Inside the membership · Fundamental Desk

The macro read on gold — and the line where it's wrong.

One thesis, one regime, and a falsifiable contract that says exactly what would prove it wrong. Across eleven instruments, refreshed on a schedule and on the events that actually move them.

Start your free 7-day trial →No card · 7 days free
Metals2/2 live
  • FRESHXAU/USDGOLD neutral
  • STALEXAG/USDSILVER bearish
Crypto2/2 live
  • WATCHBTC/USDBITCOIN neutral
  • FRESHETH/USDETHEREUM bullish

The live desk board, 27 August 2026.

Most macro commentary tells you what happened. The desk tells you what it would take to be wrong.

The output

One thesis. One regime. One line where it breaks.

This is a real thesis the desk generated for XAU/USD on 27 August 2026 — not a mock-up. Every number in it was pulled from live data, and the invalidation contract underneath is the part most macro writing never commits to.

27 Aug 2026, 07:01 UTC · FRESHXAU/USD
neutralConviction
Invalidation · the line
Invalidate above4732.0Invalidate below4600.0Watch zone4640.0 – 4700.0Max age6h
Regime

Real yields elevated at 2.32% but overridden by DXY weakness at 99.17 and a structural bid; gold consolidates below 4730 into Jackson Hole and Warsh remarks.

The thesis

Bias is neutral into Jackson Hole. Structurally gold is unambiguously bullish — price at 4658 sits above the H4 50 EMA (4604.87), D1 21 EMA (4423.81) and D1 50 EMA (4335.42), with DXY at 99.17 and a persistent official-sector bid overpowering real yields at 2.32%. Tactically, though, the last three H4s printed lower highs from the 4730 sweep. With Warsh at 16:15 UTC and sticky-PCE risk on the tape, committing directionally before the event is the wrong trade.

Active catalysts
  • Jackson Hole / Fed Chair Warsh (16:15 UTC)pull · neutraleffect · neutral
  • Sticky PCE narrativepull · bearisheffect · muted
  • Soft US yields overnightpull · bullisheffect · bullish
  • Structural central-bank bidpull · bullisheffect · bullish
Invalidation contract · falsifiable
  • invalidateH4 close above 4730 — resumes trend, neutral thesis wrong
  • invalidateWarsh explicitly hawkish — flips bearish via real-yield channel
  • watchDXY reclaims above 100.00 — removes primary bullish support
  • watchUS 10y real yield > 2.45% — tests CB-bid override
Why this conviction
  • Structural trend (bullish) and near-term price action (lower highs from 4730) disagree — genuine balance
  • Jackson Hole / Warsh in under 10 hours makes directional commitment low-quality
  • Prior desk streak: bearish tracked then flipped neutral; last two neutral calls saw price whipsaw ±0.5%
Data gaps
  • fed_funds_futures_implied_dec unavailable — cannot verify Fed pricing shift into Warsh

Real generated output, 27 August 2026. Yours updates through the trading day.

The board

Eleven instruments. Not just gold.

Metals, the FX majors and the two majors in crypto — each with its own bias, its own conviction score and its own invalidation contract, so you can see where the macro agrees with itself and where it doesn't.

Metals2/2 live
  • FRESHXAU/USDGOLD neutral
  • STALEXAG/USDSILVER bearish
FX Majors7/7 live
  • FRESHEUR/USDEURO neutral
  • FRESHGBP/USDSTERLING bearish
  • FRESHUSD/JPYYEN bullish
  • FRESHUSD/CHFFRANC neutral
  • WATCHAUD/USDAUSSIE bullish
  • FRESHUSD/CADLOONIE bullish
  • WATCHNZD/USDKIWI neutral
Crypto2/2 live
  • WATCHBTC/USDBITCOIN neutral
  • FRESHETH/USDETHEREUM bullish

The live board, 27 August 2026.

How it thinks

A conviction score you can argue with.

The desk publishes its reasoning, its uncertainty and its own recent record — so a low conviction number means something rather than hiding behind hedged language.

A falsifiable contract

Every thesis names the levels that invalidate it, a watch zone, a trend filter and a maximum age. It expires rather than quietly rotting.

It grades its own calls

Past theses are scored against what price actually did, and that record feeds the next read. When the desk has been whipsawed, it says so.

Conviction, not hedging

Uncertainty is expressed as a number out of ten rather than as weasel words. A 0.42 read is openly a low-conviction read.

Data gaps are published

When an input is missing, the desk lists it instead of writing around it. You see what it couldn't check.

How it works

Two screens, then you're reading

  1. STEP 01

    Pick an instrument

    Eleven tiles, each showing its current bias and conviction at a glance. Start with the one you actually trade.

  2. STEP 02

    Read the thesis

    The regime, the drivers in play, and the contract that says where it's wrong — written to be read in a couple of minutes.

  3. STEP 03

    Send yourself the PDF

    One tap emails you the full thesis as a two-page PDF, so you can keep it beside the chart while you work.

Why it works

The why behind the move, without ten news tabs.

The desk exists to explain the macro backdrop you're trading into, and to be honest about how confident it is. It is fundamental analysis and education — never a signal, a price target, or an instruction to buy or sell. What you do with the read is entirely your decision.

Read today's gold thesis in full.

Full access for 7 days, free. No card to start.

Start your free 7-day trial →

Fundamental analysis only — not a signal, target, or instruction to buy or sell. The contract above states precisely where this thesis is wrong.