FX News

Market-moving events, session conditions, and macro context — written for traders who run systems, not predictions.

Follow pairs by starring them on the Signals page.
Tag: calendar metadata Clear

Friday, November 13, 2026 · 1 notes

Four Calendar Routes, Three HTTP 200s, Zero Rows. And the Issuer's Own Current-Release URL Is Serving Last Month.

At 11:37 UTC, one hundred and thirteen minutes before US October producer prices, we went to four separate routes for today's American calendar — two aggregators, a third publisher's calendar, and the issuing statistical agency itself. Three returned HTTP 200. All four returned zero rows. Three of those four are successes by every criterion a pipeline checks: status code fine, content length plausible, parse succeeds, row count zero — and a row count of zero has two causes that look identical from outside, no events today or your route broke. Agreement among empty results is not corroboration; the only thing that converted those zeros into a finding was a release we knew existed. The sharper half inverts our own standing rule to prefer the issuer to an aggregator: the Bureau of Labor Statistics producer-price release sits at a stable unversioned current-release address, and two hours before this month's release it serves last month's — with a title naming its month, a headline figure, a core figure, both annual rates, a revision footnote, an embargo line with date, hour and zone, and a next-release line giving today at 8:30 Eastern. Every field populated, every field right, the document simply wrong. Scrape it into a field called latest producer prices and you have stored a correct number under a wrong label, the defect class we have logged more than two dozen times, arriving from the one direction we had not seen. The check nobody runs is free: an embargo date in the past plus a next-release date today or earlier brackets the page as superseded. Tonight's consensus — headline plus 0.4 against plus 1.3, about 6.6 year-on-year against 6.3, core plus 0.3 against plus 0.7, core 5.0 against 5.0 — now has three readers, none of them a working calendar and none of them the issuer. The expiry board's ninth consecutive failure is arithmetic, not luck: the only publisher carrying strikes with sizes posts at about 12:40 UTC, our runs read at 23:00, 06:00 and 11:30, and this one misses by 63 minutes on today's clock — leaving a real fifty-minute window between the board and the release that our schedule sits entirely outside. And our own 10.40 per cent December figure is 23.9 hours old against one reading today of 14.5 per cent, four points higher, whose claim of zero a week earlier we are cutting while keeping the level.

Thursday, November 12, 2026 · 1 notes

fed fomc

December Is Already Fully Priced, So Six Fed Speakers Can Only Move a Ten Per Cent Tail. And the Issuer Says 16:15, Not 16:30.

A quarter-point December hike is fully discounted on the one feed anybody names, which means six Federal Reserve speakers between 14:00 and 23:30 UTC cannot move the base case at all — only a half-point tail priced near ten per cent, and a January meeting nobody has priced. The wire carrying that schedule prints three December figures from CME FedWatch, and they are one observation in three dresses: 25 basis points certain plus a ten per cent chance of a further 25 gives 27.50, the page says 27.6, and solving backwards gives a half-point probability of exactly 10.40 per cent — the same figure we published yesterday from a different route. Store all three as separate inputs and your system counts one number three times. Waller's half-point condition has already fired: he said in London on 10 November that it may be needed if October core CPI came in near 0.3 per cent or higher, and core CPI printed 0.3. The market pays about ten per cent for it. We also settle a dispute we published this morning, and nobody who agreed was right: two aggregators put Goolsbee at 16:30, one put him at 16:15, and the Chicago Fed's own page says 10:15 am CT, which with daylight time over is 16:15 UTC. The lone dissenter matched the issuer. The same page calls the event a moderated virtual discussion with a named journalist, where one aggregator calls it a speech — a scheduled event with no document to read. And the day's only US government release is framed by calendars that are broken in three adjacent rows: three publishers give consensuses of 203,000, 191,000 and 195,000, a 12,000 spread, while the Labor Department's own 5 November release shows 189,000 belongs to the week ending 31 October and 1,718,000 to the week ending 24 October, both filed a week forward, and a four-week average sitting in a level field. The one-week lag between initial and continuing claims is built into the release by design, and a vendor that aligns them to one date has deleted it.

Tuesday, November 10, 2026 · 1 notes

Seven Date Fields Failed This Morning. Four of the Events They Describe Fire Before Tonight.

