Past Results

Catalysts whose event has now passed.

Sell The News

Win rate100%4 of 4 traded cards · 2 left out

10 cards

Measured from

Details

PassedThu, 24 Sept 2026, 19:00

USDMXNUS Dollar / Mexican Peso

Banxico decides for the first time since the Fed narrowed the peso carry

Mixed

17.5111

+500 to +1200 pips, which is +0.29% to +0.70%, if the Board opens the door to one more cut, and -400 to -800 pips, which is -0.23% to -0.46%, if it leans against core inflation

+0.89%Low

+1.46%High

USDMXN result

After the event it went to +0.89% at its lowest, and later to +1.46% at its highest, against the close price before Banxico rate decision.

The news wasPositive

Measured fromChart

Banxico held at 6.50% in a unanimous vote and repeated its prior "appropriate" language almost word for word, the neutral scenario's mechanics exactly. The peso still weakened sharply: USDMXN closed the 24 hours after the decision up 2.94%, more than the card's own dovish-case ceiling, driven less by the statement itself than by broad dollar strength and a hotter-than-forecast headline inflation print sitting alongside cooler core inflation, plus Banxico's language about not mechanically following the Federal Reserve.

What we said	What happened	Result
Catalyst confidence 99 out of 100: the decision is announced on schedule.	Confirmed. Banxico's Board announced its decision at 13:00 Mexico City time (19:00 UTC) on 24 September as scheduled, holding the overnight rate at 6.50% in a unanimous vote, the third consecutive hold.	Correct
Neutral scenario: "another unanimous hold with the same wording... reacts for less than an hour... the most likely single outcome."	The vote and wording matched almost exactly, unanimous, and the statement reiterated the same line that it will be appropriate to maintain the reference rate at its current level. The price reaction did not stay small, though: USDMXN closed the 24 hours after up 2.94%, far beyond a "less than an hour" reaction.	Mechanics correct, price magnitude wrong
Bullish scenario (weaker peso) trigger: "the statement drops the line that the current level is appropriate or a member dissents in favour of a cut."	Neither happened. The "appropriate" line was repeated verbatim and the vote was unanimous, 6-0 rather than a dissent. The peso still weakened, and by more than the bullish scenario's own expected range, without the specific trigger the card said would cause it.	Trigger did not fire, direction happened anyway
Bearish scenario (stronger peso) trigger: "mid-month inflation beats 3.3% and the Board stresses core inflation and the narrower gap to the Fed, ruling out further cuts."	Half right. Headline inflation did beat 3.3%, coming in at 3.42% against a 3.37% forecast and up from 3.26%. But core inflation, the figure the card said the Board actually watches, eased to 3.79% from 3.93% and came in below the 3.81% forecast, and the Board's own language leaned toward not reacting mechanically to the Federal Reserve rather than stressing the narrower gap.	Inflation trigger partly fired, Board language did not follow it
Our Expected Move +500 to +1200 pips (+0.29% to +0.70%) dovish, or -400 to -800 pips (-0.23% to -0.46%) hawkish.	Overtaken by roughly two and a half to four times the dovish range's own top end. The 24-hour close was 17.72177, up 2.94% (roughly 5,058 pips) from the signal price, despite the statement itself not being the dovish trigger described.	Far underestimated
Main risks: "the dollar has driven this pair all week and can drive it again on the day, so a clean hold may still see a move that has nothing to do with Mexico."	This reads as the best explanation for the size of the move. Coverage describes the peso sliding toward its worst week since March around this decision, and Banxico's own statement was, by design, a repeat of prior wording; the size of the reaction looks driven by broad dollar strength and the Board's language on not mechanically following the Fed rather than by any surprise in the vote itself.	Correct
Technical resistance: the 30-day high at 17.2402, then the 100-day average at 17.25246, then the 200-day average at 17.35389; a close above the 200-day "would turn the chart bullish and say the year's peso rally is over."	All three were cleared decisively. The pair traded as high as 17.76749, more than 40 pips above the 200-day average, confirming exactly the technical break the card said would end the year's peso strength.	Correct

Price Table	Rate (USDMXN)	Change from signal	Change from lowest (regular session)	Change from close before Banxico rate decision
After signal
Price at analysis (Asia session)21 Sep 2026, 23:59	17.21597	—	+0.13%
Lowest price (Asia session)22 Sept 2026, 00:09	17.19290	−0.13%	—
Highest price (New York session)24 Sept 2026, 18:05	17.73158	+3.00%	+3.13%
Close price before Banxico rate decisionWed 23 Sept 2026, 20:59, close	17.51110	+1.71%	+1.85%
After Banxico rate decision (24 Sep 2026, 19:00 UTC)
Price after Banxico rate decision (New York session)24 Sept 2026, 19:00	17.67300	+2.65%	+2.79%	+0.92%
Lowest price (New York session)24 Sept 2026, 19:00	17.66700	+2.62%	+2.76%	+0.89%
Highest price (New York session)24 Sept 2026, 20:27	17.76749	+3.20%	+3.34%	+1.46%
Close price after Banxico rate decisionThu 24 Sept 2026, 20:59, close	17.72177	+2.94%	+3.08%	+1.20%
Current price	—	—	—	—

Recorded after the event and not edited since. Past moves are not a forecast of future ones.

PassedWed, 30 Sept 2026, 23:50

USDJPYUS Dollar / Japanese Yen

The Tankan hit an eight year high and the yen fell anyway, because firms expect less inflation rather than more

Positive

157.41

+30 to +60 pips, which is +0.19% to +0.38%, if the survey confirms the drop firms forecast in June

+0.66%High

−0.12%Low

USDJPY result

After the event it went to +0.66% at its highest, and later to −0.12% at its lowest, against the close price before the September Tankan and the Summary of Opinions.

The news wasMixed

A sixth straight quarterly improvement in large manufacturer sentiment and a Summary of Opinions in which members argued for accelerating the pace both make a further rise easier to justify, which is a reason to hold yen and so a reason for this pair to fall. Against that, firms cut their one year inflation expectations to 2.6% and their five year to 2.5%, large non-manufacturers fell for the first time in five quarters, and nothing in the Summary pointed at October, all of which let the Bank take its time. Both halves are real, which is why a survey at an eight year high sold the yen rather than bought it.

Measured fromChart

The September Tankan put large manufacturer sentiment at 24, its highest since March 2018, and firms cut their inflation expectations a year ahead to 2.6% from 2.7%. Published at the same minute, the Summary of Opinions showed a board arguing for more rises rather than one content to wait, but it gave no sign of a move in October, so bets on back to back rises came off. The yen fell through Tokyo and the London morning, and USDJPY closed the reaction window at 158.067, up 65.7 pips or 0.42% from the rollover before the release. Most of that was the dollar: EUR/USD fell 0.77% across the same window and the yen finished stronger against both the euro and sterling.

What else moved the price

The release itself was a clean yen move and the day was not. In the twenty minutes after 23:50 UTC, USD/JPY rose 24.5 pips, EUR/JPY 26.8, GBP/JPY 29.5 and AUD/JPY 14.4, while EUR/USD moved a tenth of a pip, so every yen cross went the same way and the dollar leg did nothing. That stayed true to the high at 07:28 UTC, by which point EUR/JPY was 79.3 pips and GBP/JPY 103.3 pips above their pre-release closes. From there the dollar took over. The United States manufacturing survey printed 54.5 against 55.0 expected at 14:00 UTC and the pair made the low of its whole window at 14:33, down 68.3 pips in half an hour, before the same survey prices paid component at 77.9 against 72.3 expected helped turn it round. Across the full window EUR/USD fell 0.77% to 1.12422, and the yen ended it 62.8 pips stronger against the euro and 21.7 pips stronger against sterling. So the 65.7 pips this card is measured on is mostly a dollar move with a Tankan move inside the first eight hours of it. Two further items sat on the same day. Japan final September manufacturing PMI was confirmed at 54.1, a six month low, forty minutes after the Tankan. And the Japanese prime minister said her policies would lift confidence in the yen, having told the American president its undervaluation was a problem, two months after the United States Treasury bought yen alongside Japan in an operation estimated at up to 26.3 billion dollars.

