002 · Section 02 >>>>>>>
We aren't building eyes. We are building cognition.

02 —
Research

The Quant Ledger is our translation and application layer: how raw macro inputs, positioning, vol structure, and validated research become actionable opportunity framing — when a setup matters, what would invalidate it, and how we would execute if it holds.

The analytical toolkit used by institutional desks —dealer gamma, Fed liquidity vectors, COT z-scores, full IV surface — decoded for practitioners who want to build edge, not dependency. Every signal. Every source. Named.

Choose your route — dive deeper
02.1 >>>>>>>

Research
Categories

The 20 public-facing categories we publish into — gold-framed, sourced, and repeatable. Scroll the rail, tap to read.

Public-facing research categories · 20 — scroll, tap to read
Category 01 / 20
Commitment of Traders

Managed-money net positioning z-scored against 52-week history. Crowding and capitulation in gold futures, quantified — the positioning input that directly shapes how GC trades.

02.2
Application Layers
Each layer feeds the next — macro liquidity sets the regime context, positioning confirms it, vol surface reveals dealer mechanics, and technical structure defines execution.
Layer 01 — Foundation

Macro Liquidity

Fed balance sheet mechanics, RRP balances, SOMA holdings, and TGA dynamics synthesized into a directional USD liquidity score — updated weekly via OpenBB state vector.

SOMARRPEFFRBeige BookTGAH.4.1
Layer 02 — Structure

COT & Sentiment

Managed money net position z-scored against 52-week history. ETF flows in GLD/IAU tonnes. CRB momentum. Central bank net purchases lagged one month. Composite bullish sentiment index.

COT mm_net_zETF FlowsCRB IndexCOMEX Delivery
Layer 03 — Volatility

Options Surface

Full IV surface across all strikes and maturities. VRP as directional regime signal. Dealer gamma, vega, and charm exposure. Skew and kurtosis (tail-risk shape). Put-call parity implied dividend curve.

GEXVRPSkewCharmIV Surface
Layer 04 — Execution

Order Flow

Footprint, TPO, volume profile and VWAP deviation. CVD divergence as primary microstructure signal. Absorption measurement, sweep detection, book delta, stacked imbalances.

CVDFootprintTPOSweepVWAP
02.3
What This Is
Product definition, role, and core function — so you know exactly why you are subscribing.

The Quant Ledger is the research publication layer of the Ellipsys stack. It captures what the system learns through data, backtesting, execution review, feature research, model failure, regime behavior, and validation work — then translates that learning into durable research assets.

Not a daily newsletter. Not a signal product. Not a generic blog. It is the proof layer, credibility layer, and long-tail intellectual property layer.

“This person is not just posting market opinions. This system studies, validates, rejects, archives, and learns.”
The reaction every post is built to earn
Backtest studiesWalk-forward validation notesExecution studiesFeature engineering notesRegime detection researchModel postmortemsFailed hypothesis archiveWhite papers from profitable backtestsOriginal alpha research
02.4
What We Publish
Standard research artifacts — durable, repeatable deliverables, not one-off posts. Every artifact follows a required structure.
Artifact
Feature Engineering Note

Published when a feature shows useful behavior even without a full strategy attached.

Feature definitionInput sourcesTransformationExpected mechanismRegime sensitivityFailure conditions
Artifact
Execution Review

Published when live or simulated execution reveals implementation lessons worth keeping.

Setup contextExecution plan vs. actualSlippage and costsTiming qualityRisk behavior
Artifact
Regime Study

Published when a signal, feature, or strategy changes behavior under different macro or market regimes.

Regime definitionClassification rulesTime-in-state analysisFeature behavior by regimeStrategy behavior by regime
Artifact
Validated Research Note

Published when a hypothesis passes the full evidence threshold — the white-paper tier.

Hypothesis and mechanismWalk-forward validationRobustness checksRegime-conditioned resultsWhy it may be trustedWhy it may still fail
02.5
The Evidence Bar
Sentinel evidence requirements — every strategy or signal study reports all of it, or it does not publish.
Nothing publishes without this
Required in every strategy and signal study, without exception:
  • Hypothesis
  • Feature set
  • Instrument
  • Timeframe
  • Sample window
  • Costs and slippage assumptions
  • Train and test construction
  • Walk-forward validation
  • Robustness checks
  • Regime-conditioned behavior
  • Trust or distrust rationale
02.6
Research Session Pipeline
A complete Quant Ledger post is generated from a full research session — thirteen steps, hypothesis to archive.
01 /
Hypothesis formulation
02 /
SEV data pool definition
03 /
Dataset manifest
04 /
Feature manifest
05 /
Strategy specification
06 /
Sentinel backtest
07 /
Walk-forward validation
08 /
Prism tear sheet
09 /
Sensitivity surface
10 /
Failure mode analysis
11 /
Trust or distrust explanation
12 /
Publication draft
13 /
Archive entry
Scroll →
02.7
The Engines
The machinery behind every artifact — research runs in Jupyter notebooks, Codex, and Claude Code, then passes through these three.
Feature layer
Parallax

Turns raw market data into research-ready structure: market profile, auction context, order-flow summaries, and session-aware feature frames consumed by every study we publish.