We set out to build an event map for the rest of today — US CPI at 13:30 UTC, the New York option cut at 15:00, a Dallas Fed speech, and a 42 billion dollar ten-year auction at 18:00 — and instead found seven date or time fields that are wrong, missing, stale or contradictory, four of them attached to events that fire before tonight. Three dated expiry routes all returned client errors at the correct slot, which kills yesterday's "wrong slot" explanation and makes this the fifth session running with no board; we say plainly that we do not know where the optionality sits and refuse to invent a pin. One calendar lists the ten-year auction with a 5.30% prior yield while the cash ten-year trades near 4.86 — 42.0 basis points away, and 2.0 from Monday's thirty-year close, which is a reconciliation hypothesis and not a conviction. Logan is still 14:35 or 15:50, now two-to-one toward the later time but on forward calendars, which are statements of intent rather than reaction notes, so a 14:30 to 16:00 blackout is the only setting correct under both. A live blog puts a 13:30 release at 09:00. A search-topping auction record is a 12 May page with no date on it anywhere, which is why the date check fails where the level check works. And a syndicated article whose slug encodes 10 November 06:49 carries metadata saying 9 November 22:49 — eight hours apart, on the publisher whose slug we have treated as a timestamp for eight years. That is not a wrong number; it is a rule with a dependency we had not checked.

Monday, November 9, 2026 · 1 notes

US CPI BLS

Tomorrow's CPI Has No Year-Over-Year Number. One Calendar Is Carrying a 3.9% Consensus for It.

Tomorrow's US October CPI, at 13:30 UTC, will not carry a year-over-year rate for the headline index.

The BLS says so on its own page, last modified 16 September: it will not publish 12-month percent changes for October 2026 where October 2025 data are missing, and for most series October 2025 was never collected during the 1 October to 12 November 2025 lapse in appropriations. One calendar verified this morning is carrying 3.9% as the market consensus for that field anyway, with no caveat. Four annual rates now exist for a month that has none — 3.9%, 3.74% from a Cleveland Fed nowcast, 3.8% from one reader's flat-October arithmetic, and a 3.9% that is explicitly a comparison against September 2025 — each built on a different synthetic October. A filter computing actual minus consensus either reads the null as zero and sees a 3.9-point miss, skips the week's biggest release, or silently adopts a vendor's stand-in. The field that will print is the monthly change, where named forecasts run 0.19% to 0.3% — all of which round to 0.2 or 0.3. We also correct ourselves: seven hours ago we called 15:00 today a window you can safely not have and said Goolsbee had no published time. It holds four items plus Goolsbee at 15:00, confirmed on two calendars, one of which states every time with no time zone at all.

Thursday, November 5, 2026 · 1 notes

The Bank's Own Page Said the Decision Had Not Happened, Two Minutes After It Had.

At 12:02 UTC, two minutes after the Bank of England published its November decision, the Bank's own canonical page for that decision still said it had not happened — title ending "(to be published at 12pm)", body reading "will be published on 5 November 2026". The Monetary Policy Report page did the same. Appending a query string to the identical path returned the full release: Bank Rate 4%, an 8–1 vote, twenty-four paragraphs of minutes. One cache-buster was not enough — the first variant also served the placeholder, so two appendages against one path in one minute gave two different documents. A filter keyed to issuer confirmation would have stayed shut through the move it was built for, with no error code to branch on. The same event went wrong a second way in our own notes: an automated read of a Japanese publisher's 24-hour calendar block put the Bank of England at 22:30 Japan time, ninety minutes late, and in the same minute as US claims and productivity — a plausible cluster rather than an implausible orphan, so a plausibility check passes it. A slower second read of the same block gave 21:00, which is what the issuer says. Neither error was a source being wrong: both were introduced between the page and us, and an extraction fault survives adding a second source because the second source goes through the same extractor. Today's real stack in UTC, the thin consensus figures behind it, and the one sizing fact that matters: one-week dollar–yen implied volatility at 6.23%, the lowest since 22 October, on a tenor that covers tomorrow's payrolls.

Wednesday, November 4, 2026 · 1 notes

EURUSD

Three Readers Put Euro Inflation at 10:00. The ECB's Own Calendar Says 15:00.

Euro-area flash inflation for October prints today, and three independent readers put it at 10:00 UTC — a vendor calendar printing 10:00 AM under a column headed GMT, an event page printing 11:00 CET, and a Tokyo publisher printing 19:00 Japan time, which are the same instant. The ECB's own weekly calendar carries a euro-area HICP flash estimate at 15:00 with no time zone printed, and no conversion reconciles the two. The likeliest reading is that they are different objects: the ECB row is named "euro area seasonally adjusted HICP flash estimate", two extra words that are the whole story, and a news window seeded from the central bank's calendar opens four hours after the number that moves EUR/USD. On that one page two of three rows carry a bare clock time and the third carries 14:20 CET. Eurostat confirms the date in the header of its previous release and gives no hour; its own release calendar returns no table. Separately, the consensus-column trap we filed on Monday has its second instance in three days: the same vendor prints Consensus 4% and a 3.9% house forecast as two columns of one row, which is most likely the whole ten-basis-point "dispute" across readers. And two calendars list Lane as a speaker today — the ECB says it is a laudation at a prize conferral in Berlin, not a policy speech, while the panel likelier to carry rate-path content is tomorrow in Zurich on the Bank of England's day. September's headline was 3.8% against 2.5% core, with energy at 18.8% year-on-year.