What we said	What happened	Result
Catalyst confidence 100 out of 100: the September Tankan and the Summary of Opinions from the meeting that raised rates to 1.25% are both published at 23:50 UTC on 30 September 2026, which is 08:50 on 1 October in Tokyo.	Both documents were published at that minute. The Tankan covered 9,104 firms with a 99.4% response rate over a responding period that ran from 26 August to 30 September. The first rate struck after the release was 157.45.	Correct
The large manufacturers index beats the 17 firms forecast for themselves in June, and the market reads the survey against that forecast rather than against the 22 the June survey actually printed.	The index came in at 24, up from 22, a sixth straight quarterly improvement and the highest since March 2018. That is seven points above the 17 firms had forecast and one point below the 25 the market expected. Midsize firms rose to 23 from 17 and small firms to 14 from 9. Large non-manufacturers went the other way, to 35 from 37, their first quarterly fall in five quarters and a point below the 36 forecast.	Correct on the headline figure
Expected direction positive, meaning a higher USDJPY and a weaker yen. Updated after the event to read that the yen fell by as much as 0.5% to 158.21 per dollar and was the weakest major currency of the day.	The direction is right and the description of it is not. Measured the way a direction call is settled, from the rollover before the release at 157.41 to the one after it at 158.067, the pair rose 65.7 pips or 0.42%. From the card signal rate of 157.369 it is 69.8 pips or 0.44% higher. The reaction high was 158.452 at 07:28 UTC on 1 October, 104.2 pips or 0.66% above the pre-release close, so the pair went further than the 158.21 the card named rather than stopping there. And the yen was not the weakest major of the day: it was the weakest through Tokyo and the London morning, but by the Thursday rollover EUR/JPY had fallen 62.8 pips and GBP/JPY 21.7 pips from their own pre-release closes, which is a yen stronger against both.	Direction correct, the figures and the ranking wrong
Our expected move: -40 to -100 pips, which is -0.25% to -0.64%, if large manufacturers come in near 20 or better and the Summary of Opinions shows members wanting to keep going, and +30 to +60 pips, which is +0.19% to +0.38%, if the survey confirms the drop firms forecast in June. The downside is the larger range because it is the direction the survey leans.	The conditions on the downside branch both fired and the pair went the other way. Large manufacturers came in at 24, well clear of "near 20 or better", and the Summary of Opinions did show members wanting to keep going: one said the Bank will need to accelerate the pace of rate rises if signs of an upward deviation in prices appear, another said it is desirable to bring the policy rate closer to its approximate goal relatively soon. On the card own rules that is -40 to -100 pips. The pair settled +65.7 pips, +0.42%. The branch whose conditions did not fire, the +30 to +60 pips one, is the branch the outcome landed in, and it overshot the top of it by 5.7 pips.	Incorrect. The right range was reached through the wrong trigger
Price-reaction confidence 88 out of 100, on the grounds that the reaction has already happened and both its size and its direction are known.	A reaction did happen and it was quick: the pair was 24.5 pips higher twenty minutes after the release and 33.7 pips higher after an hour, and it kept going to 158.452 by 07:28 UTC. But the figures the confidence was staked on, 0.5% and 158.21 and weakest major of the day, are not what the feed carries, so the confidence was high in a description that was partly wrong.	Partly correct
Action: Watch. The lean is toward a stronger yen, the Tankan rarely moves the pair much on its own, and a reader who wants the trade should wait for the two documents rather than guess the survey.	This is the best call on the card. The lean toward a stronger yen was wrong, and Watch is what kept that from mattering. A reader who waited for the documents saw the pair rise 24.5 pips in twenty minutes and had the whole Tokyo session to work out that the stated lean had failed.	Correct
How much is priced in: low. The pair rose 2.58% over the week before the card, which is the market pricing a slow Bank of Japan, and that leaves the yen positioned for weak evidence so a strong survey has room to move it.	The reasoning did not survive the nine days between the signal and the release. The pair reached 159.034 on 24 September, 1.06% above the signal rate of 157.369, then fell to 156.37 on the morning of 30 September, 0.63% below it, and closed the evening before the release at 157.41, four pips from where the card had quoted it. The week the card was written into was a round trip, not a trend, so there was no stretched position for a strong survey to unwind. The survey was strong and the pair moved 65.7 pips, which is about one ordinary day.	Incorrect reasoning, roughly right conclusion
Bullish scenario, meaning a weaker yen: the survey confirms the drop firms forecast in June, with large manufacturers near 17 and non-manufacturers near 28, the Summary of Opinions shows most members content to wait, and the pair tests the 50 day average at 158.11.	Every condition failed and the price target was met anyway. Large manufacturers came in at 24 rather than near 17, non-manufacturers at 35 rather than near 28, and the Summary of Opinions was hawkish rather than content to wait. The pair still tested the level: it reached 158.452 at 07:28 UTC, clearing 158.11 by 34 pips, and settled at 158.067, four pips below it.	Right target, wrong reasons
Neutral scenario, described as the most likely single outcome: the survey lands between the June actual and the June forecast, around 19 or 20 for manufacturers, the Summary of Opinions is balanced, and the pair moves by less than an ordinary day and goes back to following the dollar.	The survey was nowhere near 19 or 20 and the Summary was not balanced, so the premise failed. The second half of the sentence is the best single description of the day on the card. The settled move of 0.42% is 1.04 times the 0.400% average daily move the card quoted, which is an ordinary day almost exactly, and the pair did go back to following the dollar: EUR/USD fell 0.77% across the window while EUR/JPY and GBP/JPY both finished below their pre-release closes.	Premise wrong, outcome right
Bearish scenario, meaning a stronger yen: large manufacturers come in at 21 or better in line with the Reuters survey, firms inflation expectations tick up, the Summary of Opinions shows members wanting to continue, the market brings the next rise forward, and the pair falls to the 20 day average at 156.12045 and toward the one week low at 154.418.	Two of the three conditions fired and the pair did the opposite. Large manufacturers printed 24, comfortably at 21 or better, and the Summary of Opinions showed members wanting to continue. Only inflation expectations failed: firms put consumer prices a year ahead at 2.6%, down from 2.7%, with the three year reading unchanged at 2.6% and the five year down to 2.5% from 2.6%. The pair did not fall to 156.12045 or anywhere near it. The lowest rate struck in the whole reaction window was 157.217, at 14:33 UTC on 1 October, 19.3 pips below the pre-release close.	Incorrect
Result needed for a big move: a rise in firms inflation expectations together with a strong manufacturers index. Either alone moves the pair a little, and nothing else in the release can take the pair beyond the ranges above.	This held, and it is the sharpest piece of analysis on the card. The release delivered one half, a strong index, and withheld the other, with expectations a tenth lower. The pair moved a little: 65.7 pips on the settle, against a 0.400% average day of about 63 pips. Bloomberg Economics read the survey the same way, its chief Japan economist saying the results reduce the likelihood of consecutive rises in October, which is the mechanism the card had named in advance.	Correct
What else lands that day: crowded. The Summary of Opinions lands at the same minute and can point the other way, and later the same day the United States publishes its Institute for Supply Management manufacturing survey, which moves the dollar leg of this pair more reliably than the Tankan moves the yen. By the New York open the pair will be trading the dollar again.	All three held and the last one held to the minute. The Summary did point the other way, hawkish against a survey the market read as dovish. The ISM printed 54.5 against 55.0 expected at 14:00 UTC, and the pair made the low of its entire reaction window at 14:33 UTC, falling 68.3 pips from 157.9 at 13:59 to 157.217 inside that half hour. And the pair did go back to trading the dollar: from 14:33 UTC to the rollover EUR/USD fell another 42 pips to 1.12422 while the pair recovered to 158.067, so the whole New York recovery is a dollar move. The one item the card missed was Japanese: the final September manufacturing PMI, confirmed at 54.1 from 54.9, a six month low, published forty minutes after the Tankan.	Correct
Main risks medium. The largest is that a correct read on the survey produces no move. The Summary of Opinions may point the other way. The United States manufacturing survey can reverse the Tokyo move in minutes. Intervention talk from Tokyo with the pair close to 160 is a standing risk in the same direction as this card, which cuts the downside of being wrong but also makes attribution harder.	The first three were right, and the ISM risk was right to the half hour. The intervention line was the one that was understated rather than wrong. It is written as talk at a level the pair had not reached, when the United States Treasury had actually bought yen alongside Japan in August 2026, an operation estimated at up to 26.3 billion dollars. On 1 October the Japanese prime minister said her policies would lift confidence in the yen and that she had told the American president the currency undervaluation was a problem, and the pair rose anyway. That is a live driver pointing against this card on the day of its own event, not a standing risk at 160.	Mostly correct, intervention understated
What the market may be missing: the exchange rate assumption. Large firms planned their fiscal year around 152.57 yen per dollar in June and the pair has traded weaker than that throughout, which lifts exporter profits and is part of why the survey came in high. Second, the dissent: a rise in firms inflation expectations would answer the two members who said the economy was not strong enough, most directly.	Both halves worked, one as a forecast and one as a diagnosis. The September survey has firms assuming about 154 yen to the dollar for the fiscal year, and the pair traded between 156.37 and 159.034 over the nine days before the release, weaker than the assumption throughout, which is the exporter tailwind the card described. The dissent point the card had already conceded went the way it conceded: expectations fell rather than rose, so sentiment improved while the inflation case weakened, and that combination is what let the board wait.	Correct
Technical bias bearish. Price is below three of the four moving averages, the 50 day at 158.11 sits under both the 200 day at 158.77392 and the 100 day at 159.7521, and the week bounce is heading straight into that cluster. A daily close above 159.7521 would turn the chart bullish. Support is the 20 day at 156.12045, then 154.418 and 153.411.	The bias was wrong and its own invalidation level was never reached, which is the worst combination to be in. The pair rose across the window and closed it at 158.067, above the 20 day and the signal rate. It also went well past the cluster the card expected to cap it: it reached 159.034 on 24 September, through both the 50 day and the 200 day, and the reaction high of 158.452 cleared the 50 day again. The invalidating close above 159.7521 never came, so the chart read was never formally beaten, and it was still pointing the wrong way for nine days. On the support side the 20 day at 156.12045 held by 25 pips at the 30 September low of 156.37, and 154.418 and 153.411 were never approached.	Incorrect
Impact medium, on an average daily move of 0.400%. The downside range is 0.6 to 1.6 times an ordinary day and the upside range 0.5 to 1 times one. Estimated duration 1 to 3 days: the survey is absorbed within the Tokyo session, but a change in how soon the market expects the next rise carries through to the end of the week.	Medium was the right size. The settled move of 0.42% is 1.04 times the quoted average day, and the high to low range of the reaction window, 158.452 down to 157.217, is 123.5 pips or 1.96 times one. On duration the first half is wrong and the second is right so far: the move was not absorbed within the Tokyo session, it extended into the London morning, where the high was set at 07:28 UTC. The last rate the feed carries, 157.942 at 23:59 UTC on 1 October, is three hours past the rollover and still 53.2 pips above the pre-release close, so most of the move was holding at the point this result was written.	Impact correct, duration partly correct
The Summary of Opinions showed a board broadly content to wait, which is what the market had already assumed. That reading is what the card gives as the reason a strong survey sold the yen.	The document says the opposite. Most of the opinions argued for following September rise with further increases, one member said the Bank will need to accelerate the pace if prices deviate upward, and another said it is desirable to bring the policy rate closer to its approximate goal relatively soon. The pushback in the room came from the government side, where a Cabinet Office representative urged the Bank to examine the cumulative effects of past rises. What damped the market was narrower than a board content to wait: the Summary gave no sign of a move in October specifically, so bets on back to back rises came off while the case for more rises stayed intact.	Incorrect

