D aggregate) — and how to read them when they disagree.
A GEX chart that does not state its basis cannot be verified. You have no way to tell whether that "major positive gamma" line is where customers hold gamma, where today's aggressive flow piled up, or where net GEX peaks. Mathematically these are three different quantities, and on a live tape they frequently sit at three different strikes.
HermesGEX makes the basis explicit: every key level shows which ledger it is reading, switching takes effect immediately, and the chart keeps the evidence (the pill carries a basis suffix). This chapter covers that system.
Three orthogonal axes
Every Hermes key level is defined by three questions at once:
| Axis | Question | Values |
|---|---|---|
| Perspective | Whose book is this? | Customer / Dealer |
| Measure | Computed from what data? | Customer gamma / Net-flow gamma / Net GEX |
| Expiry | Which contracts? | 0DTE / Next-day (1DTE+) / 90D full aggregate |
The axes are orthogonal — changing expiry does not change perspective, and changing perspective does not change the measure. Settle all three before you read the chart; it is the precondition for any GEX claim to be checkable.
Why perspective matters. Customer and dealer are the two sides of one trade: gamma the customer buys is gamma the dealer sells. Same number, opposite sign. Mixing the two is the single most common source of confusion in GEX discussion — see Common misconceptions.
+Γ / −Γ: three "real gamma" bases
Click the +Γ / −Γ pill on the main chart to cycle bases. The three are not three algorithms over one dataset — they are three separate datasets:
| Basis | Pill tag | +Γ reads | −Γ reads | Meaning |
|---|---|---|---|---|
| Customer gamma (default) | none | Customer long-gamma concentration | Customer short-gamma concentration | Where gamma sits on the customer book — the industry's "major positive/negative gamma" |
| Net-flow gamma | ᶠ | Peak call gamma | Peak put gamma | Gamma peaks after millisecond-level aggressor classification nets out the imbalance — where today's new flow is landing |
| Net-GEX gamma | ᴳ | Most-positive net-GEX strike | Most-negative net-GEX strike | Volume-weighted net-GEX peak — identical to the major call wall C1 / put wall P1 |
Under the net-GEX basis, +Γ is C1 and −Γ is P1. If you have both +Γ and the call wall on and they overlap, nothing is double-drawn — you are simply on that basis. Under the default customer basis, a gap of several strikes between them is the normal case.
Missing data is left blank, never guessed
The net-GEX basis comes from the main chain and is always available. The net-flow basis only exists when the upstream State family carries it. When it is absent Hermes does not draw the level and does not substitute an approximation — a blank means "this basis has no verifiable data right now".
D+ / D−: two net-delta bases
The D+ / D− net-delta extremes cycle the same way, but only across two states:
| Basis | Pill tag | D+ | D− |
|---|---|---|---|
| Dealer (default, upstream native) | none | Peak dealer positive net delta exposure | Peak dealer negative net delta exposure |
| Customer | ᶜ | Takes the dealer negative peak | Takes the dealer positive peak |
This follows from the data itself: the upstream delta field is defined as net dealer exposure and carries no separate customer-classified series. The customer view can only be derived by flipping the sign — swap the two peaks and you are done; there is no independent third dataset. Gamma gets three states and delta gets two not because a feature is missing, but because that is what the data supports.
Expiry: three independent settings
Expiry is the axis most often overlooked and the one that most often produces "the level I see isn't the level you quoted". Hermes splits it across three places that never interfere with each other:
Expiry of the heat trail
Which contracts the main-chart heat band reads per-strike gamma from: 0DTE (expiring today) / Next-day (1DTE+) / 90D (full aggregate to 90 days). When true per-strike gamma is unavailable, the 90D view falls back to the call/put-wall basis and is marked on the chart as a long-dashed, hollow-pill long-horizon reference.
Expiry of the distribution band
The right-hand distribution band picks 0DTE / next-day / aggregate on its own, independent of the heat trail — so you can overlay 90D structural distribution on a 0DTE heat trail and see whether a level reflects today's money or further-dated structure.
Per-slot expiry in the sub-panes
Each of the three sub-pane slots chooses 0DTE or next-day (1D) individually. It is implemented as an explicit metric key (e.g. "DEX Flow · 1D") rather than a global switch, so one screen can compare the same metric across two expiries.
Agreement and divergence: how to read it
The point of the basis system is not to pick the "best" one — it is to see whether the three tell the same story.
| What you see | Structural meaning | What to do |
|---|---|---|
| Customer / net-flow / net-GEX land on the same or adjacent strikes | Positioning, today's new flow and net exposure point at one price; high structural consensus | Treat it as a firm boundary; cross-check with Confluence |
| Net-flow diverges clearly from customer | New flow is relocating — the old book sits at A, fresh orders hit B | Watch how fast B builds; if it persists, B usually becomes the new boundary |
| Net-GEX (C1/P1) sits far from the customer gamma level | Volume basis and positioning basis disagree, typical around rolls and expirations | Favour the nearer expiry; cross-check the OPEX playbook |
| 0DTE and 90D levels coincide | Short-dated money is stacked on long-dated structure | The strongest magnet / barrier candidate |
Divergence is itself information. Three bases pointing at one price gives you consensus; three bases pointing at three prices gives you evidence that structure is shifting. Treating divergence as a data bug and hunting for "the right one" is the most common way to misuse this system.
Where the switches live
| What you want | Where |
|---|---|
| Switch +Γ/−Γ basis | Click the +Γ/−Γ pill to cycle, or pick it directly in Indicator Dock → Gamma → Basis |
| Switch D+/D− basis | Click the D+/D− pill, or Indicator Dock → DEX flow → D basis |
| Switch heat-trail expiry | Indicator Dock → Heat |
| Switch distribution expiry | Indicator Dock → Distribution |
| Switch sub-pane expiry | Indicator Dock → Panes, per slot |
Basis selection is shared globally within a browser: cycle the pill on the main chart and the distribution band, heat trail and replay all follow the same basis — no chart ever runs two conventions at once.
Next
Frequently asked questions
Why does the "major positive gamma" level move for the same ticker?
Because it depends on the basis. HermesGEX exposes three bases for +Γ — customer gamma (where customer long gamma concentrates), net-flow gamma (the peak call gamma of classified order-flow imbalance) and net-GEX gamma (the most-positive net-GEX strike, which equals the major call wall C1). They are three different datasets, so landing on different strikes is normal, not a data error.
Are +Γ and the call wall (C1) the same thing?
Only under the net-GEX basis, where +Γ = C1. Under the default customer-gamma basis, +Γ comes from customer long-gamma concentration while C1 comes from the volume-weighted net-GEX peak, and the two can sit several strikes apart.
Why does D+/D- have two bases when +Γ/-Γ has three?
The upstream delta series is already a net exposure measured from the dealer's side, with no separate customer-classified field, so the customer view can only be obtained by flipping the sign. Gamma, by contrast, has both a customer-classified field and an order-flow-classified field, which is why it carries three bases.
Step-by-Step GEX Reading
60-second reading SOP — each step tells you where to look in HermesGEX and which mantra to apply
Orderflow — DEX / GEX / Convexity + Net Metrics + AggDEX
Three lenses on every incoming order (direction · gamma · vol), three summary metrics for whole-book posture, one cumulative line for intent