Tuesday, November 3, 2026 · 2 notes

JOLTS job openings

Job Openings Beat by 420,000 and the Issuer Called It Little Changed.

September US job openings printed 7,523 thousand against consensus figures clustered near 7.1 million — a beat of roughly 420,000, above every estimate in one publisher's survey — and the issuer describes the month as "little changed at 7.5 million." Both are true, and they come apart because the same release rewrote August in five fields: openings up 111,000 to 7.2 million, hires up 79,000, total separations up 105,000, layoffs up 127,000 to 1.8 million, and quits down 22,000. A calendar still holding the 7,079 August figure published on 29 September computes +444 thousand on the month; the issuer's own arithmetic is +333. One briefing carried that stale 7.079 prior, and a stale 3.066 quits prior, on the morning of the release. Yesterday we published that three readers were 34 thousand apart on this consensus and gave the spread a clause of its own; the print missed all three by twelve times that, and a wider reader set today — 7.10, about 7.164, 7.1 unattributed and 7.13 — did not matter either. The "layoffs lowest since March 2025" claim is single-sourced and scored against the August figure this same release moved up by 127,000. Quits printed 3,365 thousand against one publisher carrying two different quits consensus figures on one page. Tonight: ADP at 13:15, where one page gives three different priors and two contradictory consensus statements, and ISM Services at 15:00 in the same minute as the New York cut, issuer-confirmed.

Tonight's Event Starts at 23:00 and Has No End Time.

The first American polling stations close at 23:00 UTC tonight and the last at 06:00 UTC tomorrow in the western Aleutians, with nine waves in between — and after all of them the result may still not exist: control of the House took eight days to establish in 2024, and the only dated resolution step on any calendar for one of tonight's Senate races is an Alaska ranked-choice tabulation on 18 November. Two independent readers now agree on all eleven closing boundaries, which closes a window this desk published twice off a single mid-October page. A third publisher is an hour out: its earliest closing is midnight GMT in Florida, Georgia and Virginia, which is the second wave, so a desk keyed off it opens its handling sixty minutes after the first results exist. 435 House seats and 35 of 100 Senate seats are up, and neither explainer states a control threshold. What a system faces is not a thirty-minute event with a known clearing time but an elevated gap regime beginning at a known minute and ending on an unknown date — and the option market agrees, pricing about 0.86% for today against 1.85% out to 6 November, nine-day implied at 22.46 over a 21.15 spot index, November futures above December. New Zealand labour data prints at 21:45, seventy-five minutes before the first poll closes. And one calendar, published yesterday afternoon, still records the midterm dates as remaining to be confirmed while quietly carrying a petroleum inventory row an hour out of place.

Monday, November 2, 2026 · 1 notes

Six Rows Print at 15:00 Today. The 16:15 Fed Speaker Is Moderating.

At 15:00 UTC today one calendar lists six rows — the October ISM manufacturing survey, three of its sub-indexes and September construction spending — and the 10am New York option cut falls in the same minute, with the final manufacturing PMI fifteen minutes earlier and nothing at all at 13:30. The release time is now four-sourced across four zones at 15:00 UTC, so the hour is closed; the clustering is the hazard, and a system that counts calendar rows rather than releases will read six events where there are two. Tonight's prior is also closed at 54.5, two-sourced, which shuts a single-sourced figure our Systems desk published twelve hours ago — but one dedicated event page still carries 55.6 per cent in its Prior field, marked July 2026, three releases stale, because the page was last updated on 25 August; its own sidebar further down has the August print right. Scored against that page a 54.6 print is a 1.0-point miss instead of a one-tenth beat. The 54.6-versus-54.7 consensus spread may not be a disagreement at all: one vendor prints Consensus 54.6 and TEForecast 54.7 as two labelled columns of the same row, so check which column your readers sampled before calling a tenth dispersion. Construction spending has three expected values — 0.5 from a broker, 0.6 on one vendor's indicator page and 0.4 on that same vendor's calendar. And the 16:15 Federal Reserve row that two calendars call a speech is, per the Chicago Fed's own agenda, Austan Goolsbee moderating a BIS deputy general manager's keynote at an invitation-only conference under the Chatham House Rule with no media access — while the Board's own calendar lists no speech that day and two releases no FX calendar carries.

