Today’s New York cut carries 7.2 billion euros of dollar–euro optionality across four strikes, and every one of them is above the market. The board has more than doubled in size overnight and moved further away from spot. That is the second consecutive day the euro board has been positioned entirely out of reach, and it is the opposite of what a rule reading “large expiries today” expects. While we are here: the structural finding we published yesterday about the same board was, for three of the four currencies, arithmetic.
The board, as published
One wire, timestamped 08:56 UTC, for the 10am New York cut at 14:00 UTC. Euro–dollar: 1.1150 for 1.9 billion euros, 1.1200 for 1.6 billion, 1.1215 for 2.5 billion, 1.1250 for 1.2 billion — 7.2 billion in total, against a spot one reader put near 1.1120 at 08:27 UTC and a seventeen-month low at 1.1113 tested this morning. The nearest strike is roughly thirty pips up. The 2.5 billion at 1.1215 is the largest single strike on the entire board, in any currency, and it sits about ninety-five pips above the market.
Dollar–yen: 157.15 for 810 million dollars, 157.90 for 650 million, 158.00 for 1.1 billion, 158.50 for 902 million, 159.00 for 612 million — 4.074 billion, with spot just under 158.00 on a morning reading. The 1.1 billion at 158.00 is effectively on the market. Dollar–Canada carries 1.039 billion across 1.4000 and 1.4050. Australian dollar: 0.7005 for 1.1 billion and 0.7050 for 891 million, 1.991 billion in total.
Two housekeeping notes, because they are the kind of thing that costs somebody a figure. The notionals are single-sourced; we could not open the second publisher’s board for a fifth consecutive session. And the page’s own headline says 28th October while its URL says 29th October — the content is today’s and the slug is tomorrow’s. If you scrape that site by URL date you will file today’s board under the wrong day.
The euro board has been walked away from twice, and the size is going the wrong way
Yesterday the euro board was 3.35 billion across four strikes, all above a 1.1142 spot, the nearest thirty-three pips up. Today it is 7.2 billion across four strikes, all above a spot near 1.1120, the nearest thirty pips up. The notional has rather more than doubled while the market has moved further underneath it.
The reading we would offer is that this is residue rather than a magnet. These are strikes placed when the euro was higher, expiring where the euro is not, and a bigger board of unreachable strikes is evidence of how far the currency has travelled rather than evidence of anything about today. The euro is heading for a 1.70 per cent monthly loss against the dollar on one reader’s count, the weakest developed-market currency of October.
The operational consequence is narrow. If your system treats aggregate expiry notional as a single state variable — large board today, expect pinning, reduce size — it has now been wrong about the euro on two consecutive days in the same direction, while being roughly right about the yen. The variable that carries information is not the size of the board. It is where the board sits relative to spot, and whether anyone still has a reason to defend it.
We said the market had walked away from yesterday’s nearest strike. It had traded through it four hours earlier
At 11:42 UTC yesterday this desk published that all four euro strikes were above a 1.1142 spot, the nearest by thirty-three pips, and called them “three and a quarter billion euros the market has walked away from”. The spot clause was true. The walking-away clause was not something we had established.
Our own note from that morning records a wire’s 10:20 GMT session range of 1.1135 to 1.1177 — two pips above the 1.1175 strike, printed before we published the sentence. A second publisher, writing at 16:02 GMT, describes the pair as having broken support at 1.1175 to 1.1190 during the session. So the nearest strike was not abandoned territory; it was the level the euro had just fallen out of, hours before the cut.
The correction is small in pips and not small in method. “Above spot” is a statement about one instant. “Walked away from” is a statement about a session, and it needs the session’s range, which we had in our own notes and did not look at. Today’s version of the same claim is better supported only because the gap is thirty pips to a market making seventeen-month lows in the other direction, and we will say that rather than pretend the method improved.
Yesterday’s structural finding was, for three currencies out of four, a convention restated
We also published that not one of the four expiry boards expires after the event in its own currency, and presented it as a structural observation. It is worth separating what survives from what does not.
The 10am New York cut is 14:00 UTC every single day. The Federal Reserve announces at 18:00 and always has. So “the yen board expires before the Fed” is not a discovery about today’s positioning; it is two fixed conventions subtracted from each other, and it will be true of every Fed day for as long as both conventions hold. The same goes for today’s dollar–Canada board, which dies one hour and forty-five minutes before the Bank of Canada at 13:45, and for the euro board relative to tomorrow’s European Central Bank decision.
What survives is the part that crossed sessions: 4.969 billion Australian dollars of optionality expiring ten and a half hours before an Australian quarterly inflation print, in a different session, on a different day’s board. That one was informative because the gap was not a convention — it was a choice about which day’s expiry you hold. Today’s Australian board, 1.991 billion, no longer has an event behind it at all.
We dressed a subtraction as a finding. The test we should have run before publishing is the one we will run from now on: if the interval between two timestamps is the same every week, it is a convention, and the only interesting question is whether anybody has arranged their book around it.
How much of tonight is priced, and by whose reading
On today’s decision, the readers agree for once. One publisher citing the CME tool gives 92.8 per cent for a hold; a wire at 09:58 UTC says unchanged is priced above 90 per cent; a third reader, also citing the CME tool, says about 91 per cent. That closes a figure this desk carried single-sourced for six runs. A hold at 3.75 to 4.00 is the consensus outcome and nothing in the day disputes it.
December is where it comes apart, and interestingly so. One reader citing the CME tool puts a December hike at 84.4 per cent. Another, as of roughly 05:00 UTC, cites the same tool for 82.5 per cent “probability of higher rates by year-end”. A Tokyo morning note gives just under 90. The first two nominally read the same instrument, and the second figure should be the larger of the pair — higher rates by year-end includes a hike today as well as in December — and it is the smaller. These are intraday numbers taken at different moments and that may be all it is; we are not calling either wrong. We are saying that if your sizing depends on whether December is 82 or 90, you are depending on a quantity that two readings of one tool do not currently agree on.
One more number to mark, because it belongs to the same family. Yesterday’s durable goods orders printed 1.8 per cent on the month. The consensus against which that is a beat was 1.6 per cent on one reader, 1.5 on another, and 1.4 on the Japanese schedule we carried — while the figure this desk had carried as the consensus was 1.8, which is the print. The surprise is therefore zero, two, three or four tenths depending on whose table you read. Underneath it, orders excluding transportation rose 0.5 against an expected 0.6, core capital goods orders rose 0.5 after 1.6, and August’s headline was revised up to plus 0.1 from the minus 0.1 in our file. The revision is larger than the smallest surprise on offer.
What this does not tell you
The expiry notionals come from one publisher and we could not corroborate them; the second source has failed to open for five consecutive sessions. We have no intraday spot reading later than about 08:30 UTC, so every level in this article is two to three hours stale by the time you read it, and the yen figure is a description rather than a quote. We did not reach the issuer for anything here.
Nothing in this article says the euro strikes will not be reached, or that the 158.00 yen strike will pin. The 1.1 billion at 158.00 sits on the market four hours before a Federal Reserve decision, which is precisely the configuration in which an expiry stops mattering. And the durable goods consensus spread is a statement about vendor tables, not about manufacturing: the print is the print, and it beat every published forecast we found.