Price Table	Rate (USDJPY)	Change from signal	Change from lowest (regular session)	Change from close before the September Tankan and the Summary of Opinions
After signal
Price at analysis (Asia session)21 Sept 2026, 23:59	157.369	—	+0.64%
Sat 26, Sun 27 - no trading
Lowest price (Asia session)30 Sept 2026, 02:28	156.370	−0.63%	—
Highest price (New York session)24 Sept 2026, 15:34	159.034	+1.06%	+1.70%
Sat 26, Sun 27 - no trading
Close price before the September Tankan and the Summary of OpinionsWed 30 Sept 2026, 20:59, close	157.410	+0.03%	+0.67%
After the September Tankan and the Summary of Opinions (30 Sept 2026, 23:50 UTC)
Price after the September Tankan and the Summary of Opinions (Asia session)30 Sept 2026, 23:50	157.450	+0.05%	+0.69%	+0.03%
Lowest price (New York session)01 Oct 2026, 14:33	157.217	−0.10%	+0.54%	−0.12%
Highest price (London session)01 Oct 2026, 07:28	158.452	+0.69%	+1.33%	+0.66%
Close price after the September Tankan and the Summary of OpinionsThu 01 Oct 2026, 20:59, close	158.067	+0.44%	+1.09%	+0.42%
Current price	—	—	—	—

Recorded after the event and not edited since. Past moves are not a forecast of future ones.

PassedThu, 17 Sept 2026, 11:00

GBPUSDBritish Pound / US Dollar

Bank of England holds, and the vote split is the only part nobody has priced

Mixed

1.3378

+60 to +110 pips, which is +0.44% to +0.81%, on a fourth hawkish vote, and -50 to -90 pips, which is -0.37% to -0.67%, on a softer split

+0.20%High

−0.32%Low

GBPUSD result

After the event it went to +0.20% at its highest, and later to −0.32% at its lowest, against the close price before the BoE decision.

The news wasPositive

Measured fromChart

The Bank of England held Bank Rate at 3.75% and the vote was unchanged at six to three, with the same three members wanting 4%. The summary was tougher than July, warning that policy may have to tighten if the energy shock spreads, but sterling fell anyway because traders had been hoping for a fourth vote.

What else moved the price

Both legs of this pair were repriced the day before the Bank spoke. UK inflation for August was published on the morning of 16 September at 3.1%, up from 2.9%, and the Federal Reserve raised its target range to 3.75% to 4.00% that evening, so the pair arrived at 1.33780, already 147 pips below the signal rate. The decision itself was sterling and not the dollar: in the twenty minutes after it GBP/USD fell 52 pips while EUR/USD fell 6, and EUR/GBP rose 30 pips. Over the whole day EUR/USD finished higher, so the dollar is not what took this pair to a July low.

What we said	What happened	Result
The Bank of England publishes its decision, summary and minutes together at 11:00 UTC on 17 September 2026, and is expected to hold Bank Rate at 3.75%. Confidence in the schedule is 99 out of 100.	Published at 11:00 UTC as scheduled, with the summary and minutes released alongside the decision. Bank Rate was held at 3.75%.	Correct
The decision itself is not the event. The event is the vote: three of the nine members already want 4.00%, and whether that number grows is what the market is actually trading.	The vote was the event and it did not move. The Committee split six to three again, with Huw Pill, Megan Greene and Catherine Mann the same three preferring 4.00%, and no member voting for a cut. The market read the count within seconds and sterling fell on it.	Correct
Mixed. A fourth member joining the three who want 4.00% is a hawkish surprise and lifts sterling. A retreat to seven or eight votes for no change, or softer wording about the inflation path, sends it the other way.	Neither trigger fired. The count was unchanged and the wording hardened rather than softened: the summary said policy may have to tighten if second-round effects from the energy shock crystallise, and noted tentative signs that the loosening of the labour market had bottomed out. The pair fell anyway, finishing 21 pips or 0.16% below the rollover before the decision, so the vote count did not decide the direction the way the card said it would.	Partly correct
Our expected move: +60 to +110 pips, which is +0.44% to +0.81%, if a fourth member votes to raise, and -50 to -90 pips, which is -0.37% to -0.67%, on a softer split or softer wording.	Both ranges were conditional on a change in the vote or the language, and neither happened, so neither range was ever a forecast about this outcome. What the unchanged split produced was 55 pips, or 0.41%, of decline inside twenty minutes, from 1.34045 in the announcing minute to 1.33490, and 44 pips of that was back by 12:30 UTC.	Neither range triggered
Neutral scenario: the Committee holds six to three with familiar language, the minutes repeat the balance of risks, and the pair returns to trading the dollar. Given how compressed volatility is, this outcome still produces a short sharp move as the positions taken before the decision are closed.	This is the scenario that occurred, and it got the first half right. The hold came six to three and the short sharp move happened exactly as described. The pair did not then return to trading the dollar: sterling kept sliding on its own to 1.33350 at 15:47 UTC and closed the day at 1.33570, while EUR/USD finished the same day higher than it started.	Partly correct
Price-reaction confidence 62 out of 100, with the uncertainty being that the dollar leg was repriced the night before by a larger central bank, and a strong reaction to the Federal Reserve can leave this pair trading the previous day for another session.	The risk the card was worried about is not what happened. In the twenty minutes after the decision GBP/USD fell 52 pips while EUR/USD fell 6, and EUR/GBP rose 30 pips, so the move was sterling rather than the dollar. The pair moved on its own leg, for the reason the card doubted it would.	Correct
Main risks are medium, and the main one is that nothing happens: a fully expected hold with an unchanged split is a non event, and compressed volatility means the move can be over in minutes.	Half of that is what occurred. The hold and the split were both unchanged, and the first move was mostly retraced within ninety minutes. It was not a non event: the pair went on to a July low in the afternoon and closed lower on the day.	Partly correct
What the market may be missing: the two decisions are one trade. The Federal Reserve is expected to take the US range to 3.75% to 4.00%, above Bank Rate for the first time in this cycle, and the Bank of England has one morning to answer it.	That is what settled the week. The Fed raised to 3.75% to 4.00% on 16 September, the Bank of England held the next morning, and the pair was already at 1.33780 before the Bank spoke, with 147 of its 168 pip fall from the signal rate gone on the crossing alone.	Correct
The second thing is the volatility. Realised volatility is 3.93% against 6.95% in the preceding window, the average daily move is about 21 pips, and moves out of compressed volatility are usually larger than the recent range implies.	Correct, and by a wide margin. The reaction window ran from 1.34045 to 1.33350, a range of 70 pips, and 55 of them came in the first twenty minutes, against an ordinary day of about 21 pips.	Correct
Technical bias neutral. A daily close below 1.34823, the low of the month, turns it bearish and would mean the rate gap opened by the Federal Reserve was being priced without an answer from the Bank of England.	That is the reading the chart gave and the meaning the card attached to it. The level went on the Fed rather than on the Bank: the pair closed at 1.33780 on 16 September, already below the month low, and at 1.33570 after the Bank held.	Correct