Friday, October 30, 2026 · 1 notes

The 13:00 Print Already Happened in Four German States. The 12:30 Print Moves to 13:30 on Monday.

Germany publishes October consumer prices at 13:00 UTC and four of its state statistical offices have already published theirs, every one of the four lower: Bavaria 2.8 per cent year-on-year from 3.2, North Rhine–Westphalia 2.7 from 3.3, Lower Saxony 2.9 from 3.3, Baden-Württemberg 2.6 after a three-tenths fall. So the 13:00 window guards a release whose direction was public eight hours earlier. The fifth state is the one to doubt: Saxony is reported as having "remained at 2.8", which asserts a September reading of 2.8, while a second publisher has September Saxony at 3.40 — and the other four priors cross-check to the tenth, so the suspect field is a word rather than a number. The consensus is also two objects: 3.0 per cent national year-on-year on one vendor against a Reuters poll at 3.1 for the harmonised rate. The second mechanism arrives Sunday. The United States ends daylight time on 1 November, three-sourced, so every American release moves an hour later in UTC while nothing European moves — and the issuer's own label, "Eastern Time", names a zone and not an offset, so the string "08:30 AM" is identical on the October and November pages while the instant is not. Payrolls on 6 November is 13:30 UTC, not 12:30. Our own week-ahead had October CPI on 13 November; the issuer has it on the 10th and the 13th is PPI. Euro-area Q3 GDP printed 0.4 against a 0.2 consensus — a beat that is also a deceleration from 0.6 — with France at minus 0.1 and Ireland at 5.5.

Wednesday, October 28, 2026 · 1 notes

AUD

The Issuer Says 4.3. Two Vendors Say 3.9 Against a Consensus of 4.2. The Surprise Changes Sign.

The Australian Bureau of Statistics says the monthly consumer price index rose 4.3 per cent in the twelve months to September, up from 4.0 per cent.

Two publishers put that same row at 3.9 per cent against a consensus of 4.2, with the prior correctly at 4.0 — so a system that scores surprises by subtracting consensus from actual recorded a 0.3-point miss where the issuer gives a 0.1-point beat. The sign is opposite and nothing in the row looks broken, because 3.9 is not a random error: it is the quarterly annual figure, which both tables also carry correctly one row away. Yesterday this desk published that the 3.9-versus-4.2 consensus dispute was two releases in one minute rather than a disagreement, and that a mislabelled series would show no symptom because the two trimmed means agreed at 3.6. The print produced three symptoms in twenty minutes. On the only measure the Reserve Bank of Australia targets, one publisher headlined 3.6 and dead in line while the other headlined 3.7, above forecasts, hike case firming — same release, same minute, opposite verdicts — and the Bureau's own 3.6 matches the first publisher's quarterly row and neither publisher's monthly row. August's prior was revised to 3.5 and both vendors still carry 3.6. We publish the band and refuse the tenth. None of it moved the price: swaps went above 80 per cent for a November hike, Bullock said inflation will run above the August forecasts, and the Australian dollar rallied and stalled.

Monday, October 26, 2026 · 1 notes

EUR

We Corrected 09:00 to 09:30 for the Ifo. It Printed at 09:00, and the Issuer's Schedule Was the Wrong Document.

At 06:15 UTC today this desk corrected its own week-ahead table, moving the German Ifo release from 09:00 UTC to 09:30 on the strength of the Ifo Institute's own forward schedule, which gives 10:30 local time for today and every month out to December 2027. The index printed at 09:00. The correction was the error. Two Japanese publishers settle it from a zone with no summer time: a scheduled-release note at 17:46 Japan time lists the Ifo for 18:00, which is 09:00 UTC, and a second publisher warned of the release at 17:50 and published the reaction at 18:27 Japan time - 09:27 UTC, three minutes before the hour the issuer claims the data comes out. So the aggregator we convicted this morning of storing 10:00 local in all nine rows of its history was right in all nine, and the document that was wrong was the issuer's. The rule we published at 06:15 - for a release hour go to the issuer and nowhere else, because two sources agreeing is not the test and one of them being the issuer is - was falsified two hours and forty-five minutes later by the event it came from. The print was 90.2 against four different published consensus figures: 89.8, 89.9, 90.0 and 90.1, a beat of 0.4, 0.3, 0.2 or 0.1 depending only on which field your system holds, with one publisher carrying two of them. And the suppression rule we shipped at 23:14 last night would have suppressed a 0.3-point beat, because the 89.9 consensus equals September's 89.9 print exactly. It is a false positive and we now have the print instead of the argument.

