外汇资讯

能撼动行情的事件、各时段的盘面状况与宏观脉络。写给用系统交易、而不是靠预测的人。

在信号页面给货币对加 ☆ 即可关注。
标签: NFP 清除筛选

2026年11月8日星期日 · 1 篇

USD

Payrolls Printed 165,000. Our Error Band Stopped at 126,000, and the Miss We Built It From Was Revised in Half.

US October payrolls printed 165,000 against the 65,000 consensus we reported on Thursday, and 106,000 above the 59,000 a fifth reader gave on Friday.

It beat the most bullish house in our file, RSM US at 95,000, by 70,000. Worse for us: we published an error band of 4,000 to 126,000, derived from September's 61,000 miss, and the print landed 39,000 above its top — while the same release revised September from 29,000 to 60,000 and August from 133,000 to 186,000, cutting the miss our band was built on in half. The band was too wide and too low at once, because it was centred on the consensus and the consensus was what was wrong. The three-month average we gave as context on Friday, 51,000, is now 137,000: one release moved a trend measure by a factor of 2.7. Both wage fields went hawkish — 0.4% monthly to $38.00 and 3.1% annual, where we had flagged the fields disagreeing as the likely outcome. And 55% of the beat is one sector: professional and business services at +70,000 against its own 15,000 twelve-month average, with health care another 11,000 above trend. Against it, part-time for economic reasons rose 339,000, more than twice the headline gain, on an unchanged 34.4-hour week. Canada printed in the same minute: −4,700 and unemployment at 6.7%, which Statistics Canada declines to call a six-month high even though three wires do, over a private-sector fall of 53,000 against a public-sector rise of 50,000.

2026年11月5日星期四 · 1 篇

USD CAD

Four Readers Agree Payrolls Is 65,000. Last Month They Agreed on 90,000 and It Printed 29,000.

US October payrolls print at 13:30 UTC and, for the first time in weeks, this desk cannot find a consensus dispute: four independent readers carry 65,000 to the thousand.

That unanimity is worth nothing. They are not four surveys — one of them names the source, the median of Bloomberg contributors, which is why the other three agree. A cluster of identical figures with one upstream source has the information content of that one source. One month ago the same unanimity sat at 90,000 and September printed 29,000: a 61,000 miss, 68% of the forecast it busted and 94% of the entire level everyone is quoting this morning. Repeat that error in either direction and today lands near 126,000 or 4,000 — opposite trades, both inside one month's demonstrated error, against a three-month average of 51k. The named houses are 40,000 apart around the median they all quote: one bank at 90k, another at about 50,000 with unemployment ticking down to 4.1% rather than holding 4.2%. Canada's employment report prints in the same minute, 13:30 UTC, forecast +12,200 against a −68,300 prior — an 80,500 swing asked for in one month — so a filter keyed to "Nonfarm Payrolls" or to USD alone covers half of what fires. And the field that decides December has two signs: consensus wants average hourly earnings to triple month-on-month, 0.1% to 0.3%, while the annual rate falls from 3.0% to 2.9%, so an exactly-in-line print reads as acceleration on one field and deceleration on the other.

2026年10月4日星期日 · 1 篇

payrolls NFP

We Stopped Counting Consensus Values at Five. There Were Seven. The Print Was Below All of Them.

September payrolls printed 29,000.

The seven consensus values this desk collected ran from 84,000 to 100,000, so every one of them was too high — the closest, a named bank's 60,000, by 31,000. The miss against the 27-analyst median was 61,000, which is 0.68 times the mean absolute error we published for this release a week ago. Two harder findings: the Bureau's own trailing twelve-month average moved from 31,000 to 45,000 between consecutive releases, so the denominator in our 2.9-times-trend claim was not a fact; and the 4.2 per cent unemployment figure we read and declined to publish, under a rule we wrote the same morning, was the one that printed. July now has three vintages and has changed sign twice.

交易市场的反应,而不是新闻标题
资讯告诉你的是盘面什么时候变了。请结合财经日历,并先确认最新信号,再去改设置。