Price Table	Rate (GBPUSD)	Change from signal	Change from lowest (regular session)	Change from close before the BoE decision
After signal
Price at analysis (Asia session)13 Sept 2026, 23:59	1.35251	—	+1.18%
Lowest price (Asia session)17 Sept 2026, 02:56	1.33670	−1.17%	—
Highest price (Asia session)14 Sept 2026, 00:09	1.35282	+0.02%	+1.21%
Close price before the BoE decisionWed 16 Sept 2026, 20:59, close	1.33780	−1.09%	+0.08%
After the BoE decision (17 Sept 2026, 11:00 UTC)
Price after the BoE decision (London session)17 Sept 2026, 11:00	1.33910	−0.99%	+0.18%	+0.10%
Lowest price (New York session)17 Sept 2026, 15:47	1.33350	−1.41%	−0.24%	−0.32%
Highest price (London session)17 Sept 2026, 11:00	1.34045	−0.89%	+0.28%	+0.20%
Close price after the BoE decisionThu 17 Sept 2026, 20:59, close	1.33570	−1.24%	−0.07%	−0.16%
Current price (New York session)18 Sept 2026, 20:59	1.33950	−0.96%	+0.21%	+0.13%

Recorded after the event and not edited since. Past moves are not a forecast of future ones.

PassedFri, 18 Sept 2026

USDJPYUS dollar against the Japanese yen

Bank of Japan September rate decision, a quarter point rise to 1.25% is 84% priced after the yen surge

Positive

155.97

+60 to +140 pips, which is +0.39% to +0.91%

+0.11%Low

+1.34%High

USDJPY result

After the event it went to +0.11% at its lowest, and later to +1.34% at its highest, against the close price before the BoJ decision.

The news wasMixed

The rise to 1.25% was delivered and the statement committed the Bank to keep raising rates, which is a reason to hold yen and therefore a reason for this pair to fall. Against that, two members voted to hold, August core inflation had just printed at 1.7% and has now been under the 2% target for eight months, and Ueda refused to name a pace or an interval, all of which favour the dollar side. Both halves are real, which is why the decision reads as a rise nobody has to buy the next one on.

Measured fromChart

The Bank of Japan raised its policy rate to 1.25%, the highest in 31 years, but two of the nine board members voted to hold and Governor Ueda said there is no set interval between rises. Traders who had bought the yen for a faster tightening cycle sold it back, so the yen fell on the day its central bank raised rates.

What else moved the price

The card called the day crowded and named the two things that would compete with the decision, and it was half right about them. Japan's August inflation at 23:30 UTC was a non event for the rate, worth 23 pips in the hour after it, but quarterly expiry Friday did show up: the whole New York afternoon went the other way and took 140 pips off the high before the week rolled. The larger competing event was the one the card flagged separately. The Federal Reserve raised its range on 16 September and that was a dollar move, worth 119 pips to this pair while EUR/USD fell 82 pips, and the pair therefore arrived at the Bank of Japan already 190 pips above where it had closed the Friday before. The decision itself was a yen move rather than a dollar one: in the twenty minutes after it EUR/JPY rose 88 pips and GBP/JPY 103, while EUR/USD moved under four pips, so the dollar leg explains almost none of the reaction.

What we said	What happened	Result
The Bank of Japan holds a two day meeting on 17 and 18 September 2026 and decides on the second day whether to raise the uncollateralised overnight call rate from 1.00% to 1.25%. Confidence in the schedule is 99 out of 100.	The meeting ran to schedule and the Bank raised the rate to 1.25% on 18 September, effective 24 September. That is the highest Japanese policy rate since 1995.	Correct
The decision itself is not the whole event. One member voted for 1.25% at the 31 July meeting, and the question is whether the rest of the board joins that member.	They joined. The board voted seven to two to raise, and the two who dissented did so the other way: Toyoichiro Asada and Ayano Sato both wanted to hold, Asada because inflation excluding fresh food was still under 2% and the economy could not be called strong, Sato because conditions had not accelerated enough to justify the timing. The count is what the market traded, and it traded it against the yen.	Correct
Overnight index swaps price roughly 84% for a quarter point rise, having risen from about 75% to 80% at the start of September, and the market base case is therefore that USD/JPY falls.	The swaps were right about the decision and the base case was wrong about the pair. The rise arrived as priced and the pair went up, not down: it finished the day at 156.873 against 155.968 at the rollover before the decision, a rise of 91 pips or 0.58%.	Partly correct
Expected direction positive, meaning USD/JPY rises, against a market base case that it falls.	Correct, and against the consensus. Measured the way a direction call is settled, from the rollover before the news at 155.968 to the one after it at 156.873, the pair rose 91 pips or 0.58%, which is the dollar stronger and the yen weaker on the day its central bank raised rates. From the card signal rate of 154.37 the pair is 250 pips or 1.62% higher.	Correct
Our expected move: +60 to +140 pips, which is +0.39% to +0.91%.	Underestimated on the reaction and right on the settle. From the rollover before the decision the pair reached 158.065 at 09:45 UTC, a rise of 210 pips or 1.34%, which overshoots the top of the range by half as much again. It did not hold: the pair gave back 119 pips into the New York afternoon and rolled the week at 156.873, up 91 pips or 0.58%, which is inside the range.	Underestimated
The Bank publishes no release time in advance. At recent meetings the decision has arrived between 03:00 and 06:00 UTC, with the press conference at about 06:30 UTC.	The press conference began at 06:30 UTC, which is 15:30 in Tokyo, exactly as the card said. The decision itself came earlier than the window the card gave: the pair was flat inside fifteen pips all morning and then travelled 47 pips inside the 02:54 UTC minute, six minutes before the earliest hour named.	Partly correct
Japan's August inflation is published at 23:30 UTC on 17 September, hours before the decision. Core inflation excluding fresh food at or above 2.0%, against 1.8% in July, would make the hawkish version much easier for the board to write, and a soft print gives the doves a line to use.	The print was soft. Core inflation excluding fresh food came in at 1.7% against a 1.8% consensus and 1.8% in July, its eighth month under the 2% target, with headline inflation at 1.9%. The doves used it: Asada cited exactly that measure in his dissent. The pair took it quietly, drifting from 155.939 at 23:00 UTC to 156.171 by midnight, a rise of 23 pips.	Correct
Bullish scenario: the Bank raises to 1.25% and frames it as a step rather than the start of a run, the Governor declines to point at the next meeting, and the pair recovers toward the 155.30 to 156.00 area as the short dollar position built over the last six sessions is closed. A weak Japanese inflation print that morning makes this outcome more likely.	This is the scenario that happened and the card read the event right. The rise came with no pace attached: Ueda said there is no set interval between rises and that the possibilities depend on prices, and the Bank published no new forecasts, its next Outlook Report being due in October. The soft inflation print arrived as the card said it would. What the scenario got wrong is where the pair would be. The recovery zone it named had already been passed before the Bank spoke, and the reaction carried the pair to 158.065 rather than to 156.00.	Correct scenario
Main risks are High, and the main one runs against the call: reporting says a faster pace is under discussion, and if that is confirmed the pair does not bounce, it breaks. The Federal Reserve decision on 16 September changes the dollar leg first, and a surprise there is a bigger move in this pair than the Bank of Japan is likely to produce.	Both readings held. No faster pace was confirmed and the pair did not break. The Fed did move first and it was the larger of the two: the pair rose 119 pips or 0.77% on 16 September, from 155.098 to 156.289, against the 91 pips the Bank of Japan produced two days later.	Correct
Liquidity around the announcement is thin in the way that matters. The decision often lands while London is asleep and New York is closed, so the first move happens on the fewest participants of the day and is frequently reversed once Europe arrives.	The opposite happened. Europe extended the move rather than reversing it: the pair was at 157.079 an hour after the decision and made its high of 158.065 at 09:45 UTC, in the London morning. The reversal came later and from New York, which took the pair back to 156.667 by 17:59 UTC on a quarterly expiry Friday.	Incorrect
What the market may be missing: a rise to 1.25% barely touches the rate gap. The Federal Reserve midpoint is 3.625% and a Japanese rise takes the gap from 2.625 to 2.375 percentage points, and if the Federal Reserve also raises on 16 September the gap returns to 2.625 and nothing has changed at all.	That is exactly what happened, and the card was alone in saying it. The Federal Reserve raised its range to 3.75% to 4.00% on 16 September, a midpoint of 3.875%, and the Bank of Japan went to 1.25% two days later. The gap is 2.625 percentage points, the same figure it was before either central bank met, after the largest week of policy movement of the year.	Correct
Technical bias bearish, disagreeing with the positive direction call. A close above 156.00 would be the first sign this fall is over, and if the decision passes without a change of pace and the pair still cannot close above 156.00, the chart is right and this call is wrong.	The chart lost, on the test the card set itself. The pair closed above 156.00 on 16 September at 156.289 and again after the decision at 156.873, and it never went near the 152.88 low the bearish case needed. The lowest rate traded after the signal was 153.12, at the Sunday reopening on 13 September, and the pair spent the whole of the following week above it.	Incorrect

