Education
Plain-English guides to the frameworks behind the analysis.
Every market analysis note in our research thread relies on a framework: OIS, term premium, the dollar smile, COT positioning, the Beveridge curve, the SEP. Here is the reference shelf. Each piece teaches the math, names the gotchas, and points back at the analysis where the framework is being applied today.
- Read →
A plain-English guide to event windows: how to tell what a catalyst actually moved.
Written after the Wednesday October 7 close, using that session as the live example. Our plan scored hits on the dollar and gold after hawkish FOMC minutes, but from 2:00 PM to the close DXY moved 0.04 and gold about $0.30; the moves came earlier. Close-to-close tells you what happened, not why. An event window measures a fixed slice around a release, the method academic studies of Fed announcements use. How to choose the window, what Wednesday looks like measured three ways, what windows cannot see (slow reactions, priced-in news, leaks), and how we will report them from now on.
- Read →
A plain-English guide to the trade balance and the dollar.
Written Tuesday October 6, using the August trade release ($105.6 billion deficit, Census/BEA) as the live example. The textbook says a wider deficit weakens the dollar; on release day it rarely does. Four reasons: most of the goods number is public a week earlier in the advance report; trade is a small slice of $9.6 trillion a day in FX turnover (BIS); every deficit is financed, so the price of financing matters more than the size; and causality runs both ways, since strong demand and a strong dollar both widen the gap. Where the balance does matter (GDP, the slow build of foreign-held liabilities, policy), and a conditional test for telling a harmless deficit from a worrying one.
- Read →
A plain-English guide to why a weak jobs number can fail to rally bonds.
Written Sunday October 4, using Friday October 2 as the live example. The textbook chain runs from a weak payrolls print to lower growth, lower inflation, an easier Fed and lower yields. Each link can break: the miss may already be priced, the details (revisions, why unemployment moved, wages) may disagree with the headline, the long end carries term premium that payrolls barely speak to, and an inflation-focused Fed limits how far a growth miss can travel. How to tell those explanations apart with the 2-year versus 10-year split, same-week price data and the pre-release tape, and why the grading window has to be fixed before the number comes out.
- Read →
A plain-English guide to a cool print in a hawkish regime.
A hawkish regime is not a Fed that hikes; it is a Fed with a reaction function that responds asymmetrically to the balance of risks. Under the current Warsh-led framework the September SEP already builds in a specific inflation trajectory, so a print that is cool relative to consensus but in-line with the SEP path is neutral, not dovish. Three interpretations of a cool print (cool vs consensus but in-line with SEP; cool on a single sub-component; cool across every line back-to-back with a labor-side print) and the specific real-yield decomposition (nominal flat, real up, inflation compensation down) that explains why Wednesday delivered a firmer dollar on a decisively dovish Core PCE. Historical parallels: 1994-95 mid-cycle pause, 2005-06 measured-pace terminal, 2018-19 Powell pause. When to update priors: the second-order pattern across ISM Manufacturing, Nonfarm Payrolls, and October CPI, not the single print.
- Read →
A plain-English guide to the Conference Board labor-market differential.
Inside the Conference Board Consumer Confidence composite sits a specific labor-market sub-index: percent answering "jobs plentiful" minus percent answering "jobs hard to get." The differential is a household-side perception of labor-market slack, distinct from JOLTS (employer-side) and from the unemployment rate. It leads U-3 by roughly three to four months at business-cycle turning points, materially faster than initial claims. Framework rule under the current Warsh-led regime: above plus 20 is hawkish; plus 10 to plus 20 is neutral; zero to plus 10 is dovish at the margin; below zero is recessionary. September's Conference Board release delivered a fourth consecutive monthly narrowing of the differential; the specific threshold that transmits to the FOMC's reaction function has not yet triggered. Three common errors to avoid, including reading the differential as a substitute for claims or JOLTS.
- Read →
A plain-English guide to the quarter-end positioning window.
Quarter-end delivers 40 to 80 percent above typical FX turnover. Four flow types dominate: pension rebalancing (bond-buying when equities outperform), currency-hedge rebalancing (systematic JPY or EUR buying when USD assets appreciate), the WM/R London 4pm fix (index-tracking flow), and reporting-date positioning-cleanout (traders lock in gains on outperformers). Three specific days carry specific patterns: two days out (cleanout peaks), one day out (mechanical rebalancing peaks), quarter-end itself (hedges + fix peak). The framework test for distinguishing flow from macro: if a move retraces 70 percent or more inside four trading days, it was flow. Applied to Monday’s gold crack: watch the recovery pattern into Friday.
- Read →
A plain-English guide to the sentiment-vs-spending decoupling.
Historically consumer sentiment and consumer spending moved together. Post-2022 that relationship broke: sentiment can swing between "recession-adjacent" and modestly firm while retail sales holds flat-to-firm. Three explanations that fit the data: balance-sheet resilience from 2020-2023 fiscal transfers, sentiment-as-partisan-signal post-2016, and inflation-frame vs unemployment-frame differences. Michigan versus Conference Board: Michigan more inflation-sensitive, Conference Board more labour-market-sensitive. Four data points that carry the macro read (expectations sub-index, one-year vs 5-to-10-year inflation, partisan split, high vs low income). Under a Warsh-led hawkish regime, a Michigan expectations drop above 10 percent is a specific signal to watch for cross-asset confirmation.
- Read →
A plain-English guide to weekly jobless claims.