Tuesday, October 20, 2026 · 1 notes

We Argued About 70 Basis Points. Pending Home Sales Missed by 680.

Yesterday this desk published, as a scheduling argument, that September pending home sales would land at 14:00 UTC in the same minute as the New York option cut and that its consensus could not be stated more precisely than a 70-basis-point band, 0.0 per cent at one publisher and minus 0.7 at another. The issuer has now printed: the National Association of Realtors puts the index at 66.0, down 7.3 per cent on the month and 11.3 on the year, with every region lower and the West down 9.2 monthly and 15.0 annually. The nearer edge of our band was 660 basis points from the print and the far edge 730. Worse, one of the two publishers held two consensus figures for the same release - 0.0 before it and minus 0.5 after - 50 basis points apart. The transferable finding is a ratio: 70 basis points of vendor dispersion against 680 of realised surprise, a factor of about ten, which means a news filter sized by how far apart the forecasters are would have treated this as a minor event. Housing starts in the same block missed by 6.4 per cent while building permits were almost exact, so no single number describes how uncertain housing data is. Two further calendar fields fail the same way. The Bank of Japan's own calendar shows one upcoming meeting, 29 and 30 October, and no December entry at all, so the 17 December date one aggregator publishes has no issuer behind it. And the Turkish decision time of 11:00 UTC lost its only source when we retracted the rate-cut figure that came with it.

Friday, October 16, 2026 · 1 notes

EUR

The Headline Was Confirmed. Two Components Were Not. And the Hour We Said Nobody Published Was 09:00.

Eurostat confirmed euro area September inflation at 3.8 per cent annual and 0.6 monthly, and core at 2.5 per cent per one secondary reader.

The headline did not move. Two of the four components did: energy from 18.8 in the flash to 18.9 in the final, food, alcohol and tobacco from 1.4 to 1.3, each a tenth, in opposite directions. The final also publishes contributions the flash does not - energy 1.71 points, services 1.52, goods 0.29, food 0.25, summing to 3.77 - and no secondary reader we found carried any of it. Separately, a correction: both of this desk's articles this morning said no source gave a release hour for this print. One vendor calendar did, at 09:00 GMT, and the issuer's own page metadata reads 09:00:00Z. A negative claim about publication is a claim about your own sweep. Also: three calendars and an agency schedule checked for a margin-of-error column, and none has one.

Tuesday, October 6, 2026 · 1 notes

We Said 05:35. A Second Publisher Says 06:35, and the Clock Agrees With It.

Seven hours ago this desk published that a central-bank governor's speech today was at 05:35 UTC and that the publisher's 0635 GMT and 0235 US Eastern stamps were both an hour late.

A second publisher independently stamps it 06:35 UTC, and the clock has run fifty minutes past 05:35 with no speech reported. Our conversion arithmetic was right; our choice of which field to trust was wrong, and the rule that would have worked is to go with the majority of a page's own timestamps rather than a prior about which field is reliable. The real window opens twenty-five minutes before the London open at 07:00. The October rate-rise probability now has three consistent readers at roughly one in four.

Friday, October 2, 2026 · 1 notes

euro area CPI HICP

The Headline Beat Both Our Consensus Figures. The Core We Barely Mentioned Was Exact.

Euro-area flash inflation printed 3.8 per cent against the 3.7 we published and the 3.6 one publisher carried — so the tenth of a point we logged as an open item was the same size as the smaller miss and half the larger.

Core printed 2.5, exactly the figure we published and then ignored. Which field your news filter reads decides whether today was an event at all. Payrolls at 12:30 now has five consensus values, and the July prior the market traded on has been revised from minus 23,000 to plus 21,000 — a sign flip. One calendar's ADP cell has read as a future event for 47 hours and 43 minutes.

Thursday, October 1, 2026 · 1 notes

jobless claims ISM

The Surprise Was Three Thousand Claims. The Gap Between Two Published Actuals Is Also Three Thousand.

US jobless claims beat their consensus by three thousand last night, and two published values of the actual — 197,000 and 194,000, from the same publisher — are also three thousand apart.

The prior was revised under both of them, including the 1,719,000 we published two days ago. Our own correction of the consensus was wrong by the same amount in the other direction. ISM printed at 14:00 UTC as we said, but the issuer's own press release had the timezone right all along: one institution, two labels, two surfaces. And one calendar's ADP row has now read as a future event for thirty-five hours.

Trade the reaction, not the headline
News tells you when conditions change. Pair it with the economic calendar and check the current signals before adjusting anything.