Price Table	Rate (USDJPY)	Change from signal	Change from lowest (regular session)	Change from close before the BoJ decision
After signal
Price at analysis (New York session)10 Sept 2026, 18:29	154.370	—	+0.82%
Sat 12 - no trading
Lowest price (Asia session)13 Sept 2026, 21:00	153.120	−0.81%	—
Highest price (New York session)16 Sept 2026, 20:27	156.415	+1.32%	+2.15%
Close price before the BoJ decisionThu 17 Sept 2026, 20:59, close	155.968	+1.04%	+1.86%
After the BoJ decision (18 Sept 2026, 02:54 UTC)
Price after the BoJ decision (Asia session)18 Sept 2026, 02:54	156.586	+1.44%	+2.26%	+0.40%
Lowest price (Asia session)18 Sept 2026, 02:54	156.133	+1.14%	+1.97%	+0.11%
Highest price (London session)18 Sept 2026, 09:45	158.065	+2.39%	+3.23%	+1.34%
Close price after the BoJ decisionFri 18 Sept 2026, 20:59, close	156.873	+1.62%	+2.45%	+0.58%
Current price (New York session)18 Sept 2026, 20:59	156.873	+1.62%	+2.45%	+0.58%

Recorded after the event and not edited since. Past moves are not a forecast of future ones.

PassedFri, 11 Sept 2026, 12:30

USDJPYUS dollar against the Japanese yen

United States August consumer price inflation report, the final inflation test before the Federal Reserve decision

Negative

154.43

-70 to -140 pips, which is -0.45% to -0.90%

+0.04%High

−0.78%Low

USDJPY result

After the event it went to +0.04% at its highest, and later to −0.78% at its lowest, against the close price before CPI release.

The news wasMixed

Monthly core inflation ran hot at 0.3% against the 0.2% expected, which favours the dollar and so the pair. Against that, the headline and the yearly core both landed in line, and no published scenario was satisfied.

Measured fromChart

Consumer price inflation rose 0.4% on the month and 3.4% on the year, as expected, while underlying inflation rose 0.3% on the month, hotter than the 0.2% forecast, and 2.4% on the year, as expected. July's figures were not revised, and the hotter monthly reading raised expectations for another Federal Reserve rate increase even though the dollar's initial gains reversed.

What we said	What happened	Result
Core CPI would be near 0.2% month on month and 2.4% year on year.	Core CPI rose 0.3% month on month and 2.4% year on year.	Partly correct
USDJPY would fall as cooling inflation reduced Federal Reserve tightening pressure.	USDJPY fell from 154.424 at the pre-event rollover to 153.527 at the next rollover. The pair fell, meaning the dollar weakened and the yen strengthened.	Correct
Expectations were about 0.2% monthly and 2.4% annual core CPI, with headline inflation near 0.4% monthly.	The consensus was 0.2% monthly and 2.4% annual core CPI, and 0.4% monthly and 3.4% annual headline CPI. Headline matched, annual core matched, and monthly core was 0.1 percentage point hotter.	Partly correct
July headline CPI was 0.1% monthly and 3.4% yearly, while core CPI was 0.2% monthly and 2.5% yearly.	The August release retained those July comparison figures and reported no revision to them.	Correct
USDJPY would fall 70 to 140 pips within the reaction window.	The largest verified post-release decline from the pre-event rollover was 114.6 pips, or 0.74%. The pair closed 89.7 pips, or 0.58%, lower.	Correct
The main move would occur within hours to one day.	The decline developed during the New York session and remained in place at the first rollover, eight and a half hours after the release.	Correct
The published scenarios required at least 0.3% monthly and 2.5% annual core CPI for the bullish case, 0.2% monthly core for neutral, or 0.1% or less for bearish.	Core CPI was 0.3% monthly but 2.4% yearly. This mixed result did not meet all conditions of any published scenario, and USDJPY finished lower.	No scenario matched
Energy could lift headline inflation, and the initial CPI reaction could fade during the US session.	Gasoline rose 3.9% and accounted for more than one third of the headline monthly increase. Treasury yields and the dollar initially rose, then reversed as USDJPY ended lower.	Correct
A soft core result could reverse post-payroll Federal Reserve rate repricing.	Core CPI was hotter than expected at 0.3% monthly. Federal Reserve hike odds rose from about 68% before the report to above 80% afterward.	Not triggered

Price Table	Rate (USDJPY)	Change from signal	Change from lowest (regular session)	Change from close before CPI release
After signal
Price at analysis (New York session)6 Sep 2026, 18:38	156.246	—	+2.20%
Lowest price (Asia session)08 Sept 2026, 02:24	152.880	−2.15%	—
Highest price (Asia session)06 Sept 2026, 22:41	156.283	+0.02%	+2.23%
Close price before CPI releaseThu 10 Sept 2026, 20:59, close	154.426	−1.16%	+1.01%
After CPI release (11 Sep 2026, 12:30 UTC)
Price after CPI release (New York session)11 Sept 2026, 12:30	154.404	−1.18%	+1.00%	−0.01%
Lowest price (New York session)11 Sept 2026, 13:03	153.220	−1.94%	+0.22%	−0.78%
Highest price (New York session)11 Sept 2026, 12:31	154.485	−1.13%	+1.05%	+0.04%
Close price after CPI releaseFri 11 Sep 2026, 21:00, close	153.527	−1.74%	+0.42%	−0.58%
Current price (New York session)12 Sep 2026, 18:52	153.527	−1.74%	+0.42%	−0.58%

Recorded after the event and not edited since. Past moves are not a forecast of future ones.

