One and a half years of development. Algorithms tested on several years of market data and events, and kept only where they showed a real statistical advantage. Every weekend they turn some 200 briefings into four reports and a forecast that tell you which stories move markets, and where each one stands.
The algorithms are built to take bias out of the reading. Every source is analysed on its own before anything is combined; every brief guards against confirmation, anchoring and availability bias, stays neutral between risk-on and risk-off and must argue the other side with a devil’s advocate.
The synthesis may use only the week’s collected sources and market data, never unsourced background. Figures are reconciled with the data of record, and where a story and the data disagree, the data wins and the report says so.
Knowing the story is not enough; knowing where it stands is. Every narrative is placed on its cycle, so a reader sees at a glance whether a theme is early, crowded or ending, and what would turn it.
Markets move on stories before they move on data, and every story runs through a hype cycle: discovered, spreading, dominant, cracking, one last burst of enthusiasm, the turn, the end. Story and price feed each other. Knowing the stage tells you whether a theme is early, crowded or over. Pick a stage.
A new story is forming, from a technology, a policy shift or a change in behaviour. Only specialists and contrarians talk about it, mostly in research and niche coverage. The first positions are small, prices barely move, and pullbacks shake out the few early believers.
What to watchThe gap between story and price is widest here. A good story can take a long time to be noticed, so patience matters as much as being early.
Each agency’s coverage is analysed over windows from the same day to seven days back. The leading stories are compared across Bloomberg, Reuters and Axios, merged where they describe the same theme and kept apart where the agencies disagree. The stage follows from the whole picture: how widely the story is carried, its sentiment and drivers, whether it strengthens or weakens from the oldest window to the newest, and how participants react, from disbelief to crowding.
A separate conviction review re-tests every placement against the week’s reporting and independent expert commentary. A story moves by at most one stage a week, so the cycle shows a trend, not a headline. The Wall Street Journal is read as part of the market media: its themes are validated against market data as an influence on the market’s narratives.
Markets trade expectations, and expectations are carried by stories. The same headline can mean opportunity early in a story and danger late in one, so the stage decides what a piece of news is worth. Cycles look clean only in hindsight; in real time they are noisy, which is why a disciplined reading, repeated the same way every week, matters more than any single call. Themes come and go like fashions, sometimes for fundamental reasons, sometimes simply because a newer story arrives.
Every weekend the Narrative Cycle report places the week’s leading stories on the cycle and shows how each has moved since the week before. The stages carry into the Market Outlook, so you go into the week knowing which themes are early, which are crowded and which are fading.
Weigh news by where its story stands. Good news late in a crowded story can mark a top; bad news early in a strong one is often bought.
The first price after major news becomes a reference point. When the market soon trades back through it, the news mattered less than the headline, or larger players used it to move the other way.
Central bank meetings, data releases and political events push stories from one stage to the next. Knowing the calendar and the scenarios beforehand means the reaction is decided before the news hits.
Stories travel. Growth expectations reach rates, currencies and commodities, often with a lead or a lag, so a turn in one market can announce the next.
Agency briefings, the Journal, two trader desks, targeted news monitoring and commentary from leading market practitioners, captured every day with every figure, name and quote kept verbatim.
Each outlet is read on its own across look-back windows of up to seven days: its causal logic, drivers, sentiment, bias and the narratives it ranks. Stale feeds are refused.
Rates, the agency narrative cycle, the Journal narrative and the trader desks, each under a fixed brief so issues compare week to week.
The Market Outlook is assembled, then a separate conviction review re-tests every narrative against the reporting and independent expert commentary and checks it against last week and market data.
Completeness checks, figures reconciled with the data of record, calculations computed rather than estimated. Then the author reads, edits and approves every report before the four reports and the forecast go out.
The week’s rate themes ranked and tested against the Treasury curve, slopes and tenors tabulated, three probability-weighted scenarios each challenged by a devil’s advocate.
The leading stories in Bloomberg, Reuters and Axios, merged across agencies, each placed on the seven-stage cycle with sentiment, implied price trend and confidence.
Every instrument two independent trader desks cover, with bias, levels, catalysts and trade ideas, compared day by day. Days without coverage are left out, never inferred.
The market media’s narratives validated against market data as influences on market themes, opportunities ranked by confidence with trigger and invalidation, and best, worst and likely cases for the week ahead.
A computed 30-day path for the 3-month, 2-year, 10-year and 30-year yields and the 2s10s spread, with its likely range.
Most readers go into the week with last week’s curve. Professional members go in with a 30-day path for the 3-month, 2-year, 10-year and 30-year Treasury yields and the 2s10s spread, and the range they are likely to stay in.
Fast Fourier Transform isolates the regular waves in each yield’s history, so you see where the next turn is likely to come.
A random walk accounts for the news nobody can foresee, so the path you get is realistic, not wishful.
The fan shows how far yields could stray, so you know at once what is ordinary and what is a real surprise.
What the curve has done after days like today, bear or bull, steepening or flattening, and what it means for rates and currencies.
The headline is free every week; the full forecast is for Professional members. See the plans
Produced with AI-assisted analysis: language models read, analyse and synthesise under fixed analytical briefs, only from the collected sources and market data, with automated checks between stages. The author reviews and signs off every issue before it is published, so it is not an AI-only product. Probabilities and conviction levels are analytical judgements of this process.
Market commentary for information only: not investment advice, not a recommendation to buy or sell any security, and not tailored to any reader. Scenario probabilities are analytical estimates. Levels are as of the close stated in each report.
The terms the reports use, in plain words.