The DOL weekly claims release is administrative rather than survey-based: actual filings from state UI systems, not sampled. Initial claims measure the flow into unemployment (leading indicator); continuing claims measure the stock of insured unemployed (coincident indicator). The 4-week moving average is 30 percent less volatile than the weekly print and is the specific summary measure the FOMC staff uses. Framework rule for the current Warsh-led regime: 4-week average below 210k is hawkish, 210 to 240 is neutral, above 240 is dovish. Thursday’s 202,250 4-week reading sits deep in the hawkish bucket. Three common errors: reading the weekly level as trend, ignoring the reference week, confusing initial with continuing.
- Read →
A plain-English guide to flash PMIs vs ISM.
PMIs are diffusion indices: they measure the direction of change, not the level of activity. A 58 does not mean the economy is 8 percent above trend; it means roughly 58 percent of respondents report better conditions than last month. S&P Global flash lands two to three business days before ISM. Three differences: panel composition (S&P broader, ISM larger firms), sub-index construction (ISM more employment-sensitive, S&P more orders-sensitive), release timing (flash first-mover). Three sub-indices carry the macro read: employment (hiring pace), new orders (leads output), input prices (pipeline pressure). Under a Warsh-led hawkish regime a composite above 56 is hawkish, 52-56 neutral, below 52 dovish. Wednesday’s 58.4 sits deep in the hawkish bucket.
- Read →
A plain-English guide to Existing Home Sales as a macro signal.
NAR Existing Home Sales measures closed transactions on single-family homes, condos, and co-ops. Broad coverage (roughly 90 percent of the resale market), lagged (measures closings not contracts). Quantity leads price because homes are illiquid and price-sticky: sellers pull listings before cutting; buyers walk before overpaying. Three data points carry the macro read: the SAAR level (sub-4M is five-decade low ex-COVID), months-supply of inventory (leading signal, watch trajectory), regional bifurcation (Sunbelt vs Rust-Belt tells the labour-market transmission story). Existing Home Sales does not touch Owner’s Equivalent Rent, so a soft print does not undermine the Fed’s inflation-first reaction function. Where the signal matters is in the medium-term growth dashboard.
- Read →
A plain-English guide to the week after an FOMC hike.
Week one after a Fed decision is the direct-reaction window (option-book unwinds, initial cross-asset repricing). Week two is the acceptance-and-testing window: the market has priced the surprise and is waiting for the next data anchor to either confirm or fade the trajectory. Four variables to watch (dollar extension vs consolidation, 2-year path signal, term-premium proxy, USD/JPY carry unwind). Three risks (fresh data undermines projection, oil supply shock reversal, Chair language walk-back). Historical parallels: 1994 saw 3.5 percent DXY extension over 30 sessions; 2015 saw 2.1 percent. The current 2026 restart-hike sits inside that range.
- Read →
A plain-English guide to quad witching with a macro catalyst.
Quad witching (third Friday of March, June, September, December) delivers 25 to 45 percent elevated volume, elevated intraday range, and reduced end-of-day directional conviction. Three pin-risk mechanics: gamma hedging unwind, pin risk on specific strikes, dividend arbitrage flow. When a macro catalyst lands on the same session (BoJ decision on September 18): FX and rates carry cleaner signal than equities (which are subject to pin mechanics); Monday morning re-pricing is where the real read lives; vol collapses across the following Mon-Wed as the market reprices to a normal regime. The three signals to log Friday (FX close clean, rates close clean, equity close treated as approximate).
- Read →
A plain-English guide to the Fed-then-BoJ 48-hour window.
A Fed decision Wednesday followed by a BoJ decision Friday compresses the pricing window: the market has to close positioning from meeting one and re-open into meeting two without a full intervening session of price discovery. USD/JPY is the only G10 pair where both central banks matter equally. The four outcome combinations decompose the setup: (1) BoJ hike + hawkish + unanimous (double-compression, USD/JPY reverses violently, ~15 percent probability given Sep 17 setup), (2) BoJ hike + measured + split (hike-but-yen-falls, USD/JPY 157-158.50, ~45 percent), (3) BoJ hold + hawkish hint + split hint (~25 percent), (4) BoJ hold + dovish + unanimous (yen crushed, ~15 percent). Ueda communications playbook.
- Read →
A plain-English guide to restarting a hiking cycle.
A hike inside an existing cycle is a repricing of the trajectory. A hike that restarts a cycle after a long hold is a repricing of the entire policy regime. Four historical parallels: 1994 (Greenspan Fed hiked 25bp after 21-month hold, 300bp of hikes over following 12 months, DXY +6 percent, gold -15 percent), 1999-2000 (12-month hold, 150bp over next year), 2015-2018 (7-year hold, initially slow pace, DXY -9 percent in 2016 as pace disappointed), 2026 (Warsh unanimous 25bp after 26-month hold, SEP median jumped 30bp to 4.10 percent, no cuts through 2027). Three signals the market reprices in first two weeks: front-end, term-premium proxy, dollar in week two.
- Read →
A plain-English guide to FOMC-eve gamma compression.
On the day before an FOMC decision, dealers hedging option books run short gamma positions across DXY and rates. Short-gamma hedging pushes dealers to sell into strength and buy into weakness, which dampens the tape response to unrelated catalysts by 30 to 50 percent. Three signatures: catalyst reactions dampened, implied vol premium widening into the close, two-way ranges collapse. Wednesday reaction is amplified by 20 to 40 percent relative to a no-compression baseline because dealers have to reverse the hedging. The Warsh-first-FOMC wrinkle: press conference tenor vol premium wider than statement tenor, because the market has no calibration for his cadence.
