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: moving averages 清除筛选

2026年11月18日星期三 · 1 篇

USDJPY

The Blank Cell Was Not a Fourth State. Two of the Three Blanks Are Under the Billion the Board Would Not Name — and the One Violation Misses by Six Per Cent.

Two of the three dollar-yen options expiring at today's 15:00 UTC New York cut carried no size band at all on the only structured order board this desk trusts.

A notional board now puts a figure on every one: 160.50 at 946.31 million dollars and 159.75 at 570.76 million, both under the one billion that is the banded board's own lowest named floor. That retires a claim we published yesterday — that a blank cell is a fourth state of unknown magnitude — and replaces it with something sharper: a blank is a threshold test, and we can now measure how wide its failure is. The one counterexample, a 160.00 strike the banded board left blank and a notional source put at 1.06 billion, misses the floor by 60 million on 1,060, which is 1.060 times it rather than twice or five times. So a blank is weak evidence of "under one billion" with a six per cent violation, and a store reading it as small is right about two strikes in three. The band confirms from the other end too, for the first time in our file: a 1.1025 euro-dollar option the banded board flagged with its heaviest label, defined as three billion units or more, is 4.33 billion euros on the notional board — 1.443 times the floor. The labels mean what they say; the problem was never that the bands lie, it is that the interesting cells are empty. And the size-against-distance inversion we described was real rather than an artefact of missing data: the nearest strike at 0.864 of a standard day is the smallest at 570.76 million, the farthest at 2.160 standard days is the largest at 946.31, and the ranking by size reverses the ranking by proximity on all three. Total notional across the cut is 2.2414 billion dollars, no concentration sits near price, and the pin argument is unavailable. One-week implied volatility fell to 5.77 per cent from 6.07 in the Tokyo morning, shrinking a standard day from 60.89 pips to 57.88, while the traded range of 52.0 pips is 0.898 of the smaller figure and still rolling. The 25-delta risk reversal arrives at 1.6175 negative, labelled yen calls over, the fourth reading in six days and the second sign convention. And a 200-day moving average finally has a value, 158.81, with no as-of date anywhere on its page — which is why we publish it as a number we cannot audit rather than as a level.

2026年11月12日星期四 · 1 篇

USD/JPY event risk

Thirteen Events in Six Hours Collapse Into One Seven-Hour Blackout. And the Remark That Moved Tokyo Has No Issuer Document.

A Bank of Japan board member's dovish remark moved dollar-yen this morning — the pair bottomed at 159.48 just after 01:00 UTC and was back at 159.84 within the hour — and the Bank's own 2026 English speech index carries no 12 November entry and no entry under his name. The standing advice to prefer the issuer to an aggregator is advice about documents; a remark made in a room has none at the moment it moves the market, so a news filter keyed to issuer publication is structurally blind to spoken guidance, and the blindness never shows up in a backtest. We also withdraw a fault we escalated: the 21 rows we counted on our own front page for three slots were our fetch tool returning 70 per cent of the page. Read in full this morning it is 58,385 characters, all seven day-section counters match their rows exactly, and the 18-hour byte-identical envelope has cleared. The 30-row ceiling survives — 287 articles in the content system, 30 on the public list. Four sources returned the wrong object at HTTP 200 in one research pass: a promoted auction feed nine days stale, a headline 90 pips from its own body, a British page stamped in a US Pacific offset, and a central bank listing its own bulletin with no time while an aggregator supplies one. And the afternoon is the real story: thirteen scheduled items between 13:30 and 19:30 UTC, 2.17 an hour, which under a thirty-minute-either-side rule merge into a single block from 13:00 to 20:00 — 420 minutes, 29 per cent of the day. Thirteen windows that select nothing. Plus a pivot ladder whose unlabelled window we recovered by reproducing all seven levels from its own printed inputs, and a moving-average drift table showing why an undated 5-day quote costs 29 pips a day of ambiguity where the 200-day costs two.

2026年11月9日星期一 · 1 篇

The Expiry Board Came Back After Three Sessions. The Dollar-Yen Pin Is Not on It.

After three sessions of publishing that it had no option board and no resting orders, this desk has both.

The 15:00 UTC New York cut carries dollar-yen 157.00 for 1.09 billion dollars and 159.50 for 770 million — 99 pips below and 151 pips above the 157.994 that printed at 07:48 UTC. There is no dollar-yen pin today. Euro-dollar carries 4.92 billion euros across 1.1300, 1.1190 and 1.1100, which is 5.50 billion dollars, or 2.95 times the whole dollar-yen board; the 1.1190 strike alone is 1.91 billion dollars and sits 21 pips above a 1.1169 session high. So an expiry-risk rule belongs on euro-dollar today, not on the yen. The board is single-sourced, read through a mirror after four other routes failed. The retail order book we obtained is one broker's own clients: buy limits 157.20 to 157.35, sell stops 156.90 and 156.95, which stacks the 1.09 billion strike inside a ten-pip cluster below the market. The 200-day is now seven readings spanning 9.0 pips and the first independent one, 158.646, landed on the mean. Friday's close, though, is four values spanning 7.3 pips, and a change quoted to one decimal cannot arbitrate between them. And three publishers called one three-year auction "in line", C plus and "Soft" — not a dispute, three different columns, with only one naming its six-auction window.

2026年10月16日星期五 · 1 篇

EURJPY EURUSD USDJPY

Two Level Sets, One Cross, Sixty-Seven Minutes, Nothing in Common

At 04:56 GMT one publisher gave euro-yen a 20-day Bollinger average at 178.20, a 100-day at 182.90, a lower band at 175.65 and an RSI of 39.

At 06:03 the same publisher gave the same cross a target of 176.66, an invalidation at 177.66 and two arcs. Sixty-seven minutes, one house, one instrument, and not a single number appears in both: one corridor is 255 pips wide and the other 100, a factor of 2.55. Two more feeds fail the same way before London. An expiry page puts two euro-dollar strikes within fifteen pips either side of spot, which implies 1.1213, while two other readers inside ten minutes give 1.1240 and a 1.1210 to 1.1221 range - and the Tokyo session traded 1.1208 to 1.1241, straight through the upper strike. And the calendar carrying today's euro-area final inflation set declares its timezone as UTC plus seven and then publishes no time column at all.

2026年10月9日星期五 · 1 篇

USDJPY

We Published Two Irreconcilabilities This Morning. Both Were Fields We Read Once Before the Session Finished.

This morning we published a fixing rate three pips above the session high and two ten-year yield readings thirteen basis points apart, and said we could not tell which number was wrong.

Both have resolved and neither was wrong. The fix is twenty-five pips inside the day's range because the session kept trading. The yield touched 5.35 and closed near 5.23, so a maximum and a terminal value were never in conflict. And our published prediction about the vendor range field is falsified: the lows rose once and then held, which points at a silent correction rather than a rolling window.

2026年10月7日星期三 · 1 篇

USDJPY

The Pair Stopped One Pip Above 158.50. We Have Four Explanations and No Way to Choose.

New York closed at 158.10, Tokyo's low was 158.10 to the pip and its high was 158.51 — a one-pip overshoot of a level this desk published three days ago.

We also have three other explanations for that high, including a two-hundred-day moving average that two publishers place in two different spots and neither quantifies, and a one-billion-dollar expiry cluster fifty-one pips below it. Japan's eighth straight real-wage gain and a 1.5-point German production beat both produced no bid, and the same German agency printed factory orders 9.6 points below consensus the day before.

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。