Reflex Research API Documentation

Reflex Research provides programmatic access to derived and measured datasets.

The focus is on clean, reproducible market structure signals, not raw exchange feeds. The derived data is from options market pricing and volatility, and the measured data contains historic and current information on job openings within companies. We focus on providing high quality data on a limited number of companies, rather than raw data for an entire exchange. Currently for the derived data we focus on the S&P500 and the top 25 companies in the index. For the measured data, at least the top 25 companies in the index are covered, with more being added frequently.

The options market and the underlying stock market are tightly linked, using information from options can yield great insight into the underlying stock market.

Breeden–Litzenberger Probability Density Function

Breeden–Litzenberger (BL) probability density function describes the risk-neutral distribution of the underlying price at a specific option expiry, as implied by the full listed options market.

Rather than relying on historical returns or parametric assumptions, BL extracts this distribution directly from option prices across strikes. In practical terms, it answers a single, highly actionable question; Given today’s option prices, what range of outcomes is the market assigning probability to at expiry, and how is that probability distributed?

The result is a forward-looking, market-implied probability density, conditioned on one expiry.

What BL captures and why options matter:

Options encode state-contingent pricing. Deep out-of-the-money options reflect tail risk, while skew and curvature across strikes reveal asymmetries in market expectations.

By using the entire option surface, BL incorporates:

• Downside crash insurance pricing.
• Upside convexity demand.
• Volatility skew and smile effects.
• Concentrated liquidity and positioning at specific strikes.

This makes BL fundamentally different from models based on realised volatility or fitted distributions. It is forward-looking by construction and anchored to where capital is actively being priced.

Practical trading and risk applications:

BL PDFs are not forecasts in a traditional sense. Their value lies in understanding what the market is pricing, and how that pricing changes over time.

Common applications include:

• Tail risk analysis.
• Measuring how much probability mass is assigned to extreme outcomes.
• Skew and convexity monitoring.
• Identifying shifts in distribution shape before they appear in spot or realised volatility.
• Scenario stress testing.
• Evaluating portfolio exposure under market-implied scenarios rather than arbitrary shocks.
• Relative value and strategy design.
• Comparing implied distributions across expiries or against internal models to inform option structures.

Importantly, BL does not need to be “correct” in a real-world sense to be useful. The edge comes from understanding how the market is pricing risk, not from predicting the future.

Gamma Exposure (GEX)

Gamma Exposure (GEX) describes how option gamma is distributed across strikes for a given expiry, weighted by open interest. It provides a structured view of where convexity is concentrated in the options market, and how sensitive price dynamics may be around those levels.

Rather than attempting to infer dealer positioning or flow direction, the Reflex Research GEX dataset focuses on observable, mechanically defined quantities derived directly from listed options.

At its core, GEX answers a focused structural question; Where in price space is option convexity concentrated, and how sensitive is the market to underlying price movements at those levels?

What GEX captures and why gamma matters:

Gamma governs how rapidly option delta changes as price moves. In practice, this determines how non-linear exposure builds around certain strikes, and where small price changes can have outsized effects.

By aggregating option gamma weighted by open interest, GEX captures:

• Where convexity is concentrated in the option chain.
• Which strikes are most sensitive to small price movements.
• Regions where price movement may accelerate or dampen due to nonlinear effects.
• Structural differences between call-heavy and put-heavy regions.

Calls and puts are treated symmetrically, and no assumptions are made about who holds the positions. The dataset reflects what exists in the option chain, not who is presumed to be on the other side.

This keeps GEX grounded in what can be measured, rather than what must be inferred.

Practical trading and risk applications:

GEX is best understood as a market structure indicator, not a directional signal. Its usefulness lies in contextualising price behaviour rather than predicting it.

Common applications include:

• Identifying high-convexity price regions.
• Highlighting levels where price movement may accelerate or stabilise.
• Contextualising volatility regimes and regime shifts.
• Strike-level risk mapping for stops, targets, and sizing.
• Informing options strategy construction around key convexity zones.
• Comparing convexity structure across expiries to identify changes in market focus.

As with BL, the edge does not come from GEX being “right” in a predictive sense. It comes from understanding how the options market is structurally arranged at a given moment, and how that structure evolves as price moves.