- Read →
A plain-English guide to a pre-FOMC weekend.
The FOMC blackout window is a self-imposed rule: from the second Saturday before the meeting through the Thursday after, no Reserve Bank president, no governor, and no committee member speaks publicly on policy. Three specific signatures on a pre-FOMC Monday: volume compresses 20 to 35 percent below trailing average, ranges tighten 25 to 40 percent, positioning skews get set. Four signals the Monday tape gives about the Wednesday move: direction lean, range dispersion, rates-vs-dollar coherence, vol-of-vol read. The specific setup on Sep 14, 2026 (committed and quiet, 28 percent vol premium) means Wednesday reaction will be proportional to SEP-median surprise, not violent.
- Read →
A plain-English guide to a CPI-plus-Michigan Friday.
When CPI at 8:30 AM ET and Michigan preliminary at 10:00 AM ET land the same day, the market gets two orthogonal signals inside a ninety-minute window: hard-data inflation (CPI) plus soft-data consumer perceptions of realised and expected inflation (Michigan). The Fed weights the two together, not independently, because persistence is a function of both. The 2x2 matrix of hot-or-cool-CPI by anchored-or-unmoored-expectations gives four setups with distinct tape signatures. September 11 delivered the "hot CPI + unmoored expectations" case (~15 percent of hot-CPI Fridays). Reading the ninety-minute window: full extension on CPI vs half-extension-plus-big-second-push vs reversal on Michigan.
- Read →
A plain-English guide to final-demand PPI-to-CPI transmission.
Wholesale prices transmit to consumer prices through three specific channels each with its own lag and pass-through elasticity. Channel 1: direct pass-through in goods (30-60 day lag, 0.60-0.85 elasticity). Channel 2: intermediate-goods embedded cost (90-180 day lag, 0.20-0.35 elasticity). Channel 3: services pass-through via input costs (portfolio management fees direct to PCE, healthcare services 60-90 day lag). Reading a diesel-driven print: Jazan supply shock only partially in the August CPI print; larger effect lands in September CPI (October release). Core-goods-vs-services split more informative than headline. Four questions to ask on a hot-goods PPI day.
- Read →
A plain-English guide to a pre-PPI tape.
Producer Price Index for Final Demand: measures average change in selling prices received by domestic producers. Three headline aggregates (goods, services, construction); the desk watches core "final demand less foods, energy, and trade services". PPI matters for two reasons: (1) some components feed core PCE more directly than CPI does (portfolio management fees, healthcare services); (2) PPI is a lead indicator on cost pressure that has not yet flowed to the consumer. A pre-PPI Wednesday tape has three consistent features: position-clearing on front-end, DXY consolidation, gold and silver ranges compressing. Elasticity of DXY move to surprise: linear inside +/- 0.3 percentage point band, quadratic outside it. Three signals to read on the Wednesday close.
- Read →
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 →
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 →
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 →
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 →
A plain-English guide to reading ISM plus JOLTS as a single signal.
First-Tuesday-of-the-month 10:00 AM ET compressed release: ISM Manufacturing and JOLTS land in the same one-minute window and the market treats them as one signal. Three reasons the fusion is the reading convention (shared window, complementary labour coverage, both are Fed reaction-function inputs). Four archetypal combined signatures (both firm aligned-hawkish, both soft aligned-dovish, ISM firm plus JOLTS soft mixed, ISM soft plus JOLTS firm stagflationary) each with characteristic tape response magnitudes 1.5x to 2x sum. Five diagnostic dimensions to classify the print. Warsh-era reading convention: aligned-firm reinforces the September FOMC hike case, aligned-soft pushes back, mixed cases hold the setup. Three common misreads: elevating Construction Spending to a third signal, reading Prices Paid as a demand signal, ignoring JOLTS revisions.
- Read →
A plain-English guide to month-end FX rebalancing flows.
Month-end rebalancing is a stack of four sub-flows (equity-side, bond-side, sovereign-fund and central-bank reserve, corporate-treasury dividend and coupon) that converge on the WMR 4:00 PM London fix. Direction depends on relative asset-class performance during the month: US-equity outperformance produces dollar-selling; underperformance produces dollar-buying; large DXY moves during the month can override. Three interaction patterns with macro flow (same-direction reinforcement, opposite-direction cancellation, fix-driven overshoot and retrace); today Aug 31 delivered the opposite-direction cancellation shape with post-Warsh flip-bid against equity-outperformance rebalance-offer. Four measurable tape signatures for identifying the rebalance footprint in the 3-4 PM London window.
- Read →
A plain-English guide to a Chair keynote flipping the distribution.
A single Fed Chair keynote can flip the probability distribution for the next FOMC by 100+ percentage points across modal outcomes. Happens once every 12-18 months globally. Distinguish from a magnitude-surprise (modal outcome held, weight shifted). Five tape signatures on a flip session (cross-asset amplification well above eve-gamma projection, front-end leads belly and long-end 2-3x, cross-asset consistency across all pairs, positioning-driven overshoots with 20-40 percent retrace inside 90 minutes, implied vol collapse alongside realised extension). Four-week second-order pattern: Week 1 crowd unwind, Week 2 data-flow test, Week 3 distributional re-calibration, Week 4 next-catalyst focus. Five re-calibration questions for updating the framework probability weights. Base rates: flip extends 55 percent, flip fades 30 percent, flip fails and inverts 15 percent.
- Read →
A plain-English guide to keynote-eve gamma compression.