PassedWed, 16 Sept 2026, 18:00

EURUSDEuro against the US dollar

United States Federal Reserve rate decision and economic projections

Negative

1.1543

-60 to -110 pips, which is -0.52% to -0.95%

+0.03%High

−0.72%Low

EURUSD result

After the event it went to +0.03% at its highest, and later to −0.72% at its lowest, against the close price before the Fed decision.

The news wasNegative

Measured fromChart

The Federal Reserve raised its target range to 3.75% to 4.00%, its first increase since 2023, and the vote was unanimous. Its new projections showed most officials expecting at least one more rise this year, and the dollar strengthened against the euro.

What else moved the price

US advance retail sales for August landed at 12:30 UTC the same day, 1.2% against the 0.8% expected, five and a half hours before the decision. The pair was about 5 pips higher a quarter of an hour later, so the beat is not what moved it. The Fed window itself was a broad dollar move rather than anything on the euro side: between the last rate before the decision and the rollover after it, GBP/USD fell 70 pips, USD/JPY rose 130 pips and USD/CHF rose 76 pips. A reader crediting this move to the euro leg would have it the wrong way round.

What we said	What happened	Result
The Federal Reserve decision, projections and press conference happen on 16 September 2026 at 18:00 UTC, with 99% confidence in the schedule and about a 60% probability of a 25 basis point increase, against the 58.4% the market priced after the August employment report.	The FOMC announced on time at 18:00 UTC and raised the target range by 25 basis points to 3.75% to 4.00%, its first increase since 2023. The vote was unanimous at 12 to 0. Futures pricing had firmed to roughly 60% by the meeting, so both the schedule and the probability were read correctly.	Correct
Negative. EUR/USD falls, meaning the euro weakens against the US dollar, because the Fed is slightly more likely to raise while the 25 basis point ECB increase on 10 September is already heavily priced.	EUR/USD fell, which is the dollar stronger. The pair went from 1.15430 at the rollover before the decision to 1.14610 at the rollover after it, 82 pips or 0.71% lower, and it sits 152 pips or 1.31% below the signal rate of 1.16130.	Correct
Priced in: medium. A Fed increase is slightly more than half priced while the ECB increase is roughly 89% priced, so the dollar has far more room to reprice than the euro does.	That is how it resolved. The ECB raised its deposit rate to 2.50% on 10 September and the euro took no lasting support from it, while the Fed increase six days later still moved the pair. The new projections put the median year-end rate at 4.1% for both 2026 and 2027, against 3.8% and 3.6% in June, which is the repricing the card said was available.	Correct
Our expected move: -60 to -110 pips, which is -0.52% to -0.95%, if the Fed raises rates and keeps further tightening possible.	The reaction to the decision was 82 pips, or 0.71%, from the rollover before it to the rollover after it, which is inside the range. The largest fall within the window was 87 pips, from 1.15470 in the announcing minute to 1.14600 at 20:25 UTC.	Correct
Bearish scenario: EUR/USD would probably fall if the Fed raises the target to 3.75% to 4.00% and keeps another increase possible, preserving the US rate advantage even after the expected ECB hike.	That is the scenario that occurred, on both halves of it. The Fed raised to 3.75% to 4.00% and 16 of the 18 participants pencilled in at least one further increase, four of them two, so the door was left open exactly as the card described, and the pair fell.	Correct scenario
Estimated duration 1 to 3 days. A one-time hike could fade within a session, while a hike combined with projections showing further tightening could begin a multi-day repricing of US yields and the dollar.	It did not fade. The pair finished the announcement day at 1.14610, a pip off the low of the window, and the last rate of the week on 18 September was 1.14840, still 129 pips or 1.11% below the signal rate.	Correct
Main risks are high, and US CPI on 11 September could change the rate probability sharply: a monthly core reading at or above 0.3% would strengthen the case for a hike, while 0.1% or lower would support a hold.	Core CPI rose 0.3% in August against the 0.2% expected, published on 11 September, and hike pricing firmed from there into the meeting. Headline CPI was 0.4% on the month and 3.4% over the year, with gasoline up 3.9% and worth over a third of the monthly increase.	Correct
What the market may be missing: a Fed hike is more credible than the calm EUR/USD price structure suggests, with three July votes for one already on the record, although wage growth of 3.1% gives officials a reason to wait.	The committee went further than the card argued. The increase was unanimous at 12 to 0 rather than carried by a narrow majority, and the slower wage growth held nobody back.	Correct
Technical bias neutral, which disagrees with the negative catalyst call, and a decisive daily close below 1.15650 would confirm a breakdown and open 1.15000.	The decision settled that disagreement the way the catalyst call wanted. The pair was already under the level going in, at 1.15430 on 15 September, and the Fed took it through 1.15000 to 1.14610.	Correct

Price Table	Rate (EURUSD)	Change from signal	Change from lowest (regular session)	Change from close before the Fed decision
After signal
Price at analysis (New York session)04 Sept 2026, 20:59	1.16130	—	+0.79%
4 days with no trading
Lowest price (New York session)14 Sept 2026, 13:57	1.15220	−0.78%	—
Highest price (New York session)09 Sept 2026, 12:11	1.16540	+0.35%	+1.15%
Sat 12, Sun 13 - no trading
Close price before the Fed decisionTue 15 Sept 2026, 20:59, close	1.15430	−0.60%	+0.18%
After the Fed decision (16 Sept 2026, 18:00 UTC)
Price after the Fed decision (New York session)16 Sept 2026, 18:00	1.15259	−0.75%	+0.03%	−0.15%
Lowest price (New York session)16 Sept 2026, 20:25	1.14600	−1.32%	−0.54%	−0.72%
Highest price (New York session)16 Sept 2026, 18:00	1.15470	−0.57%	+0.22%	+0.03%
Close price after the Fed decisionWed 16 Sept 2026, 20:59, close	1.14610	−1.31%	−0.53%	−0.71%
Current price (New York session)18 Sept 2026, 20:59	1.14840	−1.11%	−0.33%	−0.51%

Recorded after the event and not edited since. Past moves are not a forecast of future ones.

PassedFri, 2 Oct 2026, 12:30

USDCADUS Dollar / Canadian Dollar

The September jobs report tests a labour market resting on one strong month

Negative

1.4221

-50 to -90 pips, which is -0.36% to -0.65%, on a figure at or below the twelve month trend

−0.16%Low

+0.28%High

USDCAD result

After the event it went to −0.16% at its lowest, and later to +0.28% at its highest, against the close price before the jobs report.

The news wasNegative

Measured fromChart

September payrolls rose by just 29,000, unemployment edged up to 4.2%, wage growth slowed and July and August were revised down by a combined 60,000. The report weakened the dollar at first, but the Canadian dollar was weaker still as oil fell and the pair recovered the whole move within an hour.

What else moved the price

There was no competing Canadian release, but the Canadian dollar entered Friday near an 18-month low after eight losing sessions. Brent fell about 2.7% and West Texas Intermediate about 3.7%, while the US-Canada interest-rate spread and concerns about Canadian growth kept weighing on the loonie; those forces overrode the report's initial dollar-negative reaction.