Market Profile / TPOOrder-flow summariesCVD framesSession structureNotebook-ready tables
Validation engine
Sentinel

Event-driven backtesting with honest execution simulation — realistic fills, session-aware spreads, slippage, and a walk-forward harness. Nothing publishes without passing through it.

Event-driven simulationRealistic fill modelingWalk-forward harnessOut-of-sample checksDrawdown kill-switch
Scoring layer
PRISM

Scores every validated run and renders the verdict: pass, warn, or reject. Tear sheets, sensitivity surfaces, and plain-language trust or distrust explanations.

Pass / Warn / Reject verdictsTear sheetsSensitivity surfacesScore breakdownsTrust / distrust rationale
System inputs the Ledger receivesFeature ingestion outputsBacktesting and validation resultsExecution attributionRegime labelsFailed hypothesesModel postmortemsAnomalies worth formal studySentinel evidence packetsPrism tear sheets
Boundary

Market Codex is a separate product. It may pull from some of the same sources and share some of the same meaningful information, but the two are not coupled. The Quant Ledger is the gold research publication and translation layer — nothing else.


02.8 >>>>>>>

Five Instruments.
One System.

Options surface anchor · full decomposition on GC
GC
COMEX Gold Futures
Trend
3,047
▲ 34.40 · +1.14%
Primary benchmark for the desk. Regime, liquidity, and options pressure are mapped here first, then checked across MGC, XAU/USD, GLD, and GDXU for implementation and public-market confirmation.
Gamma Wall3,100
Support2,980 / 2,920
VRP-1.4%
MGC
Micro Gold Futures
Watch
3,047
▲ 33.80 · +1.12%
Same underlying as GC at smaller contract size. We use it for sizing, session liquidity, and slippage versus full-size prints — not as a separate “thesis,” always in lockstep with GC.
Vs GC0.03% tracking
LiquidityUS / EU overlap
SpreadStable
XAU/USD
Spot Gold CFD
Bullish
3,041
▲ 29.20 · +0.97%
24-hour spot reference for gap risk and headline-driven moves. Cross-checked against GC for basis, session anomalies, and broker-specific behavior before any live read is trusted.
Asia+0.31%
London+0.22%
USAligned
GLD
SPDR Gold Shares ETF
Flow
292.4
▲ 2.1 · +0.72%
Institutional and retail access vehicle. Flows and holdings changes are read as slow-moving conviction and funding pressure alongside futures positioning — still one leg of the same gold book.
Holdings Δ+4.2 t
Premium / NAVTight
Vs GCConfirm
GDXU
MicroSectors Gold Miners 3x ETN
Vol
42.8
▲ 2.4 · +5.9%
Leveraged miners beta for stress-testing momentum and vol-of-vol around the same macro gold narrative — used sparingly for convexity context, not as a second primary.
Beta vs GC~2.1x
DecayMonitor
BookSatellite

02.9 >>>>>>>

Every Conclusion
is Traceable.

No black boxes. Every signal maps to a named, primary government, exchange, or regulatory source.

Layer 01
Federal Reserve
SOMA HoldingsH.4.1 ReleaseNY Fed RRPEFFRBeige BookFed Balance SheetForeign Bank HoldingsFedWatch CME
Layer 02
Treasury & Rates
Daily Par Yield CurveAuction ResultsSIFMA Fixed IncomeCorporate Bond StatsFreddie Mac PMMSS&P MBS IndexCredit Spreads
Layer 03
Positioning & Flows
CFTC COT ReportsETF Flows (GLD/IAU)ICI MMF DataBarchart Spec/CommCOMEX DeliveryCRB IndexCapital & Country Outlooks
Layer 04
Economic Data
FREDBLS LaborBEA (GDP/PCE)EIA EnergyCensus HousingTradingEconomicsBrookingsHousehold Consumption/GDP
Layer 05
Markets & Options
CME QuikStrikeFull IV Surface (GC)Open Interest CFTCSector SPDRsLipper AlphaMarketAuxEDGAR (10K/10Q)
Layer 06
Physical Gold
COMEX DeliveryLBMA SpotSGE Au9999Shanghai–London BasisCentral Bank Net PurchasesMine Supply/Surprises

02.10 >>>>>>>

Education

The Quant
Ledger

We exist because retail practitioners deserve the same analytical frameworks institutional desks use — not simplified summaries, not signals to follow blindly, but the actual ingestion-scoring-regime pipeline, taught completely.

Every post teaches the framework alongside the conclusion. Every data source is cited. Every OpenBB script is explained. We don't give you fish. We teach the entire quantitative ocean.

5
Gold instruments
GC MGC XAU GLD GDXU
24+
Primary data
sources integrated
20
Public research
categories
4
Analytical layers
per deep-dive
© 2026 THE QUANT LEDGER · EDUCATIONAL RESEARCH ONLY · NOT INVESTMENT ADVICE