The day before a Fed Chair keynote (or ECB Sintra opening, IMF Governor speech, BoJ policy address) delivers a specific tape shape: FX and rates volume 20-40 percent below trailing, implied vol rising while realised falling, intraday moves that pop on news and fade inside 30 to 90 minutes, bid-ask spreads widening in the last two hours, range compression 30-50 percent versus trailing, cross-pair correlation tightening toward 1.0. Three mechanisms combine (paired-risk positioning, discretionary-flow thinness, dealer gamma absorption). Three keynote-day resolution shapes with base rates 55 / 25 / 20 percent (amplification, fade, two-way spike). Three exception cases that break the framework (concurrent same-day catalyst, Chair pre-announces thesis, keynote cancellation).
- Read →
A plain-English guide to compressed macro release days.
A compressed release day is any session where two or more scheduled prints hit inside the same 30-minute window. Compression happens when the default calendar collides with a fixed-holiday shift, a Fed communication window, or a legally mandated deadline. Wednesday August 26, 2026 delivered the archetypal case: BEA moved the Q2 GDP second and July Core PCE together to clear Jackson Hole. Two prints in one window amplifies the signal by 1.5 to 2 times through positioning gamma compression, not just release timing. Four archetypal cases (aligned, dominant-print, ambiguity, in-line) each have a characteristic tape response; five diagnostic dimensions tell them apart. Today scored Case 4 (in-line) with a modest tilt to Case 2 through the headline PCE overshoot.
- Read →
A plain-English guide to supply-shock vs growth-shock bond rallies.
A 6bp decline in the 10-year looks the same on the tape regardless of driver, but supply-shock rallies fade in one to two weeks while growth-shock rallies extend for four to eight. Five diagnostic dimensions (break-even decomposition, coincident commodity move, cross-asset direction, dollar behavior, OIS repricing magnitude) tell the two archetypes apart. The Fed reads them differently: supply-shock deflation is looked through, growth-shock disinflation produces policy response. Risk assets respond in opposite directions: supply-shock is risk-positive, growth-shock is risk-negative. Today's Tuesday August 25 tape scores four-of-five supply-shock signatures on the Brent-driven yield rally.
- Read →
A plain-English guide to a pre-symposium positioning week.
The three-to-five sessions before a Fed Chair keynote at Jackson Hole (or ECB Sintra, IMF meetings) follow a recognisable shape because the same accounts position around the same fixed event. Four phases: Monday-Tuesday squaring, Wednesday macro-buffer, Thursday gamma-compression, Friday keynote-day. Measurable footprint: FX and rates volume runs 20-30 percent below trailing average through Wednesday, then jumps 50-100 percent above on keynote-day; option-implied vol rises even as realised falls; positioning gamma compresses. Compressed release calendars (like Wednesday August 26, 2026: BEA moved GDP second + Core PCE together) amplify surprise response 1.5-2x. Historical precedents: Powell 2020 (AIT continuation), Powell 2022 ("pain" reversal), Powell 2018 (neutral fade).
- Read →
A plain-English guide to reading a Jackson Hole speech.
Jackson Hole (last weekend of August) is the most-anticipated Fed communication outside FOMC meetings. Three reasons speeches carry more weight than typical Chair appearances (timing 2-3 weeks before September FOMC, academic-conference format encourages new frameworks, global central-banker audience). Six speech dimensions to watch (thesis/framework, forward guidance language, inflation-employment weighting, financial-conditions references, dispersion acknowledgments, data-dependency). Three tape-response phases and magnitude bands. Historical examples: Powell 2020 AIT, Powell 2022 "pain," Volcker 1979.
- Read →
A plain-English guide to regional Fed surveys vs national ISM.
Six regional Fed surveys (Empire State, Philly Fed, Richmond, Dallas, Kansas City, Chicago). Historical correlation to subsequent ISM: 0.55-0.65 for the composite, 0.4-0.5 for Philly Fed alone. Three reasons regional surveys can mislead (sample bias, geographic concentration, methodology differences). How markets read a big regional surprise (initial response then follow-through reduction, subsequent regional confirmation, ISM decisive). Aug 20 Philly Fed 47.4 vs 25.0 test framework.
- Read →
A plain-English guide to gold breaking multi-year highs.
Multi-year-high breaks in gold have three-phase dynamics. Phase 1 (Days 0-1) initial break with technical amplification. Phase 2 (Days 2-7) consolidation and back-test; ~65% first-attempt breaks hold. Phase 3 (Days 8-30) extension to next major level or reversal; ~55% reach next level within 3 months. Four supporting-factor check (Fed reaction-function shift, central-bank diversification, geopolitical premium, inflation stickiness); three of four active for current setup. Framework for August 19 break above $4,500.
- Read →
A plain-English guide to reading retail earnings as macro signal.
Five companies matter most: Walmart, Home Depot, Target, Lowe's, TJX. Three signals to extract: comparable sales trend (above +3% strong, below -1% weak), management guidance (raised/maintained/lowered/withdrawn), category-specific commentary (essentials vs discretionary vs trade-down). Aggregating the week produces highest-signal aggregate: aligned soft = decisive consumer-weakness signal. Home Depot Tuesday -0.6% sets tone; Target Wednesday and Walmart Thursday will confirm.
- Read →
A plain-English guide to pre-FOMC-minutes positioning patterns.