VIXnD

VIXnD represents the market-implied forward volatility over a specific horizon, derived directly from listed option prices rather than historical returns or model extrapolation.

Conceptually, it answers a focused question; What level of volatility is the options market currently pricing for the next n days, conditional on today’s information?

Unlike spot VIX, which is anchored to a fixed 30-day maturity and limited to the S&P 500, VIXnD generalises the concept of forward-looking implied volatility to arbitrary horizons, aligned with specific option expiries. The same methodology is also applied to individual stocks.

What VIXnD represents:

The Reflex Research VIXnD dataset is computed using the same methodology as the Chicago Board Options Exchange (CBOE) VIX, but evaluated over user-defined horizons.

Key characteristics include:

• Derived entirely from listed option prices.
• Forward-looking by construction.
• Horizon-specific rather than calendar-fixed.
• Anchored to discrete option expiries.
• Independent of realised or historical volatility.

In effect, VIXnD answers “what volatility is priced now for this horizon”, not “what volatility has been observed”.

Why a horizon-specific volatility matters:

Most trading decisions are horizon-dependent. A one-week position and a three-month position face fundamentally different risk profiles, even if spot volatility is unchanged. By aligning implied volatility to a specific horizon, VIXnD allows users to:

• Compare short-dated versus longer-dated risk pricing.
• Detect compression or expansion in forward volatility.
• Separate spot volatility moves from term-structure effects.
• Anchor strategy risk to the actual holding period.

This is particularly useful when volatility term structure is steep, inverted, or unstable.

Practical trading and risk applications:

VIXnD is best interpreted as a pricing reference, not a signal generator. Typical applications include:

• Volatility regime identification.
• Strategy calibration relative to market-implied risk.
• Relative value analysis across horizons or against internal forecasts.
• Risk budgeting using forward-looking volatility measures.
• Event risk assessment at specific horizons.

As with other implied measures, the informational value lies in how VIXnD changes over time, and how it compares to alternative volatility references.

Interpretation caveats:

VIXnD does not attempt to:

• Predict realised volatility.
• Smooth or model the volatility surface.
• Infer positioning or flow.
• Correct for risk premia.

Like all implied volatility measures, it embeds risk premia, supply–demand imbalances, hedging pressure, and liquidity effects, particularly at short horizons.

It should therefore be interpreted as a market state descriptor, not a forecast. Its usefulness comes from understanding how volatility is being priced across horizons, and how that pricing evolves through time.

Jobs

This section is still under construction.

How the datasets fit together

Reflex Research provides a set of complementary datasets designed to describe market expectations, market structure, and underlying economic activity, all derived from observable data rather than narrative or discretionary interpretation.

The options-derived datasets, Breeden–Litzenberger (BL), Gamma Exposure (GEX) and VIX-n-D, form a coherent view of how risk is being priced in financial markets. BL describes where the market assigns probability at expiry, GEX describes where convexity and sensitivity are concentrated across price levels, and VIXnD describes how much volatility is being priced over specific forward horizons. Together, they provide a structured, forward-looking representation of expectations, asymmetries, and nonlinear risk embedded in option prices.

The Jobs dataset operates at a different layer. Rather than describing market structure, it captures real economic signals derived from corporate hiring activity. Changes in job postings, role composition, and hiring intensity provide insight into company-level and sector-level dynamics that are often visible well before they appear in earnings, guidance, or macro releases.

Viewed together, these datasets allow users to connect economic activity to market pricing, and market pricing to risk and positioning. This enables workflows that move beyond single-indicator trading toward a more complete understanding of what the market is pricing, where it is sensitive, and what underlying conditions may be driving those expectations.

Reflex Research does not attempt to collapse these signals into a single model or narrative. Instead, it provides clean, well-defined primitives that users can combine according to their own frameworks, strategies, and time horizons.

What this site is

• Introduction to the metrics.
• Practical Python examples.
• Minimal theory, maximum usability.
• Copy-paste-friendly code.

What this site is not

• A marketing website.
• A trading tutorial.
• A replacement for the API schema.

For endpoint inspection and schema details, use the Swagger UI exposed by the API.