What we said	What happened	Result
Catalyst confidence 99 out of 100: the Bureau of Labor Statistics publishes the September employment report at 12:30 UTC on 2 October.	The Bureau of Labor Statistics published the report at 12:30 UTC on 2 October as scheduled. Payrolls rose by 29,000 and unemployment rose to 4.2%.	Correct
Expected direction mixed: the direction could not be known before the release, but a figure at or below the 12-month trend would weaken the dollar and lower USDCAD.	The report was clearly dollar-negative: 29,000 jobs was below both the card's 31,000 trend threshold and the 90,000 Reuters consensus, unemployment rose, wages slowed and both prior months were revised down. USDCAD fell first, then reversed and finished the reaction window higher, meaning the Canadian dollar weakened more than the US dollar.	Data read correct, settled direction wrong
Market expectations were neutral and clustered near 50,000, close to the 12-month trend rather than August's 162,000.	The direction of the consensus was right, a sharp slowdown from August, but its level was materially understated. Economists polled by Reuters expected 90,000, while a separate Dow Jones survey put the consensus at 84,000. The actual 29,000 was below every estimate in the survey cited by CIBC.	Partly correct
The key question was whether August's 162,000 was an exception to a weak trend, and a large downward revision could reverse the meaning of a decent September headline.	August was revised down by 29,000 to 133,000 and July by 31,000 from a gain of 21,000 to a loss of 10,000. September then added only 29,000, leaving July and August together 60,000 lower than first reported and confirming that August had overstated the underlying pace.	Correct
A figure at or below the 12-month trend would produce a fall of 50 to 90 pips, or 0.36% to 0.65%, in USDCAD.	The condition fired at 29,000, but the forecast move did not. The pair's post-release low was 23 pips below the pre-release rollover, and by the Friday rollover it was about 30 pips, or 0.21%, higher. The full reaction range was 63 pips, but it finished in the opposite direction.	Incorrect
The bearish scenario for USDCAD was payrolls at or below 31,000, a sharp downward August revision, or unemployment rising from 4.1%; any of those would make the dollar give back part of its September gain.	All three parts fired: payrolls were 29,000, August was revised from 162,000 to 133,000 and unemployment rose to 4.2%. USDCAD dipped immediately but recovered the entire fall within the hour and finished higher, so the scenario read the release correctly and the pair incorrectly.	Conditions correct, rate reaction wrong
A big move needed payrolls outside zero to 100,000 or an August revision larger than 50,000; inside those bounds the reaction should fade within the session.	September payrolls stayed inside that band at 29,000 and the August revision was 29,000, although the two-month revision was 60,000. The first dollar-negative move did fade within an hour, and the pair's 63-pip reaction range stayed inside the 90-pip ceiling.	Correct
The day was clear for the pair because no central bank met and Canada's employment report would not arrive until 9 October, leaving only the dollar leg to move.	There was no Canadian data release or central-bank decision, and Canada's jobs report remained due on 9 October. The claim that only the dollar leg would move was too strong: oil fell sharply and domestic growth and Bank of Canada expectations were already weighing on the Canadian dollar.	Calendar correct, market read incomplete
The main risks included elevated oil supporting the Canadian dollar, while the Canada-US tariff dispute could weaken it independently of payrolls.	The commodity risk mattered but changed sign. Brent fell about 2.7% and West Texas Intermediate about 3.7% on Friday, removing support from the Canadian dollar. The loonie entered the release after eight losing sessions, with Canadian growth concerns and a wide US-Canada rate gap already working against it.	Correct risk, wrong oil direction
The technical bias was neutral inside 1.37539 to 1.39406, and a daily close above 1.39406 would turn it bullish.	The neutral setup had already failed well before the catalyst. USDCAD broke above 1.39406 during the signal window, reached 1.42624 before the report and rolled over the day before it at 1.42210. The card arrived at the event after a bullish break rather than inside its old range.	Bias invalidated before the event
Impact high and duration one to three days, with most of the move in the first five minutes and an initial overshoot likely to retrace as the report was read.	The shape was right but the scale was not high. USDCAD fell on the release and recovered the entire move within an hour, but its 63-pip high-to-low range was modest beside the large payroll surprise. Persistence beyond Friday cannot yet be measured because the market is closed for the weekend.	Timing correct, impact overstated

Price Table	Rate (USDCAD)	Change from signal	Change from lowest (regular session)	Change from close before the jobs report
After signal
Price at analysis (Asia session)13 Sep 2026, 23:59	1.38679	—	+0.04%
Lowest price (Asia session)14 Sept 2026, 00:06	1.38620	−0.04%	—
4 days with no trading
Highest price (New York session)01 Oct 2026, 15:34	1.42624	+2.84%	+2.89%
Close price before the jobs reportThu 01 Oct 2026, 20:59, close	1.42210	+2.55%	+2.59%
After the jobs report (02 Oct 2026, 12:30 UTC)
Price after the jobs report (New York session)02 Oct 2026, 12:30	1.42183	+2.53%	+2.57%	−0.02%
Lowest price (New York session)02 Oct 2026, 12:30	1.41980	+2.38%	+2.42%	−0.16%
Highest price (New York session)02 Oct 2026, 19:03	1.42612	+2.84%	+2.88%	+0.28%
Close price after the jobs reportFri 02 Oct 2026, 20:59, close	1.42506	+2.76%	+2.80%	+0.21%
Current price (New York session)02 Oct 2026, 20:59	1.42506	+2.76%	+2.80%	+0.21%

Recorded after the event and not edited since. Past moves are not a forecast of future ones.

PassedFri, 2 Oct 2026, 09:00

EURUSDEuro / US Dollar

Euro area inflation lands on a pair at its monthly low, hours before US payrolls

Positive

1.1242

+20 to +45 pips, which is +0.17% to +0.39%, if headline reaches 3.4% or core rises above 2.5%

−0.21%Low

+0.38%High

EURUSD result

After the event it went to −0.21% at its lowest, and later to +0.38% at its highest, against the close price before inflation flash estimate.

The news wasPositive

Measured fromChart

Euro-area inflation accelerated to 3.8% in September from 3.2%, while core inflation rose to 2.5%; the complete release is due on 16 October. The euro did not respond to the hot flash estimate, and weak US payrolls later reversed the morning loss.

What else moved the price

The card called the day conflicting, and the later US jobs report did overwrite the euro-area reaction: payrolls rose by only 29,000, unemployment reached 4.2%, wages slowed and prior months were revised down. EURUSD rose 53.5 pips, or 0.48%, from just before payrolls to its later high and retained 20.2 pips, or 0.18%, at the rollover. The earlier decline was euro-wide rather than a hidden dollar move, with EURGBP down about 17 pips, EURJPY about 31 pips and EURCHF about 35 pips between 08:59 and 12:29 UTC.

What we said	What happened	Result
Catalyst confidence was 98 out of 100 that Eurostat would publish the September flash estimate on 2 October at 09:00 UTC.	Eurostat published the flash estimate on 2 October as scheduled. It put annual euro-area inflation at 3.8% in September, up from 3.2% in August, and scheduled the complete September release for 16 October.	Correct
Positive. The card said a hot reading was more likely after Spain printed above expectations and energy prices stayed high, while EURUSD was positioned for a softer result.	The inflation call was right but the rate call was wrong. Headline inflation reached 3.8%, above the 3.6% consensus reported after the release, and core inflation rose to 2.5% from 2.4%. EURUSD gained only 2.4 pips at its best before payrolls and was 18.6 pips, or 0.17%, lower immediately before that second release.	Release direction correct; price direction wrong
Our expected move was +20 to +45 pips, or +0.17% to +0.39%, if headline inflation reached 3.4% or core inflation rose above 2.5%.	The headline trigger fired by a wide margin, but the euro-area release produced no move inside the forecast range. From the first post-release rate to the last minute before US payrolls, the high was only 2.4 pips, or 0.02%, higher and the low was 31.4 pips, or 0.28%, lower.	Wrong; the positive range was not reached before payrolls
Bullish scenario: headline inflation reaches at least 3.4%, core rises above 2.5%, services carries more of the increase and EURUSD bounces toward the 100-day average at 1.1519963.	Only part of the data scenario occurred. Headline inflation reached 3.8% and services inflation rose to 3.2% from 3.0%, but core rose only to 2.5%, not above it. The pair did not bounce toward 1.1519963 and instead traded down to 1.12190 before payrolls.	Scenario partly occurred; price outcome wrong
Neutral scenario: headline inflation lands between 3.1% and 3.3%, core stays at 2.4% and the pair waits for US payrolls.	The data did not fit the neutral scenario: headline inflation was 3.8% and core was 2.5%. The price response nevertheless looked neutral for its first half hour before a euro-wide decline developed.	Did not occur
Bearish scenario: headline inflation falls toward 3.0% with core flat or lower, weakening the case for another ECB increase and sending the pair toward 1.13534.	The data did the opposite: headline and core inflation both rose. The pair still broke below 1.13534 before the release and was already at 1.12422 at the prior rollover, so the price target was reached for reasons other than the bearish inflation scenario.	Data scenario did not occur; price level broke before the event
A move in core inflation was needed for a move beyond the forecast ranges because the headline was largely an energy story the ECB could look through.	Core inflation rose one tenth to 2.5% rather than moving above the card's 2.5% threshold. Energy accelerated to 18.8% from 14.3% and accounted for most of the headline jump, while the immediate EURUSD response stayed far below the positive range.	Correct
The day was conflicting because the US September employment report would land three and a half hours later, could move EURUSD several times as much and could erase the euro-area reaction.	Payrolls rose by only 29,000 against a consensus near 90,000, unemployment rose to 4.2%, earnings grew 0.1% on the month and July-August revisions removed 60,000 jobs. EURUSD rose 53.5 pips, or 0.48%, from the minute before payrolls to its later high, reversing the morning decline and moving far more than the inflation release did.	Correct
Estimated duration was within hours because the US employment report would reset the pair that afternoon.	That is what happened. The inflation release produced almost no first reaction, the pair then fell into 12:30 UTC, and weak payrolls reversed it; by the rollover EURUSD was 9.8 pips, or 0.09%, above the prior rollover despite having been 23.2 pips below it just before payrolls.	Correct
Technical bias was bearish, with a daily close below 1.13534 confirming a breakdown and a close above 1.1519963 needed to end the immediate selling.	The bearish chart call was right. Between the signal and the release the pair fell as low as 1.12140 and the prior rollover was 1.12422, both below 1.13534; it never approached the 1.1519963 recovery level in the reaction window.	Correct
The euro was oversold near its 30-day low, so merely avoiding disappointment could be enough for a bounce.	Avoiding disappointment was not enough. Even a large headline beat and a rise in core inflation produced only a 2.4-pip upside before the pair resumed falling across EURGBP, EURJPY and EURCHF as well as against the dollar.	Wrong
The main risks were high because payrolls could erase the euro move, the flash estimate could later be revised, and an energy-led headline could be read differently from underlying inflation.	All three cautions remain relevant. Payrolls did erase the morning move, energy rose 18.8% and drove the headline while core was 2.5%, and Eurostat labels these figures estimates until the complete 16 October release.	Correct