Three patterns before minutes release: Pattern I Confirmation positioning (extends in direction of press-conference reading), Pattern II Uncertainty-hedging (reduces positioning, elevates volatility premium), Pattern III Contrarian positioning (moves opposite direction from tape). Response asymmetry: Pattern I creates hawkish-surprise vulnerability. Current setup Pattern I with modest II element; hawkish-minutes surprise would produce disproportionate reversal. Three extraction points: dissent language strength, balance-of-risks language, data-threshold language.
- Read →
A plain-English guide to consumer confidence inflection signals.
Single-month Michigan Sentiment drops of 5+ points are inflection signals. Historical follow-through distribution: 45% retail contraction, 35% Fed pivot rescue, 20% recession within 12 months. Three inputs identify inflection vs noise: breadth of decline (both sub-indexes falling), demographic breadth (broad-based), consistency with Conference Board. Friday drop 55.2→51.0 (4.2 points at upper end of moderate signal, approaching inflection). Two of three inflection signals clean; overall reading: inflection with high confidence.
- Read →
A plain-English guide to profit-taking after a multi-session rally.
Distinguishing profit-taking (normal consolidation, trend eventually resumes) from reversal (change in underlying driver, full retracement) after a multi-session rally. Three-signal check: volume (below-average = profit-taking, above-average = reversal), cross-asset alignment (isolated = profit-taking, aligned = reversal), sub-session pattern (window-specific = profit-taking, sustained = reversal). Two-day rule: whether the next session extends or reverses is the decisive input. Base rates for each pattern.
- Read →
A plain-English guide to reading in-line data in a primed market.
In-line data prints are not automatically non-events. In a primed market (positioning already moved decisively in one direction), in-line prints can produce material continuation moves via secondary interpretation channels. Four response patterns: consolidation with continuation (most common), extension via secondary interpretation, partial retracement of primed direction, delayed hawkish reversal. Wednesday CPI landed into Pattern C/A primed-dovish; response fit Consolidation with Continuation.
- Read →
A plain-English guide to pre-CPI positioning patterns.
CPI releases produce among the largest scheduled market moves. Four pre-CPI positioning patterns: Aligned Conviction (aggressive positioning), Balanced Uncertainty (modest range-bound), One-Sided After Shock (post-NFP dovish extension), Hedging Surge (options-market visible). Reading the pattern predicts reaction magnitude asymmetry: primed markets react more to contrary prints than to confirming prints. Current setup Pattern B/C mixed.
- Read →
A plain-English guide to MOF verbal intervention.
MOF verbal spectrum: routine commentary (0 signal), directional (5 pips), concern language (20-40), threat language (50-100), denial language (100-200; precursor to physical). Verbal operates in both directions: physical July 30 defended yen weakness (163+), verbal August 10 defended yen strength (157-). Identifies source rank, media placement, timing. Follow-through: verbal-only decays over 2-3 weeks absent physical or fundamental follow-up. Implicit MOF trading band of 158-163 now operational.
- Read →
A plain-English guide to reading a negative NFP print.
Three types: Type 1 (crisis-driven, magnitude 100k+ negative), Type 2 (inflection-point, modest with revisions), Type 3 (statistical noise). Six-signal check separates them (prior revisions, wage growth, unemployment move, participation, ADP alignment, JOLTS alignment). August 7 print reads Type 2 with 5-of-6 signal confirmation. Type 2 implications: follow-through expected, Fed response likely, recession risk 40% within 12 months. Base rates for tape response: gold +2-5%, DXY -100-250 pips, 10Y -10-25bp over 5-10 sessions.
- Read →
A plain-English guide to oil supply shocks by type.
Four types of oil supply shock: Type A single-vessel tanker incident (3-7% Brent), Type B multi-vessel/route-interference (8-15%), Type C state-actor blockade (20-40%), Type D permanent infrastructure damage (proportional to bpd loss). Four inputs identify type: vessels/facilities count, actor state vs non-state, duration expectation, response chain. Thursday incident was Type A; 50-70% typical retracement within 5-10 sessions.
- Read →
A plain-English guide to outsized single-session gold moves.
Gold sometimes moves 3-10x what the catalyst would predict. Three amplification channels: real-yield decoupling ($3-4 per bp of real-yield change), positioning capitulation (short-covering cascades), central-bank momentum flow (CB diversification demand rising with the trend). Wednesday +$205 decomposition: ~25-30% real-yield, ~20-40% positioning, ~20-30% CB momentum. Follow-through: 3-phase pattern over 5-10 sessions; 40% extend, 35% consolidate, 25% retrace. Friday NFP is decisive input.
- Read →
A plain-English guide to reading NFP week: ADP, Claims, NFP.
NFP week begins Wednesday with ADP employment, continues Thursday with Weekly Initial Claims, closes Friday with NFP. This piece is the framework for reading the sequence: what ADP actually measures (private-sector, 25M workers), where it diverges from NFP (government, small business, seasonal adj), what Claims signals (weekly high-frequency labor read), what NFP's four numbers do (headline, AHE, unemployment, revisions), and how the three prints interact through the week.
- Read →
A plain-English guide to ISM Manufacturing vs Services.
ISM Manufacturing (first business day of month) and ISM Services (third business day) both diffusion indexes; both use 50 as expansion/contraction boundary. Manufacturing is leading indicator; Services is coincident with the cycle. Sub-indexes (New Orders, Employment, Prices Paid, Supplier Deliveries) carry more signal than the headline. When the two diverge (one above 50, one below), the divergence typically persists 2-4 months before re-converging. Framework for the current Manufacturing 49.2 + Services expected 52.0 divergence.
- Read →
A plain-English guide to six-month annualized inflation.
