Live, today's note FX, macro, ~6 min read

Tuesday: DXY broke 99. 10Y at 4.81. USD/JPY collapsed to a 7-month low.

Gold extended another $47 (+1.01%) to a fresh cycle high $4,651. Phase 3 of the multi-year-high framework is now roughly one session from the $4,700 interim target. DXY at 98.84 tested but did not reclaim the 99 handle. 10Y softened 3.4bp to 4.7011 (first pullback after five sessions of term-premium extension). The BEA moved Q2 GDP second and July Core PCE together to Wednesday Aug 26 (8:30 AM ET); Chair Warsh's first Jackson Hole keynote lands Friday Aug 28 at 10:00 AM ET. Consensus PCE 2.90% YoY: a print in the 2.85-2.95% band lets the pre-symposium regime hold; above 3.00% flips the pre-Warsh setup.

08 September 2026 · By Igor Komolov

Read the note See the engine

DXY, daily close

10 Aug to 08 Sept

98.84

-0.60% wk

Source: ICE DXY, daily close. Annotation by TradingFuse.

Recent research

All notes, RSS

Reference 8 Sep · I.K. · 9 min

A plain-English guide to when higher yields stop supporting the dollar.

The yield-differential model (higher UST yields pull dollar higher) works most of the time and breaks in three specific regimes. Regime 1: the yield move is term-premium rather than real-rate, so foreign capital does not rebalance into the dollar; signature is long-end steeper than front-end and TIPS breakevens widening. Regime 2: foreign policy repricing narrows the forward differential even when spot US yields rise; signature is a sharp move in the single pair with the policy news, other DXY components lagging. Regime 3: a crossflow shock (oil, terms of trade, reserves policy) changes the balance-of-payments arithmetic for the paired currency. The three regimes usually run together; separating them is the diagnostic skill. Tuesday September 8 delivered roughly 40/55/5 across the three.

Read

Reference 4 Sep · I.K. · 9 min

A plain-English guide to why a hot NFP might not move the dollar.

A hot NFP with the front-end pricing the hike but the FX not extending is a specific and recurring footprint. Three mechanisms produce it. Mechanism 1: dollar-long positioning is already stretched (CFTC net above 70th percentile), so the marginal dollar-buyer is a leveraged fast-money desk that already owns dollars. Mechanism 2: flip-week Week 2, where the aggressive dollar-long positioning that would have amplified a hot print is already in place; the print corroborates rather than creates. Mechanism 3: a crossflow story (BoJ-hawk, ECB-hawk, oil supply) offsets one leg of the DXY basket. The signature diagnostic is the ratio of front-end move (basis points) to DXY move (pips); a standard NFP prints 1.5-2.5 pips per basis point, a priced-in-rates-only print prints 0.3-0.7. Friday September 4 fits the pattern with 2018, 2022, and 2023 parallels.

Read

Macro 4 Sep · I.K. · 8 min

Friday: NFP +162k blew out the flip. DXY yawned, gold gave back.

August NFP printed +162,000 versus +53,000 consensus (a 200 percent-plus upside surprise) with unemployment holding at 4.1 percent, wages +0.3 percent MoM, and revisions adding +55k to prior two months. Fed hike odds for September 15-16 FOMC moved from roughly 50-55 percent to about 60 percent. Yet DXY closed only 16 pips higher at 99.16; EUR/USD moved zero pips; gold sold $62; 10Y up 2.2bp to 4.784 percent. The classic in-the-price footprint: front-end priced the print, FX did not because the flip has already run its Week 1 crowd-unwind. Working thesis moves to hawkish-hike 60 percent; the next swing factor is CPI Friday September 11.

Read

Reference 3 Sep · I.K. · 9 min

A plain-English guide to NFP-eve inside a Fed-hike-flip week.

NFP-eve inside a Fed-hike-flip week is the specific case where two overlapping compression shapes (post-flip crowd-unwind Week 1 and pre-NFP standard eve) run simultaneously. Five signatures identify the doubly-compressed shape: volume 20-30 percent below trailing, implied vol rising steeply while realised collapses, data-print responses smaller than the same prints outside a flip-week, cross-pair correlation tightening toward 1.0, session close unusually close to the 4:00 PM London fix. Three Friday resolution shapes with base rates 45 / 30 / 25 percent; the Case 3 soft-print response magnitude is 2-3x the Case 1 corroboration response because the compressed flip positioning has to unwind alongside the standard NFP-response flow.

Read

Macro 3 Sep · I.K. · 8 min

Thursday: claims and ISM Services corroborated the flip. Pre-NFP gamma compression.

Weekly Claims for week ending Aug 29 firm; ISM Services PMI in line with 54.3 consensus and Employment sub-index above 50. Gold bid off the $4,300 to $4,400 back-fill zone for the third session, closing $4,420 (+0.84 percent). DXY held 99.545 with a 25 percent retrace of Wednesday post-ADP high. 10Y softened 1bp to 4.78 percent as the front-end priced tomorrow NFP rather than today prints. Case for Phase 4 rebase-and-restart on gold gaining evidence. NFP-eve inside a flip-week is a doubly-compressed session: volume 20 percent below trailing, realised vol under 4 vol points, data-print responses smaller than the same prints would trigger outside a flip-week. Setup: hawkish-hike 55 percent, invalidation triggers if NFP prints below 40k with unemployment ticking to 4.3.

Read

Reference 2 Sep · I.K. · 9 min

A plain-English guide to ADP signal quality inside a hike-flip week.

ADP inside a flip-week has three jobs (standard ADP-to-NFP anchor; first cross-check on the flip labour-market story; positioning signal on how far the crowd-unwind has run) versus one outside a flip-week. Signal quality is worse in noise but better in information; net matters more but harder to read. Four archetypal flip-week ADP prints (firm-corroborates, ambiguous-with-offset, soft-without-revision, aligned-soft-with-ISM) each with different response magnitudes and different implications for the flip four-week base rates. Three tape-signature diagnostics (front-end direction and magnitude, DXY-vs-gold response, two-day volume trajectory) classify which of the four shapes today print delivered. Three common misreads: treating the miss as decisive when it comes with a firm revision, extrapolating Wednesday response to Thursday, ignoring the Beige Book on the same afternoon.

Read

All research · Reference explainers

The engine, MetraFuse

A multi-agent research system, with an editor on the door.

MetraFuse continuously ingests market structure, regime conditions, positioning data, and the economic calendar. It drafts. An analyst edits. The byline carries a human name. Nothing is published without sign-off.

Inputs are dated and stored. Every chart is reproducible.
No autoposting. Every note is reviewed and initialled.
Methodology is documented and versioned.

Read the methodology

SCHEMATIC, v2026.05

Market structureorder-flow, breadth

Regimevol, correlation

PositioningCOT, funding, ETF

Calendarmacro, earnings, supply

↓ agents

MetraFuse, core 4 agents, 1 editor

research_brief() . check_priors() . draft_note() . cite_sources()

↓ human review

Weekly note

Reference update

Methodology note

Portfolio

Each site is its own publication, published by us.

All sites

howdoesforexwork.com

Live

How Does Forex Work?

A plain-English reference for how the foreign exchange market works: currency pairs, pips, leverage, and risk.

142 pages . last reviewed 2026-05-21

cornusd.com

Live

CORN/USD

Daily corn-market analysis. CBOT front-month price paired with CFTC managed-money positioning, framed as falsifiable setups with explicit invalidation.

9 pages . last reviewed 2026-06-27

in development

Coming soon

More sites are on the way

We are expanding the portfolio to other markets and topics, each as a genuine standalone publication. See the portfolio for current status.