Price Table	Rate (EURUSD)	Change from signal	Change from lowest (regular session)	Change from close before inflation flash estimate
After signal
Price at analysis (Asia session)21 Sept 2026, 23:59	1.14633	—	+2.22%
Sat 26, Sun 27 - no trading
Lowest price (New York session)01 Oct 2026, 17:18	1.12140	−2.17%	—
Highest price (Asia session)22 Sept 2026, 05:09	1.14790	+0.14%	+2.36%
Sat 26, Sun 27 - no trading
Close price before inflation flash estimateThu 01 Oct 2026, 20:59, close	1.12422	−1.93%	+0.25%
After inflation flash estimate (02 Oct 2026, 09:00 UTC)
Price after inflation flash estimate (London session)02 Oct 2026, 09:00	1.12504	−1.86%	+0.32%	+0.07%
Lowest price (New York session)02 Oct 2026, 12:17	1.12190	−2.13%	+0.04%	−0.21%
Highest price (New York session)02 Oct 2026, 14:31	1.12853	−1.55%	+0.64%	+0.38%
Close price after inflation flash estimateFri 02 Oct 2026, 20:59, close	1.12520	−1.84%	+0.34%	+0.09%
Current price (New York session)02 Oct 2026, 20:59	1.12520	−1.84%	+0.34%	+0.09%

Recorded after the event and not edited since. Past moves are not a forecast of future ones.

PassedThu, 24 Sept 2026, 07:30

USDCHFUS Dollar / Swiss Franc

The Swiss National Bank sits at zero while everyone else tightens

Mixed

0.8249

+30 to +60 pips, which is +0.37% to +0.73%, on an unchanged assessment, and -60 to -120 pips, which is -0.73% to -1.47%, if zero is given an end date

−0.26%Low

+0.57%High

USDCHF result

After the event it went to −0.26% at its lowest, and later to +0.57% at its highest, against the close price before SNB quarterly monetary policy assessment.

The news wasPositive

Measured fromChart

The SNB held at 0% as expected but did not repeat June unchanged: it raised its inflation forecast for every year on the horizon and softened its intervention language, both conditions the card tied to a stronger franc. The franc weakened anyway. USDCHF closed the 24 hours after up 1.35%, well past the card's bullish-case ceiling, having already moved most of the way there before the assessment on what looks like broad dollar strength rather than anything the SNB said.

What we said	What happened	Result
Catalyst confidence 97 out of 100: the assessment is published on schedule.	Confirmed. The SNB published its assessment at 07:30 UTC on 24 September as scheduled, holding the policy rate at 0% for a fifth consecutive meeting.	Correct
Bearish scenario (stronger franc) trigger: "the conditional inflation forecast is revised up enough to put an end to zero on the horizon, or the Bank drops the intervention language in a way that reads as tolerance for a stronger franc."	Both named conditions fired in modified form. The SNB raised its inflation forecast for every year, 2026 to 0.7% from 0.6%, 2027 to 0.8% from 0.6%, 2028 to 0.8% from 0.7%, and coverage described the SNB's "intervention bias" as "dialed down," replacing June's "increased willingness to intervene" against franc appreciation with its standard, weaker wording. Neither put an explicit end date on zero, which is likely why the market did not read it as the full trigger.	Trigger partly fired, in the direction the card called bearish
Bullish scenario: "the assessment repeats June... the language on the franc is unchanged... the pair grinds toward the one year high at 0.81947."	The assessment did not repeat June unchanged, the forecast and the intervention language both moved, yet the pair did not just grind toward the one-year high, it blew through it, reaching 0.82964, about 1.2% above the level the card named as the ceiling of an unchanged outcome.	Outcome matched the bullish price target despite the bearish-leaning statement
Our Expected Move +30 to +60 pips (+0.37% to +0.73%) on an unchanged assessment, or -60 to -120 pips (-0.73% to -1.47%) if zero gets an end date or intervention language tolerates strength.	The 24-hour close, +1.35% (roughly 111 pips), nearly doubled the top of the bullish range, despite the statement content leaning toward the conditions the card tied to the bearish range instead.	Direction opposite the statement content, magnitude far exceeded either range
Main risks: "the dollar leg has already been repriced by a much larger central bank the week before, so attribution on the day will be imperfect," and global rate expectations "can overwhelm a Swiss assessment entirely."	This looks like the best explanation available. Most of the eventual move, 1.02 of the 1.35 percentage points, was already in place before the assessment even happened, in line with a broadly stronger dollar that same week; the SNB's own mildly franc-supportive signals did not visibly move the pair the other way.	Correct
Technical breakout: "a close above 0.81947, the highest close of the year... confirms the trend and says the assessment changed nothing."	The level was cleared decisively, up to 0.82964, but the assessment did change something on two of the card's own named metrics; the clean technical read and the actual policy content point in different directions even though the price outcome matched the technical one.	Level cleared, "changed nothing" reading did not match the statement
What the market may be missing: "the haven bid is entirely unspent, and it is available the moment the rate gap stops widening."	Still unspent. The franc kept weakening against the dollar even with a mildly hawkish inflation revision and a softer intervention line, reinforcing the card's own point that the rate differential is currently overwhelming every other signal, the SNB's own words included.	Correct

Price Table	Rate (USDCHF)	Change from signal	Change from lowest (regular session)	Change from close before SNB quarterly monetary policy assessment
After signal
Price at analysis (Asia session)13 Sep 2026, 23:59	0.81661	—	+0.19%
Lowest price (New York session)14 Sept 2026, 16:50	0.81510	−0.18%	—
Highest price (New York session)16 Sept 2026, 20:45	0.82650	+1.21%	+1.40%
Sat 19, Sun 20 - no trading
Close price before SNB quarterly monetary policy assessmentWed 23 Sept 2026, 20:59, close	0.82490	+1.02%	+1.20%
After SNB quarterly monetary policy assessment (24 Sep 2026, 07:30 UTC)
Price after SNB quarterly monetary policy assessment (London session)24 Sept 2026, 07:30	0.82493	+1.02%	+1.21%	+0.00%
Lowest price (London session)24 Sept 2026, 07:30	0.82279	+0.76%	+0.94%	−0.26%
Highest price (New York session)24 Sept 2026, 15:13	0.82964	+1.60%	+1.78%	+0.57%
Close price after SNB quarterly monetary policy assessmentThu 24 Sept 2026, 20:59, close	0.82762	+1.35%	+1.54%	+0.33%
Current price	—	—	—	—

Recorded after the event and not edited since. Past moves are not a forecast of future ones.

PassedWed, 14 Oct 2026, 12:30

NZDUSDNew Zealand Dollar / US Dollar

US inflation lands on a currency already sold to the floor of its range

Positive

0.5810

+40 to +80 pips, which is +0.69% to +1.38%, on a soft core reading

Pending

—