Inflation gets reported YoY; the Fed reads it as six-month annualized. The two can differ materially when inflation is turning. YoY smooths but lags 6-9 months at turning points; six-month annualized captures the current trajectory. July 2026 core PCE at 3.3% YoY but ~2.85% six-month annualized: dovish trajectory below the Fed's 2% target range within striking distance. Formula, calculation, three-month alternative, base-effect and composition-shift caveats. Framework for reading a Fed reaction function focused on trajectory, not headlines.
- Read →
A plain-English guide to reading a suspected FX intervention in real time.
Official confirmation of intervention comes days, weeks, or months after the event. Real-time reading uses five forensic signals: speed and shape (vertical drop plus partial recovery), absence of proximate catalyst, NY Fed rate check reported, cross-asset correlation decoupling (USDJPY down while 10Y up), historical timing pattern (after prolonged weakness, coincident with macro catalyst cover). All five signals passing puts intervention probability above 90%. Thursday's tape passed all five. What intervention buys, what it doesn't, three-phase follow-through pattern.
- Read →
A plain-English guide to reading a contradictory Fed message.
Fed messages become contradictory for four reasons: committee dispersion smoothed over in prepared statement, Chair inexperience with format, genuine policy uncertainty, deliberate optionality. When statement and Q&A conflict, Q&A wins the market's interpretation (unscripted, specific to hypotheticals, hedging patterns readable). Five Q&A dimensions carry highest signal: verb tense, bar-setting language, time-frame specificity, dispersion acknowledgments, referent selection. Warsh's Wednesday: four of five read dovish, explaining the market's dovish interpretation despite hawkish written text. Term-premium credibility discount showed in 30-year yield hitting cycle high.
- Read →
A plain-English guide to JOLTS vs NFP: different labor-market signals.
Both series called labor-market data. NFP is flow (net payroll change) with 2-3 week lag, released first Friday 8:30 AM ET, moves markets 40-100 pips on first minute. JOLTS is stock (unfilled job openings) with 4-6 week lag, released monthly 10:00 AM ET, moves markets 10-25 pips. NFP timely and comprehensive; JOLTS later but structural. When they align, aggregate labor signal is strong. When they conflict, muted reaction because Fed itself needs more data. Tuesday's JOLTS softening softened FedWatch hike probability into Wednesday.
- Read →
A plain-English guide to reading a geopolitical risk-premium unwind.
Geopolitical risk premiums build in discrete steps and unwind in a compressed single-session repricing. The unwind is typically 2-3x more compressed than the build because positioning is all offside simultaneously. This piece is the framework: three-phase structure (initial repricing Day 0, consolidation Days 1-5, resolution Weeks 2-4), three second-order effects (breakeven repricing, central-bank reaction function shifts, cross-asset positioning unwinds), why oil moves outsized in the Day 0 window, what to watch in Phase 2.
- Read →
A plain-English guide to reading FOMC-day tape.
The FOMC statement lands 2:00 PM ET Wednesday; press conference 2:30 PM ET; SEP (quarterly meetings only) at 2:00 PM ET alongside the statement. This piece is the framework for reading FOMC-day tape end to end. What actually lands at 2:00 PM ET (three documents); reading the statement (three signal-carrying change categories); reading the dots (median dot, dispersion, longer-run, growth-unemployment consistency); reading the press conference (opening statement variance, Q&A hedging patterns, committee-dispersion acknowledgment); three-phase tape-timing patterns; the reversal risk over Day 1-5 as post-meeting speeches reveal dispersion. Framework for the July 29-30 meeting.
- Read →
A plain-English guide to gold in a rising real-yield environment.
Gold has one primary rate-sensitive driver: real yields. This piece is the framework for reading gold when real yields decisively rise. Nominal-vs-real decomposition (four sources of nominal yield rise, only some weigh on gold); the Tuesday-through-Thursday tape as a live case study (breakevens rose then stalled, real yields rose only on Day 3); positioning amplification (Thursday's -$86 was 7-9x the mechanical prediction); four-day follow-through pattern; the three complications (central-bank demand, safe-haven flow, positioning extremes).
- Read →
A plain-English guide to reading Treasury auction demand.
Every US Treasury auction has a canonical set of four numbers: high yield vs when-issued, bid-to-cover ratio, indirect bidder share, direct bidder share. Read together they tell you whether the auction cleared with strong demand, weak demand, or somewhere in between. This piece sets out what each number means, how to combine them (strong auction, weak auction, dealer-backstop mixed, foreign-vs-domestic rotation), and the specific FX and rate implications of each configuration.
- Read →
A plain-English guide to pricing a Fed rate-hike tail risk.
CME FedWatch hike probability moved 4→16.6% Tuesday. This piece sets out what the number actually measures (fed-funds-futures implied path, not market opinion), the three channels through which it flows into cross-asset pricing (rates, FX, gold), where the pricing usually comes from (hot data, supply-side inflation shock, Fed speeches, foreign central bank action), and the three possible FOMC-day resolutions (hike happens, dovish hold, hawkish hold). Framework for reading pre-FOMC positioning.
- Read →
A plain-English guide to the Fed communications blackout window.
For the ten days before every FOMC meeting, senior Federal Reserve officials stop speaking publicly about monetary policy. This piece sets out what the blackout is (a self-imposed policy since 2005), why it exists (reduce information asymmetry, focus attention on the statement, reduce internal signaling risk), how markets behave inside it (rates thin, FX headline-driven, data prints outsized), and what typically happens on reopen (post-meeting speeches surface committee dispersion). Framework for reading the current July 18-31 window.
- Read →
A plain-English guide to shipping chokepoints and oil supply risk.
Roughly 60 percent of world oil moves by sea, funneled through five narrow passages. Hormuz (20 mbpd, no full commercial bypass), Bab-el-Mandeb / Red Sea (6-7 mbpd oil plus most Asia-Europe container flow), Suez, Malacca (25 mbpd), Turkish Straits. This piece sets out what each carries, the historical premium a threat produces, the market signals (Brent-WTI spread, options skew, tanker rates), and why the Iran-retaliation story reaches oil specifically through Hormuz and the Red Sea rather than through Iran’s own barrels.
- Read →
A plain-English guide to reading retail sales against a disinflation print.
Retail sales measures consumer demand. When it comes in firm the day after a soft CPI/PPI combination, markets have to reconcile: is inflation falling because demand is cooling, or is it falling despite demand holding up? The two interpretations point to different Fed paths. This piece sets out the four components that matter, the control-group aggregate, and how to read a firm-retail-sales-into-soft-inflation combination.
- Read →
A plain-English guide to reading PPI against CPI.
PPI arrives a day after CPI in most months and delivers the producer-side check on the consumer-price read. When PPI confirms the CPI direction, the disinflation or reinflation signal earns weight. When it contradicts, the market has to figure out which side of the supply chain is doing the moving. This piece sets out the mechanics, the components that matter, and how to read a same-week PPI-CPI combination.
- Read →
A plain-English guide to reading a soft CPI against a hawkish Fed.
A dovish inflation surprise usually gets dovish central-bank language a few weeks later. When the chair delivers hawkish language on the same day as a soft print, markets have to price the conflict. This piece sets out the three specific patterns that emerge, how each resolves historically, and what to watch for in the days following. Warsh’s Tuesday testimony against the soft June CPI is the current example.
- Read →
A plain-English guide to how oil shocks transmit to inflation and rates.
Brent surged 9% Monday on renewed US-Iran conflict over the Strait of Hormuz. Oil shocks translate to headline inflation within weeks, core inflation within months, and rate expectations within days. This piece sets out the four transmission channels, the specific lags each carries, and how markets typically price the transmission before the data confirms it.
- Read →
A plain-English guide to reading Fed Chair congressional testimony.
Tuesday brings Chair Warsh’s first congressional testimony, before the House Financial Services Committee. Testimonies are the most-watched Fed communication event outside the FOMC decision itself. This piece sets out the four sections that carry policy signal, the specific question types to watch for, and how testimony language compares to statement and minutes language across the last three chairs.
- Read →
A plain-English guide to reading a split Fed committee.
The June 2026 FOMC came out 9-8-1 on the projected direction of rates by year-end. That is the most divided committee since the taper-tantrum era. This piece sets out how to read a split committee: which votes matter, how splits resolve, the historical base rates on convergence direction, and why the median dot can be a poor summary when the underlying distribution is bimodal.
- Read →
A plain-English guide to reading FOMC minutes.
The Federal Reserve publishes minutes of its policy meetings three weeks after each decision. The minutes are the most detailed record we get of how the committee thinks — but they require careful reading. This piece sets out the six sections that matter, the language patterns that reveal dissent, and the specific words that shift market pricing when they appear or vanish.
- Read →
A plain-English guide to retrace patterns: cascade, intervention, and regime shift.
Large single-session FX moves retrace in three characteristic shapes. Cascades retrace fast and partial (30-50% within 2-3 sessions). Interventions retrace slow and shallow (20-40% over weeks). Regime shifts barely retrace at all. Reading which shape the retrace takes identifies which driver was in charge of the original move — a diagnosis the tape itself provides.
- Read →
A plain-English guide to positioning-liquidation events.
A move too big and too fast to be explained by the macro tape is a positioning event, not a fundamental one. This piece explains the mechanics: what a liquidation cascade looks like from the option-flow and stop-cluster side, how to identify one in real time, and why the direction of the eventual retrace is not the same as the direction of the initial move.
- Read →
A plain-English guide to the Bank of Japan policy toolkit.
The BoJ policy rate is the headline; it is not the tool. This piece sets out the five instruments the desk actually deploys: the QQE framework, YCC, the Rinban schedule, the JGB purchase cap, and unsterilised FX intervention conducted for the MoF. The seven pressure points in that stack, and what each signals when the Bank moves them.
- Read →
A plain-English guide to yield differentials and FX carry.
The two-year cash-rate gap is the single cleanest driver of every G10 cross. This piece explains why the swap-implied differential (not the spot-rate gap) is what matters, how carry decomposes into hedged and unhedged components, and why the 2026 yen carry has been the trade of the year.
- Read →
A plain-English guide to gold as a macro signal.
What gold actually prices: the real-yield identity, the dollar inverse, the central-bank bid, and the risk-off impulse. When these drivers agree the gold print is informative; when they disagree the disagreement is the signal. The framework that turns a single price into a regime read.
- Read →
A plain-English guide to the Taylor rule.
The published policy rule that benchmarks the federal funds rate against inflation and the output gap, the four parameter choices that matter, what the rule recommends today against the 3.4% core PCE print, and why Warsh has hinted the next framework may make explicit reference to it.
- Read →
A plain-English guide to nowcasting.
How the Cleveland Fed inflation nowcast and the Atlanta Fed GDPNow models work, what daily inputs feed them, why they tend to beat consensus forecasts by 30-40bp on release day, and how to read them against the upcoming print.
- Read →
A plain-English guide to the oil market.
Brent vs WTI, the curve structure (contango vs backwardation), what OPEC+ actually controls and what it doesn’t, the geopolitical-premium overlay, and why a $73 print after a $93 peak is the cleanest single read of the Iran de-escalation flow.
- Read →
A plain-English guide to central bank FX reserves.
What "war chest" means in practice: the $1.3 trillion Japan holds, how it is composed, what fraction is actually deployable for intervention, and the constraints that make the headline number misleading.
- Read →
A plain-English guide to real yields.
How a real yield is defined (nominal minus breakeven), the difference between TIPS-derived and survey-derived measures, why real yields drive FX more than nominal yields, and where the current US 10-year real yield at roughly 2.0% places the dollar regime.
- Read →
A plain-English guide to the carry trade.
How a carry trade is constructed, the math behind the yield differential and the FX-volatility cost, why USD/JPY is the canonical example, and the historical pattern of why carry trades end (rate convergence, vol spike, or intervention).
- Read →
A plain-English guide to monetary policy frameworks.
What a "framework" actually is, the 2020 FAIT regime and why it lasted only five years, the alternatives (Taylor rule, price-level targeting, NGDP targeting), and what Warsh’s "not well suited" comment yesterday tells us about where the framework review is headed.
- Read →
A plain-English guide to financial conditions indices.
How the Goldman, Chicago Fed, and Bloomberg FCIs are built, why the Fed treats them as the policy-transmission channel, and what a 50-basis-point move in the index actually does to real-economy outcomes.
- Read →
A plain-English guide to forward guidance.
How the Fed communicates intent through statement language, the SEP, and the press conference, the four eras of guidance (qualitative, calendar, threshold, state-contingent), and the language hooks markets actually trade.
- Read →
A plain-English guide to r-star, the neutral rate.
What r-star actually is, the three models (Laubach-Williams, Holston-Laubach-Williams, Lubik-Matthes) that estimate it, why the longer-run dot is essentially r-star plus 2 percent, and why Warsh’s drift on this number is the most durable signal in the SEP.
- Read →
A plain-English guide to inflation expectations surveys.
The four surveys central banks watch — Michigan, NY Fed SCE, the Conference Board, and the SPF — what each one asks, why the Fed weights them differently, and the "anchored vs unmoored" framing that today’s Michigan long-run jump just put on the table.
- Read →
A plain-English guide to CPI components and the supercore.
The four categories that make up the CPI basket, the OER trick that makes shelter dominate the index, why the "supercore" (services ex-housing) is the cleanest demand read, and the math behind the sub-component weights.
- Read →
A plain-English guide to inflation breakevens.
How TIPS-vs-nominal gives you a market-implied inflation expectation in one number, why the 5y5y forward is the cleanest read of long-run anchoring, and the construction math behind both, written for the CPI Wednesday and the SEP in two weeks.
- Read →
A plain-English guide to the SEP and the dot plot.
What the Summary of Economic Projections actually is, how the dot plot is assembled, the difference between the median dot and the consensus call, and why Warsh’s first SEP at the 16-17 June meeting is the most-watched piece of paper of the year.
- Read →
A plain-English guide to the NFP report.
Two surveys, one release: the establishment survey that counts jobs, the household survey that counts people, the birth-death model that adjusts for new firms, and the benchmark revisions that quietly rewrite the picture once a year.
- Read →
A plain-English guide to FX intervention.
When Japan’s MoF actually steps in, the difference between verbal and actual intervention, the 2022 and 2024 precedents, and what to watch in the price action around USD/JPY 160.
- Read →
A plain-English guide to ADP vs nonfarm payrolls.
Two payroll measures, three structural differences, and the rule of thumb that lets you translate an ADP surprise into an expected NFP reaction. The math behind why they diverge by 100k or more in a single month.
- Read →
A plain-English guide to the Beveridge curve.
The vacancy-unemployment relationship that sits behind every JOLTS release, the post-pandemic outward shift and what is reversing it, and the u* = √(uv) shorthand for reading the curve in one line.
- Read →
A plain-English guide to economic surprise indices.
How the Citigroup CESI is built, why FX desks watch it more than the prints themselves, and the math behind translating an upside ISM into the index move it produces.
- Read →
A plain-English guide to PCE vs CPI.
Why the Fed reads PCE instead of CPI, what the trimmed-mean variant adds (and why the new Chair prefers it), and how to translate a print in one measure into the other.
- Read →
A plain-English guide to the dollar smile.
Stephen Jen’s three-regime framework for the dollar, the math behind the smile shape, and where DXY actually sits today between left-side risk-off, middle-trough underperformance, and right-side US exceptionalism.
- Read →
A plain-English guide to the term premium.
Decomposing a 10-year Treasury yield into the expected path of policy and the term premium, the math from the ACM model, and why the bucket has gone from deeply negative to firmly positive in this cycle.
- Read →
A plain-English guide to COT positioning, with corn as the case.
How the CFTC Commitments of Traders report is built, the math that turns a raw net-long number into a percentile, and what May’s corn print is saying about a crowded long.
- Read →
A plain-English guide to currency correlations.
What the correlation between two markets measures, how to compute it in five lines of arithmetic, and the one cell of today’s dollar matrix that’s gone the wrong way.
- Read →
A plain-English guide to OIS.
What overnight-index swaps measure, what they do not, and how to read them against fed-funds futures.
Reference pieces are evergreen. They update when a framework meaningfully changes, not on a calendar. The Atom feed covers both analysis and reference releases.